energy and building design | energy and building design ......1.2 energy efﬁ ciency in the...

Division of Energy and Building DesignDepartment of Architecture and Built EnvironmentLund UniversityFaculty of Engineering LTH, 2008Report EBD-T--08/8

Harris Poirazis

Single and Double Skin Glazed Offi ce Buildings

Analyses of Energy Use and Indoor Climate

Lund UniversityLund University, with eight faculties and a number of research centres and specialized institutes, is the largest establishment for research and higher education in Scandinavia. The main part of the University is situated in the small city of Lund which has about 103 700 inhabitants. A number of departments for research and education are, however, located in Malmö. Lund University was founded in 1666 and has today a total staff of 5 500 employees and 40 000 students attending 140 degree programmes and 1 600 subject courses offered by 66 departments.

Division of Energy and Building DesignReducing environmental effects of construction and facility management is a central aim of society. Minimising the energy use is an important aspect of this aim. The recently established division of Energy and Building Design belongs to the department of Architecture and Built Environment at the Lund University, Faculty of Engineering LTH in Sweden. The division has a focus on research in the fi elds of energy use, passive and active solar design, daylight utilisation and shading of buildings. Effects and requi-rements of occupants on thermal and visual comfort are an essential part of this work. Energy and Building Design also develops guidelines and methods for the planning process.

xxxxx

1


Analyses of Energy Use and Indoor Climate

Harris Poirazis

Doctoral Dissertation


2

KeywordsGlazed offi ce buildings, single skin façades, double skin façades, building simulations, building performance, energy use, indoor climate, thermal environment, thermal comfort.

© copyright Harris Poirazis and Division of Energy and Building Design. Lund University, Lund Institute of Technology, Lund 2008.The English language corrected by L. J. Gruber BSc(Eng) MICE MIStructE.Layout: Hans Follin, LTH, Lund.Cover photo: Harris Poirazis

Printed by KFS AB, Lund 2008

Report No EBD-T--08/8Single and Double Skin Glazed Offi ce Buildings. Analyses of Energy Use and Indoor Climate.Department of Architecture and Built Environment, Division of Energy and Building Design, Lund University, Lund

ISSN 1651-8136ISBN 978-91-85147-23-6

Lund University, Lund Institute of TechnologyDepartment of Architecture and Built EnvironmentDivision of Energy and Building Design Telephone: +46 46 - 222 73 52P.O. Box 118 Telefax: +46 46 - 222 47 19SE-221 00 LUND E-mail: [email protected] Home page: www.ebd.lth.se

xxxxx

3

To UsoaI’ve only gone this far because you tied my shoe laces


4

Abstract

5

Abstract

The energy effi ciency and thermal performance of highly glazed offi ce buildings are often questioned. However, nowadays glazed buildings are increasingly being built around the world, because (a) there is a growing tendency among architects to use large areas of glass in the façade, often with the aim of contributing to a better view of the outside and access to daylight, (b) users often like the idea of increased glass area, relating it to a better view of the outside and a more pleasant indoor environment and (c) many companies prefer the distinctive image of themselves (e.g. transparency or openness) that a glazed offi ce building can provide. Due to insuffi cient knowledge concerning function, energy use as well as indoor environment of glazed offi ce buildings for Scandinavian conditions, a project was initiated in order to gain knowledge of their possibilities and limitations.

The aim of this thesis is to clarify and quantify how highly glazed façades affect the energy use and thermal comfort of offi ce buildings. Another aim was to validate or identify the needed improvement of building energy simulation tools, in order to ensure the precision of the simulations. Fi-nally, suggestions have been given for determining how the design can be improved with regard to energy effi ciency and thermal comfort.

The fi rst part of this project involved establishing a reference building with different single skin glazed alternatives, choosing simulation tools and carrying out simulations for the determined alternatives. As the reference building, a moderately glazed offi ce building representative of the late nineties was chosen. Using this building as a starting point, the window area to external wall area ratio was increased gradually, in order to meet a fully glazed offi ce building. Results were obtained through varying the building’s orientation, the interior layout (open plan and cell type offi ces) and the type of glazing and solar shading devices. The different building alternatives were compared with different indoor environment classifi ca-tions and a sensitivity analysis was presented regarding the occupants’ comfort and the energy used for operating the building.

In the second part of the project parametric studies were carried out regarding the performance of various double skin façade cavity alterna-


6

tives, in order to gain knowledge of the possibilities and limitations of the system’s performance. Simulations on a zone and a building level were then carried out, in order to achieve optimal integration of the system. The simulations included different glazing and shading devices for both “standard” double façades and airfl ow window modes.

The results showed that highly glazed buildings tend to perform poorly unless designed carefully, resulting in increased energy use and poorer thermal environment. For Swedish climatic conditions during winter months, windows with low thermal transmittance are essential, in order to improve the building’s energy performance and thermal comfort, es-pecially for highly glazed buildings. Low g and especially geffective values have a positive effect in lowering the cooling demand; externally placed shading or double skin facades can have this effect. Hybrid ventilated double façades can reduce the heating demand and improve the quality of thermal environment. Airfl ow windows with low E inner pane result in a radical improvement of the indoor climate, reaching the thermal comfort levels of an offi ce building with a conventional façade. In general, double skin façades result in improved energy and thermal performance of the building mainly when applied on the south façade, but their impact is limited, since the cooling demand is rather limited for Scandinavian climatic conditions. Parameters such as the impact of temperature control set points, plan type and orientation on energy use and thermal comfort were also studied.

Achieving improved building performance, when using fully glazed façades, can be a great challenge. Individual building design that takes into consideration the type of façade including the size and type of glazing, the position of shading devices, the temperature set points, the building occu-pancy and plan type can defi nitely lead to improved building performance. If this is established, even in highly glazed cases, the building performance may reach reasonable levels as to energy use and indoor climate. However, a building with low energy demand cannot be achieved by a highly glazed building in a Scandinavian climate.

Contents

7

Contents

Keywords 2

Abstract 5

Contents 7

Acknowledgements 13

How to read this thesis 15

1 Introduction 19

1.1 General 191.2 Energy effi ciency in the building sector 201.3 Energy effi cient building design 211.4 The “Glazed Offi ce Building” project 221.5 Aim of the thesis 221.6 Limitations of the thesis 231.7 Defi nitions and symbols 25

2 Background 31

2.1 General 312.2 The performance and quality of a building as a system 322.3 Building Environment 342.3.1 Design Criteria 342.3.2 Indoor Environment 352.3.2.1 Thermal Comfort 352.3.2.2 Conditions of thermal comfort 362.3.2.3 Thermal comfort and productivity 432.3.2.4 Other indoor climate parameters that infl uence the occupants’ health and productivity 452.3.3 Architectural quality 462.3.4 Environmental performance 472.3.5 Costs 48

2.4 Building technology 492.4.1 Glass in buildings 492.4.1.1 General 492.4.1.2 Basic physics of the glass 502.4.1.3 Thermal functions of the glass 52


8

2.4.2 Single skin façades 542.4.2.1 Glazing 542.4.2.2 Shading devices 552.4.3 Double skin façades 572.4.3.1 General 572.4.3.2 Classifi cation of double skin façades 572.4.3.3 Technical description of the cavity 582.4.3.4 Advantages and disadvantages of double skin façades 58

3 State of the art 63

3.1 Glazed offi ce buildings in Nordic climates 633.1.1 General 633.1.2 Layout of typical offi ce buildings 643.1.3 Offi ce buildings in Sweden 653.1.4 Energy performance of Swedish offi ce buildings 663.1.5 Glazed offi ce buildings in Sweden 68

3.2 Double skin façades 703.2.1 Building physics of the double skin façade cavity 703.2.1.1 General 703.2.1.2 Modelling approaches 713.2.1.3 Measurements – test rooms and real buildings 753.2.2 Integration of double skin façades 773.2.2.1 Contribution of double skin façades to the HVAC strategy 783.2.2.2 Examples of coupling double skin façades and HVAC 813.2.2.3 Control strategy 853.2.3 Energy performance of buildings with integrated double skin façades 863.2.4 Typical constructions - Examples of buildings 89

3.3 Building simulation software 913.3.1 Building energy simulation tools 923.3.2 Software for DSF modelling 963.3.2.1 Façade simulation software 963.3.2.2 Building simulation software 97

4 Methods 99

4.1 Generation of building alternatives 994.1.1 Reference building (30% window to external wall area ratio) 1004.1.2 Single skin alternatives (60% and 100% window to external wall area ratios) 1024.1.3 Double skin alternatives (100% window to external wall area ratio) 1044.1.3.1 Pilot study on component level using WIS 3 1044.1.3.2 Study on a zone level using IDA ICE 3.0 1104.1.3.3 Study on a building level using IDA ICE 3.0 113

4.2 Description of the studied parameters 1134.2.1 Single skin alternatives (30%, 60% and 100% glazed alternatives) 1134.2.1.1 IDA ICE 3.0 output (zone level) 1144.2.1.2 IDA ICE 3.0 output (building level) 114

Contents

9

4.2.2 Double skin alternatives 1164.2.2.1 WIS 3 simulations (component level) 1164.2.2.2 IDA ICE 3.0 Simulations (zone level) 1194.2.2.3 IDA ICE 3.0 Simulations (building level) 119

4.3 Description of the simulation tools 1194.3.1 Simulations using WIS 3 1194.3.1.1 Temperatures at the centre of each layer 1204.3.1.2 Temperatures at different heights of the cavity 1214.3.2 Simulations using IDA ICE 3.0 1244.3.2.1 General description 1244.3.2.2 Description of double façade model 1254.3.2.3 Validation of IDA ICE 3.0 Double Façade model (IEA SHC Task 34/ECBCS Annex 43) 126

5 Description of the building model 129

5.1 Description of the reference building 1295.1.1. Geometry of the building 1295.1.2. Offi ce layouts 1305.1.3 Description of building elements 1345.1.4 Special modifi cations for the simulated model 1385.1.5 Control set points for indoor air temperature 1405.1.6 Occupancy 1415.1.7 Lights 1455.1.8 HVAC 1465.1.8.1 Heating and cooling 1465.1.8.2 Ventilation 1465.1.8.3 Equivalent heat recovery effi ciency 1475.1.8.4 Use of electricity 1495.1.9 Electrical equipment 149

5.2 Description of single skin glazed alternatives 1505.2.1 Description of 60% glazed building 1505.2.1.1 Façade construction 1505.2.1.2 Window properties 1525.2.2 Description of 100% glazed building 155

5.3 Description of double skin glazed alternatives 1575.3.1 WIS 3.0 simulations 1575.3.1.1 Geometry of the “standard” double façade box 1575.3.1.2 Geometry of the airfl ow window 1575.3.2 IDA ICE 3.0 Input (zone level) 1595.3.2.1 Offi ce description (IDA ICE 3.0 - zone level) 1595.3.2.3 Geometry of the multi storey high façade 1605.3.2.4 Properties of the inner and outer skin 1605.3.2.5 Shading devices 161

5.4 Assumptions made during the calculations 163


10

6 Results and discussion 165

6.1 Reference building 1666.1.1 Energy use 1666.1.1.1 Impact of fl oor plan type 1666.1.1.2 Impact of orientation 1696.1.1.3 Impact of control set points 1696.1.2 Indoor climate on a building level 1706.1.2.1 Weighted average mean air temperatures 1706.1.2.2 Perception of thermal comfort 1746.1.3 Indoor climate on a zone level 1816.1.3.1 Mean air temperatures and potential overheating problem 1826.1.3.2 Directed operative temperatures 1866.1.3.3 Perception of thermal comfort 190

6.2 Single skin glazed alternatives (60% and 100% window to external wall area ratios) 1996.2.1 Energy use 1996.2.1.1 Impact of fl oor plan type and orientation 2006.2.1.2 Impact of windows and shading devices for the 60% and 100% glazed alternatives 2026.2.2 Indoor climate on a building level 2096.2.2.1 Weighted average mean air temperatures 2096.2.2.2 Impact of window and shading type on the perception of thermal comfort for the 60% and 100% glazed alternatives 2106.2.3 Indoor climate on a zone level 2246.2.3.1 Directed operative temperatures 2246.2.3.2 Perception of thermal comfort 226

6.3 Double skin façades 2336.3.1 Simulations on a component level (pilot study using WIS 3) 2336.3.1.1 Pre study: reducing the number of “standard” double façade alternatives 2346.3.1.2 Parametric study: infl uence of cavity geometry on system performance 2416.3.1.3 Performance of the glazing alternatives 2526.3.2 Parametric studies on a zone level (IDA ICE 3.0) 2656.3.2.1 “Standard” double façade mode (naturally ventilated cavity) 2666.3.2.2. “Standard” double façade mode (mechanically ventilated cavity) 2706.3.2.3 “Standard” double façade mode (hybrid ventilated cavity) 2746.3.2.4 Airfl ow window mode 2776.3.2.5 Impact of the “ventilated façade” concept 2826.3.2.6. Impact of shading device type 2916.3.3 Parametric studies on a building level (IDA ICE 3.0) 2956.3.3.1 Energy use 2976.3.3.2 Thermal comfort 300

6.4 Comparison of single and double skin façade building alternatives 3026.4.1 Impact of glazing size on energy use and thermal comfort of single skin buildings 3026.4.2 Impact of glazing type on energy use and thermal comfort 3046.4.3 Comparison of buildings with single and double skin façades 306

Contents

11

6.4.4 Comparison of best performing alternatives 308

7 Conclusions 311

7.1 Energy use and thermal comfort for highly glazed offi ce buildings located in Scandinavia 3127.1.1 Plan type 3127.1.2 Control set points 3127.1.3 Orientation 3157.1.4 Glazing area 3157.1.5 Glazing and solar shading type 3167.1.6 Double skin façades 3177.1.7 Best performing alternatives 320

7.2 Methods for determining energy and indoor climate performance 3217.2.1 Lessons learnt from the simulation work 322

7.3 Improving the energy and indoor climate performance of highly glazed buildings: general recommendations and further studies 323

8 Summary 327

8.1 Introduction 3278.2 Background 3278.3 Methods 3288.4 Discussion and conclusions 3308.4.1 Glazing area 3308.4.2 Glazing and solar shading type 3308.4.3 Double skin facades 3318.4.4 Other parameters that infl uence the building performance 3328.4.5 Best performing alternatives 3338.4.6 Determination of energy and indoor climate performance 334

8.5 General recommendations for improvements of highly glazed offi ce buildings 334

References 337

Appendix A 347

Appendix B 357

Appendix C 359

Appendix D 365

Appendix E 377

Appendix F 381

Appendix G 385

Appendix H 387

Appendix I 389


12

Appendix J 391

Appendix K 393

Appendix L 397

Appendix M 403

Acknowledgements

13

Acknowledgements

I wish to thank my supervisors Maria Wall and Åke Blomsterberg for their guidance and the useful advice that they gave me during the research work and for the compilation of this report. I would also like to thank my col-leagues at the Division of Energy and Building Design; especially Gunilla Kellgren for being so kind and helpful, Bengt Hellström for all the advice on technical issues, and Johan Nilsson for his support and friendship dur-ing the past years and Henrik Davidsson for just introducing the "Con-stanza" concept to me. The contribution and support of Per Sahlin and Mika Vuolle proved to be substantial for meeting the (reasonably delayed) deadlines. Special thanks to Jean Rosenfeld for believing in me and being the main driving force for starting a PhD. Along the way several people and organizations contributed to this work; I thank all of you. Finally, I would like to thank all the experts who, by making available their theses, reports and articles, have provided easy access to knowledge.

I would like to express my greatest gratitude to my parents Kaiti and Stathis for their support during all the years of my study. Last but defi -nitely not least, I wish to thank Usoa for always believing in me and for supporting my choices without considering any cost.

The project was funded by the Swedish Energy Agency and SBUF (Development Fund of the Swedish Construction Industry), and sup-ported by Skanska and WSP.

Lund, December 2007

Harris Poirazis


14

How to read this thesis

15


When writing this thesis, a main aim was to provide enough background knowledge so that a “new” reader, less knowledgeable in this fi eld, can read it. Since the fi ndings of this thesis are primarily addressed to architects, HVAC engineers, glazing/façade experts and consultants in the building industry, a common language had to be established. Thus, the fi rst three chapters were quite extensive covering different aspects of the fi eld.

In the fi rst (Introduction) chapter a brief description is given of what energy effi cient design means and why energy savings in the building sec-tor are important. Then, the contribution of the “Glazed Offi ce Building” project to the fi eld is briefl y explained and the main aims and limitations of this thesis (as a part of the whole project) are described. Finally, defi ni-tions and symbols used during the thesis are briefl y explained.

The main aim of the second (Background) chapter is to provide a theoretical background regarding different aspects of the building system and their impact on building performance. The building is described as a system consisting of the “building environment” and the “building technol-ogy”. The parameters that the “building environment” and the “building technology” comprise are described briefl y in a top down organization. Emphasis is given to the aspects studied further in this thesis (such as conditions of thermal comfort and other indoor climate parameters that infl uence the occupants’ health and productivity), while other parameters, such as architectural quality, environmental performance and costs are described briefl y.

The third chapter (State of the art) aims to inform the reader about the research already carried out in the fi eld, focusing mainly on three topics: (a) glazed offi ce buildings in Nordic climates, (b) double skin façades and (c) building simulation software. In the fi rst part a description of typical (glazed) offi ce buildings in Sweden with regard to energy performance is provided. The second part deals with building physics of the double skin façade component (such as modelling approaches of the cavity) and integra-tion techniques of double skin façades in buildings. Finally a description


16

of the available building simulation software is given focusing on those that can handle double skin façade systems. A brief reference regarding the validation of these tools is also provided.

In the fourth chapter (Methods) the generated building alternatives are described. The necessary reasoning concerning the alternatives selected for the parametric study is provided for better understanding. The (output) parameters of the simulated single and double skin façade alternatives are briefl y discussed. Finally, the tools used for the parametric studies are briefl y described and a reference to their validation is provided.

The fi fth chapter (Description of the building model) aims to describe the input as inserted in the two software tools used (IDA ICE 3.0 and WIS 3). The input concerns physical parameters of the building, properties of the building materials, occupancy, control set points, schedules and HVAC characteristics for the 30%, 60% and 100% glazed alternatives studied. Information regarding the construction of the double skin façade cavities, both on a component and on a building level, is also provided focusing mainly on geometrical characteristics.

In the sixth chapter (Results and discussion) the results of 30%, 60% and 100% single skin glazed alternatives and 100% double skin glazed alternatives are extensively studied. The results mainly focus on energy use and thermal comfort issues. A comparison between different building alternatives takes place, in order to investigate the infl uence of differently varied parameters on the building performance. The parametric study of the single skin building alternatives is focused mainly on parameters such as control set points, offi ce layout, glazing type, etc, while for the double skin façade building alternatives the glazing type and the ventilation mode are the two main issues of interest. Prior to the double skin building al-ternatives, a pilot study on a double skin cavity (on a component level) took place, in order to fi lter the simulated alternatives and reduce their number. The results of this pilot study focus on investigating the crucial (mainly geometrical) parameters that infl uence the cavity’s performance. Better understanding of the possibilities and limitations of the cavity confi guration result in an understanding as to what can be achieved on a building level. Simulations on a building level follow considering "standard" double façade mode alternatives ((a) naturally ventilated, (b) mechanically ventilated and (c) hybrid ventilated) and airfl ow windows. Optimization of the integration is the main aim of this step. Discussion concurrently with the results is carried out regarding the performance of the simulated building alternatives. Single and double skin alternatives are compared with regard to energy use and thermal comfort.

A Conclusions (seventh) chapter follows providing the reader with the main conclusions from the simulations. Suggestions in order to improve the building performance regarding energy use and thermal comfort are


17

provided. The main aim of the conclusion chapter is not only to understand the building’s response when design parameters vary but also to quantify its performance for offi ce buildings located in Scandinavia. Eighth chapter: Summary


18

Introduction

19

1 Introduction

This PhD thesis is an important part of the “Glazed Offi ce Building project”, a fi ve years project initiated in 2003, with the aim to gain knowl-edge concerning the possibilities and limitations of glazed offi ce buildings exposed to Swedish climatic conditions, mainly with regard to energy use and indoor climate. Both single and double skin glazed offi ce buildings are examined, starting with a reference building (typical offi ce building of the 90s), in Scandinavia.

1.1 GeneralThe energy effi ciency and thermal performance of highly glazed offi ce buildings are often questioned. However, nowadays more and more glazed buildings are built for the following reasons:

• there is a growing tendency among architects to use large areas of glass in the facade often with the aim of contributing to a better view of the outside and access to daylight

• users often also like the idea of increased glass area, relating it to better view and a more pleasant indoor environment. However, users do not often take into account the risk of visual and thermal discomfort that can occur due to this construction type

• companies who want to create a distinctive image for themselves (e.g. transparency or openness) often like the idea of being located in a glazed offi ce building

Deciding how to build with a fully glazed façade can be a complicated issue. The energy effi ciency and provision of an acceptable indoor climate are issues that should be considered. Other parameters to be taken into consideration during the decision and design process are visual comfort, building aesthetics, sociological and psychological determinants (such as visual and acoustical privacy), life cycle cost, etc. By prioritizing at an early


20

stage the goals to be achieved, the design team can improve the building performance and fulfi l the design requirements.

This thesis aims to compare the performance (as to energy use and indoor climate issues) of conventional and highly glazed façades. Through extensive parametric studies the impact of design parameters on building performance is studied. Optimizing energy and indoor climate perform-ance of single skin highly glazed buildings is the fi rst goal to be achieved. The proper integration of double skin façade systems is also investigated with the aim of further improvements.

When large proportions of glazing are used in the façade, it is essential to know from the early design stage what is to be achieved and how to achieve it. Clear goals and suffi cient knowledge of the calculation tools that should be used for predicting building performance are essential for a successful building design.

1.2 Energy effi ciency in the building sectorEnergy use in the building sector accounts for a large proportion of the total energy use in most countries around the world. Particularly for the EU, the building sector accounts for more than 40% of the energy use and is expanding (directive 2002/91/EC). Consequently, the building sector has a major energy-saving potential. According to the Energy Effi ciency Action Plan (2006) the largest cost effective savings potential lies in the residential and commercial building sector. The full saving potential is estimated around 27% and 30% respectively. In order to achieve this, the Buildings Directive was introduced in 2002 with the aim to reduce energy use by implementing energy conservation measures in the building sec-tor. It states that (almost all) buildings sold or rented within the EU shall have an energy performance certifi cate not older than 10 years. Moreover, public authority buildings and buildings frequently visited by the public should set an example by taking environmental and energy considerations into account and therefore should be subject to energy certifi cation on a regular basis (i.e. every 10 years).

According to the directive 2002/91/EC, the energy performance of buildings should be calculated on the basis of a methodology, which may be differentiated at regional level, taking into account climatic and local conditions as well as indoor climate environment and cost effectiveness. Some of the factors which determine building performance are thermal insulation, the heating and air-conditioning installations, the application of renewable energy sources and the design of the building. More specifi -cally, according to the 2002/91/EC directive, the energy performance of a

Introduction

21

building is described as “the amount of energy actually consumed or estimated to meet the different needs associated with a standardised use of the building, which may include, inter alia, heating, hot water heating, cooling, ventilation and lighting. This amount shall be refl ected in one or more numeric indica-tors which have been calculated, taking into account insulation, technical and installation characteristics, design and positioning in relation to climatic aspects, solar exposure and infl uence of neighbouring structures, own-energy generation and other factors, including indoor climate, that infl uence the energy demand”.

1.3 Energy effi cient building designAt this point it is crucial to clarify the fundamental elements of a success-ful low energy building design. In order to achieve that, it is essential to understand what a low energy building is. As Nilsson, (2003) described, reducing energy use is not enough to cover the concept of energy effi ciency, since the fi rst can be achieved just by switching off the heating and cool-ing systems, the lighting and the ventilation. This, however, should not be the case in practice. Thus, in order to describe the concept of energy effi cient building design, it is essential to understand what overall build-ing performance is. If the building is considered as a system with input (such as location, climate, use, etc) and output (building performance), breaking it down into the parts it consists of can help us understand the interactions and their infl uence on the building performance. In this way energy conservation improvements can be suggested without compromis-ing performance. The analysis of the parts that constitute the building performance is carried out in Chapter 2.

Another issue to be clarifi ed when low energy building design is dis-cussed is what “low” means. Since this expression can often be quite un-clear, a reference building has to be chosen for reasons of comparison. For example, if the topic is energy conservation in buildings, then residential, commercial and offi ce buildings can be compared. If the topic is energy conservation in offi ce buildings, improvements should be suggested after taking into consideration the specifi c building use; a parameter that can be studied in this case is whether it is energy effi cient to use large areas of glazing in the external façade. If the topic is energy conservation in highly glazed offi ce buildings (e.g. due to an architectural concept), then the use of large glazed areas in the external façade of the building is a requirement and the comparison should be limited to highly glazed offi ce buildings. In this case comparison improvements can be suggested (e.g.) by adding


22

a second skin to the building envelope. From the above it is clear that energy effi ciency is a relative and not an absolute measure.

1.4 The “Glazed Offi ce Building” projectThe research project “Glazed offi ce buildings – energy use and indoor climate” was initiated and funded in 2003 in order to gain knowledge of the possibilities and limitations of glazed offi ce buildings exposed to the Swedish climate, with regard to:

• energy use • indoor environment (thermal and visual comfort, indoor air quality,

acoustics, etc)• environmental performance• architectural quality and • life cycle cost

Included in the “Glazed Offi ce Buildings” project are:

• further development of calculation methods and analysis tools • improvement of analysis methodology • calculation of life cycle costs • development of advice and guidelines for design/construction of glazed

offi ce buildings in a Swedish climate • strengthening and development of competence concerning resource

effi cient advanced buildings in Sweden

A detailed description of the organization and the main goals of the project can be found in the “Introductory report for the Glazed Offi ce Buildings project” (Poirazis, 2005a).

1.5 Aim of the thesisThe aim of the research presented in this thesis is to:

• determine how the energy and indoor climate performance can be analysed

• clarify and quantify how highly glazed façades affect energy use and thermal comfort

Introduction

23

• determine how the design can be improved with regard to energy ef-fi ciency and thermal comfort

In order to achieve this, a large number of building alternatives, both with single and double skin façades, were simulated on an offi ce room and a building level.

The fi rst part of this project involved establishing a reference building with different single skin glazed alternatives, choosing simulation tools and carrying out simulations for the determined alternatives. As the reference building, a moderately glazed offi ce building representative of the late nineties was chosen. Using this building as a starting point, the window area to external wall area ratio was increased gradually, in order to meet a fully glazed offi ce building. Results were obtained through varying the building’s orientation, the interior layout (open plan and cell type offi ces) and the type of glazing and solar shading devices. The different building alternatives were compared with different indoor environment classifi ca-tions and a sensitivity analysis was presented regarding the occupants’ comfort and the energy used for operating the building.

In the second part of the project parametric studies were carried out regarding the performance of various double skin façade cavity alterna-tives, in order to gain knowledge of the possibilities and limitations of the system’s performance. Once these results were obtained, simulations on a building level were carried out, in order to achieve optimum integration of the system. A comparison of single and double skin alternatives fol-lowed, with the main focus on energy saving potential and indoor thermal quality issues.

Within the main goals of this thesis, validation and improvement of building energy simulation software was also carried out, in order to ensure the precision of the simulations. More than anything, however, this thesis aims at increasing the knowledge of how to conserve energy in highly glazed offi ce buildings without compromising the quality of the indoor environment.

1.6 Limitations of the thesisThe present report deals mostly with energy use and indoor thermal envi-ronment calculations for a virtual reference offi ce building, with different façade alternatives, in Nordic climates. More specifi cally, it deals with en-ergy use for heating, cooling, lighting, offi ce equipment, pumps and fans, and the server rooms (including their cooling). In terms of indoor thermal environment, importance has been given to the mean air temperatures,


24

the directed operative temperatures and the perception of thermal comfort (PMV and PPD values). Daylight was taken into consideration only in order to keep a minimum and maximum illuminance level intensity at the working surface; when the illuminance exceeds these levels, artifi cial light was considered. A detailed separate study analyzing daylight and visual comfort was carried out within the “Glazed Offi ce Buildings” project (Bülow-Hübe, 2007). Investment cost and LCC analysis were also carried out separately (Sjodin, 2007).

Different offi ce building alternatives i.e. different façade alternatives for offi ce buildings were simulated and analyzed for this report. A virtual refer-ence building was created, a building representative of offi ce buildings built during the ninties in Sweden. The façade of this building was changed for different glazed façade alternatives. All the building elements were chosen based on commercially available products. Some of the building design parameters were assumed to remain the same during the simulations and some were changed. The parameters that were kept the same were:

• the shape of the building• the roof, ground fl oor, interior wall and intermediate fl oor construc-

tion• external obstructions to the building• infi ltration and exfi ltration of the building envelope

Parameters that varied during the simulations were:

• building orientation• offi ce layout (cell type and open plan type)• internal loads (number of occupants and equipment)• occupancy• proportion of the glazed façade area (30%, 60% and 100% glazing)• glazing and frame type• shading devices (type and position)• thermal transmittance of the façade elements• heat recovery effi ciency• specifi c fan power• control set points (temperature, lighting)• HVAC installations

Introduction

25

1.7 Defi nitions and symbolsBuilding Envelope: The total area of the boundary surfaces of a building through which heat, light, air and moisture are transferred between the internal spaces and the outside environment (Limb, 1992).

Comfort zone: The range of indoor conditions considered acceptable by a certain proportion (e.g. usually more than 80%) of the people working or living in the space (Limb, 1992).

Daylight Factor (DF): The ratio of indoor illuminance at a given point to the simultaneous outdoors illuminance on an unobstructed horizontal surface.

Degree Day: The number of degrees of temperature difference on any one day between a given base temperature and the 24 hour mean outside air temperature for the particular location. The average number of degree days for a given period (i.e. during the heating season) is the sum of these degree days divided by the given period (Limb, 1992).

Draught: Excessive air movement in an occupied enclosure causing dis-comfort (Limb, 1992).

Energy Conservation: The deliberate design of a building or process to reduce its energy use or to increase its energy effi ciency (Limb, 1992).

Exhaust Air: The air removed from a space and not reused therein (Limb, 1992).

Glare (discomfort): The sensation of annoyance caused by high or non-uniform distributions of brightness in the fi eld of view.

Glare (disability): Caused when intraocular light scatter occurs within the eye, the contrast in the retinal image is reduced (typically at low light levels), and vision is partly or totally impeded (e.g., when the eye is con-fronted by headlights from oncoming automobiles).

Heating: The transfer of energy to a space or to the air by the existence of a temperature gradient between the source and the space of air. The process may take different forms, i.e. conduction, convection or radiation. The process is the opposite of cooling (Limb, 1992).


26

Heat Recovery Effi ciency (or Heat Recovery Effectiveness): The propor-tion of heat recovered from otherwise waste heat passing through a heat recovery system. Normally it is expressed as a percentage (Limb, 1992).

Humidity, absolute (dv): The ratio of the mass of water vapour to the total volume of the sample.

Humidity, relative (φ): The ratio of the mole fraction of water vapour in a given moist air sample to the mole fraction in an air sample saturated at the same temperature and pressure.

Indoor Climate (or Indoor Environment): The synthesis of day to day values of physical variables in a building e.g. temperature, humidity, air movement and air quality, etc, which affect the health and/or the comfort of occupants (Limb, 1992).

Illuminance (E): Expresses the amount of luminous fl ux that arrives at a surface and is measured in lux.

Illuminance and luminance distribution: A measure of the light varia-tion from a point to another point across a plane or a surface.

Luminance: Expresses the light refl ected off a surface and is measured in lumens per square meter per steradian or in candelas per square meter (cd/m²). In a way the luminance is directly related to the perceived “bright-ness” of a surface in a given direction.

Luminous Effi cacy: Refers to the ratio of total luminous fl ux emitted by a lamp to the energy used. It is expressed in lumens per watt (lm/W). According to Erhorn and Stoffel (1996),

systemlighting of power total

spaceconsidered of area flooreilluminanc surfacedesiredn

×= (lm/W)

Mechanical Ventilation: Ventilation by means of one or more fans (Limb, 1992).

Multizone: A building or part of a building that comprises a number of zones or cells (Limb, 1992).

Natural Ventilation: The movement of outdoor air into a space through intentionally provided openings such as windows and doors, or through non powered ventilators or by infi ltration (Limb, 1992).

Introduction

27

Occupancy: The time during which people are in a building, usually expressed in hours per day (Limb, 1992).

Occupant Behaviour: The pattern of activity of occupants in a building, including the number of occupants, their distribution, activities and time spent within the building, the way they interact with building facilities, such as ventilation systems, window openings, etc (Limb, 1992).

Outdoor Air: The air taken from the external surroundings and therefore not previously circulated through the system (Limb, 1992).

PMV (Predicted Mean Vote): An index of thermal sensation since it expresses the correlation between indoor environment parameters and people’s sensation of thermal comfort (ISO Standard 7730, 1984). It is a function of activity, clothing, air temperature, mean radiant temperature, relative air velocity and humidity. The scale for PMV varies from -3 (cold) to +3 (warm) (see also Section 2.3.2.2).

PPD (Predicted Percentage Dissatisfi ed): The PPD of a large group of people is an indication of the number of persons who will be inclined to complain about the thermal conditions (ISO Standard 7730, 1984). It is expressed as a percentage (see also Section 2.3.2.2).

Radiation: The transmission of heat through space by propagation of infra red energy; the passage of heat from one object to another without necessarily warming the space in between (Limb, 1992).

Single Zone: A building or a part of a building comprising one zone of uniform pressure (Limb, 1992).

Solar transmittance (Tsol): The ratio of the irradiation transmitted through the window system to the irradiation impinging on the window system.

Solar (total) transmittance (or solar factor) (g) The sum of the (primary) solar transmittance (τ) and the ratio of the part of the solar irradiation absorbed by the window system that is transferred to the room, to the irradiation impinging on the window system.

Supply Air: Air delivered to a space and for the purpose of ventilation, heating, cooling humidifi cation or dehumidifi cation (Limb, 1992).

Temperature, Ambient: The temperature of the air within a room or zone (Limb, 1992).


28

Temperature, Dry Bulb: The air temperature indicated by a dry tempera-ture sensing element (such as the bulb of a mercury in glass thermometer) shielded from the effects of radiation (Limb, 1992).

Temperature, Effective (θeff): The temperature of a still, saturated atmosphere that would produce the same effect as the atmosphere in question.

Temperature, Environmental: The temperature of the air outside a room or zone (Limb, 1992).

Temperature, Operative (θop): The operative temperature empirically combines the dry bulb and the mean radiant temperatures. The operative temperature is the temperature at which a person emits the same heat output as before, but when air temperature (θa) = radiant temperature (θr) = operative temperature (θop). θop does not have the same value for all the parts of the room (when the weighting method is used) (Peterson, 1991).

Temperature, Directed Operative: It is calculated in the same way as the operative temperature, the only difference being that the weighting is done only for the point where the occupant is placed and towards the surface of interest as shown in Figure 2.6, Subsection 2.3.2.2.

Temperature, Radiant (or Surface) (θr): Radiant or surface temperature is the temperature of an exposed surface in the environment. The tem-peratures of individual surfaces are usually combined into a mean radiant temperature.

Temperature, Resultant: It is similar to the effective temperature but it includes humidity effects.

Temperature, Wet Bulb: The air temperature indicated by a sensing element kept wet (usually by a wick), the indicated temperature thus being related to the rate of evaporation from the wetted bulb. This wet bulb temperature is used by psychrometers to measure relative humidity (Limb, 1992).

Thermal Comfort: A condition of satisfaction expressed by occupants within a building regarding their thermal environment. Since the thermal comfort condition is a subjective feeling of satisfaction, building designers attempt to satisfy as many of the occupants as possible (usually 80% or more) (Limb, 1992).

Introduction

29

Thermal Transmittance (U-value, expressed in W/m2K): The heat fl ow transmitted through a unit area of a given structure, divided by the difference between the effective ambient temperature on either side of the structure, under steady conditions (Limb, 1992). The U-value is only defi ned for dark conditions (excluding solar radiation).


30

Background

31

2 Background

The main aim of this chapter is to provide a theoretical background regarding different aspects of the building system and their impact on building performance. A description of the building system’s structure is briefl y given and the parts of the system are described. Emphasis is given to the aspects studied further in this thesis (such as conditions of thermal comfort and other indoor climate parameters that infl uence the occupants’ health and productivity), while other parameters, such as architectural quality, environmental performance and costs are briefl y described. It has to be noted that the information provided regarding the latter aspects is mainly intended for “new” readers less knowledgeable in this fi eld. It is essential to clarify that the high complexity of these issues does not allow a comprehensive description within a few pages; however, an effort has been made with the aim to bring out the properties of the parts that make up the building system and infl uence its performance.

2.1 GeneralWell defi ned requirements for the design of a building can be a key to a successful design. As Nilsson (2003) describes, a “good” building perform-ance is often based on general criteria related to aesthetic attractiveness, satisfactory operation (a building that fulfi ls its purpose as far as users are concerned) and durability. However, in practice the requirements taken into account during the design process should be defi ned in greater detail. Ensuring that everybody involved in the design process is aware of the consequences of the requirements is the fi rst step to a successful building design.

Nilsson (2003) describes the requirements as performance and quality ones. Performance requirements are the ones that must be fulfi lled; oth-erwise the building can not be used for its intended purpose. Acceptable thermal climate, air and lighting quality, restrictions on disturbances (such as draughts, noise, glare, etc) and operational reliability are considered as performance requirements. Quality requirements, on the other hand,


32

ensure improved overall performance, so that the building will become a “good” building. Examples of these requirements are energy effi ciency, an aesthetically attractive design, effi cient use of space, minimised life cycle cost, generality and fl exibility, durability and ease of maintenance. This distinction between performance and quality requirements is made, in order to clarify which parameters can be compromised to a certain extent during the design process.

A detailed description of the requirements and their impact on the building performance can be found in the “Introductory Report for the Glazed Offi ce Building Project”, (Poirazis, 2005a). Below follows a sum-mary of this report.

2.2 The performance and quality of a building as a system

Before discussing how this thesis deals with improving some of the qual-ity requirements, especially energy effi ciency and thermal comfort, it is essential to describe the main components that constitute the building’s performance. In this way a common language between the author and the reader can be established, facilitating a precise understanding of the interaction between quality (such as energy effi ciency) and performance (such as indoor environment) requirements, the aim of which is to improve the building’s overall performance.

Initially, the building can be considered as a:

• “Sub system” of the environment that it is located in; during its whole life the building has impacts related mainly to environmental and energy use issues.

• “Hyper system” (a larger system, a system that contains other ones) infl uencing the comfort and productivity of the occupants.

In Figure 2.1 a scheme of the building as a system is presented. The per-formance and quality requirements included in the “building environment” interact with each other and infl uence its performance. The interaction of “building technology” with “building environment” during an early design stage aims to improve the building performance. A successful holistic approach during the design stage of an offi ce building requires considera-tion of the performance of individual parts that constitute the building system. Moreover, successful optimization of the building's performance requires a deep understanding of the interactions between these parts, as shown in Figure 2.1.

Background

33

Building Environment

Indoor environment

Architectural quality

Costs

Energy performance

Building T echnology

Integration of passive systems

Integration of active systems

Building simulation software

Out

put

Inpu

t Environmental performance

Figure 2.1 The building system.

The design team should take into account the design constraints at an early stage of the decision making process, in order to achieve an overall approach and more accurate predictions. Thereby, unpleasant surprises resulting in an increase in the building’s life cycle cost and/or impairment of its performance as to energy use and indoor climate can be avoided. These constraints (stated as input in Figure 2.1) are:

• Climate (solar radiation, outdoor temperature, etc)• Site and obstructions of the building (latitude, local daylight availabil-

ity, atmospheric conditions, exterior obstructions, ground refl ectance, etc)

• Use of the building (operating hours, occupant density, schedule and activity, etc)

• Building and Design Regulations

It is obvious that optimum building design (maximization of the output) can not be achieved, since the overall goodness can be defi ned in different ways depending both on the design constraints and on the way that the design team prioritises its goals and needs (Andersen, 2000). In sustainable


34

building design the integration of solar technologies is a delicate matter. Optimal performance of passive or active solar systems can not be achieved unless their integration is considered at an early design stage. Their effi -ciency is highly dependent on the system’s input (such as the location and use of the building) and is directly infl uenced by the building’s shape and orientation. Their integration has impacts on the life cycle cost, on the environmental profi le, and it can be crucial for the quality of the indoor environment. Validated building simulation software can ensure proper integration, when considered at an early stage. From the above it is clear that “building technology” can interact with the “building environment” in improving the overall performance.

The main aspects studied in this thesis are the energy performance and the indoor environment. The continuous lines drawn in Figure 2.1 emphasise the interactions that were studied for the “Glazed Offi ce Build-ings” project as it will be described later.

2.3 Building Environment

2.3.1 Design CriteriaThe main goal of the “Glazed Offi ce Buildings” project was to study dif-ferent types of offi ce buildings, in order to determine if energy effi ciency can be achieved without compromising performance requirements such as the indoor environment, while maintaining reasonable LCC (investment, operating and maintenance cost), low environmental impacts and good architectural quality (see Section 2.3.3). In Figure 2.2 the (top-down) hierarchy for the “Glazed Offi ce Buildings” project is presented. The ob-jectives at the top are quite abstract but they become more specifi c as one follows the hierarchy down. The bold characters emphasize the aspects studied in this thesis.

Energy Efficient O ffice Building

Indoor Environment

Environmental Performance

Architectural Quality

Costs Integration of Solar

Technologies

Figure 2.2 Design criteria of an energy effi cient offi ce building.

Background

35

2.3.2 Indoor EnvironmentAccording to Wouters (2000), early building design is the most important step in achieving an acceptable indoor environment. The energy use of the technical installations constitutes an important part of the building performance. Thus, energy effi ciency can be considered an important part of the building design. However, low energy use design can not be the only target, since other parameters also contribute to the improved overall performance.

2.3.2.1 Thermal ComfortAchieving an acceptable indoor environment with respect to energy use is one of the most diffi cult tasks when an offi ce building is designed, especially if it is highly glazed. The main components that defi ne the indoor environment are shown in Figure 2.3. These components affect the occupants’ productivity and consequently the total economic value of the building (see Subsection 2.3.5).

I ndoor Environment

Acoustics T hermal C omfort

Visual Comfort

Psychosocial Comfort

Indoor Air Quality

Productivity

Figure 2.3 Criteria of indoor environment.

As far as this thesis is concerned, thermal comfort is the main component studied (the other parameters are investigated separately within the “Glazed offi ce buildings” project). Thermal comfort can be divided into primary and secondary factors that infl uence the quality of the thermal environ-ment, as shown in Figure 2.4 (ASHRAE Fundamentals, 2005).


36

T hermal C omfort

Primary Factors

Secondary Factors

Figure 2.4 Main aspects of thermal comfort.

2.3.2.2 Conditions of thermal comfortAs outlined above, the factors infl uencing thermal comfort are divided into primary and secondary ones. The primary factors are shown in Figure 2.5:

Primary Factors

T emperature and radiation

(dry bulb, mean radiant)

O perative and resultant temperatures

Relative humidity

Air speed and turbulence

Clothing

Perception of thermal comfort

Figure 2.5 Primary factors of thermal comfort.

• Temperature and radiation (dry bulb, mean radiant): The thermal sensation is dominated by the surrounding temperature. However, the standard dry bulb temperature is not always a suffi cient indicator for establishing a good indoor thermal environment, since it does not take into account the infl uence of radiant energy. The mean radiant temperature, however, is a more appropriate thermal comfort indica-tor, since it is a measure of the average radiation exchange between the occupant and the surrounding surfaces.

• Operative and resultant temperatures: The operative and mean re-sultant temperatures empirically combine the dry bulb and the mean radiant temperatures. The operative temperature is the temperature at which a person emits the same heat output as before, but when air temperature (θa) = radiant temperature (θr) = operative temperature (θop) (Peterson, 1991). The θop does not have the same value for all the parts of the room (when the weighting method is used).

Background

37

Near cold surfaces, the operative temperature drops since the tem-peratures are low and the angle factors large (the closer the measuring point is to a surface, the larger is the angle factor). The more the occu-pant moves away from the surface, the more the angle factor decreases and thus the smaller the effect of the cold surface becomes. Since all the surfaces of a room are weighted, the walls can have a larger effect than a cold window (although the wall temperatures can be very close to the mean air temperature). Thus, the operative temperature gives a measure of a room as a whole but it is not a suffi cient indicator for showing the impact of a cold surface on the occupant’s comfort. For this reason, the directed operative temperature is preferred. This is calculated in the same way, the only difference being that weighting is performed only for the point where the occupant is placed and towards the surface of interest as shown in Figure 2.6. The operative temperature, however, can be used when persons are moving inside the space (i.e. open plan offi ce type).

Directedoperativetemperature

Operativetemperature

Figure 2.6 Operative and directed operative temperature.

• Relative humidity: Relative humidity is the ratio of the moisture content at a certain temperature to the maximum possible moisture con-tent at that temperature (until condensation starts). Generally, humidity affects the heat loss by evaporation, which is most important at high temperatures and high metabolic rates (ASHRAE Fundamentals, 2005). However, questionnaires have shown that even in the comfort zones the relative humidity has a large impact on the perception of the thermal environment. High relative humidity means that the moisture content of clothing increases which alters their insulating properties. Usually the relative humidity in an offi ce space varies between 30% and 60%.

• Air velocity and turbulence: the sensation of thermal comfort is in-fl uenced by air velocity and the scale of turbulence. Often the increased velocity can be an advantage in an offi ce space, when the temperatures


38

are higher than within the comfort range. A typical way to increase the air velocity is to use circulation fans in the rooms. However, at other times, draughts cause discomfort due to localised cooling.

• Clothing: clothing provides thermal insulation. Thus, it has an impor-tant infl uence on acceptable temperature. The choice of clothing can alter comfort preferences by as much as 2 to 3K. The unit that expresses the clothing insulation is clo; (1clo = 0.155 m2KW-1). According to ASRAE Fundamentals (2001), “because people change their clothing for the seasonal weather, ASHRAE Standard 55 specifi es summer and winter comfort zones appropriate for clothing insulation levels of 0.5 and 0.9 clo respectively”.

• Perception of thermal comfort: It is important to clarify that the con-ditions of thermal comfort are not easy to defi ne, since thermal comfort is a very subjective value. Thus, a thermal environment that is acceptable to some people may be totally unacceptable to others. According to Andresen (2000), it is not only that people are differently dressed and have different metabolic rates, but their assessment of comfort is also infl uenced by their psychosocial environment, which can not easily be taken into account by any calculation method. The most commonly used calculation methods are the ASHRAE standard 55 (ASHRAE, 1992) and the ISO Standard 7730 (ISO, 1984). In order to describe this factor it is necessary to explain the concepts of Predicted Mean Vote (PMV) index and Predicted Percent Dissatisfi ed (PPD). Both ASHRAE and ISO standards are based on the concepts of PMV and PPD developed by Fanger (1970).

o The PMV index is a measure of thermal sensation, since it expresses the correlation between indoor environment parameters and peo-ple’s sensation of thermal comfort. It is a function of activity, cloth-ing, air temperature, mean radiant temperature, relative air velocity and humidity. As it is described in ASHRAE fundamentals, (2001) “The PMV index predicts the mean response of a large group of people according to the ASHRAE thermal sensation scale”. The ASHRAE sensation scale is presented below:

+3 = hot+2 = warm+1 = slightly warm0 = neutral-1 = slightly cool-2 = cool-3 = cold

Background

39

The Predicted Percentage Dissatisfi ed (PPD) of a large group of people is an indication of the number of persons who will be in-clined to complain about the thermal conditions. After estimating the PMV, the predicted percent dissatisfi ed (PPD) with a condition can also be estimated. Those persons not scoring +1, -1 or 0 are deemed to be dissatisfi ed. From this part, the predicted percentage dissatisfi ed (PPD) of occupants could be determined.

Liddament (1996) writes that “the immediate conclusion of this work was that it was not possible to defi ne a set of thermal conditions that would satisfy everyone. Even when the average of the predicted mean vote was zero, i.e. a neutral thermal environment, 5% of the test occupants were dissatisfi ed”. Accepting that no single environment is judged satisfactory by everybody, the standards specify a comfort zone based on 90% acceptance or 10% dissatisfi ed occupants. Thus, the upper limit for operative temperature in summer is 26°C, given 50% relative humidity, sedentary activity, 0.5 clo and a mean air velocity of less than 0.15 m/s.

Based on the PMV index, the PPD index can be calculated. The PPD index predicts the percentage of the occupants who will judge their thermal comfort unsatisfactory (corresponding to a vote below –2 or above +2). PPD as a function of PMV is shown in Figure 2.7:

Figure 2.7 PPD as a function of PMV (Source: Low energy Building Design, Pedersen, (2001)).

Andresen (2000) claims that “the PMV model is interpreted as a constant set-point for a given clothing, metabolic rate and air velocity. It does not consider any effects due to adaptation, cultural differences,


40

climate and seasons, age, sex or psychosocial attributes”. According to the author, “recent research casts doubt upon the application of steady-state heat exchange equations to what in practice is a variable environment (Clements-Croome, 1997). People are not passive recipi-ents of the environment, but take adaptive measures to secure thermal comfort. They may modify their clothing or activity, modify the lighting or solar heat gains, or modify the ventilation rate through opening of doors and windows. This suggests that a more “adaptable” method of comfort judgment is needed”.

Other parameters such as state of health, level of physical activity, gen-der, working environment and individual preferences can also infl uence the perception of thermal comfort.

Although the secondary factors of thermal comfort, apart from asymmetric thermal radiation (directed operative temperature), are not studied in the present thesis, a brief description follows.

Secondary Factors

Nonuniformity of the environment

Age Gender Adaptation

Seasonal and Circadian Rhythms

Asymmetric thermal

radiation

Draught

Vertical air temperature difference

Warm and cold floors

Figure 2.8 Secondary factors of thermal comfort.

As shown in Figure 2.8, the secondary parameters that infl uence the thermal environment are as follows

Background

41

• Non-uniformity of the environment

The non-uniform conditions that lead to local discomfort are prob-ably the most important of the above secondary factors. According to ASHRAE Fundamentals (2001), “a person may feel thermally neutral as a whole but still feel uncomfortable if one or more parts of the body are too warm or too cold”. A number of reasons can lead to a non-uniform environment. Some of these are:

o a cold windowo a hot surfaceo draught o a temporal variation of these

In buildings, asymmetric or non-uniform thermal radiation can be caused usually by poor and large windows, non-insulated walls, or improperly sized heating panels on the wall or ceiling, etc. In offi ce buildings the most common reasons for discomfort due to asymmetric thermal radiation or draught are large and/or poor windows (during the heating and cooling periods) and improperly sized, installed or operated ceiling cooling systems.

Draught is an undesired local cooling of the human body caused by air movement. ASHRAE Fundamentals (2001) describe draught as one of the most annoying factors in offi ces. Draught makes the occupants demand higher air temperatures in the room or stop the ventilation systems. This can often lead to temperatures above the comfort levels.

Vertical air temperature difference: In most of the offi ces (or generally spaces in buildings) the air temperature is not completely uniform but increases with height above the fl oor. When the gradient is suffi ciently large, local warm or cold discomfort can occur at the head and/or the feet, although the body as a whole is thermally neutral.

Warm and cold fl oors: According to ASHRAE Fundamentals (2001), “due to the direct contact between the feet and the fl oor, local discomfort of the feet can often be caused by a too-high or too-low fl oor temperature. Also, the fl oor temperature has a signifi cant infl uence on the mean radiant temperature in a room”. Often, when the fl oor is too cold, the occupants feel cold discomfort in their feet and as a result they increase the temperature level in the room. During the heating period, this can be a reason for the increase in energy use. A radiant system, which radiates heat from the fl oor, can solve this problem. A diagram of how the PPD increases due to the cold/warm ceiling and walls is shown in Figure 2.9:


42

Figure 2.9 Infl uence of ceiling and wall temperature on the PPD (ASHRAE Fundamentals, 2001)

• Age

Since metabolism decreases with age, young and old people do not always have the same preferences when it comes to thermal comfort. Often older people prefer higher ambient temperatures. However, pre-vious research showed that sometimes the thermal environment in an offi ce can satisfy both ages. The need for higher ambient temperature for older people in their homes can be explained by the lower level of activity.

• Gender

Both men and women can be satisfi ed with the same thermal condi-tions. In ASHRAE Fundamentals (2001) it is mentioned that the temperature of women’s skin and evaporative loss are slightly lower than those of men and this balances the lower metabolism of women.

Experiments proved that people can not adapt to preferring warmer or colder climates. According to ASHRAE Fundamentals (2001) “it is therefore likely that the same comfort conditions can be applied throughout the world. However in determining the preferred ambient temperature from the comfort equations, a clo value that corresponds to the local cloth-ing habits should be used”. Thus, adaptation does not really infl uence

Background

43

the preference of the occupants regarding the ambient temperature. However, people used to living or working in warm climates can more easily stand higher temperatures while maintaining the same levels of performance than people from colder climates.

• Seasonal and circadian rhythms

According to ASHRAE Fundamentals (2001) there is no difference between indoor thermal conditions of comfort during the winter and the summer. However, the preference of an occupant for thermal com-fort may change during the day since the body has a lower temperature rhythm during the early morning hours and a higher one late in the afternoon.

2.3.2.3 Thermal comfort and productivityAlthough a lot of attempts have been made to correlate productivity with indoor climate factors, no detailed studies have been made so far that can predict accurately the interaction between these two parameters. The occupants’ effi ciency is a really complicated matter, since many different parameters can infl uence it separately and together at the same time.

Wyon (2000a) describes how discomfort conditions lead to reduced productivity. The author claims that a change from 18ºC and dry to 28ºC and humid can increase the proportion of dissatisfi ed from 10 to 90%. Additionally, the thermal effect is greater for clean air than for normally polluted indoor air.

According to the IDA ICE 3.0 manual, Wyon (2000b), for operative temperature between 20 and 25°C no work is regarded as lost. Above and below these limits experiments show an average loss of 2% in perform-ance per degree.

Hanssen (2000) claims, that the ambient temperature can give the most specifi c correlation between indoor climate parameters and productivity. The author, referring to previous research by Wyon (1987), describes the correlation between air temperature and the number of accidents, pro-ductivity of manual work, fi nger dexterity, number of breaks and mental performance.

The (lower and higher) x-axes (Figure 2.10) show the activity level and the clothing (summer and winter case) insulation of the occupants. The x-axis in the middle of the diagram can be applied both for active work (1.4 met) during the summer (0.6 clo) and for sedentary work (1 met) during the winter (1 clo). As the author describes, “For an offi ce workplace it may be especially appropriate to examine the effect of air temperature on mental performance and manual work”.


44

Figure 2.10 Correlation between air temperature and the number of accidents, productivity of manual work, fi nger dexterity, number of breaks, and mental performance (reference: Wyon, 1986).

It is obvious that it is not always easy to evaluate and improve the thermal conditions in order to increase the comfort and productivity of the oc-cupants. Advanced computer simulation programs often predict

• operative temperatures• temperature swings• relative humidity and air velocity

in a building with “predictable” occupants. However, the behaviour and adaptability of occupants in reality are far harder to predict. A quite prac-tical approach in order to ensure acceptable indoor thermal environment could be to ensure that the air temperature and the “cold draught” from windows stay within acceptable limits.

Background

45

2.3.2.4 Other indoor climate parameters that infl uence the occupants’ health and productivity

• Indoor air quality

According to Nathanson (1995) the quality of the indoor environment depends on the interaction between the site and the

o climateo building systemo potential contaminant sources (e.g. furnishings, moisture

sources, work processes and activities, outdoor pollutants, etc) o building’s occupants

The HVAC system is designed to provide thermal comfort, distribute outdoor air to the occupants, remove odours and contaminants, or dilute them to acceptable levels.

The infl uence of the indoor environment on the comfort and health of the occupants has been the focus of research for many years. However, according to Hanssen (2000) there is relatively little research examining the total effect of indoor air quality on human wellbeing, employee performance and productivity at work.

• Visual comfort

When designing an offi ce building attention should be paid to the visual comfort of the occupants. Although electricity savings through daylight utilisation may not be so impressive (compared for example with those for heating or cooling), correct distribution of light may create a more pleasant indoor environment and thus improve the mood and productivity of the occupants. In “Sustainable Building Technical Manual” (Public Technology Inc and US Green Building Council, 1996), it is pointed out that “daylighting creates healthier and more stimulating work environments than artifi cial lighting systems and can increase the productivity up to 15%. Daylighting provides also changes in light intensity colour and views that help support worker productivity”. Surveys mention that 90% of the occupants prefer to work close to the window, having a view to the outside.

The comfortable visual environment depends on vision, perception and what we want to see in different room confi gurations and for differ-ent activities. The absence of sensation of physiological pain, irritation or distraction is the main aim, when the visual properties of an indoor environment are optimized. Christoffersen (1995) writes in his PhD


46

thesis that “visual perception is an active, information-seeking process, partly conscious and partly unconscious, involving many mechanisms in a cognitive process interpreted by the eye and the brain”.

When the shape and position of windows for an offi ce building are designed, attention should be paid to the provision of visual comfort for the occupants, without forgetting the importance of solar shading to avoid overheating. The visual function parameters which determine good visibility and pleasant indoor environments are the:

o illuminance, luminance (level and ratios) and daylight factoro distribution - Uniformity of light across a surfaceo glareo direction

2.3.3 Architectural qualityThe architectural design (essential for achieving architectural quality) of an offi ce building has great impact on both energy use during the occupation stage and provision of comfort for the occupants. The building’s shape and location, the façade’s orientation and the integration of passive or active solar systems are some of the parameters that infl uence the performance of the building in terms of energy use. On the other hand the occupants’ perception of indoor environment is closely related to their sociological needs, psychological state, and individual differences infl uencing directly their comfort and productivity (Poirazis, 2005a). Thus, it is clear that many parameters have to be examined at the early design stage in order to succeed architectural quality (i.e. energy effi cient design and attractive working environment).

In order to achieve an energy effi cient design, different parameters have to be considered and combined carefully. The location and orientation of the building, the area of the building’s skin in relation to the volume of space enclosed, the type of façade (proportion of glazing, structure, etc) are crucial for the building’s performance. In highly glazed offi ce buildings, the building’s skin is obviously more sensitive to the outdoor environment. High solar gains during the summer may lead to overheating problems during the summer, increasing the energy demand for cooling. Furthermore, the greater the area of building skin in relation to the volume of space enclosed, the more the building is infl uenced by heat exchanges at the skin. Thus, it is a general rule that a square fl oor plan is thermally more effi cient than a rectangular one because it contains less surface area over which to lose or gain heat. On the other hand, this building form is less effi cient in terms of daylighting and passive solar heating and cooling.

Background

47

The shape and orientation of the building have also great impact upon wind driven air infi ltration through the envelope.

Obviously more parameters than an acceptable thermal and visual en-vironment should be considered in order to achieve an attractive working atmosphere. Parameters related with the perception of indoor environment (such as privacy, interaction levels, territoriality and crowding) may have a great impact on the occupants responses regarding mainly sociologi-cal, psychological (such as visual and acoustic privacy and aesthetic) and physiological determinants. A more detailed description is provided by the “Design Guide for Interiors”, (U.S. Army Corps of Engineers, 1997).

2.3.4 Environmental performanceAnother quality requirement that should be taken into account in a holis-tic approach of a building design is the environmental performance. The impacts of buildings on the environment are diverse. Not only during the construction stage, but also during the occupation and demolition stage, the building interacts with the environment in different ways.

As a physical structure, the building is composed of different elements. These elements are extracted, manufactured, assembled, maintained, demolished and fi nally disposed of. The total environmental impact of the materials used is the sum of the impacts caused in each of the above stages. As a “living part” the building has inputs (energy use, services) and outputs (CO2 emission, wastes, etc).

The environmental performance of a building depends (a) the con-struction (sum of the performance during the manufacturing stage of the materials, the transportation stage and the erection of the building), (b) occupation/maintenance stage (according to Adalberth (2000) for modern apartment buildings the 70-90% of the total energy use (during the life cycle) is used during the occupation stage) and (c) the refurbishment and demolition stage of an offi ce building (obviously less important than the ones described above).

The study of a whole life cycle of a product or a process (in this case a building) from the extraction of raw materials to the disposal or recon-struction is called Life Cycle Analysis (LCA). At this point, it is necessary to set out the advantages and disadvantages of this method and to make clear the importance of its use during the design stage of a building.

The main benefi t of the LCA is that it provides information on the environmental performance of different building concepts in a really accessible format. On the other hand, the lack of consistent and peer reviewed international level databases of Life Cycle Inventories of build-ing related products is one of the main problems which limits the use of


48

LCA. However, an LCA limited to the period from construction until deconstruction is feasible.

The holistic approach during the design stage of a building involves variables that often interact with each other. Impacts on the surround-ing environment, energy needed for the construction and maintenance of the building, use of recycled building components and integration of solar technologies that improve the environmental profi le of the building are some of the factors that defi ne the environmental performance of the building.

The main reason for using an LCA is to insert it as part of the decision making process in order to optimize the effi ciency of the system over the life span of the building. Since a lot of parameters are involved, the com-plexity increases and the goals should be prioritized from the early design stage in order to make the process more clear.

2.3.5 CostsIn the existing literature three ways of calculating the cost of a building are given.

• Investment costs: consideration of the investment cost only• Life cycle costs: Cost over the whole life of a building• Total economic value: It goes further than life cycle costing, since it

also includes more “hidden” building related costs or profi ts such as the productivity of the workers in the building.

According to Fuller and Petersen (1996) energy conservation projects provide excellent examples for the application of LCCA. On the other hand, the total economic value is time restrictive, since it often involves very detailed calculations that require detailed input.

For the ongoing “Glazed Offi ce Buildings” project the Life Cycle Cost (LCC) has been calculated (Sjödin, 2007). The productivity of the occu-pants (such as lost working hours) is also calculated in a simple way taking into account thermal comfort criteria (Poirazis, 2005b). The reason for the decision to calculate the LCC is that calculation of only the invest-ment costs for the suggested alternatives does not provide information on the energy, environmental and maintenance costs, which is crucial when it comes to energy effi cient offi ce buildings. The integration of solar technologies with alternative heating, ventilating, and air conditioning (HVAC) systems and strategies can provide a wide variety of models (some of which are considerably more energy effi cient than others), interesting to study in a life time prospective.

Background

49

2.4 Building technologyAt an early stage the “building technology” should be involved in the build-ing design process (Figure 2.1) with the aim of improving the building’s performance. By integrating new technologies into the building process, the complexity of the system increases. In these cases the implications should be considered and the evaluation process should start from the beginning, considering the problem as new.

As an example, in order to clarify the foregoing, a solution regarding thermal comfort problems of highly glazed buildings can also be provided by proper integration of double skin façade systems (integration of passive systems). In this way the provision of a more uniform thermal environ-ment can be achieved (performance requirement), by controlling the cavity temperatures to avoid cold walls and temperature asymmetries.

From the above it is clear that the interactions between the constituents (performance and quality requirements) of the building environment and the implications of building technology are a complicated and deli-cate matter. Prioritizing the main goals and the quality requirements to be fulfi lled, deciding about the trade off values and creating a common understanding between the different sides of the design team is the fi rst step to an improved building performance. The complexity of the system requires continuous focus on the main goals to be achieved without in any case compromising the performance requirements. The higher risk of facing unexpected problems, when newer and more complicated tech-nologies are integrated, should be considered and a gradual integration process should be adapted.

2.4.1 Glass in buildings

2.4.1.1 GeneralThe main function of architectural glass is to transmit daylight. Increased glazing areas that admit daylight were achieved at the beginning of the nineteenth century, when the development of framed building structures, suspending the heavy load bearing wall, liberated the window area (But-ton and Pye, 1993). Glass and the steel skeleton came together as key elements in the modern architectural movement, in order to increase natural daylight, transparency, health and social well-being. For the last two centuries glass has had its own material developments, which brought new aesthetic qualities to architecture. The method of suspending glass assemblies as a curtain of glass leads to the full liberation of glass from the building structure. This liberation, however, was followed by numerous


50

performance demands of the building skin. According to Button and Pye (1993) today, the building structure is no longer of primary architectural importance; of equal design consideration and equal fi nancing investment are heating, lighting and air conditioning services.

During the last decades there has been much research on variable transmission glass and new technologies in photochromic, thermochromic and electrochromic panes. This research is driven by the desire for energy savings and improved occupant comfort. The development of coated and body tinted glass aims to cover needs for appearance, view (in and out), daylight, passive solar gains, solar control and thermal insulation; laminated glass provides durability, fi re resistance, safety and explosion and bullet impact protection.

2.4.1.2 Basic physics of the glass

Radiation Electromagnetic radiation is the principal energy source, which provides both heat and light. The complete electromagnetic spectrum is delineated in Figure 2.11. The intermediate part of the spectrum, which extends from a wavelength of approximately 0.1 to 100 µm and includes part of the UV and all of the visible and infrared (IR) is considered as thermal radiation and is related to heat transfer.

10-5 10-4 10-3 10-2 10-1 1 10 102 103 104

X rays

Ultraviolet

Visible

Thermal

Infrared

Microwave

Gamma rays

Figure 2.11 Spectrum of electromagnetic radiation.

All bodies emit and absorb energy in the form of electromagnetic radiation. The thermal radiation emitted by a surface varies for different wavelengths. The term “spectral” is used in order to refer to the nature of this depend-ence. The spectral distribution depends on the nature and temperature of the emitting surface. To properly quantify radiation heat transfer, we must be able to treat both spectral and directional effects.

Background

51

Surface absorption, refl ection and transmissionWhen the irradiation interacts with a semitransparent object (such as glass), portions of this radiation may be refl ected, absorbed or transmit-ted, as shown in Figure 2.12. From a radiation balance of a medium, it follows that:

Gλ= Gλ,ref+ Gλ,abs+ Gλ,tr

Irradiation (Gλ)

Reflection

(Gλ, ref)

Absorption (Gλ,abs)

Transmission (Gλ, tr)

Figure 2.12 Refl ection, absorption and transmission of solar radiation in glass.

Refl ectance occurs when the surface of a material refl ects an incident beam of light and it expresses the fraction of incident radiation refl ected by the glass. Refl ection can be specular, diffuse or a mixture of the two.

Specular refl ection (a)

If a material surface is microscopically smooth and fl at, such as fl oat glass, the incident and refl ected light rays make the same angle with a normal one to the refl ecting surface, producing specular refl ection.

Diffuse refl ection (b)

If a material has a rough surface, that is if it is not microscopically smooth, diffuse refl ections will occur. Each ray of light falling on a small particle of the surface will obey the basic law of refl ection but, because these particles are randomly oriented, the refl ections will be randomly distributed. A perfect diffusely refl ecting surface would in practice refl ect light equally in all directions, giving a perfect matt fi nish.


52

θ

θ

a b

Figure 2.13 Specular and diffuse refl ection.

Absorptance expresses the fraction of incident radiation absorbed. In other words, absorptance is that part of the incident light which is lost in the body of the glass, increasing its temperature.

Transmittance expresses the fraction of incident radiation directly transmitted through the glass. Transmittance is that part of the incident light remaining after refl ection and absorption. Transmitted light is subject to modifi cation by refraction, diffusion and colouring.

2.4.1.3. Thermal functions of the glass

Glass and thermal comfortWindows infl uence occupant comfort by:

• heat gain or heat loss through the glass, which either raises or lowers the room air temperature.

• radiation exchange between the glass and occupant.• increased convection close to window, thus increased cold down

draught during winter.

The glass surface temperature infl uences the thermal comfort of an oc-cupant close to the window, due to the heat loss produced by long wave radiation exchange between the occupant and the window. During winter, the glass temperature is often lower than the other room surfaces, producing a loss of heat from the occupant’s body surface by long wave radiation and contribution to cold discomfort. During summer, the glass temperature can be higher than that of the other surfaces. The long wave radiation in combination with any shortwave solar radiation received through the

Background

53

glass is absorbed by the occupant and contributes to a sensation of hot discomfort.

Heat lossHeat loss is quantifi ed through the U value (thermal transmittance) measured in W/m2K. Thermal transmittance is the rate of loss of heat per square metre under steady state conditions for a temperature difference of one degree (Kelvin or Celsius) between the inner and outer environments separated by the glazing. Heat loss can be also quantifi ed by the thermal resistance (R=1/U).

According to Button and Pye (1993) there are three stages of heat loss through glass products:

• between the internal glass surface and the room surfaces• through the glass product• between the outdoor environment and the outer glass surface

Heat loss to the internal glass surface: Heat is lost to the internal glass surface from the room, whenever the glass surface is at a lower tempera-ture than the internal air temperature and the room surface temperature, through:

• exchange of long wave radiation between the glass surface and the room surfaces.

• convection/conduction from the room air moving over the surface of the glass.

Usually, the heat loss by radiation exchange is the greater heat loss (unless the glass surface has a low emissivity coating).

Heat loss through the glass product: Generally, glass is a very poor insu-lating material. However, due to technological developments there are a number of ways to decrease heat loss through glazing as described in Subsection 2.4.2.1.

Heat loss from the outer glass surface: The fi nal stage of heat loss is from the outer glass surface. As with the inner glass surface, heat transfer oc-curs by long wave radiation exchange and by convection-conduction. The balance and magnitude of heat transfer depend on the temperature of the surrounding outside surfaces and the sky temperatures. With clear skies, sky temperatures can be extremely low; this effect is demonstrated by the formation of dew and frost on surfaces exposed to clear skies due to their cooling below ambient air temperature.


54

2.4.2 Single skin façadesNowadays, there is a growing interest in using highly glazed facades in commercial buildings. The trend of covering large portions of the façade or even the entire façade by glass has its origin in Europe and is expanding to other regions. As with many other architectural trends, understanding and improving the building performance of highly glazed buildings is very important. Prior simulation studies have shown that it should be technically possible to produce an allglass façade with suffi cient energy and indoor climate performance although it is not a simple challenge (Lee, et al., 2002). This Subsection gives a brief background to the problems often met and the solutions given, in order to improve the building’s performance.

When glazed façades are designed, several devices are often imple-mented, in order to keep the heat losses low and to avoid undesired heat gains through solar radiation (during summer). According to Compagno (2002) the two main criteria when designing a fully glazed façade are the number of glazing skins incorporated in the design (single or multiple skin façades) and the positioning of shading devices.

2.4.2.1 GlazingIn energy effi cient design the proper selection of glazing elements is prob-ably the most complex task. Glazing and window design are two areas in which great technical developments have occurred over the last years. In order to achieve good window design, it is essential to fi nd the balance between demands which are often confl icting such as passive heating and cooling functions, e.g. allow solar gains but avoid excessive solar heat, provide suffi cient daylight without causing glare, allow controllable ven-tilation into the building but keep out the noise, allow visual contact with the surroundings but ensure acceptable privacy levels (A Green Vitruvious, 1999). This Subchapter focuses on the thermal insulation that glazing can provide and suggests a number of ways to decrease heat loss through it.

Single glazing provides relatively little resistance to loss of heat, since the glass is a poor insulator. To decrease the thermal transmittance, a second pane of glass separated from the fi rst pane by an air space can be added. This layer of enclosed air provides extra thermal resistance to long wave radiation exchange.

The incorporation of an air space provides several opportunities for increasing the thermal resistance of glazing:

• increasing the width of the air space: by increasing the width of the air space, extra resistance is provided. There is a limit due to convec-

Background

55

tion within the air space, which occurs at about 15mm width, after which little extra thermal benefi t is obtained. Adding a third pane of glass to give a second air space provides further improvement.

• Incorporating low emissivity coatings: The use of a low emissivity (low E) coating on the glass makes it possible to reduce the long wave radiation exchange between the panes. The higher insulating effect (lower U value) provided by a Low E coating in a double glazed unit is due to the high refl ectance of long wavelength radiation. In cold climates the higher temperature of the inner glass surface of double glazed units using Low E coating diminishes the effect of long wave radiation, which causes discomfort near the window.

• Using gases of lower conductivity: Sealed Low E double glazed units may contain gases with lower thermal conductivity than air such as argon, providing further decrease in U value.

• Evacuating the air space: the air space may be fully or partially evacuated.

The properties of glass, such as solar shading and emissivity infl uence the transmission through the glass (Carlsson, 2005). Drastic changes can be obtained by applying a coating on the glass. Coatings can infl uence the range of transmitted radiation and its absolute level. The coatings can be refl ective and selective.

Effi cient solar shading can be obtained by refl ective coatings. Increased refl ection results in reduced total transmission. Currently, the total solar energy transmittance (g value) for a sealed double glazed unit can be varied between 0.2 and 0.7 with a daylight transmittance between 0.3 and 0.8 W/m2K.

Lower U values can be obtained with coatings of low emissivity. The emissivity can be reduced from 0.87 to 0.04. The infrared radiation can be reduced to 20 %, without lowering daylight transmittance below 0.75. This type of coating is selective, as it allows transmittance of the main part of daylight, but has a high refl ectivity of the infrared radiation. Currently the U value (middle of the glass) for a sealed double glazed unit can be varied between 2.8 and 1.1 W/m2K. In modern offi ce buildings sealed double glazed units are preferably used, often with a refl ective coating.

2.4.2.2 Shading devicesIn order to achieve a certain level of solar transmittance through the single skin façade, solar control glass is often used. However, since the properties of this glass are fi xed, they restrict useful solar gains during cold months and they can reduce daylight levels. Thus, by providing additional adjust-


56

able shading devices the building performance can be further improved. Some of these devices are:

• exterior shading devices: The main advantage of these devices is that the heat resulting from the radiation from the device itself remains out of the building, keeping the cooling load levels lower during summer months. The main disadvantage, however, is that they are exposed to the effects of weather, often resulting in high mainte-nance and cleaning costs. If the exterior shading is movable, then the low solar transmittance effect could be limited to the summer months, when cooling is needed; if they are fi xed, they have similar effect as the solar control glass.

• interior shading devices: This type of shading device is less effective, since the radiation absorbed by the devices stays in the room raising the cooling demand. However, the cleaning and maintenance of these devices is much simpler than with exterior ones. Additionally, internal shading can provide the “clean façade” look, which is quite often required by architects.

• intermediate shading devices: Shading devices placed in between the panes of glass are less common in offi ce buildings. Costs associated with cleaning are lower than with the exterior ones but maintenance may be more expensive, mostly when the electric motors are also incorporated inside the cavity (Compagno, 2002). The increase in temperature between the panes due to absorption by the shading should be considered in order to avoid cracking of the glass due to the dramatic temperature increase.

A comprehensive study of several aspects related to solar shading devices has been carried out within the “Solar Shading” project at the Division of Energy and Building Design, Department of Architecture and Built Envi-ronment, Lund University. The project included the following tasks:

• determination of the primary and total solar energy transmittance (g value) of shading devices through measurements;

• development of an advanced computer program (Derob-LTH) and a user-friendly design tool (ParaSol) to predict the impact of shading devices on energy use in buildings;

• parametric studies as a basis for the development of design guidelines aimed at architects, engineers and consultants;

• measurement of the daylight transmittance and interior illumi-nance/luminance conditions in rooms equipped with shading devices.

Background

57

2.4.3 Double skin façades

2.4.3.1 GeneralThe double skin façade is a system consisting of two glass skins (single or double) placed in such a way that air fl ows in the intermediate cavity. The distance between the skins usually varies from 0.2 m up to 2 m. For protec-tion and heat extraction reasons the solar shading devices are placed inside the cavity. The ventilation of the cavity can be natural, fan supported or mechanical; the origin and destination of the air can also vary depending on the location, the use and HVAC strategy of the building.

The advantages of double skin façades compared with single skin facades are improved acoustic insulation, protection of shading devices and provision of natural ventilation in the offi ce spaces. However, energy reduction and provision of an improved indoor thermal environment can also be achieved, when these are designed and integrated properly. Due to the additional skin, a thermal buffer zone is formed which reduces the heat losses and enables passive solar gains. During the heating period, the solar preheated air can be introduced inside the building providing natural ventilation with a good indoor climate retained. On the other hand, during the summer overheating problems are often referred to when the façade is poorly ventilated (Poirazis, 2004). Different confi gurations can result in different ways of using the façade, proving the fl exibility of the system and its adaptability to different climates and locations.

A detailed description of this system can be found in the Literature review for “Double skin façades for offi ce buildings” by Poirazis (2004).

2.4.3.2 Classifi cation of double skin façadesThe most common way to categorize the system is according to the type (geometry) of the cavity, as described below.

• Multi storey: In this case no horizontal or vertical partitioning exists between the two skins. The air cavity ventilation is provided via large openings at the bottom and top of the cavity.

• Corridor: The cavity is partitioned horizontally for acoustical, fi re security or ventilation reasons.

• Box window: In this case horizontal and vertical partitioning divides the façade into smaller and independent boxes.

• Shaft box: In this case a set of box window elements are placed in the façade. These elements are connected via vertical shafts situated in the façade. These shafts ensure an increased stack effect.


58

2.4.3.3 Technical description of the cavityThe most common types of glass panes used for double skin façades are described below.

The internal skin is often a thermal insulating double glazed unit (when the air enters the cavity from outdoors). The panes are usually toughened or unhardened fl oat glass. The gaps between the panes are fi lled with air, argon or krypton.

The external skin is often a toughened (tempered) single pane of glass. Sometimes it can be a laminated glass instead.

Lee et al. (2002) claim that the most common exterior layer is a heat-strengthened safety glass or laminated safety glass. The second interior façade layer consists of fi xed or operable, double or single pane, casement or hopper windows. Low emittance coatings on the interior glass façade reduce radiative heat gains to the interior.

Oesterle et al. (2001) suggest that for a higher degree of transparency, fl int glass (glass with high refraction and low dispersion) can be used as the exterior layer. Since the number of layers and the thickness of the panes are greater than in single skin construction, it is really important to maintain a “clear” façade, if transparency is the goal. The main disadvantage in this case is the higher construction costs, since fl int glass is more expensive than the normal glass.

The shading devices are usually horizontal louvres placed inside the cavity for protection. In the existing literature, there is no extended des-cription concerning the material and the geometry of the shading devices used for double skin facades. However, it is mentioned that in large scale projects it is useful to investigate the material and positioning of the glass and shading devices inside the cavity. It is also worth considering proper combination of these two elements in order to succeed in attaining the desired indoor operative temperatures.

2.4.3.4 Advantages and disadvantages of double skin façades

The advantages and disadvantages outlined in the existing literature of the double skin façade system are briefl y described below:

• Advantages

Lower construction cost compared with solutions that can be provided by the use of electrochromic, thermochromic or photochromic glass (the properties of which change according to climatic or environmental conditions).

Background

59

Acoustic insulation: In the view of some authors sound insulation can be one of the most important reasons to use a double skin façade. Reduced internal noise levels inside an offi ce building can be achieved by reducing both the transmission from room to room (internal noise pollution) and the transmission from outdoor sources i.e. heavy traffi c (external noise pollution). The type of double skin façade and the number of openings can be really critical for sound insulation concerning the internal and the external noise pollution.

Thermal insulation: During the winter, the external additional skin pro-vides improved insulation by increasing the external heat transfer resistance. The reduced air fl ow and the increased temperature of the air inside the cavity lower the heat transfer rate on the surface of the glass, which leads to reduction of heat losses.

During the summer, the warm air inside the cavity can be extracted by mechanical, fan supported or natural ventilation. Certain façade types can cause overheating problems. However, a completely openable outer layer can solve the overheating problem during the summer months, but will certainly increase the construction cost.

Night time ventilation: During the hot summer days, when the external temperature is higher than 26˚C, the interior spaces may easily become overheated. In this case, it may be energy saving to pre-cool the offi ces during the night using natural ventilation. In this case, the indoor tem-peratures will be lower during the early morning hours, providing thermal comfort and improved air quality for the occupants.

Energy savings and reduced environmental impacts: In principle, double skin façades can save energy when properly designed. Often, when the conventional insulation of the exterior wall is poor, the savings that can be obtained with the additional skin may seem impressive.

Better protection of the shading or lighting devices: Since the shading or lighting devices are placed inside the intermediate cavity of the double skin façades, they are protected from both the wind and rain.

Reduction of the wind pressure effects: The double skin façades around high rise buildings can serve to reduce the effects of wind pressure.

Transparency – Architectural design: In almost all the literature, refe-rence is made to the desire of architects to use bigger proportions of glazed surfaces.


60

Natural ventilation: One of the main advantages of the double skin façade systems is that they can allow natural (or fan supported) ventilation. Dif-ferent types can be applied in different climates, orientations, locations and building types in order to provide fresh air before and during the working hours. The selection of double skin façade type can be crucial for temperatures, air velocity, and the quality of the introduced air inside the building. If designed well, the natural ventilation can lead to a reduc-tion in energy use during the occupation stage and improve the comfort of the occupants.

Thermal comfort – temperatures of the internal wall: Since the air inside the double skin façade cavity is warmer than the outdoor air during the heating period, the interior part of the façade can maintain temperatures that are closer to the thermal comfort levels (compared with single skin facades). On the other hand, during the summer it is really important that the system is well designed, so effi cient heat extraction ensures that the temperatures inside the cavity do not increase dramatically, leading to high oprative temperatures.

Fire escape: Claessens and De Hedre mention that the glazed space of a double skin façade may be used as a fi re escape.

• Disadvantages

Higher construction costs compared with a conventional façade. For example, the additional skin increases the weight of the construction, which increases the cost.

Additional maintenance and operating costs: When the double skin and the single skin type of façade are compared, it is easily seen that the double skin type has higher cost regarding construction, cleaning, operat-ing, inspection, servicing, and maintenance.

Fire protection: It is not yet very clear whether or not the double skin façades can be positive regarding the fi re protection of a building. How-ever, some authors refer to possible problems caused by the room to room transmission of smoke in case of fi re.

Reduction of rentable offi ce space: The width of the intermediate cavity of a double skin façade can vary from 20 cm to several metres. This results in the loss of useful space. Often the width of the cavity infl uences the properties inside it (i.e. the deeper the cavity, the less heat is transmitted by convection when the cavity is closed) and sometimes the deeper the cavity, the greater the improvement in thermal comfort conditions next to the

Background

61

external walls. Thus, it is quite important to fi nd the optimum depth of the façade, to be narrow enough so as not to lose space, and deep enough so as to make it possible to use the space close to the façade.

Overheating problems: If the double skin façade system is not properly designed, the temperature of the air in the cavity may increase, overheat-ing the interior space.

Increased air fl ow velocity inside the cavity, mostly in multi storey types. The possibility of important pressure differences between offi ces is men-tioned in the case of natural ventilation via the cavity.

Daylight: The double skin façades are similar to other types of glazed facades (i.e. single skin façade). However, Oesterle et al., (2001) describe that double façades reduce the quantity of light entering the rooms as a result of the additional external skin.

Acoustic insulation: It is possible that sound transmission problems (room to room or fl oor to fl oor) may arise if the façade is not designed properly.


62

State of the art

63

3 State of the art

3.1 Glazed offi ce buildings in Nordic climates

Subchapter 3.1 is based on personal communication with Åke Blom-sterberg at the Division of Energy and Building design, Department of Architecture and Built Environment, Lund University.

3.1.1 GeneralThe offi ce buildings as known today are likely to retain their validity in the foreseeable future. Although there has been, and still is, a dramatic development of the infrastructure for communications (mobile phones, laptops, e-mails, etc.), the change in offi ce practice is far less dramatic (Kleibrink, 2002). Nowadays, the exchange of information between and within organizations is very often achieved by e-mail and the activities are increasingly dominated by discussions and fl ow of information at all levels. High interaction levels may however lead to reduced occupant productivity, due to acoustic disturbance. The increase in teamwork and communication, however, can lead to activities and persons disturbing each other, especially if the spatial organisation is not appropriate. The plan of an offi ce environment establishes the privacy (both acoustic and visual) level at which the offi ce functions. Therefore, it is still necessary to perform concentrated individual work in undisturbed surroundings. Modern offi ce work is characterized by quick changes between these two types of activity, so the challenge today is to provide for a combination of individual work and teamwork, while also providing for fl exibility for unforeseen developments.

Moreover, in new constructions the conventional envelope of offi ce buildings tends to be replaced by a highly glazed one that can lead to a pleasant visual indoor environment. The recent trend of transparent buildings is often initiated by architects, in order to provide more day-light and view to the occupants. Depending on the task of the occupants this can increase their productivity (as stated in the Sustainable Building


64

Technical Manual in 1996, productivity can increase by up to 15% when daylight is provided instead of artifi cial lighting). In some cases, however, the occupants can often feel distracted or even annoyed when they can be seen from outside.

3.1.2 Layout of typical offi ce buildingsThere are at least four different concepts of offi ce layout:

• unit or cell-type offi ce• open-plan and group offi ce• combination offi ce• the so-called “business club”

The cell-type offi ce is the most traditional form, where single or double rooms are located along artifi cially lighted corridors. A single person offi ce is very good for concentrated work, but does not promote informal com-munication between colleagues. Typical sizes for a single room are: width 1.35 m by depth 4.20 m, width 2.7 m by 4.20 m, width 3 m by depth 3.6 m or width 3 m by depth 5 m. The corridor can be 2 m wide.

The open-plan and group offi ce with hundreds of persons working was designed to encourage communication. Often, many of these offi ces have after some time been divided by head height cupboards and plants into almost cell-type offi ces. What some people perceive as a disadvantage is the lack of individual control of indoor climate and lighting. For routine processing work requiring a high degree of informal communication this type of offi ce can be preferable.

The combination offi ce, developed in Scandinavia at the end of the seventies, combines the advantages of the cell-type offi ces and open-plan offi ces, while avoiding the disadvantages. The workplaces are located in cell-type spaces along the façade and are separated from the indirectly lit internal zones by room high glazed walls. Every workplace has access to discussion areas, direct visual contact with the outside world and the means to control the climate individually. The common space in the middle serves a number of employees and offers communal services like meeting areas, copiers, printers, facilities for coffee breaks. The glazed wall provides sound attenuation for the workplaces, while allowing visual contact with colleagues. The size of the individual workplaces/rooms is similar to the cell-type ones mentioned above. The common space can be 4.8 m deep.

An offi ce should meet the following four criteria:

• Flexibility: workplaces should be standardized, but should also be adaptable to individual needs.

State of the art

65

• Functional effi ciency: the working space should meet physiological requirements, ergonomic standards and statutory requirements, in order to support optimal working conditions.

• Contact quality: the working spaces should contribute to the transpar-ency of the activities, carried out in the offi ce and encourage commu-nication and synergetic effects between employees and departments

• Corporate culture: the message conveyed by the working spaces should promote the employees’ identifi cation with the company and its pro-ducts and communicate company values internally and externally.

Typical depths for offi ce buildings are

• 12 to 13 metres for a cell-type offi ce• 13. 5 to 15.5 metres for a combination offi ce

3.1.3 Offi ce buildings in SwedenOffi ce buildings account for a signifi cant proportion of the fl oor area in non-industrial buildings in Sweden. The total fl oor area in offi ce build-ings is approximately 30 million m2 (usable area to let) (SCB, 2001). The completed fl oor area is almost evenly distributed over the decades, but with less construction during the last decade (as shown in Table 3.1).

Table 3.1 Floor area, millions m2, broken down by year of completion (SCB, 2001).

Year …-1940 1941-1960 1961-1970 1971-1980 1981-1991 1991-… Sum

Floor area 7,1 4,6 4,9 5,4 5,4 2,8 30,2

Most (65%) of the existing offi ce buildings are rather small, between 200 and 1000 m2. Many offi ce buildings are between 1000 and 5000 m2 and some are bigger than 20 000 m2 (see Table 3.2).

Table 3.2 Number of offi ce buildings within a certain range of fl oor area, m2 (SCB, 2001).

Floor area (m2) 200-999 1000-4999 5000-19999 20000-… Sum

Number of buildings 11 505 4 719 1 415 170 17 809


66

Most offi ce buildings are equipped with a heating system. However, mechanical cooling systems are also becoming rather common in offi ce buildings. Most of the buildings are equipped with a mechanical ventila-tion system, usually a system with mechanical supply and exhaust air. The newer ones often have heat recovery on the ventilation.

The energy source for heating offi ce buildings can be oil furnace, district heating, electricity, local district heating, gas or biomass. The most com-mon source is district heating (71%), as shown in Table 3.3.

Table 3.3 Area of offi ce buildings by type of heating (SCB, 2001).

Oil District Electric Local Gas Oil + el Biomass Other Sum furnace heating district heating

Heated area 2,4 23,4 2 0,2 0,6 0,4 0 4,1 33,1million, m2

Heated 7 71 6 1 2 1 0 12 100area, %

3.1.4 Energy performance of Swedish offi ce buildings

The majority of the offi ce buildings are heated by district heating. Among these buildings, those built between 1961 and 1970 have the highest use of district heating energy, 144 kWh/m2a (see Table 3.4). Most of the heat-ing is space heating, hot service water accounts for 2 – 7% of the heating (Nilson, 1996).

Table 3.4 Use of district heating in Swedish offi ce buildings, broken down by year of completion (SCB, 2001).

Year of …-1940 1941-1960 1961-1970 1971-1980 1981-1991 1991-… Average completion

Use of district 137 133 144 112 91 116 122heating kWh/m2a

If cooling is included, then the buildings constructed between 1961 and 1970 have the highest energy use, 156 kWh/m2a (see Table 3.5).

State of the art

67

Table 3.5 Average energy use for space heating, district cooling and electric-ity for cooling in Swedish offi ce buildings (SCB, 2001).

Year of …-1940 1941-1960 1961-1970 1971-1980 1981-1991 1991-… Averagecompletion

Use of district 146 143 156 127 114 127 135heating kWh/m2a

An analysis of energy use for heating and use of electricity in premises showed that the heating energy use has been reduced, while the total use of electricity has increased, during the last decades (Energiboken, 1995). The reduction in heating energy use is due to the improved thermal insulation (lower thermal transmittance) and introduction of heat recovery on the exhaust air fl ow, required by the building regulations. New premises have a lower total use of energy than older ones, but a higher share of use of elec-tricity (see Figure 3.1). In new offi ce buildings, i.e. those built after 1980, the use of electricity often accounts for 70% of the use of energy (Nilson, 1996). The previous building regulations, before 2006, did not have any real requirement for the use of electricity or the total energy use.

0

20

40

60

80

100

120

140

160

180

200

220

240

260

280

0 10 20 30 40 50 60 70 80 90 100 110 120 130

Electricity (kWh/am²)

Hea

ting

(kW

h/am

²)

Typical office buildings Refurbished office buildings New office buildings

1970

1990

1980

Figure 3.1 Relation between use of energy for space heating and use of electric-ity in Swedish offi ce buildings, as a function of year of completion (Energiboken, 1995).


68

The reduction in heating demand has in many cases taken place at the expense of an increased use of electricity. Redistribution between these two energy sources has taken place both in new construction and refurbish-ment. There are several reasons for this, e.g.

• poor knowledge as to the actual use of electricity in buildings.• increased use of offi ce appliances (PCs, printers, servers, copiers, etc)• the building regulations have focused on heating demand• no life cycle perspective is applied

There is quite a variation in the energy use of offi ce buildings as shown in Table 3.6.

Table 3.6 Energy use in offi ce buildings (REPAB, 2003).

kWh/m2year District Electricity Electricity Electricity Total Total(non-residential heating (fans, (lighting, cooling electricity energy usearea) pumps etc.) PC etc.)

Low 80 10 35 15 60 140Normal 125 18 50 30 98 223High 205 30 80 50 160 365

There is clearly an energy saving potential, especially with regard to use of electricity for ventilation, cooling, lighting. Targets have to be specifi ed for the use of electricity for fans, pumps, lighting etc and for the cooling demand. The users have to buy energy effi cient appliances (e.g. PCs). Im-portant savings can be achieved by adapting the operation of the HVAC system to the actual activity in the building and optimising the operation of the building with regard to ventilation, heating and cooling. The users can contribute by improving their behaviour with regard to the use of lighting, PCs etc.

3.1.5 Glazed offi ce buildings in SwedenEspecially during the nineties, highly glazed offi ce buildings were built in Scandinavia, some of them with single and others with double skin façades. This has been made possible by technical development regarding the construction and physical properties of glass.

During the last years architects have developed an interest in applying the technology of double skin façades in Scandinavia. Buildings with dou-ble skin façades built in Sweden are described in several literature sources. A report on requirements and methods for double skin façades has been

State of the art

69

produced by Carlson in 2003. A literature review on double skin façades for offi ce buildings has also been written by Poirazis (2004a), describing the main aspects of the system and presenting buildings from Scandinavia, Finland, Germany, United Kingdom, Belgium, etc. Buildings described from Sweden are the Kista Science Tower, The Nokia House Kista, Arlanda airport, the ABB Business centre and the GlasshusEtt.

Before describing further the properties of double skin façades, it is use-ful to understand why offi ce buildings with fully glazed façades are being built. Architecturally an airy, transparent and light building is created, with more access to daylight than in a more traditional offi ce building (Svensson, 2000). Furthermore, the main argument for constructing double skin glazed façades is that a decision has been made in the fi rst place to build a glazed building because of the transparent appearance (Svensson, 2001). The individual arguments, compared with a single skin glazed façade, are noise reduction, natural light, possibility to open windows, protected solar shading, burglary protection, night ventilation, preheated supply air, ad-ditional heat during winter and removal of solar energy via the double skin sustainable construction. This type of building enables ventilation to be adapted to the different seasons and often some kind of hybrid ventilation. It is also claimed that offi ce buildings with double skin façades can result in a reasonable energy use and a reasonable indoor climate.

There is a lack of knowledge in the building trade in Sweden concern-ing the design of highly glazed buildings and the calculation of energy use, thermal comfort and the infl uence of different technical solutions on these buildings.

When highly glazed offi ce buildings are designed, exact copies of build-ings located in other climates should be avoided. Adaptation to Swedish requirements for energy use and indoor climate, as well as adjustment to Swedish climate and Swedish engineering (building and HVAC), is necessary.

The complexity of building and HVAC systems requires a compre-hensive view. Energy, comfort and costs must be analysed for Swedish conditions. The low outdoor temperatures and solar gains during winter in Sweden can result in low temperatures of the inner layer and hence thermal discomfort (draughts, a non-uniform indoor climate, etc). The low altitudes of the sun during the winter can cause visual discomfort due to glare problems, mostly close to the façade. For deep buildings the daylight level can be low in the core of the building, although the façade is fully glazed.


70

3.2 Double skin façadesThe main aim of this Subsection is to provide necessary background in-formation from previous research carried out in the fi eld of double skin façades. Since the evaluation of the methods used (in order to determine the performance of double skin façades) is a part of this thesis, approaches and methods used for modelling and measuring the cavities are briefl y described as found in the existing literature. Moreover, a description of previous research regarding proper building integration of double skin façades is provided, with the emphasis on different HVAC strategies and the contribution of double skin façades for an optimized performance. Background information regarding energy simulations of double façades that were carried out on a building level is also provided. Finally, examples of buildings and typical constructions used are given.

3.2.1 Building physics of the double skin façade cavity

In the following subchapters a brief description is given of the model-ling approaches and the measurement methods used (as described in the literature). Although it is not the main aim of the thesis to develop a model for predicting the performance of the double skin façade cavity, a basic understanding of the physical model (air fl ows, temperatures etc.) and the methods used was required in order to evaluate the results, when validating the simulation tools used. Detailed description of the model-ling methods can be found in the chapter “Double Skin Façade Modelling Approaches” of the report “Double Skin Façades; A literature review” (Poirazis, 2006).

3.2.1.1 GeneralAccurate prediction of the airfl ow rates in a ventilated cavity is essential for successful modelling of the double skin façade cavity. Before describing each of the air fl ow simulation methods that different authors have used, it is essential to briefl y describe the two possible ways of ventilating the double skin façade cavity. As Shiou Li (2001) describes double skin façades can be both naturally and mechanically ventilated. Although, according to the author, natural ventilation can provide an environmentally friendly at-mosphere, it is not without risk. Unless the ventilation is designed properly the solar heat gain within the façade cavity will not be removed effi ciently and will increase the cavity temperature. For the naturally ventilated cavi-

State of the art

71

ties the air is brought into the cavity and exhausted by two means: wind pressure and/or the stack effect. According to the author the wind effect is dominant all the year round. As the author mentions, natural ventilation systems in urban environments may also experience signifi cant problems of noise transmission and pollution and may result in uncomfortable indoor environments in extreme weather conditions.

The mechanically ventilated double skin façades often use an underfl oor or overhead ventilation system in the building to supply or exhaust the cavity air to ensure good distribution of the fresh air. In this case the air is forced into the cavity by mechanical means. This air rises and removes heat from the cavity and continues upwards to be expelled or re-circulated. The mechanically assisted ventilation systems allow the building to be sealed, thereby providing more protection from traffi c noise than naturally ventilated systems. In areas with severe weather conditions or poor air quality, the mechanically assisted ventilation system can keep conditions in the buffer zone nearly constant to reduce the infl uence of the outdoor air to the indoor environment.

If the airfl ow in the DSF cavity is mechanically driven, the amount of air passing the cavity is known; in the case of a naturally ventilated cav-ity, however, it is necessary that the fl ow is calculated. In addition, the airfl ow in the cavity can be limited to maintain the necessary airfl ow rate in the room. Knowledge of air temperature in the double façade cavity is essential if one wants to maintain a comfortable indoor environment, especially when the cavity air is directly used for ventilation indoors. The air temperature and the airfl ow in the cavity are interrelated parameters and one can not be estimated properly without the other. Knowledge of the fl ow regime is also essential for prediction of the air temperature and air fl ow.

3.2.1.2 Modelling approachesThis part is based on the “DSF Modelling Approaches” subchapter of the Literature review written by Poirazis (2006).

Nowadays, building simulation software and developed mathematical models vary over a wide range of complexity. The simplest model is de-scribed by a few equations and the most complex one is the CFD model solving the conservation equations for mass, momentum and thermal energy.

According to Champagne (2002), in the HVAC fi eld there is a need to validate a proposed design to ensure proper performance. The two methods typically used are experimental or numerical. According to the author the experimental values are very reliable, when performed in a


72

controlled environment; however, there are several major drawbacks to this approach since, for instance, it is expensive and time consuming. Numerical approaches such as computational fl uid dynamics (CFD) are informative and when applicable can also save time and money.

At the same time, Hensen (2002) categorizes building simulation ap-proaches by level of resolution into macroscopic and microscopic. Accord-ing to the author the macroscopic approaches deal with entire building systems, indoor and outdoor conditions over some periods, while micro-scopic approaches use much smaller spatial and time scales. The building simulation software is normally related to the macroscopic approaches, while the CFD has the microscopic technique which is usually restricted to the steady state condition. The macroscopic (network) method is more suitable for the time series considerations.

Another direction is taken by Djunaedy, et al. (2002), which categorizes the main air fl ow modelling levels of resolution and complexity as:

• Building Energy Balance (BEB) models that basically rely on airfl ow guesstimates.

• Zonal Airfl ow Network (AFN) models that are based on (macro-scopic) zone mass balance and inter-zone fl ow pressure relationships; typically for a whole building.

• CFD that is based on energy, mass and momentum conservation in all (minuscule) cells that make up the fl ow domain; typically a single building zone.

Hensen, et al. (2002), explain that although airfl ow is obviously an im-portant issue for building performance assessment, the development of its treatment in modelling methods often lags behind the treatment ap-plied to the other important issues, such as energy fl ow paths. Nowadays emphasis has been given to airfl ow simulations which mostly focus on the following two approaches:

• Computational Fluid Dynamics (CFD) based on conservation equations for mass, momentum and thermal energy for all nodes of a two- or three-dimensional grid inside or around the object under investigation. The CFD approach is applicable to any thermo fl uid phenomenon; however, in building physics applications, there are several problematic issues, such as the amount of necessary comput-ing power, the nature of the fl ow fi elds and the occupant-dependent boundary conditions. This has often led to CFD applications being restricted to steady-state cases or very short simulation periods.

• The network method, in which a building is treated as a network of nodes representing rooms, parts of rooms and system components,

State of the art

73

with inter-nodal connections. According to the authors, the as-sumption is made that for each type of connection there exists an unambiguous relationship between the fl ow through the component and the pressure difference across it. Conservation of mass for the fl ows into and out of each node leads to a set of simultaneous, non-linear equations, which can be integrated over time to characterize the fl ow domain.

The position of Park, et al. (2003), Gertis (1999), Hensen, et al. (2002), and the work of many other researchers indicates that it is very diffi cult to fi nd a simple model that would describe the DSF performance appropri-ately. As explained in Hensen, et al. (2002) predicting the performance of a double skin façade can be quite diffi cult, since highly transient parameters such as cavity temperature, ambient temperature, wind velocity and di-rection, transmitted and absorbed solar radiation and angles of incidence govern the main driving forces.

Manz and Frank (2005), point out that the thermal design of buildings with the DSF type of envelope remains a challenging task. As yet, there is no software tool that can accommodate all the following three modelling levels: optics of layer sequence, thermodynamics and fl uid dynamics of DSF and building energy system. The complexity of the prediction task is the main reason for the long lasting research and application of simplify-ing techniques. The iterative approach of the network method became the reason to distinguish the three main issues in the DSF modelling:

• Optical element – responsible for the optical properties of the DSF materials

• Heat transfer element– responsible for the heat transfer processes in the DSF

• Flow element – responsible for the motion of the fl uid in the DSF

In various network methods these elements are defi ned differently in terms of nomenclature. In some methods they even stay undefi ned. The elements (the physical processes behind them) infl uence each other and, as has been argued, together they govern the main heat and mass transfer processes in the DSF. Several researchers (Saelens, Faggembauu, van Paassen, Di Maio, Manz and others) suggested the separation of the fl ow element and the heat transfer element in the predictions, which makes for better accuracy of wind infl uence predictions and advanced calculations of the convective and radiative heat transfer (Saelens, 2002).

Saelens (2002), performed an investigation of an accuracy change with stepwise enhancement of the network model. The diagram, depicted in


74

Figure 3.2, represents the stepwise change in the network models starting from the simplest case (a) – a single zone model.

Figure 3.2 Diagram of the different models with raised shading device, (Saelens, 2002).

According to the author:

• SZ (single zone) model: the cavity is represented by a single node. Radiation and convection in the cavity are combined. The heat transfer through the cavity surfaces is described by a single U-factor. The solar radiation is inserted in the air node and the cavity surface temperatures are not calculated.

• (SZRC) model: the previous model is somewhat improved, since the radiation and convection in the cavity are treated separately. The absorbed solar energy is inserted in the cavity layers and is a function of the angle of incidence.

• (AL) model: A further improvement consists of accounting for the temperature gradient along the height of the cavity. In order to allow an analytical solution, a temperature profi le is chosen with a linear temperature gradient.

• (AE) model: An exponential temperature gradient is assumed as a further improvement for the analytical model.

• (NUM) model: A numerical model is developed, which is based on a cell centred fi nite volume method. As an improvement over the other models, the radiation heat transfer in the cavity is treated more correctly and shadowing is taken into account.

The CFD code is able to perform many tasks that the network modelling will never achieve. However, some of the CFD features are too sophisti-cated and unnecessary for the design stage (e.g. the grid distribution of the velocity, temperature, dissipation of energy etc., obtained when the CFD modelling is performed) and, as mentioned above, the CFD modelling is often restricted to steady state simulations. According to the authors, whose

State of the art

75

works are mentioned in this section, there is a steadily growing experience in CFD modelling in general and in CFD modelling of DSF, but still there are a number of issues which are considered to be problematic in practice (Hensen, et al., 2002; van Dijk and Oversloot, 2003; Ding, et al., 2004; Jaroš, et al., 2002; Chen, 1997):

• amount of necessary computer power• complex fl ow fi elds• uneven boundary conditions• necessity to validate the results and the diffi culties to achieve

satisfaction with validations• the need for users to have advanced knowledge

A more detailed description of developed network and CFD models that can be found in the existing literature is given in the report “Double Skin Façades; A literature review report” (Poirazis, 2005).

3.2.1.3 Measurements – test rooms and real buildingsIn this section measurements made both in test rooms and in real build-ings are described.

Saelens and Hens (2001) in “Experimental evaluation of Airfl ow in Naturally Ventilated Active Envelopes” describe the most common meas-urement techniques for calculating the air fl ow rates in both naturally and mechanically ventilated active envelopes. The airfl ow in ducts and cavities can be determined by measuring:

• the pressure difference across an orifi ce, nozzle or venturi tube• the air velocity using anemometers• the air fl ow directly using tracer gas techniques

In the same paper the airfl ow through naturally ventilated active envelopes has been experimentally analysed. The authors proposed a method to de-termine the airfl ow through the cavity by means of the pressure difference over the lower ventilation grid. From the pressure difference over the lower ventilation grid, the airfl ow rate through the cavity was determined from the pressure characteristic of the active envelope. The method has been verifi ed by tracer gas measurements and proved to be reliable.

Saelens, referring to Onur et al. (1996) in his PhD thesis writes that for mechanically ventilated cavities, the airfl ow rate can be determined by measuring the pressure difference across an orifi ce placed in the exhaust duct. However, this method is less suited for naturally ventilated cavities. As Saelens describes in “Experimental evaluation of Airfl ow in Naturally


76

Ventilated Active Envelopes” (2001), “the driving forces are usually small and because of the high fl ow resistance of the orifi ce, the fl ow in the cavity would be too much affected. Furthermore, it would be diffi cult to fi nd a suitable place for the orifi ce as no exhaust duct is available”.

Saelens (2001) after studying reports of Park et al. (1989) and Faist (1998), described a second method to estimate the airfl ow rate by meas-uring the air velocity with anemometers. The author concluded that the determination of the airfl ow rate from velocity measurements seems obvious, but is likely to produce erroneous results, since the velocity in a naturally ventilated channel is not uniform and is infl uenced by lowering or raising the shading device. Furthermore, according to the author, there is no guarantee that the resulting velocity vector is perpendicular to the reference surface. Detailed information about the velocity vectors may be obtained by placing an array of individual velocity measuring points, which may however affect the development of the airfl ow in the cavity. Hence, determining the airfl ow rate in naturally ventilated active envelopes from measured velocities is a less recommendable method.

A third, less common method, is the use of tracer gas measurements (Ziller (1999); Busselen and Mattelaer (2000)). Tracer gas techniques such as the constant concentration, constant emission and tracer dilution method (Raatschen, 1995 and ASHRAE, 1997) make it possible to determine the airfl ow rate in both naturally and mechanically ventilated active envelopes without interfering with the driving forces.

In “Modelling of air and heat transport in active envelopes”, Saelens, Carmeliet and Hens (2001) compare (using measurements) fi ve models, of varying complexity, of a mechanically ventilated active envelope. The authors claim that radiation and convection in the cavity have to be modelled separately in order to obtain reliable results. According to the authors, for an accurate prediction of active envelope performances, the vertical temperature profi le has to be implemented properly (e.g. by an exponential expression). A sensitivity study performed with the numerical model reveals that the air temperature at the inlet of the cavity, the airfl ow rate, the distribution of the airfl ow in the cavity and the angle of solar incidence are the governing parameters.

Saelens (2002) describes in his thesis measurements carried out at the Vliet test building (two one storey high multiple-skin façades and a tradi-tional envelope). According to the author there are two main aims of the measurements: (a) the measurement set-up is used to extend knowledge of the thermal behaviour of multiple skin façades and (b) the data is used to evaluate modelling assumptions and to derive and check relationships for modelling parameters.

The author compares different models for the convective heat transfer coeffi cient with the measurements. Additionally, the measurements are

State of the art

77

used to evaluate the numerical model and to assess the reliability of models with different levels of complexity. Finally, the data are used to assess how the inlet temperature should be determined.

Shiou Li in 2001 wrote an MSc thesis which proposes a protocol for experimentally determining the performance of a south facing double glazed envelope system. The protocol was applied to an experimental study of a south-facing, single story double glazed ventilated system. In order to achieve that, two modular full-scale double glazed window models with naturally or mechanically assisted ventilation were constructed and monitored. The main goal was to develop and apply the test protocol by monitoring and analyzing the thermal performance of double façades. By using this test protocol the author claims average cavity heat removal rate approximately 25% higher for the active system when compared to the naturally ventilated one. Also, the passive system has a higher temperature difference between the indoor glass surface and the indoor air than the active system.

3.2.2 Integration of double skin façadesThe integration of the double skin façade systems in offi ce buildings is crucial for thermal performance and energy use during the occupation phase. Stec & Paasen (2003) presented a paper in which they describe different HVAC strategies for different double skin façade types. Accord-ing to the authors, the integration procedure of double skin façades in the building should include (a) defi ning the functions of the double skin façade in the building, (b) selecting the type of the double skin façade, its components, materials and dimensions of the façade that fulfi l the requirements, (c) optimizing the design of the HVAC system to couple it with the double skin façade, and (d) selecting the control strategy to supervise the whole system.

The authors briefl y introduce the concept of different cavity depths and describe its infl uence on the air temperatures inside the cavity. According to them, the dimensions of the façade together with the openings deter-mine the fl ow through the façade; narrower cavities result in higher fl ow resistance and smaller fl ow through the cavity and a higher increase in air temperature in the cavity. The authors conclude that (a) in the cold period it is more suitable to use narrow cavities to limit the fl ow and increase the cavity temperature and (b) in the hot period the double skin façade should work as a screen for the heat gains from radiation and conduction. It is diffi cult to claim in general whether the narrow or deep cavities will perform better because in one case the cavity temperature and in the other case the temperature of the blinds will be higher.


78

Examples concerning the infl uence of different depths on the properties of the cavity are shown in “Second Skin Façade Simulation with Simulink Code” by Di Maio and van Paassen in (2000). In “Modelling the Air Infi l-trations in the Second Skin Façade” in (2001) the same authors conclude that “narrow cavities are more useful, because they can deliver a higher and hotter air fl ow compared to the air fl ow delivered by wide ones”.

3.2.2.1 Contribution of double skin façades to the HVAC strategy

As Stec et al. (2003) describe, an HVAC system can be used in the follow-ing three ways in a double skin façade offi ce building:

• full HVAC system (the double façade is not a part of the HVAC) which can result in high energy use. The user can select whenever he/she prefers mechanically controlled conditions inside or natural ventilation with the use of the double skin façade).

• limited HVAC system (the double façade contributes partly to the HVAC system or plays the major role in creating the right indoor climate). In this way the double façade can play the role of:

o pre-heater for the ventilation airo ventilation ducto pre-cooler (mostly for night cooling)

• no HVAC. The double façade fulfi ls all the requirements of an HVAC system. This is the ideal case that can lead to low energy use.

During the heating periods the outdoor air can be inserted from the lower part of the façade and be preheated in the cavity (Figure 3.3). The exterior openings control the air fl ow and thus the temperatures. Then, through the central ventilation system the air can enter the building at a proper temperature. During the summer, the air can be extracted through the openings from the upper part of the façade. This strategy is usually applied to multi storey high double skin façades.

State of the art

79

AHU

Winter Summer

Figure 3.3 Double skin façade as a central direct pre-heater of the supply air.

During the whole year, the double skin façade cavity can work only as an exhaust duct without the possibility of heat recovery for the HVAC system (Figure 3.4). It can be applied both during winter and summer to the same extent. The main aim of this confi guration is to improve the insulation properties in the winter and to reduce the solar radiation heat gains during the summer. There are no limitations to individual control of window opening.

AHU

Winter/Summer

Figure 3.4 Double skin façade as an exhaust duct.


80

It is also possible to use the double skin façade as an individual supply of the preheated air (Figure 3.5). This strategy can be applied in both the multi-storey and box window types. An exhaust ventilation system im-proves the fl ow from the cavity to the room and to the exhaust duct. Extra conditioning of air is needed in every room by means of VRV system or radiators. This solution is not applicable for the summer conditions, since the air temperature inside the cavity is higher than the thermal comfort levels. Also in this case there are no limitations to individual control of window opening.

Box window Multi-storey

Figure 3.5 Double skin façade as an individual supply of the preheated air.

Finally, the double skin façade cavity can be used as a central exhaust duct for the ventilation system (Figure 3.6). The air enters through the lower part of the cavity and from each room. The supply ventilation system stimulates the fl ow through the room to the cavity. Heat can be recovered by means of heat pump or heat regenerator at the top of the cavity. Because the air in the cavity is not fresh air, the windows cannot be operable.

State of the art

81

AHU

Winter

Figure 3.6 Double skin façade as a central exhaust duct for the ventilation system.

As Stec et al. (2003) describe, generally supply façades couple better with the winter systems in which their preheating properties can be used. The exhaust façade is more effi cient in cooling the cavity in the summer. Problems arise when one façade needs to couple both of the periods, in which case the construction must be adjusted for summer and winter conditions.

3.2.2.2 Examples of coupling double skin façades and HVAC

In 2001, van Paassen and Stec wrote a paper “Controlled Double Façades and HVAC” that deals with the preheating aspects of double skin façades. The authors claim that for the winter period the most signifi cant parameter should be the heat recovery effi ciency. The main aim of the paper was to show the usability of the cavity air for ventilation purposes. According to the authors, it is possible to defi ne by simulation how the heat recovery ef-fi ciency depends on the outside conditions, the dimension of the cavity, the area of inlet and outlet for outside air and the height of the building.

For the simulations, the authors chose the following four double skin façade types:

1. Double skin façade with controlled airfl ow through the cavities (Figure 3.7). The façade is a multi-storey façade with no opening junctions that allow the air to be extracted out. There is only one inlet for the ventilation airfl ow at the bottom of the façade. It is controlled by an air damper such that the air supply to the cavity


82

is just enough for ventilating all the rooms above. The controlled trickle ventilator delivers the desired airfl ow to each room (80 m3/h)

Figure 3.7 Coupling DSF and HVAC; Controlled air fl ow in the cavity.

2. There are no open junctions on each fl oor, no controlled airfl ow in the cavity and open dampers in this system (Figure 3.8). Ad-ditionally, the upper part of the façade is open allowing the air to be extracted.

State of the art

83

Figure 3.8 Coupling DSF and HVAC; Uncontrolled air fl ow in the cavity.

3. There are open junctions between the outside and the cavity on each fl oor, which cause heat exchange between air inside the cav-ity and outside air. The main airfl ow is the same as in the second system (Figure 3.9). The authors claim that this should be the best system for summer time when cooling is required, but due to the open junctions preheating of the cavity air will be much lower than in the other systems with closed junctions.


84

Figure 3.9 Coupling DSF and HVAC; Open junctions in each fl oor.

4. There are open junctions on each level, but the storeys are separated from each other (Figure 3.10). Consequently each storey creates its own system. The authors claim that in practice this can be the most convenient system since the same module can be used on each storey and the problems due to large temperature gradients at different levels in the cavity can be avoided (on each storey there is more or less the same temperature in the cavity).

Figure 3.10 Coupling DSF and HVAC; Each storey is separated.

State of the art

85

The conclusions drawn by the authors were that:

• The dimensions of the cavity, (height and width) have the great-est infl uence on the heat and fl ow performance in the double skin façade and hence they are the most important parameters in design-ing the double skin façade.

• High-rise buildings with very narrow cavities may not ensure the airfl ow in the cavity needed for ventilation purposes.

• In general, a double façade with airtight junctions and proper air-fl ow control in the cavity is an interesting pre-heater for ventilation air. In a four storey building with cavity width of 0.2 m an overall heat recovery effi ciency of 40% can be obtained. According to the authors this effi ciency can be increased to 72% if the ventilation fl ow inside the cavity is properly controlled. A disadvantage is the vertical temperature gradient inside the double façade. It gives lower comfort or higher cooling capacities at higher fl oors.

• Splitting the cavities of high rise buildings into separate parts by combining for example four storeys with their own inlets and outlets can be essential. If this is done for each fl oor the effi ciency can drop to 35%.

• In order to use the double façade for night cooling and for heat recovery, controlled dampers in the open junctions are needed. During cooling periods they should be fully open.

3.2.2.3 Control strategyA crucial point when integrating double skin façade systems in buildings is to defi ne a control strategy that allows the use of solar gains during the heating period and provides acceptable thermal comfort conditions dur-ing the whole year. In the case of cavities with all year round mechanical ventilation, there is a high risk of overheating the offi ces during the sum-mer months, when the design of the double skin façade is not coupled properly with the strategy of the HVAC system. According to Stec et al. (2003) this system allows the outside conditions to infl uence the indoor climate. As the authors describe, an effi cient control system can manage rapidly changing outside conditions. Successful application can only be achieved when the contributions of all the devices can be synchronized by an integral control system.

According to the authors, the control system of the building should take into consideration the following principles:

• the occupants should be able to infl uence everything, even if their intervention wastes energy.


86

• energy saving can be achieved when the control system takes maxi-mum advantage of the outside conditions before it switches over to the air conditioning system.

• all control systems must be focused on realization of the required comfort with the lowest energy use.

• during the unoccupied period the control system is focused only on the energy savings, while during the occupied period it must also be focused on comfort.

According to the authors, the main tasks that the control system has to fulfi l are to: (a) keep the right level of temperature inside the building, (b) supply suffi cient amount of ventilation air to the building and (c) ensure the right amount of light inside the building.

3.2.3 Energy performance of buildings with integrated double skin façades

A complete study of energy performance was presented by Saelens, Car-meliet and Hens in “Energy performance Assessment of Multiple Skin Façades” in 2003. The authors claim that only few combinations of Multi Storey Façade modelling and building energy simulation are available. Ac-cording to the authors, “most of these papers are restricted to only one MSF-typology. Müller and Balowski [1983] analyse airfl ow windows, Oesterle et al [2001] give a comprehensive survey of double skin façades and Haddad and Elmahdy [1998] discuss the behaviour of supply air windows”.

In the above paper the authors focus on the energy saving objectives of three Multi Storey Façade typologies used in a single offi ce. The MSF-model is coupled with TRNSYS. According to the authors, “to simulate the energy demand of the offi ce, a cell centred control volume model, describing the MSF, is coupled to a dynamic energy simulation program. The results of the energy simulations are compared and confronted with the objectives found in literature”.

The authors focus on one storey high solutions:

• a conventional façade with an insulated glazing unit (IGU)• a naturally ventilated double skin façade (DSF)• a mechanically ventilated airfl ow window (AFW)• a mechanically ventilated supply air window (SUP)

The reduction of the transmission losses, the possibility of recovering the transmission losses by the airfl ow, the position of the shading device sheltered from climatic conditions and the ability to remove the absorbed solar heat are the most commonly mentioned energy advantages.

State of the art

87

The authors conclude that it is possible to improve the building’s energy effi ciency in some way by using multiple skin façades. However, most typologies are incapable of lowering both the annual heating and cooling demand. This can be achieved only by (a) combining typologies or (b) changing the system settings according to the particular situation. This implies that sophisticated control mechanisms are essential, in or-der to make multiple skin façades work effi ciently throughout the year. Furthermore, the authors conclude that the energy performance strongly depends on the way the cavity air is used. In order to correctly evaluate the energy effi ciency of multiple skin façades, it is imperative not only to study the transmission gains and losses but also to take into account the enthalpy change of the cavity air and to perform a whole building energy analysis.

Gratia and De Herde (2004), claim that there are still relatively few buildings in which double-skin façades have actually been realized, and there is still too little experience of their behaviour in operation. For this reason the authors chose to study the natural ventilation in a multi-sto-rey double skin façade using the TAS software. The study was made on a building level for a sunny summer day. The authors have analyzed the behaviour of the double skin façade for various conditions, with the main focus on the impact of the double skin orientation and the impact of wind orientation and degree of wind protection.

Hendriksen, Sørensen, Svensson and Aaqvist wrote a paper that focuses mostly on the heat loss, indoor climate and energy aspects of double skin façades. Examining four different cases of double skin façades, they provide useful information concerning daylight, climate and energy aspects. The fi rst case is with simple double glazing and the other three with D.S.F. as described below:

• Simple double glazing• Double inner - single outer glazing• Single inner – double outer glazing• Double inner - double outer glazing

According to the authors, when a single layer of glazing is added to a double low-E glazing in a double skin façade construction the reduction in heat loss expressed by the U value is modest (<20%). Introducing an extra double low-E glazing can reduce the heat loss by approximately 50%. The authors conclude that a traditional window façade offers better conditions regarding heat loss than a fully glazed or a double skin façade, due to the reduced heat loss from the non-transparent parts of a traditional façade.

Saelens, Blocken, Roels and Hens presented a paper at the IBPSA 2005 Conference regarding the importance of optimizing multiple skin façades


88

in terms of energy performance. The paper aims at optimizing the energy performance of a single storey highly glazed multiple skin façade by chang-ing the settings of the façade and the HVAC. Moreover, the authors men-tion that energy effi ciency is an important argument for choosing Multiple Skin Façades (MSFs) as façade concept. MSF-systems are presented as being superior to traditional façades during both the heating and cooling seasons (for the climate examined). They claim that two main principles can be distinguished: (a) MSFs may reduce the transmission losses in winter and the gains in summer and (b) MSFs can either expel the return air to avoid overheating or reuse the return air, in order to use the absorbed solar energy and recover some of the transmission losses.

The three multiple-skin façades considered were (a) a mechanically ventilated airfl ow window, (b) a naturally ventilated double-skin façade and (c) a mechanically ventilated supply window. Their performance is compared with the performance of a traditional cladding with exterior and interior shading devices. The considered cases are shown in Figure 3.11.

Figure 3.11 Schematic representation of the multiple skin (Saelens et al., 2005).

State of the art

89

The authors compare the base case traditional façades against the optimized MSFs and conclude that by implementing control strategies the energy effi ciency of all façade systems is signifi cantly improved. For the naturally ventilated skin façade, controlled openings are benefi cial and a simple open-close strategy is suffi cient. The authors claim that controlling the airfl ow rate is a successful approach to lower the cooling demand for all façade typologies. The supply window has the highest potential to benefi t from the optimization techniques. It is able to considerably reduce the heating demand while providing an acceptable cooling demand. The traditional façade with exterior shading device, however, still provides the best solar protection. The double skin façade is also able to effi ciently control the cooling demand but is limited in improving the heating demand. The airfl ow window is capable of signifi cantly lowering the heating demand but still suffers from high cooling demands.

3.2.4 Typical constructions - Examples of buildingsIn this part examples of buildings with double skin façades are given, in order to describe typical constructions. Examples of buildings located in Germany, Finland and Sweden are described in detail in the tables in Appendix A, while general conclusions are drawn below. The reason for choosing buildings located in Finland and Sweden is to describe typi-cal constructions for the climate studied in this thesis. The examples in Germany can provide additional knowledge since the reasoning behind the design and integration of double skin façades seems to be more “so-phisticated”. The main aim is to investigate (according to the existing literature) whether there are typical constructions corresponding to certain climatic conditions (countries) and to give the reason for the selection of façade in each case. The parameters this study focuses on are (a) the type of façade, (b) the ventilation strategy, (c) the geometry (mainly depth) of the ventilated cavity and (e) the glass and shading devices used. The cases mentioned in this part are described in detail in the literature review report “Double Skin Façades for Offi ce Buildings” by Poirazis (2004).

When the buildings with double skin façades in Germany are compared with the ones located in Finland and Sweden, it can be noticed that in most of the cases in Sweden and Finland the cavity is multi storey high while in Germany box window, shaft box and corridor type façades can be found.

The ventilation strategy of the cavity follows the same tendency as described above. In most of the cases in buildings located in Finland and Sweden the openings close during the winter for extra insulation and open during summer for heat extraction reasons. In Germany, however,


90

the ventilation strategy of the cases studied differs. Provision of natural ventilation in the rooms during spring and autumn months is often achieved; in extreme weather conditions solutions were given so as not to compromise the (thermal) comfort of the occupants. In two of these cases it was stated that by using the preheating mode of the cavity, the building can be naturally ventilated during long periods of the year (70-75% for the Düsseldorf City Gate and 50-60% for the ARAG 2000 Tower). In other cases (e.g. GSW Headquarters) the warm air inside the cavity is returned to the central plant for heat recovery purposes. Generally, in the mentioned cases the ventilation can be natural, mechanical or even fan supported; in some cases the airfl ow (and consequently the air temperature) inside the cavity is regulated in such a way that natural ventilation can be suffi cient for long periods during the year.

The depth of the cavity varies in all the cases described above. The narrowest cavities are in Galleries Lafayette and Posdamer Platz 1 (approxi-mately 0.2m deep), both located in Germany, while the deeper one is the in the Korona building in Finland (2m). In the examples given the cavities in Finland tend to be somewhat deeper although there is not any specifi c reasoning for that; the opposite could be expected since deeper cavities are required in warmer climates for more effi cient heat extraction.

A typical construction of a double skin façade in Germany consists of a single outer and a double inner glazing unit. The outer pane of glass is in most cases an 8 or 12mm toughened pane; usually the glazing is a clear pane, while in some others (e.g. Victoria Life Insurance Buildings) a laminated solar control glass is used as an outer pane . In some of the cases studied the outer skin is openable for more effi cient heat extraction. The inner skin is in almost all the cases a low E glazing unit. Similar numbers and types of panes (single outer and double inner with low E skin) are used for the cases in Sweden. In Finland, however, the cases mentioned differ. In some cases the construction is similar to that in Germany (single outer and double inner skin; e.g. Martela, Itämerentori, Nokia Ruoholahti, Nokia K2, etc). In some other cases the number of panes in the inner and outer skin is larger. For example, the façade of the Sanomatalo building consists of a triple pane inner skin (toughened 6mm inner, toughened 4 mm intermediate and solar control 6mm outer) and a double pane outer one (toughened and laminated 6+6 mm panes); the construction of the Sonera façade is similar (triple inner and double outer). The façade of the JOT Automation Group building consists of a triple inner envelope (6 mm solar control glass (outer), 4 mm clear glass (middle), 4 mm clear glass (inner)) and a single outer one (10 mm tempered, green solar protective glass pane); the construction of the Radiolinja façade is similar (triple inner and single outer). Unfortunately, values such as thermal and solar

State of the art

91

transmittance of the inner or outer skin for the façades mentioned above could not be found in the literature.

In almost all the cases (regardless of country or construction) the shad-ing devices, usually venetian blinds, are placed inside the ventilated cavity (in most cases closer to the outer skin).

From the constructions studied it appears that it is hard to describe a “typical” double skin façade construction when performance evaluation for each building is not accessible. The façade constructions described above (and in more detail in the literature review report “Double Skin Façades for Offi ce Buildings” by Poirazis (2004)), can however be considered a good starting point for analyzing the tendencies, when deciding the cases to be simulated in this thesis.

3.3 Building simulation softwareThe use of simulation tools during the design stage can help the designer improve the overall building performance. The system building – instal-lations can be optimised with regard to indoor climate and energy use. Different alternatives can be studied, compared and optimized in terms of energy use and indoor environment at a low cost (avoiding full scale experiments), since the performance can be analysed and predicted at an early stage. The simulations can also be used to predict the energy use and indoor climate of an existing building (in the case of a refurbishment project).

According to Jacobs and Henderson (2002) the building simulation software can be divided into the following categories:

Practitioner Design Tools: Software used by architects, engineers, and other practitioners to automate common tasks that are part of the day-to-day design process (often integrated into CAD environments).

Whole Building Energy Analysis Tools: Software that predict annual energy use (and often operating costs) by simulating operating condi-tions. These detailed hour-by-hour building simulation tools are used by some practitioners as part of the design process and for energy code compliance.

Energy and Environmental Screening Tools: Software focusing on the economic and environmental impacts of using new energy-effi cient technologies in buildings.

Specialized Analysis Tools: Software used most often for research pur-poses and include technically accurate simulation models; they are often


92

developed for academic study of a building science problem. These tools focus on building or system performance details.

3.3.1 Building energy simulation toolsOver the last 50 years a great number of dynamic Building Energy Analysis tools has been developed. These tools provide the user with indicators es-sential for the building performance such as energy use, temperature and costs. Below, some of these tools that were considered during the early stage of the “Glazed Offi ce Buildings” project are presented and briefl y described. This description is based on the report “Constructing the Ca-pabilities of Building Energy Performance Simulation Programs” written by Crawley, Hand, Kummert and Griffi th in 2005.

BLAST: The BLAST program was developed by the USA CERL and the University of Illinois. The BLAST (Building Loads Analysis and System Thermodynamics) software is a set of programs that aims to predict the energy consumption, energy system performance and cost of buildings. The tool contains the major subprograms: (a) Space Loads Prediction, (b) Air System Simulation and (c) Central Plant. BLAST can both estimate the annual energy performance and perform peak load (design day) cal-culations for mechanical equipment design.

BSim 4: BSim 4 was developed by the Danish Building Research Institute (SBI) in 2004. It is a user friendly tool used for energy design of buildings and moisture analysis. The tool comprises several modules: (a) tsbi 5: a combined transient thermal and transient indoor humidity and surface humidity simulation module, (b) SimView: graphic model editor and input generator, (c) SimLight: tool for daylight analysis conditions in simple zones, (d) XSun: graphical tool for analysis of direct sunlight and shadowing, (e) SimPV: a simple tool for calculating electrical yield from PV systems, (f ) NatVent: a simple tool for one zone natural ventilation and (g) SimDxf: a simple tool for importing CAD drawings.

DEROB-LTH: DEROB-LTH was originally developed at the Numeri-cal Simulation Laboratory, University of Texas, Austin, USA. Since the beginning of the 1980, the program has been further developed at Lund University, under the name DEROB-LTH. Building energy simulation tool used to explore the complex dynamic behaviour of buildings for different designs. The behaviour is expressed in terms of temperatures, heating and cooling loads and different comfort indices. The form of the building can be modelled in a fl exible way. The model for assessing the solar insulation of building surfaces is detailed and includes the infl uence of different types

State of the art

93

of shading devices. The window model has been improved and accurately calculates the properties of a window package. The simulation uses a time step of one hour and calculates values in response to hourly values for climatic data, internal loads and airfl ows. DEROB-LTH is very good at calculating the energy balance regarding solar energy, taking into account transmittance, absorptance, and refl ectance in and out from a volume (and to adjacent volumes).

DOE-2.1E: DOE-2 was developed by Lawrence Berkeley National Labora-tory. The DOE-2.1E software has been widely used for the past 25 years to predict the hourly energy use and energy cost of buildings. The tool consists of one module for translation of input (BDL processor) and four simulation subprograms (LOADS, SYSTEMS, PLANT and ECON). The LOAD subprogram calculates the sensible and latent components of the hourly or cooling load for each constant temperature space, taking into account weather and building use patterns. The output of LOADS is the input for SYSTEMS, which handles the secondary systems (calculates the performance of fans, coils and ducts). The output of SYSTEMS is airfl ow and coil loads. PLANT calculates the behaviour of boilers, chillers, cooling towers, etc, according to the secondary systems cooling and heating loads. Finally the ECONOMICS subprogram calculates the cost of energy.

ECOTECT: ECOTECT is a highly visual and interactive building design and analysis tool, covering thermal, energy, lighting, shading, acoustics and cost aspects, developed by Square One. The main aim of the tool is to allow designers to take a holistic approach to the building design (rather than sizing a HVAC system), by providing continuous interactive and visual feedback. The software is entirely designed and written by architects and is intended mainly for use by architects.

EnergyPlus (Version 1.2.2): EnergyPlus is a tool based on the most popular features and capabilities of BLAST and DOE-2.1E which aims to provide an integrated (simultaneous loads and systems) simulation for accurate energy, temperature and comfort predictions. The EnergyPlus module calculates the system response of heating and cooling systems (with variable time step) providing more accurate space temperature prediction, crucial for system and plant sizing, occupant comfort and occupant health calculations. EnergyPlus has two basic components: (a) a building systems simulation module and (b) a heat and mass balance simulation module. The building systems simulation manager handles communication between the heat balance engine and various HVAC modules and loops, such as coils, boilers, chillers, pumps, fans, and other equipment components. The heat balance module manages the surface and air heat balance modules


94

and acts as an interface between the heat balance and the building systems simulation manager. The heat and mass balance calculations are based on IBLAST (a research version of BLAST).

ESP-r Version 10.1: ESP-r has been under development for more than 25 years. It follows the pattern of “simulation follows description” where additional technical domain solvers are invoked as the building and system description evolves. It works with third party tools such as Radiance to support higher resolution assessments as well as interacting with supply and demand matching tools. Although ESP-r has a strong research herit-age (e.g. it supports simultaneous building fabric-network mass fl ow and CFD domains), it is being used as a consulting tool by architects, engineers and multidisciplinary practices and as the engine for other simulation environments.

IDA ICE Version 3.0: IDA Indoor Climate and Energy (IDA ICE 3.0) is a building energy simulation program for the simulation of energy use for heating, cooling, lighting etc., thermal comfort and indoor air quality in buildings. The tool is a multi-zone dynamic energy simulation program, which in a detailed manner takes into account HVAC equipment, which is simulated as well. The tool is based on a general simulation platform for modular systems, IDA ICE 3.0 Simulation Environment. In IDA ICE 3.0, physical systems from several domains are described using symbolic equations, stated in either or both of the simulation languages Neutral Model Format (NMF) and Modelica. The user defi nes the tolerance control solution accuracy, allowing complete isolation of numerical errors from modelling approximations.IDA ICE 3.0 offers 4 types of interfaces for different user categories:

• Wizard level: leading the user through the steps of building a model for a specifi c type of study.

• Standard level: the user is expected to formulate a model using domain specifi c concepts and objects (zones, radiators, windows, etc).

• Advanced level interface: the user is able to browse and edit the mathematical model of the system.

• NMF and or Modelica programming: for developers.

It can be used to provide an integrated (simultaneous loads and systems) simulation for accurate energy, temperature and comfort predictions.

PowerDomus Version 1.5: PowerDomus is a whole building simulation tool for thermal comfort and energy use analysis. The tool visualizes the

State of the art

95

solar path and inter-building shading effects and provides reports with graphical results of zone temperature and relative humidity, PMV and PPD, thermal loads statistics, temperature and moisture content within user-selectable walls-roofs, surface vapour fl uxes and daily-integrated moisture sorption-desorption capacity. PowerDomus is a user friendly software that has the aim of being used by a large number of users.

Tas Version 9.0.7: Tas is a software that simulates the dynamic thermal performance of buildings and their systems using different modules. The core module is the Tas Building Designer, a dynamic building simulation tool with integrated natural and forced airfl ow. The module has a 3D graphics based geometry input that includes a CAD link. Tas has been used for over 20 years in the UK and around the world. It has a reputa-tion for robustness, accuracy and a comprehensive range of capabilities. Developments are regularly tested against ASHRAE, CIBSE and ISO/CEN standards.

TRNSYS Version 16.0.37: TRNSYS is a software with a modular structure that was designed to solve complex energy system problems by breaking the problem down into a series of smaller components. The components can be simple (such as a pump or pipe) or more complicated (such as a multi-zone building model). These components are confi gured and assem-bled using the TRNSYS simulation studio integrated visual interface, and the building input data is entered through the visual interface TRNBuild. The simulation engine then solves the system of algebraic and differential equations that represent the whole system. The time steps usually consid-ered by the program are 1 hour or 15 min but can also achieve 0.1sec time steps. Apart from the detailed multizone building model, the library of TRNSYS includes components commonly found in thermal and electrical energy systems (such as solar thermal and photovoltaic systems, low energy buildings and HVAC systems, renewable energy systems, etc).

TRNSYS, DOE 2.1E, IDA ICE 3.0, and ESP-r were validated within IEA Task 22 (Building Energy Analysis Tools). The test cases developed for the RADTEST (Radiant Heating and Cooling Test Cases by Achermann and Zweifel, 2003) were based on the ENVELOPE BEST TEST from IEA Task 12 (Judkoff and Neymark, 1995). TRNSYS, DOE 2.1E, IDA ICE 3.0 were empirically validated by Travesi, Maxwell, Klaassen, and Holtz (2001), as described in the IEA report “Empirical Validation of Iowa Energy Resource Station Building Energy Analysis Simulation Models”. DOE 2.1 D, BLAST 3.0, ESP and TRNSYS have been validated by the International Energy Agency Building Energy Simulation Test (BESTEST) and Diagnostic Method (Judkoff and Neymark, 1995).


96

3.3.2 Software for DSF modellingIn this part a description is given of the available software for double skin façades based on the report “Ventilated double façades” written by Flamant, et al. (2004). The authors distinguish these tools into: (a) com-ponent simulation software and (b) building simulation software. The component simulation software simulate the thermal, energetic and visual behaviour and performance of the façade component while the building simulation software simulate the whole building (façade included) in order to predict the thermal dynamic behaviour of the building, indoor temperatures, energy use , etc.

3.3.2.1 Façade simulation softwareWIS 3: WIS (Window Information System) is a freeware software devel-oped to calculate the thermal and solar characteristics of window systems and components. One of the unique elements of this tool is the combina-tion of glazing and shading devices, with the option of free or forced cir-culation between both. This makes the tool suited to calculate the thermal and solar performance of double skin façades. WIS performs calculation of the transfer of short wave radiation for all angles of incidence, but is unable to perform dynamic calculations. It is possible to model natural convection, caused by stack effect, but wind induced convection is not covered. The WIS algorithms are based on international standards such as the ISO Standard 15099.

BISCO/TRISCO/VOLTRA: These tools were developed by Physibel and aim to model heat transfer of building elements. The unique feature of this software group is the potential to perform thermal calculations in combination with thermal bridging effect with the components. In greater detail, BISCO calculates two dimensional steady state heat transfer in objects with any shape. TRISCO performs the same kind of calculation but for three dimensional objects, while VOLTRA is an extension for time dependent boundary conditions of the steady state TRISCO. It is possible to model forced ventilation (ventilation fl uxes should be defi ned a priori), but not natural ventilation. Control system and building modelling are not considered in the tool. These three software tools can calculate both temperature distribution along the cavity and heat loss.

State of the art

97

3.3.2.2 Building simulation softwareCAPSOL: CAPSOL is a user friendly tool that calculates multi zone tran-sient heat transfer. The building and the environment are considered as a set of zones between which heat fl ux occurs due to conduction, convec-tion, radiation and ventilation. Solar transmission is angular-dependent. The characteristics of shading devices have to be calculated with software (i.e. WIS) and can then be introduced into CAPSOL. The convective heat transfer coeffi cient is considered constant. It is possible to set up a model where the vertical thermal stratifi cation can be taken into account. Mechanical ventilation can be modelled as one of the control options, while natural ventilation requires coupling of CAPSOL with the ventila-tion model (e.g. COMIS).

TRNSYS: TRNSYS is briefl y described in subsection 3.3.1. When simulat-ing thermal behaviour of a building, TRNSYS can manage airfl ows, but does not calculate them. In order to do that TRNSYS must be coupled with COMIS, which has been completely integrated into TRNSYS; this can be done manually or with the commercial tool TRNFLOW. The win-dow model in TRNSYS uses output data from the WINDOW 5 software where each glazing absorbs and refl ects a part of incoming solar radiation, depending on the glazing material and the incident angle. The convective heat transfer coeffi cient is not necessarily constant, but it varies according to fl ow regime in the cavity.

ESP-r: ESP-r is briefl y described in subsection 3.3.1. The important aspect of this tool is its ability to perform modelling at different levels of resolution (one or more zones in the building can be associated with the 3D CFD domain). In the thermal, airfl ow and lighting domains all heat and mass transfer processes are solved simultaneously at each time step of simulation. The modelling of venetian blinds in ESP-r can be complex.

TAS: TAS is briefl y described in subsection 3.3.1. This tool calculates the thermal performance of buildings and their systems. It incorporates a module which is capable of performing dynamic building simulation with integrated forced and natural fl ow, arising from wind and stack effects. The internal convection heat transfer coeffi cient may vary from hour to hour. Transmission and absorption characteristics of transparent constructions depend on the angle of incidence.

IDA ICE 3.0: IDA ICE 3.0 is an object-oriented simulation environment intended to handle a wide range of different types of simulation problems. The systems to be simulated are decomposed into subsystems that are


98

modelled using the Neutral Model Format (NMF). These models are then connected together to form a simulation model of the original system. The equation solver can solve most systems of differential-algebraic equations that can be formulated in NMF. Both steady-state and dynamic prob-lems can be dealt with, as well as discontinuities in the solutions (Bring, A. et al., (1999). A double façade is modelled in a special function as an external window outside an internal window and a wall. The air inside the cavity of the double façade has one temperature node. The air in the double façade is connected to the environment by leak functions at the top and bottom.

According to Flamant, et al., 2004, the six kinds of simulation software were analyzed, and according to the authors (each describing the software he/she used): “The software TRNSYS, ESP-r and TAS are powerful transient energy simulation programs and are able to simulate a ventilated double façade, the building, the HVAC systems and strategies in a certain extent. These pro-grams can make the coupling between thermal and airfl ow models. Neverthe-less, all these programs face similar obstacles regarding the level of resolution necessary to model some major thermodynamic fl ow paths in ventilated double façades. Time and experience are required in order to use properly these quite complex software. The software CAPSOL shows less functionalities than the three previous ones but this software can be recommended for specifi c points of interest due to its facility of use.

The software WIS combines a user-friendly interface with the most advanced calculations of thermal and solar properties of window and façades. The WIS algorithms are based on international (CEN, ISO) standards, but WIS also contains advanced calculation routines for components or conditions where current standards do not apply… Finally, BISCO, TRISCO and VOLTRA belong to a series of software aimed at modelling the heat transfer of building details using the energy balance technique. These programs are well adapted to calculate the interaction between the glass skins and the shading layer in combination with the thermal bridging effect of the subcomponents around the VDF (ventilated double façade) …” Flamant, et al. (2004) avoid giving any recommendation as to which is the most suitable tool for DSF modelling, as this depends on the objectives for setting up the model and the user’s experience to work with the simulation tools.

Methods

99

4 Methods

In this thesis a parametric study has been carried out focusing on energy use and indoor climate. Offi ce buildings with 30%, 60% and 100% window to external wall area ratios have been simulated, in order to calculate the energy demand during the occupation stage. Moreover, parameters such as control set point for indoor air temperatures, building orientation, plan type, façade, window and shading device type, etc, have been varied in order to study their impact on the building in terms of energy use and thermal comfort. The single and double skin alternatives studied as well as the parameters varied are described in detail below, in Sections 4.1 and 4.2. The simulation tools used for this study, WIS 3 and IDA ICE 3.0, are described in section 4.3. WIS 3 was used for a pilot study of double skin façade components for steady state boundary conditions, and IDA ICE 3.0 for parametric studies on a building level for an entire weather year.

4.1 Generation of building alternativesIn this part, a description of the methodology of the parametric study is given. Different building alternatives were simulated, in order to study the impact of design parameters on energy use and thermal comfort perfor-mance. This study is divided into three main parts:

• establishment and simulation of the reference building alternatives (30% window to external wall area ratio)

• simulations of the highly glazed (60% and 100% window to external wall area ratios) single skin alternatives

• simulations of the highly glazed (100% window to external wall area ratio) double skin alternatives; (a) pilot study on a component level and (b) simulations on a zone and building level

The main aim of this section is to present the generated alternatives, but also to provide the reasoning behind this selection.


100

4.1.1 Reference building (30% window to external wall area ratio)

Initially, a 30% glazed (virtual reference) offi ce building was created (see Chapter 5). The building has 6 storeys and is of rectangular shape. Two plan types, three control set points and three orientations were considered. The 18 generated alternatives were compared with each other, in terms of energy use and quality of thermal environment both on building and zone level. The simulations of the reference building were carried out for different:

• plan typeso cello open

• building orientationso north-south (short façade)o east-west (short façade)o northwest-southeast (short façade)

• control set points for indoor air temperature:o strict (22 – 23°C)o normal (22 – 24.5°C)o poor (21 – 26°C)

The façade construction of the reference building remained the same for all the simulated alternatives discussed in this section. The main focus of the parametric studies of the reference building is a sensitivity analysis re-garding the plan type, orientation and control set points and later, a cross reference comparison between the glazed alternatives. The 18 reference building alternatives are shown in Figure 4.1.

Methods

101

Ref

eren

ce B

uild

ing

(SS-

30%

)

Cel

l Typ

e (S

S-30

%-C

ell)

Ope

n Pl

an T

ype

(SS-

30%

-Ope

n)

Nor

th-S

outh

(S

S-30

%-O

pen-

NS)

N

orth

-Sou

th

(SS-

30%

-Cel

l-NS)

N

orth

-Sou

th 4

5º

(SS-

30%

-Cel

l-NS4

5)

East

-Wes

t (S

S-30

%-C

ell-E

W)

Nor

th-S

outh

45º

(S

S-30

%-O

pen-

NS4

5)

East

-Wes

t (S

S-30

%-O

pen-

EW)

Stric

t Con

trol

(SS-

30%

-Cel

l-N

S-st

rict)

Nor

mal

Con

trol

(SS-

30%

-Cel

l-NS-

norm

al)

Poor

con

trol

(SS-

30%

-Cel

l-N

S-po

or)

Stric

t Con

trol

(SS-

30%

-Cel

l-EW

-stri

ct)

Nor

mal

Con

trol

(SS-

30%

-Cel

l-EW

-no

rmal

)

Poor

con

trol

(SS-

30%

-Cel

l-EW

-poo

r)

Stric

t Con

trol

(SS-

30%

-Cel

l-N

S45-

stric

t)

Nor

mal

Con

trol

(SS-

30%

-Cel

l-N

S45-

norm

al)

Poor

con

trol

(SS-

30%

-Cel

l-N

S45-

poor

)

Stric

t Con

trol

(SS-

30%

-Ope

n-N

S-st

rict)

Nor

mal

Con

trol

(SS-

30%

-Ope

n-N

S-no

rmal

)

Poor

con

trol

(SS-

30%

- O

pen-

NS-

poor

)

Stric

t Con

trol

(SS-

30%

- Ope

n-N

S45-

stric

t)

Nor

mal

Con

trol

(SS-

30%

- Ope

n-N

S45-

norm

al)

Poor

con

trol

(SS-

30%

- Ope

n-N

S45-

poor

)

Stric

t Con

trol

(SS-

30%

-Ope

n -E

W-s

trict

)

Nor

mal

Con

trol

(SS-

30%

-Ope

n-EW

-nor

mal

)

Poor

con

trol

(SS-

30%

-Ope

n -E

W-p

oor)

Figure 4.1 Tree diagram for reference building alternatives (SS=single skin façade, NS=Nort-South).


102

4.1.2 Single skin alternatives (60% and 100% window to external wall area ratios)

Seven alternatives were initially created for the parametric studies with 60% and 100% window to external wall area ratios. For these alternatives different window and shading device types and positions were applied and for each one 3 control set points and 2 plan types were simulated. The 42 (60% glazed) generated alternatives were compared on a build-ing level. The same number of alternatives was simulated for the 100% glazed alternatives and conclusions have been drawn both on building and zone level. A fl ow chart for the selection of the alternatives, as the fi rst step of the parametric study’s methodology, is set out below (Figure 4.2). Commercially available window types in general use were chosen, with visual transmittance higher than 0.5 to ensure acceptable indoor daylight availability. The selection was made based on personal communication with Dr. Helena Bülow-Hübe (Division of Energy and Building Design, Lund University).

Reference building 30% Glazed Building Triple clear glazing (2+1)

(Ug=1.85, g=0.691, Tsol=0.579, Uf=2.31)

Building alternative 1 60%, 100% Glazed Building

Triple clear glazing (2+1) (Ug=1.85, g=0.691, Tsol=0.579,

Uf=2.31)

Increase of the window area

Building alternative 2 60%, 100% Glazed Building 4+30+4-12Ar-Ot4 (1+2)

(Ug=1.14, g=0.584, Tsol=0.44 Uf=1.6) Intermediate blinds

Improved U-value of the glazing and frame

Decreased number of panes, Internal blinds


6Hbl-12Ar-4 (~brilliant 66) (Ug=1.14, g=0.354, Tsol=0.297

Uf=1.6)

Same window type

Same % of window area

Same Ug value, Uf value

Same Ug value, Uf value, blinds Decreased g value, Tsol


6Hbl-12Ar-4 (~brilliant 66) (Ug=1.14, g=0.354, Tsol=0.297 Uf=1.6)

Internal screens


6Hbm-12Ar-4 (~brilliant 50)

(Ug=1.14, g=0.277, Tsol=0.221,Uf=1.6)

Internal blinds


4SN-15Ar-4 (OptithermSN)

(Ug=1.14, g=0.584, Tsol=0.221,Uf=1.6)

Same Ug value, Uf value, blinds Increased g value, Tsol

Same window

type

Internal screens


6Hbl-12Ar-4 (~brilliant 66) (Ug=1.14, g=0.354, Tsol=0.297 Uf=1.6)

External blinds

Same window

type

External horizontal

louvres

Same Ug value, Uf value, Different shading position

Figure 4.2 Generation of 60% and 100% glazed building alternatives.

Methods

103

The fi rst building alternative works as a “bridge” between the 30% and 60% glazed buildings. The Uglazing and Uframe values were kept the same, in order to study the impact of (larger) window area on energy use and indoor climate. The number and type of panes, the type and positioning of shading devices, etc, are the same as for the reference building. The total U value of the window, however, is not the same, since glazing and frame areas are different (Af = 32% for the long and Af = 19% for the short façade for the reference building, Af = 27.7% for the 60% and Af =14.3% for the 100% glazed building), as described in Chapter 5.

The window type of the second alternative was changed (lower Uglazing and Uframe values), in order to get a more realistic solution for a highly glazed building. The frame properties were chosen according to recommen-dations of Schüco International and the glazing system 4-30-4-12Ar-Ot4 (2+1) (Optitherm) was chosen from Pilkington’s Glass Catalogue (2004). The type and position of the shading devices remained the same.

The triple glazed unit of the second alternative was replaced by a double glazed one with the same Uglazing (Brilliant 66 according to Pilkington’s Glass Catalogue). The solar factor (g value) of the third alternative was decreased to 0.354 (from 0.584 of the second alternative). The intermediate venetian blinds were replaced by internal ones (the properties of the blinds remain the same) increasing the geffective. This window type was considered as typical alternative used for a 60% and 100% glazed building.

In the fourth alternative the thermal transmittance of the window remained the same, while the solar factor (g value) decreased even more, by up to 0.277, in order to study its impact on the cooling demand (Brilliant 50 according to Pilkington’s Glass Catalogue). The number of panes, the position and type of shading devices remained the same with the third alternative.

In the fi fth alternative the thermal transmittance of the window re-mained the same, while the solar factor (g value) increased up to 0.584. The window type used in this case was a double glazing unit 4SN-15-4 (OptithermSN according to Pilkington’s Glass Catalogue), with similar properties to the second alternative; the venetian blinds, however, were placed internally. This case was selected in order to further investigate the infl uence of g and geffective values on the heating and cooling demand (the position of shading devices infl uences the geffective values).

In the sixth alternative internal screens (Hexcel 21136 Satine Blanc 101 according to Parasol) were placed instead of venetian blinds. The window construction was identical with the third alternative, since this was considered to be the one more often used. This alternative makes it is possible to investigate the infl uence of different types of shading device on energy use and indoor environment.


104

In the last alternative, fi xed horizontal external louvres have replaced the internal screens (according to Schüco’s recommendations). The window construction was again identical with the one of the third alternative.

4.1.3 Double skin alternatives (100% window to external wall area ratio)

The simulations of double skin façade alternatives are carried out in the following 3 stages:

• Pilot study: simulations on a component (double skin façade cavity) level. The study is carried out in order to (a) gain knowledge regar-ding the impact of design parameters on the system’s performance and (b) select glazing and geometry characteristics for the further simulations on zone and building level. The simulations were car-ried out using WIS 3 and also ISO standard 15099 for steady state boundary conditions.

• Study on a zone level: All year round simulations were carried out on a single offi ce zone (as described in detail below). The main aim of this step is to evaluate the performance of different double skin façade alternatives for Swedish climatic conditions. The parameters varied were the glazing type, the façade mode and the ventilation strategy. The one zone model gives fl exibility and speed for the simulations (compared with the simulations on a building level) making possible to run a large number of alternatives. On the other hand, in this way only comparative studies can be achieved; absolute values of the energy and indoor climate performance can be obtained only by simulations on a building level. The software used for this study was IDA ICE 3.0.

• Study on a building level: All year round simulations were carried out for the whole building. The aim of this step is to obtain absolute values regarding the building’s performance for further comparisons with the single skin glazed alternatives. The software used was IDA ICE 3.0.

4.1.3.1 Pilot study on component level using WIS 3In this chapter parametric studies of different double skin façade alterna-tives were carried out, in order to estimate the infl uence of key design parameters on the performance of a double skin façade component. The main aim was not to calculate absolute values of the system performance

Methods

105

but instead to “get a feeling” of its possibilities and limitations. The focus of this study is the performance of the double skin façade cavity, while the optimization of the system’s integration was studied later in this thesis using the IDA ICE 3.0 software.

In general the performance of a double skin façade cavity is highly de-pendent on the climatic conditions and thus the location of the building. The orientation of the façade, the building use and the HVAC strategy, can also be crucial for the façade design. The parameters varied were the cavity geometry, the number and type of panes used, and the HVAC mode of the façade for certain climatic conditions. Once knowledge was gained of what can be achieved with different double skin façade confi gurations and the extent to which certain design parameters infl uence the system’s performance, then optimized façade integration was studied (second and third steps as described above). It becomes obvious that the conclusions obtained from the present study are independent of location (since the climatic conditions were predifi ned) and as a result, the output is not li-mited to Nordic climates. The optimization of double skin façades in terms of energy use and thermal comfort, however, also involves the integration process (highly dependent on the building location) which will be studied in the next steps (carried out for Swedish climate).

Due to the large amount of input and output, the parametric studies were carried out selectively, with the main emphasis on the possibilities and limitations of the system.

The parameters varied during this study were:

• outdoor climatic conditions (typical summer and winter day and night, extreme summer day)

• mode of the double skin façade system (standard double façade and airfl ow window mode, as explained below)

• type of ventilation inside the cavity (natural, mechanical)• geometry of the façade (height and depth of the cavity)• type of glazing (number and type of panes)• type and position of shading devices inside the cavity

Outdoor climatic conditionsThe double skin alternatives were studied for a typical summer day and night, a typical winter day and night and an extreme summer day (in order to study potential overheating problems occurred by the high inner pane temperatures). The outdoor and indoor air temperatures and convection coeffi cients set as input in WIS 3, were decided after discussions with Dr. Bengt Hellström (Division of Energy and Building Design, Department


106

of Architecture and Built Environment, Lund University, Sweden) and are presented in Table 4.1.

Table 4.1 Indoor and outdoor environment for the WIS 3 simulations

Name Outdoor Indoor Direct Convection Convection temperature temperature solar coeffi cient coeffi cient (°C) (°C) radiation outdoors indoors (W/m2) (W/m2K) (W/m2K)typical summer day 20 24 500 8 2.5typical winter day 0 23 300 20 3.6extreme summer day 30 25 900 8 2.5

Façade mode and type of ventilationThe two modes considered in this study were the “standard double façade” and the “airfl ow window” mode. In the “standard double façade” mode air always enters the cavity from outdoors. The cavity can be (a) closed during the heating season for increased thermal insulation and opened during the cooling season for heat extraction purposes (naturally ventilated cavity) or (b) used for preheating the air supplied in the AHU (as supply air) during the heating season and extracted during cooling periods (mechanically ventilated case). For the “airfl ow window” mode the air always enters the cavity from indoors (offi ce space) as exhaust air all year round. The aim is to improve the inner glass temperatures for extreme winter and extreme summer conditions. During the heating season the cavity air ends up in the AHU for heat recovery purposes, while during the cooling season the air is extracted to the outside.

Geometry of the façade (height and depth of the cavity)Two double skin façade construction types were considered: a multi storey type (10, 20 and 30 m high) and a box window type (3.5 m high). Since the software used for the current simulations (WIS 3) is two dimensional, the cavity width does not infl uence the output of the simulations. For all the cases the width was assumed to be 3.5 m and the depth varied from 0.2m to 1.6m, which is a typical range.

Type of glazing (number and type of panes)For the “typical double façade” mode the thermal barrier (double glazing unit) is required as inner layer, while in the “airfl ow window” cases, it is required as outer one. The glazing unit alternatives used for the simulations and the reasoning for selecting them are described below.

Methods

107

Standard double façade modeFor the “standard” double façade mode the outer skin consists of a single pane and the inner skin of a double one (as described above). The venti-lated cavity is between the intermediate and external panes. According to the recommendation of WSP consultant Diana Avasoo, seven different glazing combinations were initially considered for this construction type as shown in Table 4.2. A detailed description of the glazing types, the façade construction and the input parameters follows in Subchapter 5.3.

Table 4.2 Glazing combinations for the “standard” double façade mode.

Case External Gap Intermediate Gap Internal pane (800mm) pane (12mm) pane

A Clear pane Ventilated Clear pane Air Clear pane 8 mm 4 mm 4mm

B Clear pane Ventilated Clear pane Argon Low E 8 mm 4 mm 4 mm

C Clear pane Ventilated Optigreen1 Argon Clear pane 8 mm 6 mm 4 mm

D Optigreen1 Ventilated Clear pane Argon Clear pane 8 mm 4 mm 4 mm

E Optigreen1 Ventilated Clear pane Argon Low E 8 mm 4 mm 4 mm

F Clear pane Ventilated Solar control + lowE2 Argon Clear pane 8 mm 6 mm 4 mm

G Solar control +lowE3 Ventilated Clear pane Argon Low E 8 mm 4 mm 4 mm

1 body tinted solar control glass2 soft coated glass3 hard coated glass

In the fi rst case (A) three clear panes were applied. This case was considered as a basic case and was chosen mainly for reasons of comparison between the single and double skin façade reference building.

In the second case, the clear inner pane was replaced by a low E coated one (with the same thickness), while the intermediate and outer panes were kept the same. This case is considered as a common solution in existing buildings.

In the third case a solar control coated pane was placed as an interme-diate pane. The third and fourth cases have the same pane types. In the fourth alternative, however, the solar control pane with increased thickness (8 mm instead of 4mm) was placed as outer pane. Since the solar control


108

(coated) panes have soft coatings they can not be used as single panes. This is the reason why a body tinted pane was selected in the fourth case.

The fi fth case is a combination of the second and fourth cases. A low E coated pane is used as internal one and a body tinted solar control pane as outer. This is a more advanced solution with lower thermal and solar transmittance.

The sixth case has clear inner and outer panes, while the intermediate pane is a solar control + low E (soft coated) one. The main difference be-tween this and the fi fth case is that instead of applying the two coatings in the outer and inner panes, both of them were applied on the intermediate one. When the solar control coating is applied on the intermediate pane there is a possibility of thermal failure due to overheating; nevertheless this case was considered, in order to study the impact of coating position on the façade behaviour (two façades (E and F) with similar properties and different performance).

Finally, in the seventh case the advanced glazing used in the sixth case was placed as an outer pane; this time, however, it is hard coated. A low E pane is also placed as inner pane instead of the clear pane used in the sixth case in order to achieve lower thermal transmittance.

Airfl ow window modeFor the airfl ow window mode the outer skin consists of a double pane and the inner skin of a single pane. The ventilated cavity is between the internal and intermediate panes. According to the recommendation of WSP consultant Diana Avasoo, seven different glazing combinations were initially considered for this construction type as shown in Table 4.3. A detailed description of the glazing types, the façade construction and the input parameters follows in Subchapter 5.3.

Methods

109

Table 4.3 Glazing combinations for the Airfl ow Window mode

Case External Gap Intermediate Gap Internal pane (12mm) pane (800mm) pane

A Clear pane Air Clear pane Ventilated Clear pane 8 mm 4 mm 4mm

B Clear pane Argon Low E Ventilated Clear pane 8 mm 4 mm 4 mm

C Clear pane Argon Optigreen1 Ventilated Clear pane 8 mm 6 mm 4 mm

D Optigreen1 Argon Clear pane Ventilated Clear pane 8 mm 4 mm 4 mm

E Optigreen1 Argon Low E Ventilated Clear pane 8 mm 4 mm 4 mm

F Solar control + lowE2 Argon Clear pane Ventilated Clear pane 8 mm 8 mm 4 mm

G Solar control +lowE2 Argon Clear pane Ventilated Low E3

8 mm 4 mm 6 mm

1 body tinted solar control glass2 soft coated glass3 hard coated glass

The fi rst (reference) case was considered as in the “standard” double façade mode, the one with three clear panes.

In the second case the clear intermediate pane was replaced by a low E coated one (with the same thickness), while the inner and outer panes were kept the same. This case is equivalent to the second case of the “standard” double façade mode.

The third and fourth cases are identical with the ones used for the “standard” double façade mode. For the third case a solar control pane was placed as an intermediate one while the inner and outer panes were kept the same (clear). In the fourth case the solar control (body tinted) pane was placed as an external pane.

The fi fth case is again similar to the one of the “standard” double façade mode. A low E coating is used for the intermediate pane and a body tinted solar control pane is used as an outer one. The low E glazing is placed as intermediate pane (instead of internal pane), since the thermal barrier is needed to the external skin.

In the sixth case the solar control (soft coated) with low emissivity gla-zing unit has been placed as external pane (slightly thicker than the one of the “standard” double façade mode for safety reasons), while the inner and intermediate panes were kept clear.


110

Finally, the seventh case is exactly the same as the one in the “standard” double façade mode (the advanced coated glazing was in this case soft and was placed as an outer pane, and a low E pane was placed as inner pane).

Position of shading devices inside the cavityThe position of shading devices inside the cavity was also changed dur-ing this study, in order to investigate its infl uence on the inner pane temperatures.

4.1.3.2 Study on a zone level using IDA ICE 3.0The aim of the parametric studies of different double skin façade cavities (carried out in WIS 3) was to estimate the infl uence of key design param-eters on the performance of the double skin façade component. The aim of the simulations carried out in IDA ICE 3.0, described in this Section, is to study the effi cient integration in terms of energy use and thermal comfort of double skin alternatives selected from the WIS 3 simulations. In order to achieve this goal, the study was carried out both on a zone and a building level. On a zone level the main aim was to optimize the combination between glazing and shading devices and controls for the cavity dampers for each ventilation strategy and façade orientation. The parametric study was carried out on a zone level only for box window cases (for both modes).

In order to tune the control set points of the dampers for different dou-ble skin façade alternatives (in the cases of naturally ventilated and hybrid cases, as explained below), four orientations were considered: north, west, south and east. For the “standard” double façade mode three ventilation strategies were considered:

• naturally ventilated cavity: open cavity during the cooling periods for extracting the heat from the cavity and closed during winter for increasing the temperature of the inner layer. The exhaust air from the offi ce space is used for heat recovery in the AHU.

• mechanically ventilated cavity: air (equal to the supply air for each zone) enters the cavity from outdoors and is driven to a mixing box. This mixing box mixes the cavity air and the outdoor air, in order to meet the set point of the supply air temperature. If further heating or cooling is needed, then it is provided by the coils of the AHU. The cavity air was used as supply air into the zones. The exhaust air from the offi ce space is used for heat recovery in the AHU.

Methods

111

• hybrid ventilated cavity: open cavity during the cooling periods (as in the naturally ventilated cases), and use of the outlet air as supply during heating periods (as in the mechanically ventilated cavity). The exhaust air from the offi ce space is used for heat recovery in the AHU.

For the airfl ow windows the cavity is mechanically ventilated all year round. The exhaust air from the offi ce zone enters the cavity and is driven to the AHU for heat recovery purposes.

The glazing alternatives considered for this study are:

• “standard” double façade mode:

o case A: Selected as a reference case. The fi rst case has the high-est thermal and total solar transmittance due to the tree clear panes

o case D: This case has the solar control body tinted pane as an exterior layer and clear intermediate and inner panes.

o case E: In this case both a solar control body tinted pane (ex-ternal) and a low E pane (intermediate) were applied.

o case F: This alternative has the lowest thermal and total solar transmittance. Only the intermediate pane is an advanced one (low E and solar control coatings), while the inner and external ones are clear

• airfl ow window mode:

o case A: The fi rst (reference) case was considered as in the “stand-ard” double façade mode, the one with three clear panes

o case D: A solar control (body tinted) pane was placed as an external one, while the inner and intermediate panes are clear

o case E: A low E coating was used for the intermediate pane and a body tinted solar control pane was used as the outer one

o case F: In the sixth case the solar control (soft coated) glazing unit of low emissivity has been placed as external pane, while the inner and intermediate panes were kept clear

o case G: In this case an advanced coated glazing was placed as outer pane and a low E pane was placed as inner pane. This case was selected to fi nd out if further decrease of the U value (compared with case F) results in energy savings


112

For the standard double façade mode 96 alternatives in total were simu-lated (as shown in detail in Figure 4.3), and for the airfl ow window cases 40. More details regarding the properties of the glazing and the shading devices considered are provided in Chapter 5.

Dou

ble

Skin

Faç

ades

(ID

AIC

E 3.

0 Si

mul

atio

ns)

Airf

low

win

dow

mod

e

Mul

ti st

orey

hig

h B

oxW

indo

w

4 or

ient

atio

ns

3 ve

ntila

tion

stra

tegi

es

4

glaz

ing

type

s 2

shad

ing

devi

ce ty

pes

Box

Win

dow

tota

l num

ber o

f si

mul

atio

ns=

96

Opt

imiz

ed c

ontr

ol se

t poi

nts f

or d

ampe

rs

4 or

ient

atio

ns

5 gl

azin

g ty

pes

2 sh

adin

g de

vice

type

s

tota

l num

ber o

f si

mul

atio

ns=

40

P

aram

etric

stud

y:

Para

met

ric

study

:

Nat

ural

ly v

entil

ated

cav

ity

Opt

imiz

ed c

ontro

l set

po

ints

for d

ampe

rs

3 ve

ntila

tion

stra

tegi

es (n

atur

al,

mec

hani

cal,

hybr

id)

1 ve

ntila

tion

stra

tegy

(mec

hani

cal)

Zone

le

vel

Bui

ldin

g le

vel

“Sta

ndar

d” d

oubl

e fa

çade

mod

e

Figure 4.3 Methodology of IDA ICE 3.0 simulations.

Methods

113

4.1.3.3 Study on a building level using IDA ICE 3.0The aim of the simulations carried out on a building level is to study the performance of the building in terms of energy use and thermal comfort. Absolute values are calculated, in order to carry out comparisons with the single skin glazed offi ce building alternatives.

As shown in Figure 4.3 the cases simulated on a building level are the:

• naturally ventilated multi storey double façades• naturally ventilated box windows double façades• mechanically ventilated box windows double façades• hybrid ventilated box window double façades• airfl ow windows (mechanically ventilated)

The two best performing (in terms of energy use) glazing alternatives studied on a zone level were selected for further simulations on a building level. The reasoning for this selection is given in Subchapter 6.3.3.

4.2 Description of the studied parameters

4.2.1 Single skin alternatives (30%, 60% and 100% glazed alternatives)

The parameters studied in IDA ICE 3.0 regarding the reference building (30% window to external wall area ratio) and the highly glazed single skin buildings (60% and 100% window to external wall area ratios) concern mainly the (a) energy use for heating, cooling, equipment, pumps and fans and lighting and (b) indoor climate (mean air temperatures, directed operative temperatures and perception of thermal comfort i.e. Fanger’s comfort indices).

The energy use, the weighted average mean air temperatures and the perception of thermal comfort are examined and compared on a building level. On a zone level the parameters studied are the mean air temperatures, the directed operative temperatures, the Predicted Mean Vote and the Predicted Percentage of Dissatisfi ed. The parametric studies on the zone level were carried out only for zones of the cell plan offi ce building. The directed operative temperatures were calculated for the cell type, while the operative temperatures were calculated for the open plan.


114

4.2.1.1 IDA ICE 3.0 output (zone level)Because of the large number of simulations, the parametric study on a zone level is carried out only for the cell type 30% and 100% glazed buil-dings. The large number of simulated zones provides suffi cient informa-tion regarding:

• mean air temperatures• directed operative temperatures• PMV and PPD values

4.2.1.2 IDA ICE 3.0 output (building level)The comparison on a building level of the 30%, 60% and 100% glazed buildings can be complicated, due to the large amount of simulated alterna-tives and the different performance criteria involved (energy use, thermal comfort, etc). In order to study the impact of each orientation, plan type, set point or façade element on the building performance, one parameter at a time was changed and the rest remained the same. For example, if the impact of the control set points (strict, normal or poor) on the energy use was examined, then all the alternatives with the same plan types and orientations were compared with each other.

The performance parameters examined on a building level were:

• energy use for o heatingo coolingo ventilationo pumps, fans, etco lighting

• weighted (average mean monthly) air temperatures for the working area: For each building zone the mean air temperatures were calculated by IDA ICE 3.0 (for every hour). These results were used to calculate the average monthly mean air temperatures for each zone. The weighted average mean air temperature is the sum of the mean air temperatures of each zone multiplied by the size and number of zones divided by the total fl oor area (excluding the corridors) as shown in Equation 4.1.

total

zoneszonezonemean,mean A

)NA(TT

∑ ××=

Equation 4.1 Weighted Average Mean Air Temperature.

Methods

115

where>

Tmean,zone is the monthly average zone temperatureAzone is the zone areaNzones is the number of the zonesAtotal is the total fl oor area

The limitation of the weighted average mean monthly air temperatures is that no information is given as to the variation over time or space.

For the cell type offi ce building, only the occupied zones were in-cluded in the calculations (offi ces and meeting rooms, as described in detail in Chapter 5). The corridor (zone 11) was excluded, since the impact on the occupants’ comfort is limited. However, for the open plan type the whole area was considered for the calculations, since all this area was considered as working space. All the calculations (exclu-ding those for energy) were carried out for a middle fl oor (in IDA ICE 3.0).

• number of hours between certain (weighted) average mean air temperatures for working space: This output gives information similar to the weighted average mean air temperatures for the working area. However, this is more a quantitive indicator compared with the previous output, since the previous one only gives monthly averages and does not provide any information concerning the variation of the indoor temperatures during the year.

• weighted average PMV: As already described in Subsection 2.3.2.2, the Predicted Mean Vote is a qualitative indicator of the perception of thermal comfort. Likewise, the weighted average PMV is a monthly average PMV value of each occupant multiplied by the number of occupants in each zone and the number of identical zones, divided by the total number of occupants. This factor can basically show whether or not the controls are proper for certain occupancy, since it provides information for all the year.

• number of working hours for certain average PPD: As already des-cribed in subsection 2.3.2.2, the Predicted Percentage of Dissatisfi ed is more a quantitative indicator of the perception of thermal comfort. In a similar way weighted average PPD is the sum of the PPD of each occupant multiplied by the number of occupants in each zone and the number of identical zones, divided by the total number of occupants.

This output is more a quantitative indicator for the perception of thermal environment in the working space. Although it does not show


116

whether the occupant feels warm or cold, it provides the necessary in-formation to classify an indoor thermal environment, both on a zone and on a building level.

4.2.2 Double skin alternativesThe simulations for the double skin façade alternatives were carried out at three levels: (a) component level (pilot study aiming to reduce the number of glazing alternatives initially considered using WIS 3), (b) zone level (parametric study focusing on the impact of the façade mode and construction on energy use and of a specifi c offi ce zone for different orientations, using IDA ICE 3.0) and (c) building level (parametric study investigating the impact of glazing type on the system’s performance (natu-rally, mechanically and hybrid ventilated double façades and mechanically ventilated airfl ow windows), using IDA ICE 3.0). The main parameters studied at these 3 levels are briefl y described below.

4.2.2.1 WIS 3 simulations (component level)The main aim of this study is to better understand the possibilities and limitations of different double skin façade constructions. This pilot study can be considered as a fi rst step for optimizing the system’s performance and integration on a building level. During the calculations carried out on the double skin façade level, several parameters were varied (façade mode, geometry of the cavity, size of the openings, glazing type and shading device position inside the cavity – when applied). In this way, their impacts mainly on the air fl ow, the air temperature at different heights of the cavity and the surface temperature of different panes were investigated.

Due to the large number of varied parameters a detailed methodology was developed with the aim to decrease the number of simulations (Figure 4.4). A description of the methodology is given below and, when necessary, the reasoning is set out.

Methods

117

Dou

ble

Skin

Faç

ades

“Sta

ndar

d” d

oubl

e fa

çade

mod

e A

irflo

w w

indo

w m

ode

Nat

ural

ly v

entil

ated

cav

ity

Mec

hani

cally

ven

tilat

ed c

avity

Mec

hani

cally

ven

tilat

ed c

avity

Pr

elim

inar

y st

udy:

Se

lect

ion

of g

lazin

g co

mbi

natio

ns

Para

met

ric

study

: in

fluen

ce o

f cav

ity

geom

etry

on

syst

em’s

pe

rfor

man

ce

Perf

orm

ance

of t

he

glaz

ing

alte

rnat

ives

Influ

ence

of

cavi

ty h

eigh

t and

de

pth

on a

irflo

ws

Influ

ence

of c

avity

he

ight

and

dep

th o

n te

mpe

ratu

re p

rofil

e

Influ

ence

of %

of o

pene

d ar

ea o

n ai

rflow

s and

te

mpe

ratu

re p

rofil

e

Influ

ence

of w

eath

er

cond

ition

s on

optim

al d

epth

Influ

ence

of

glaz

ing

on

optim

al d

epth

Influ

ence

of s

hadi

ng d

evic

es

posit

ion

on th

e ai

r and

the

inne

r pan

e te

mpe

ratu

res

Air t

empe

ratu

res

(mul

ti st

orey

hig

h)

Surf

ace

tem

pera

ture

s (b

ox w

indo

w)

Prel

imin

ary

stud

y:

influ

ence

of d

epth

on

air

and

surfa

ce

tem

pera

ture

Surf

ace

tem

pera

ture

s O

utle

t air

te

mpe

ratu

res

Air

tem

pera

ture

s Su

rfac

e te

mpe

ratu

res

Figure 4.4 Methodology of WIS 3 simulations.


118

“Standard” double façade modeAs described in detail in the Subsection 4.1.3.1, initially seven glazing combinations were considered for each mode. However, due to the large amount of simulations a preliminary study was carried out (see Figure 4.4) for the “standard” double façade mode, and the number of glazing alternatives was reduced from seven to four. For the airfl ow window mode, on the other hand, all seven alternatives were studied.

Infl uence of cavity geometry on system performance (naturally ventilated cavity): Generally, studying a naturally ventilated cavity is more complicated than studying a mechanically ventilated one, since the air-fl ow rate depends on the characteristics of the cavity such as glazing type, geometry of the cavity, etc. For this reason a further parametric study was carried out for the naturally ventilated double façade mode, in order to study the impact of construction on the system performance (see Figure 4.4). The parameters studied were:

• infl uence of cavity height and depth on airfl ows• infl uence of cavity height and depth on air temperature profi le• infl uence of opened area on airfl ows and air temperature profi le• infl uence of weather conditions on optimal depth• infl uence of glazing on optimal depth• infl uence of shading device position on the air and the inner pane

temperatures

Performance of the glazing alternatives (naturally and mechanically ventilated cavity): After investigating the infl uence of geometry on the performance of the naturally ventilated cavities, different glazing alterna-tives were evaluated. The outputs of these simulations were the airfl ow and air temperature profi le along the cavity and the inner pane’s surface temperatures. A similar study was carried out for the mechanically venti-lated “standard” double façade alternatives.

Airfl ow window modeFinally, for the airfl ow window cases, all seven glazing alternatives were considered. As shown in Figure 4.4, for the airfl ow window mode only mechanical ventilation was considered. A brief parametric study was car-ried out to investigate the infl uence of cavity depth on the inner pane surface temperature and the outlet air temperatures. This was followed by an evaluation of the seven glazing alternatives.

Methods

119

4.2.2.2 IDA ICE 3.0 Simulations (zone level)The output of the parametric study on a zone level was mainly focused on (a) energy demand for heating and cooling the offi ce (zone behind the double façade) and (b) monthly average PMV.

The main aim of the parametric studies on a zone level is to evaluate the simulated alternatives, in order to select the ones for further simula-tions on a building level. Since it takes a long time to simulate such a complicated and detailed building model, this screening is essential for the optimization of its performance.

4.2.2.3 IDA ICE 3.0 Simulations (building level)The two parameters studied and compared on a building level are the energy use and the number of hours with certain PPD values. The energy use (absolute) values obtained from the building level serve comparison purposes with the single skin alternatives, while the PPD values give a more “quantative” way of indoor climate comparisons.

4.3 Description of the simulation tools

4.3.1 Simulations using WIS 3The software used for calculations of the double skin façade cavity is WIS 3. According to the WIS 3 User’s Guide “WIS is a European software tool for the calculation of the thermal and solar properties of commercial and innovative window systems on the basis of known component properties and thermal and solar/optical interactions between the components.

One of the unique elements in the software tool is the combination of glazings and shading devices, with the option of free or forced air circulation between the components. This makes the tool particularly suited to calculate the thermal and solar performance of complex windows and active facades”.

WIS 3 has been built in algorithms based on international (CEN, ISO) standards, such as the ISO Standard 15099 (2003).

As already mentioned the two main outputs of the simulations are:

• temperatures at the centre of each layer.• vertical air temperature profi le along the ventilated cavity.


120

4.3.1.1 Temperatures at the centre of each layerThese values were calculated by WIS 3 and they refer to the centre of each layer. A short description of the calculated values follows.

T1: temperature of the outer glass (outer surface – average temperature) Tglass1: average temperature of the outer glass (average temperature)T2: temperature of the outer glass (inner surface – average temperature)Tgap1: temperature of the air between the outer and the intermediate glass (in

the centre of the cavity)Tgap1-(a): temperature of the layer between the outer glass and the blind (in the

centre of the cavity)Tgap1-(b): temperature of the layer between the blind and the intermediate glass

(in the centre of the cavity) Tbo: temperature of the outer surface of the blinds (average temperature)Tblind: temperature of the blinds (average temperature)Tbi: temperature of the inner surface of the blinds (average temperature)T3: temperature of the intermediate pane (outer surface – average tempera-

ture)Tglass2: average temperature of the intermediate glass (average temperature)T4: temperature of the intermediate pane (inner surface – average tempera-

ture)Tgap2: temperature of the layer between the intermediate and the inner glass

(in the centre of the cavity)T5: temperature of the inner pane (outer surface – average temperature)Tglass3: average temperature of the inner glass (average temperature)T6: temperature of the inner pane (inner surface – average temperature)

T6

T5

Tgap2

T4

T3

Tgap1

T2

T1

Figure 4.5 Case without venetian blind.

Methods

121

T6

T5

Tgap2

T4

T3

Tgap1-(b)

Tbi

Tblind

Tbo

Tgap1-(a)

T2

T1

Figure 4.6 Case with venetian blind.

4.3.1.2 Temperatures at different heights of the cavity

General caseAccording to ISO Standard 15099 (2003), by assuming that the mean velocity of the air in the space is known, the temperature profi le and the heat fl ow may be calculated by a simple model.


122

Height h

Air temperature Tgap i (h)

Average airtemperature, Tgap i

Air flow j v,i

H0,i

Hi

Average surfacetemperature, Tave i

Outlet airtemperature, Tgap i_out

Inlet airtemperature, Tgap i_in

Figure 4.7 Air fl ow in the gap of a window system.

Due to the air fl ow through the double skin façade cavity, the air tempera-ture in the cavity varies with the height (see Figure 4.7). The temperature profi le depends on the air velocity in the space and the heat transfer coef-fi cient to both layers. The air temperature profi le in the space i is given by:

i0,h/H

ingap_i,iav,iav,gap_i e)T(TT(h)T −⋅−−=

Equation 4.2 Temperature profi le along the cavity

where:

Tgap_i(h): is the temperature of the air in gap i at position h, (in m); H0,i: is the characteristic height (temperature penetration length, in m);Tgap_i,in: is the temperature of the incoming air in gap i, (in °C);Tav_i: is the average temperature of the surfaces of layers i and i+1, given by

equation:

2

)T(TT

1f_ib_iav_i

++=

Equation 4.3 Average temperature of the surfaces of layers i and i+1

Methods

123

where:

Tav_i : is the average temperature of the surfaces of layers i and i+1, (in °C)Tb_i : is the temperature of the surface of layer (pane, fi lm or shading) i,

facing the cavity i, (in °C)Tf_i+1: is the temperature of the surface of layer (pane, fi lm or shading) i+1,

facing the cavity i, (in °C)

The characteristic height (H0,i) of the temperature profi le is defi ned by the following equation:

i

icv,

ipii0, V

h2

scH ⋅

⋅

⋅⋅=

ρ

Equation 4.4 Characteristic height

where:

ρi : is the density of the air at temperature Tgap_j , in (kg/m3)cp : is the specifi c heat capacity, in (J/kgK)si : is the depth of the cavity i, in (m);Vi : is the mean velocity of the air fl ow in the cavity i, in (m/s);hcv,j : is the heat transfer coeffi cient for ventilated cavities, in (W/m2K).

Case with no venetian blind in the cavity If there is no shading in the cavity, then Tav is the average value of T2 and T3. When the incoming air in the cavity is at 0°C, it is assumed that the air density is 1.25 and when the incoming air in the cavity is at 25°C the air density is 1.18. The heat transfer coeffi cient for the ventilated cavity is calculated by the following equation:

4V2hh ccv +=

Equation 4.5 Heat transfer coeffi cient for the ventilated cavity

where:

hc : is the heat transfer coeffi cient for the cavity if it is not ventilated (given from WIS 3)

V : is the mean velocity of the air fl ow in the cavity in (m/s)


124

Case with blinds in the cavity In the cases that between the outer and the intermediate pane shading is applied, then two different temperatures are calculated: the one between the outer pane (or inner pane, depending on where the ventilated cavity is situated) and the blind, and the other between the blind and the in-termediate pane. In the fi rst case the temperatures Tbi and T2 were used for Tav and in the second case the temperatures Tbi and T3 were used for Tav. For the cases with ventilated cavity no air exchange between the two subcavities was assumed.

4.3.2 Simulations using IDA ICE 3.0

4.3.2.1 General descriptionBefore a building thermal simulation tool was chosen, certain perfor-mance criteria were developed. The program was to have the following features:

1. A dynamic building simulation tool2. User friendly interface3. Multi-zone capability4. Simple natural ventilation features5. Simulation of HVAC systems typical for offi ce buildings6. Reasonably accurate simulations of different shading devices7. Possibility of adding new simulation modules developed by the user

e.g. a double skin façade module8. Good support9. Reasonably well spread among researchers and consultants in Swe-

den10. Known outside Sweden

The software candidates were:

• Bsim2000 developed by the Danish Research Institute (SBI)• IDA ICE 3.0 developed by EQUA (Stockholm, Sweden)• DEROB LTH developed by the University of Lund • BV2 available from CIT Management AB (Gothenburg, Sweden)

– Bsim2000 has most of the above features (except for 7 and 9)– IDA ICE has all the above features (was therefore chosen for the simu-

lation of the building alternatives)

Methods

125

– DEROB has some of the above features (except for 2, 4, 5, 7 and 9)– BV2 has some of the above features (except at least 3 and 7)

IDA ICE 3.0 is a computational program for indoor climate studies of individual zones within a building, as well as energy use of an entire buil-ding (EQUA, 2002). IDA Indoor Climate and Energy is an extension of the general IDA Simulation Environment. This means that the advanced user can, in principle, simulate any system whatsoever with the aid of the general functionality in the IDA environment.

Validation tests have shown the program to give reasonable results and to be applicable to detailed buildings physics and HVAC simulations (Acherman 2000 and 2003).

4.3.2.2 Description of double façade modelAfter personal communication with Dr. Bengt Hellström (Division of Energy and Building Design, Lund University) a brief description of the IDA ICE 3.0 double façade model is given below.

The window in IDA ICE 3.0 is divided into frame and glazing. The input data for the frame is the area fraction and the U value. For the glaz-ing, the input parameters are the U value, the solar transmittance (Tsol) and the solar heat gain coeffi cient (SHGC or g) at normal incidence. Also the emittances of the outermost and innermost surfaces of the glazing are given. The solar shadings are specifi ed by coeffi cients, which, when multiplied by U, T and g of the glazing, give the total values of the gla-zing/shading system.

The surface temperatures for the frame and the glazing are calculated from heat balance equations. Absorption of solar irradiation in the gla-zing is assumed to occur only at the innermost pane of the window and the absorbed energy rate is calculated from the difference between the g and the T values of the glazing (with or without solar shading) and the U value.

The double façade is modeled as an external window (with or without a shading device), outside an internal window and a wall. The cavity is assumed to be closed to the outside, except for four openings. One at the top, one at the bottom, whose areas can be chosen; a third opening connects the double façade cavity with the room and, fi nally, it is also possible to have mechanical exhaust ventilation of the cavity and its fl ow rate can be chosen.

The air temperature of the double façade cavity is obtained from an energy balance equation, using convective heat exchange with the surfaces and air exchange with the outside. Temperature stratifi cation is not taken


126

into account, as the air inside the cavity of the double façade has one temperature node.

The surface convection heat transfer coeffi cients are chosen as the maxi-mum of two values, one calculated from forced convection, depending on the air speed, and one calculated from natural convection, depending on the temperature difference to the surrounding air and the slope of the surface.

The natural convection driven air exchange in the cavity is calculated from the density difference between the air in the cavity and outside air, considering the pressure drops at the inlets and outlets.

4.3.2.3 Validation of IDA ICE 3.0 Double Façade model (IEA SHC Task 34/ECBCS Annex 43)

Concurrently with the “Glazed Offi ce Buildings” project, IEA SHC Task 34/ECBCS Annex 43 (Testing and Validation of Building Energy Simulation Tools) was started. The aim of the Task was to investigate the availability and accuracy of building energy analysis tools and engineering models to evaluate the performance of innovative low-energy buildings. The scope of the Task was limited to building energy simulation tools, including emerging modular type tools, and to widely used innovative low-energy design concepts. Activities include development of analytical, comparative and empirical methods for evaluating, diagnosing, and cor-recting errors in building energy simulation software.

The objective of Subtask E (Double-Façade Empirical Tests) was to assess the suitability and awareness of building energy analysis tools for predicting heat transfer, ventilation fl ow rates, cavity air and surface tem-peratures, solar protection effect, and interaction with building services systems in buildings with double skin façades.

The validation process was carried out in two steps. First comparative test cases (Kalyanova and Heiselberg, 2005) were simulated and the results were cross compared; then empirical cases (Kalyanova and Heiselberg, 2006) were carried out and the output of the different software were compared with the measurements of the test facility. The empirical tests were led by Aalborg University (AAU), Denmark, using a new facility being constructed at AAU. Detailed description of the test facility can be obtained by (Kalyanova and Heiselberg, 2005).

The double skin façade confi gurations considered for this validation procedure are set out below (Figure 4.8):

• DSF100. All façade openings closed• DSF200. Openings open to the outside

Methods

127

• DSF300. Openings open to the inside• DSF400. Bottom opening open to outside; top opening open to in-

side• DSF500. Top opening open to outside; bottom opening open to in-

side

Figure 4.8 Double skin façade confi gurations considered for the validation tests.

Within the test cases there are a number of variations to check the infl u-ence of various parameters, including:

• driving force of airfl ow (buoyancy, wind, mechanical fan, combined forces)

• internal (thermal)/External (thermal, solar, wind) boundary condi-tions

• opening area (fully opened, opening area controlled by temperature and/or airfl ow rate)

Some of the output used for the validation of the simulation tools are listed below:

• direct and diffuse solar irradiation on the window surface • solar radiation transmitted from the outside into the DSF cavity • solar radiation transmitted from the DSF cavity into the room • energy used for cooling/heating in the room• hour averaged surface temperature of external window surface facing

outdoors and the DSF cavity• hour averaged surface temperature of internal window surface facing

the room and the DSF cavity• hour averaged fl oor and ceiling surface temperature and air tem-

perature in the room

Simulation results for comparative and empirical tests were obtained by IDA ICE 3.0 (LTH-Lund, Sweden), BSim 2000(Aalborg University, Den-


128

mark), VA114 (VABI, the Netherlands), TRNSYS-TUD (TUD, Germany) and ESPr (ESRU, UK). Conclusions of the software validation will be araible after the complitation of the IEA Task 34/ECBS Annex 43.

Description of the building model

129

5 Description of the building model

A detailed description of the building model as designed and as inserted in IDA ICE 3.0 is provided in this Chapter. The single and double skin offi ce building alternatives designed for the “Glazed Offi ce Buildings” project consist of 30% (reference case), 60% and 100% glazing and are assumed to be located in Gothenburg (Sweden) i.e. for the simulations the weather data chosen was recorded in Gothenburg in 1977, which is considered to be a representative year. There are no adjacent buildings shading. The design of the reference building was determined by the project team, with researchers from the Division of Energy and Building Design (LTH), architects and engineers from WSP and Skanska. First, detailed performance specifi cations (see Appendix B) for indoor climate and energy use were established and then typical constructions were determined for a reference offi ce building representative of construction from the late nineties. System descriptions and drawings were prepared. The reference building was presented to a reference group and agreed upon.

5.1 Description of the reference buildingThe description of the reference building concerns the real (designed) building and the simulated model of the building (input for the IDA ICE 3.0 software) made for the energy and indoor climate simulations.

5.1.1. Geometry of the buildingThe reference building is a 6 storey building as shown in Figure 5.1. In terms of geometry and installations, the fl oors 1, 2, 3, 4, and 5 are com-pletely identical. However, the fl oors 1-4 are connected (fl oor, ceiling) with other internal zones of the building, while the roof of the 5th fl oor is


130

connected to the outside and the ground fl oor is connected to the ground (i.e. there is no basement).

connected to the outside and the ground fl oor is connected to the ground i.e. there is no basement.

Figure 5.1 View of the reference building.

The height of the building is 21m, the length 66 m and the width 15.4 m. Architectural drawings (fl oor plans, cross sections, facades) are presented in Appendix C. The room height is 2.7 m and the distance between in-termediate fl oors is 3.5 m. There is a suspended ceiling. The total fl oor area is 6177 m2 (BRA usable fl oor area i.e. fl oor area inside exterior walls) and 5448 m2 (LOA non-residential/premises fl oor area). The total area (on the inside including the window area and the area covered by interior walls and intermediate fl oors) of each of the long façades is 1386 m2 and that of the short façades 327.6 m2. Each opaque (wall) area is 957 m2 and 224 m2 respectively. The window area (including the frames) is 429 m2 (30.9% of the facade) and 104 m2 (31.6%) respectively (total window area = 31%). The roof area inside the exterior walls is 1030 m2.

5.1.2. Offi ce layouts Two different common fl oor layouts were designed, one with cell-type offi ces and one with open plan offi ces. In practice an offi ce building of-ten has a mixture of these two plan types. In order to simplify the input model and thus reduce the time of simulation, it is important to create as


131

few thermal zones as possible. On the other hand, the assumptions made should not infl uence the accuracy of the results or limit the output of the simulations. For the six-storey reference building, 3 different fl oor types are assumed, each one with several thermal zones (Figures 5.2 and 5.3). The zones were chosen to represent different kinds of rooms with differ-ent orientations. Adiabatic conditions were assumed for the ceiling of the ground fl oor, for the ceiling and fl oor of the 1st fl oor and for the fl oor of the 2nd fl oor. In this way it is assumed that below and above fl oors 1, 2, 3 and 4 there are identical zones. This is partly correct, since below the 1st and above the 4th fl oor the zones are not exactly the same. However, since the temperatures on all the fl oors are very similar, the infl uence of the connections plays a minor role. For the total energy use the simulated energy use of the 1st fl oor is multiplied by the factor 4.

Figure 5.2 Cell type offi ce building as modelled in IDA ICE 3.0 showing the simulated zones.

Figure 5.3 Open plan type offi ce building as modelled in IDA ICE 3.0 showing the simulated zones.


132

A detailed description of the geometry of the zones (including their number of repetition on each fl oor), the HVAC installations, the occupancy, the equipment, the artifi cial lighting and the furniture is given in Appendix D.

• Cell type offi ce layout

The fl oor area of the cell type offi ce building as defi ned in IDA ICE 3.0 (excluding the fan room) is 6177 m2. However, the non-residential space is 5448m2. In Figure 5.4 the area of each zone type is shown.

Figure 5.4 Zone areas for cell type offi ce building (p=persons).

As shown in Figure 5.4, 44% of the building area is corridor, 51% offi ce space and only 4% meeting rooms. Figures 5.5 and 5.6 show the fl oor plans for the cell type offi ce building (ground fl oor and 1st-5th fl oors).

Figure 5.5 Ground fl oor (cell type): Input for IDA ICE 3.0.

1 2 3 4

5

678

10 11

9


133

Figure 5.6 First - fi fth fl oors (cell type): Input for IDA ICE 3.0.

• Open plan offi ce

Figure 5.7 shows the area of each zone type.

Figure 5.7 Zone areas for open plan type offi ce building.

In the same way, 3 fl oors were also considered for the open plan reference building. The fi rst fl oor has 6 zones. There is airfl ow exchange between the zones 1, 4, 8 assuming a big opening (always open door) between them. The ground fl oor is shown in Figure 5.8 and fl oors 1 to 5 in Figure 5.9.

1 2 3 4

5

678

10 11

9


134

Figure 5.8 Ground fl oor (open plan type): Input for IDA ICE 3.0.

Figure 5.9 First - fi fth fl oors (open plan type): Input for IDA ICE 3.0.

5.1.3 Description of building elements

• Thermal transmittance of the building materials

A description is given below of the properties of the building elements used for the reference building (see Table 5.1).

1

2

4

5

8 9

1

2

4

5

8


135

Table 5.1 Description of the building elements used, disregarding thermal bridges.

Building Material type Thickness Thermal Density Specifi c U-value element (from inside (m) conductivity (kgm-3) heat (Wm-2K-1) to outside) (Wm-1K-1) (Jkg-1K-1)

Gypsum board 0.013 0.18 758 840 External Mineral wool 0.1068 0.036 16 754 wall Wood (studs) 0.011 0.14 500 2300 0.27 (long Gypsum board 0.009 0.18 758 840 façade) Air gap 0.04 0.25 1.2 1006 Facing bricks 0.12 0.58 1500 840

External Concrete 0.2 1.7 2300 880 wall Mineral wool 0.145 0.036 16 754 (short Air gap 0.04 0.25 1.2 1006 0.21 façade) Facing bricks 0.12 0.58 1500 840

Gypsum 0.026 0.22 970 1090 Internal Air gap 0.032 0.17 1.2 1006 wall Light insulation 0.03 0.036 20 750 0.62 Air gap 0.032 0.17 1.2 1006 Gypsum 0.026 0.22 970 1090

Linoleum Linoleum 0.0025 0.156 1200 1260 fl oor Concrete 0.3 1.7 2300 880 1.75 Acoustic tiles 0.012 0.057 720 837

Ground Linoleum 0.0025 0.156 1200 1260 fl oor Concrete 0.1 1.7 2300 880 0.32 Expanded plastic 0.1 0.035 1000 1700

Roof Acoustic tiles 0.0125 0.057 720 837 above Concrete 0.3 1.7 2300 880 6th fl oor Mineral wool 0.2 0.036 16 754 0.16 Wood 0.02 0.14 500 2300 Under felt 0.003 0.13 930 1300

The thermal transmittance of the materials used was initially calculated by IDA ICE 3.0. However, since the thermal losses due to thermal bridges were not included in these calculations, further calculations were carried out as described in Appendix E. A comparison between the theoretical values and the practical values calculated by Swedish Building Regulations is shown in Table 5.2. Finally, it was decided that the use of the practical values, according to the calculation procedure in the Building Regulations, was preferable for the simulation of the reference building. In order to meet the requirements of the Swedish Building Regulations (overall thermal transmittance of the building), modifi cations were made in the IDA ICE 3.0 library (increase of insulation for the external walls and roof ).


136

Table 5.2. Theoretical and applied thermal transmittance of the materials used.

Building element Theoretical U-value Applied U-value (Wm-2K-1) (Wm-2K-1)

External wall (long façade) 0.27 0.32External wall (short façade) 0.22 0.25Internal walls 0.62 0.62Roof (above 6th fl oor) 0.16 0.19Ground fl oor 0.32 0.32Intermediate fl oors 1.75 1.75

The total U · A value of the building envelope does not quite meet the requirements of the Swedish Building Regulations, but the energy use for heating meets the requirements of the Building Regulations reference building.

• Windows

A description of the geometry and the properties of the windows of the reference building follows.

o Windows of the long façade (type A)

sash frame

100mm 1.3 m

1 m

Figure 5.10 Typical window in the long façade.


137

Table 5.3 Properties of window in the long facade.

Window properties Size of window 1.3 m2

Uwindow typical of 90s 2 W/m2K

Glazing properties Description Triple glazed unit. Outer 4mm clear fl oat, 30mm space, D4-12 inner IGU (Insulated Glazing Unit).

Size (Ag) 0.88 m2

Ug (Calculated with Parasol) 1.85 W/m2K

Frame properties Description Wood covered by aluminium on the outside

Size (Ag) 0.42 m2 (32% of the total window area)

Uf 2.31 W/m2K

Shading device* Description Intermediate white venetian blind placed in the 30 mm gap (at 45 degrees)

Uglazing effective 1.65 W/m2K

* Shading device: it is assumed that the venetian blind closes (100%), when the in-cident light inside the glass exceeds 100W/m2. The Uglazing effective was calculated by the Parasol software.

o Windows in the short façade (type B)

sash frame

100mm 2.7 m

1.6 m

Figure 5.11 Typical window in the short façade.


138

Table 5.4 Properties of window in the short facade.

Window properties Size of window 4.32 m2

Uwindow typical of 90s 1.94 W/m2K

Glazing properties Description Triple glazed unit. Outer 4mm clear fl oat, 30mm space, D4-12 inner IGU (Insulated Glazing Unit).

Size (Ag) 3.5 m2

Ug (Calculated with Parasol) 1.85 W/m2K

Frame properties Description Wood covered by aluminium on the outside

Size (Ag) 0.82 m2 (19% of the total window area)

Uf 2.31 W/m2K

Shading device* Description Intermediate white venetian blind placed in the 30 mm gap (at 45 degrees)

Uglazing effective 1.65 W/m2K

* Shading device: it is assumed that the venetian blind closes (100%), when the in-cident light inside the glass exceeds 100W/m2. The Uglazing effective was calculated by the Parasol software.

5.1.4 Special modifi cations for the simulated modelIn order to input the real (virtual) building into IDA ICE 3.0 some special modifi cations had to be made.

• Offi ce volume

IDA ICE 3.0 calculates the thermal losses only for the interior of a room (inside part of the internal walls, upper part of the fl oor to lower part of the ceiling, etc). Thus, in order to include the transmission losses through the external wall above the suspended ceiling and the concrete fl oor (part b in Figure 5.12), the room height was increased from 2.7 m (real room height) to 3.5 m (Figure 5.13). The same assumption was made for the internal walls. Since the construction of the internal walls is light, in the same way the internal walls were included, when the offi ce geometry was defi ned.


139

Concrete (0.3 m) Air (0.5 m) Acoustic tiles Office room (2.7 m) External wall

a

b

Transmission losses calculated by IDA ICE 3.0 for 2.7 m floor height

Office room (2.7 m)

Figure 5.12. Real offi ce model.

Concrete (0.3 m) Acoustic tiles Office room (3.2 m) External wall

Transmission losses calculated by IDA ICE 3.0 for 3.5 m floor height

c Office room (3.2 m)

Figure 5.13. Equivalent offi ce model.

• Windows

Equivalent windows were also assumed, in order to save time for the simulations. For each façade of every thermal zone only one window is assumed. The size of the equivalent window is equal to the sum of the real ones, while the proportion of the frame area to the window area remains the same. The equivalent window is placed 0.8 m from the fl oor and 15 in the middle of each zone.

• Internal boundary conditions

On each fl oor it is assumed that each zone is connected with an identical one. This was achieved, since adiabatic conditions were assumed from


140

the computing program. The same assumption was made between the fl oors. This was achieved by keeping the distance between the zones at a minimum level of 0.5 m.

• Infi ltration rates

The infi ltration rate assumed for the reference building is 0.1 ach (air changes per hour) for the whole building. The easiest way to insert infi ltra-tion into IDA ICE 3.0 is to increase the mechanical supply and exhaust air by 0.1 ach, and reduce the heat recovery effi ciency.

The effi ciency calculations are described in the HVAC Subsection 5.1.8.

5.1.5 Control set points for indoor air temperatureThree control set points were chosen for the simulations of the reference building, as shown in Table 5.5. The normal control set point is considered the standard (reference) case, since the lower and upper temperature limits meet the requirements for indoor temperatures according to practice in modern Swedish offi ces (VVS, 2000). However, the two other control set points can provide useful information concerning variation in energy use as a function of the mean air and directed operative temperature and the perception of thermal comfort.

Another parameter changed with the three control set points is the ar-tifi cial light provided at the workplace. For the strict control it is assumed that the lights are switched on according to the occupants’ schedule, regard-less of the amount of daylight inside the offi ces. For the normal and poor control set points, however, set points of 500 lux and 300 lux respectively were assumed at the workplace. The main reason that these set points were assumed is to calculate the savings in electricity for artifi cial lighting for different control set points, glazing, shading devices and proportion of glass in the building.


141

Table 5.5 Classifi cation of the indoor environment.

Classifi cation Minimum Air Maximum Air Daylight at(control set points) Temperature ºC Temperature ºC workplace (winter) (summer) (lux)

Poor 21 26 Setpoints+Schedule 300-5000Normal 22 24.5 Setpoints+Schedule 500-5000Strict 22 23 Schedule

5.1.6 Occupancy

• Occupant density

o Cell type

For the cell type offi ce building the number of occupants is shown in Table 5.6.

Table 5.6 Occupants for cell type (p=persons).

Zone type Number of Total (theoretical) Total (real) number occupants for number of of occupants (during each zone occupants for working hours) each zone type

Corner offi ces 1 22 17.6Double offi ce rooms 2 166 132.8Single offi ce rooms 1 156 124.8Meeting rooms (6p) 6 66 24.6Meeting rooms (8p) 8 8 3.2Meeting rooms (12p) 12 36 14.4Storage room 0 0 0Corridor (1st fl oor) 0 0 0Corridor (2nd-6th fl oor) 0 0 0

Total 454 319.2

The total fl oor area of the building (inside the external walls) for the number of working places (only offi ce rooms) is 18 m2/occupant. Accord-ing to personal communication with the architect involved in the “Glazed Offi ce Buildings” project, Christer Blomqvist, it was assumed that 80%


142

of the occupants are present during offi ce hours. The distribution of the occupants in the building is shown in Figure 5.14.

Figure 5.14. Occupants (real) per zone for cell type.

The density of the occupants for the cell type offi ce building is shown in Figure 5.15.

5,26,3

5,4

12,6

9,6

18,9

0123456789

10111213141516171819202122

Corner officerooms

Double officerooms

Single officerooms

Meetingrooms (6p)

Meetingrooms (8p)

Meetingrooms (12p)

Zone type

m²/

real

occ

upan

t

Figure 5.15 Density of each zone for cell type (real number of occupants, p=persons).


143

o Open plan type

The total fl oor area of the building (inside the external walls) for the number of working places (only offi ce rooms) is 15.5 m2/occupant. For the open plan type it is assumed that there is a 20% increase in the oc-cupants for the offi ces (compared with the cell type), while the density of the meeting rooms remains the same. The number of occupants is shown in Table 5.7.

Table 5.7 Occupants for the open plan.

Zone type Number of Total (theoretical) Total (real) number occupants for number of of occupants (during each zone occupants for working hours) each zone type

Typical corner zones 16 211.2 169Reduced corner zone 12 14.4 11.5Intermediate zones 24 172.8 138.2Meeting rooms 8 192 76.8Storage rooms 0 0 0

Total 590.4 395.2

The distribution of the occupants in the building is shown in Figure 5.16.

Figure 5.16 Occupants (real) per zone for open plan.


144

The density of the occupants for the open plan type offi ce building is shown in Figure 5.17.

6,3

22,421,2

16,2

0123456789

101112131415161718192021222324

Typical corner zones Reduced corner zone Intermediate zones Meeting Rooms

Zone type

m²/

real

occ

upan

t

Figure 5.17 Density of each zone for open plan (real number of occupants).

The total real increase of the occupants in the open plan type is 15.5% (for the offi ces the increase is 20%), since the design and use of the two plan types (cell and open) are not the same (for example the cell type has 2 meeting rooms on fl oors 1-5 for 6 persons each, while the open plan has 4 meeting rooms for 8 persons each for the same fl oors). Thus, an equivalent number of occupants (interior of the open plan with density of the cell type) had to be considered in order to calculate the relative increase in the occupants in the open plan.

• Occupant load

The occupants’ schedule, activity level, clothing and use of the rooms are shown in Table 5.8.


145

Table 5.8 Occupant load.

Schedule Schedule for offi ces:

08:00 -12:00, 13:00 – 17:00

Typical winter occupant schedule (01/01-31/04, 01/10-31/12):

50% working during the Christmas vacations, otherwise 100% weekends closed.

Typical summer occupant schedule (01/5-30/09):50% working during July, 75% working during June and August, otherwise 100%, weekends closed.

Schedule for meeting rooms:

10:00 -12:00, 13:00 – 15:00

Typical winter occupant schedule (01/01-31/04, 01/10-31/12): 50% working during the Christmas vacations, 100% rest, weekends closed.

Typical summer occupant schedule (01/5-30/09):50% working during July, 75% working during June and August, 100% rest, weekends closed.

Activity level Offi ce activity1 met = 108 W / occupant (1 met corresponds to 58.2 W / m2 body surface)Task: sitting, reading

Clothing For winter conditions: 1 clo For summer conditions: 0.6 clo

Use of rooms For the offi ces it is assumed that 80% of the offi ce workers are present during the working hours (since people can work at home or be absent for some other reason). For the meeting rooms it is assumed that they are used 50% of the time and used to 80% of their capacity (total 40% comparing with the offi ces).

5.1.7 LightsFor the artifi cial lighting, energy effi cient lighting (fl uorescent tubes with HF fi ttings) was assumed. For the cell plan this means an installed power of 12 W/m2 for the offi ces and the meeting rooms and desired illuminance at the desk of 500 lux (according to personal communication with Dr. Helena Bülow-Hübe, Division of Energy and Building Design, Lund University). For the corridors and the rest of the spaces an installed power of 6 W/m2 and desired illuminance of 250 lux was assumed. However, for the open plan the installed power of 12 W/m2 was assumed for all the working space. The luminous effi cacy of the lights is set to 41.7 lm/W.


146

5.1.8 HVAC

5.1.8.1 Heating and coolingThe heating was provided by water radiators, while the cooling by chilled beams as described in Appendix D. For the water radiators proportional control and for the chilled beams PI control was assumed.

5.1.8.2 VentilationThe airfl ow rates were set after discussions with the HVAC engineer Lars Sjöberg.

• For the cell type: The air is supplied in the offi ces and extracted from the corridors. For the offi ces there is a CAV (constant air volume) control, supplying 10 l/s air (typical design value for cell type plan offi ces) for each person. For the meeting rooms VAV (variable air volume) CO2 control is assumed.

• For the open plan: In this case the air is supplied and exhausted from the offi ce space (since there is no separation between offi ces and corridors). The supply air for each person is assumed to be 7 l/s (normal design value for open type plan offi ces) for the offi ce space (CAV control) and a VAV CO2 control is assumed for the meeting rooms.

In order to keep the supply and exhaust air fl ow rates balanced in IDA ICE 3.0, it was assumed that the air is supplied in the rooms through the Air Handling Unit and passes to the corridor through transfer air devices (simulated as air leakage paths) to be exhausted. Thus, the amount of the exhaust air (from the corridor) is equal to the mechanical and natural ventilation of the offi ces. However, since the exact amount of air sup-plied in the meeting rooms is not known (due to the VAV CO2 control), in order to keep the balance between the total supply and exhaust air, it was assumed that the amount of air supplied is also exhausted from the meeting rooms.

The infi ltration and exfi ltration of the whole building was assumed to be 0.1 ach. This rate corresponds to 2.7 m room height. For the meeting rooms (VAV CO2 control), however, since the mechanical supplied air in the meeting rooms can not be increased manually (assuming that it is natural ventilation) the infi ltration of the meeting rooms was added to the infi ltration of the offi ces depending on the size of each offi ce. For the cell type building there is a similar problem with the corridor. Since there is only exhaust air, the infi ltration of the corridor was added to the offi ces.


147

It is important to consider an assumption that was made for the Air Handling Unit’s effi ciency; the easiest way to insert infi ltration in IDA ICE 3.0 is to increase the mechanical supply and exhaust air by 0.1 ach, and reduce the Heat Recovery Effi ciency. However, since the mechanical ven-tilation of the meeting rooms is not known, it is assumed that the airfl ow for each one depends on the schedule and the number of occupants.

The theoretical heat recovery effi ciency for the AHU is 60%. For the cell type, the effi ciency drops to 53.8%, due to the natural ventilation, as discussed in Section 5.14.

The supply air temperature of the AHU varies with the outdoor tempera-ture, which is also inserted in IDA ICE 3.0 as shown in Figure 5.18:

Figure 5.18 Set point for supply air temperature (IDA ICE 3.0 input).

More detailed description of the ventilation rates of each zone is given in Appendix F.

5.1.8.3 Equivalent heat recovery effi ciencyThe infi ltration rate decided for the reference building is 0.1 ach for the whole building. As already mentioned, the easiest way to insert infi ltra-tion in IDA ICE 3.0 is to increase the mechanical supply and exhaust air by 0.1ach, and reduce the heat recovery effi ciency. The theoretical value of the effi ciency was 60%, and the mechanical air fl ow (excluding the natural one) was:


148

o 3997 l/s for the cell plano 3307 l/s for the open plan

The total airfl ow (including natural airfl ow) was:

o 4459 l/s for the cell plano 3521 l/s for the open plan

Therefore, the practical value for the heat recovery effi ciency is 53.8% and 52.4% respectively.

The AHU is on from 5:30 till 22:00 during weekdays (100%) and weekends (50%). However, there is natural ventilation (infi ltration and exfi ltration) during the rest of the hours. Thus, the off value (natural ven-tilation during the hours that the AHU is not working) is set to 0.1038 for the cell type according to the following equation:

100

efficiency recovery heat100nventilatio total

nventilatio natural

value off−

=

Equation 5.1. Off value for the AHU .

Likewise, for the open plan the off value (natural ventilation during the hours that the AHU is not working) is set to 0.1314 (heat recovery effi -ciency for the off value is 0, since the off value for the heat exchanger is set equal to 0). The on value during the weekends is calculated as follows:

nventilatio total

nventilatio naturalnventilatio mechanical 50%value on

+=

Equation 5.2. On value for the AHU

The schedule of the AHU is presented in Table 5.9.


149

Table 5.9 Schedule of AHU unit

AHU Properties Cell type Open plan

Air Handling Week days: Weekends: Week days: Weekends:Unit schedule 06:00 – 20:00 08:00 – 17:00 06:00 – 20:00 08:00 – 17:00

On Value 1 0.552 1 0.5679

Off Value 0.1038 0.1314

5.1.8.4 Use of electricityRegarding the electricity for ventilation, specifi c fan power of 2.5 kW/m3s was assumed, which is representative value of a well performing building of the nineties.

5.1.9 Electrical equipmentIn this chapter, a brief description of the electrical equipment used is given. For the cell type offi ce building, the corner offi ces (1 occupant) are equipped with 1 PC (125 W), 1 printer (30 W) and 1 fax (30 W). The double and single offi ces are equipped only with PCs (2 and 1 respectively). No electrical equipment is assumed for the meeting rooms. Four copiers (500 W), 4 printers and 2 faxes are placed in each corridor for general use. The annual energy use of equipment for the cell type offi ce building is 22 kWh/m2.

For each fl oor of the open plan offi ce building it is assumed that there is 1 PC (30 W) per occupant, while the printers (8 units of 30 W and 4 units of 50 W), the faxes (8 units of 30 W) and the copiers (4 units of 500 W) are mainly used by everybody. No equipment was assumed for the meet-ing rooms. The annual energy use of equipment for the open plan offi ce building is 21 kWh/m2. The lower annual energy use for the equipment of the open plan alternatives is inconsistent with the higher density. As one would expect, the larger number of occupants would increase the need for equipment use; however, the common use of equipment and the more and bigger meeting rooms of the open plan decrease this need.

The schedule assumed for the use of the equipment is from 08:00-12:00 (80%), from 12:00-13:00 (15%) and from 13:00-17:00 (80%) for a typical workday. During the Christmas vacations and July 50% of the typical use was assumed and 75% during June and August. During the weekends no use of equipment was assumed. The number and schedule of the units were decided after personal comunication with WSP architect


150

Christer Blomqvist. The energy used by the offi ce equipment is suggested by Wilkins (2000).

For both (cell and open) plan types a server room was assumed with energy use of 175 kWh/occupant (Jensen, 2003). Thus, the annual energy use is 10 kWh/m2 for the cell and 13 kWh/m2 for the open plan type. The cooling of the server room is also included in the energy use calculations. Energy use of 87.5 kWh/occupant was assumed for both plan types. The energy use is 5 kWh/m2 for the cell and 6kWh/m2 for the open plan type. The server rooms were not modelled by IDA ICE 3.0 (unlike, the energy use for the fans and pumps).

5.2 Description of single skin glazed alternatives

5.2.1 Description of 60% glazed buildingThe 60% glazed building is identical with the reference one, with the only difference that the glazed area increases from 30% to 60% as shown in Figure 5.19. A detailed description of the façade construction follows.

Figure 5.19 60% glazed alternative.

5.2.1.1 Façade construction

• Façade with internal or intermediate shading devices


151

A view of the façade of the 60% glazed building (with intermediate or internal shading devices) is shown in Figure 5.20. The 60% window area refers to a fl oor height of 3.5 m (exterior façade). The windows can be opened for natural ventilation of the building.

The total window area of a single offi ce is 5.04 m2 out of 8.4 m2 wall area (60%). The glass area is 3.6 m2 (72.5% of the window area) and the frame area 1.4 m2 (27.5% of the window area; see Figure 5.20).

1880mm

650mm

50mm

60mm

60mm

50mm

450mm

580mm

1130mm

openable

openable openable

1200mm

2400mm

1150mm

300mm

Figure 5.20 60% glazed façade construction (with intermediate or internal shading devices).

• Façade with external horizontal louvres

The main difference of the façade for the alternative with fi xed horizontal louvres is that 2 of the 3 windows are not openable, as shown in Figure 5.21. This results in a higher glass to window area ratio. Additionally, the Uf of the frame for the openable window is 1.8 W/m2K instead of 1.6 W/m2K. The total window area of a single offi ce remains 5.04 m2 out of 8.4 m2 wall area (60%). The glass area is 4.3 m2 (84.7% of the window area) and the frame area 0.7 m2 (15.3% of the window area). The open-able frame is 0.27 m2 (38.5% of the frame).


152

2000 mm

650 mm

50 mm

50 mm

450 mm

700 mm

1130 mm

openable

1200 mm

2400 mm

1150 mm

300 mm

Figure 5.21 60% glazed facade with external fi xed horizontal louvres.

5.2.1.2 Window propertiesThe window properties for the 60% glazed alternatives are shown in Tables 5.10 and 5.11. The rationale for the choice of alternatives is given in Subsection 4.1.2. For the given Uglazing and Uframe the Uwindow was calculated according to the following equation:

frameglazing

frameframeglazingglazingwindow AA

AUAUU

+

×+×=

Equation 5.3 Thermal transmittance of the window.

Uframe includes the edge losses of the glass. For the 7th alternative there are two frames used in the façade as shown in Figure 5.21. The U-value was calculated as a weighted average. The window properties (glazing and frame) for the 60% glazed alternatives are shown in Table 5.10.


153

Table 5.10 Window properties (glazing & frame).

Building Uwindow Uglazing Uframe Aglazing Aframealternative (W/m2K) (W/m2K) (W/m2K) (m2) (m2)

1 1.97 1.85 2.31 3.6 1.4 2 1.27 1.14 1.6 3.6 1.4 3 1.27 1.14 1.6 3.6 1.4 4 1.27 1.14 1.6 3.6 1.4 5 1.27 1.14 1.6 3.6 1.4 6 1.27 1.14 1.6 3.6 1.4 7 1.26 1.14 1.6 1.8 3.82 0.89 0.27 1.65 1.16

Table 5.11 sets out the impact of the shading devices for different glazing alternatives. The “effective” value refers to the cases in which the shading devices are used. The default value (set point) of IDA ICE 3.0 was used for the simulations. This value corresponds to a maximum limit of 100 W/m2 on the inside of the glass. Above this value the shading devices are used 100% (45° slat angle).

The Uglazing, g, Tsol, Ugl. effective, ggl.effective and Tgl,effective (effective solar transmittance) are calculated for perpendicular ray angle “standard-ized” with the “Parasol” software (Wall and Kvist 2003), while Tvis is taken from Pilkington Catalogue (2004). When the glazing properties are inserted in IDA ICE 3.0, however, the Tvis is not needed, since the software calculates the daylight availability taking into account the direct solar transmittance and not the visual one.

For the case with external louvres the solar and direct solar transmit-tance used are monthly average values calculated by Parasol software. The reason for this additional calculation is that since the horizontal external louvres are fi xed (and thus operating all year round) monthly solar factor values can better approximate the reality compared with the yearly g value that IDA ICE 3.0 uses for the calculations. Using the mentioned comput-ing program the gsystem and the Tsystem for every month were calculated as shown in Table 5.12. More information about the glazing units (e.g. daylight transmittance, commercial names of the panes, etc) is given in Table 4.2. These monthly values were inserted in IDA ICE 3.0 (schedule of external shading devices).


154

Table 5.11 Impact of shading on glazing alternatives.

Building Uglazing g Tsol Ugl. effective ggl.effective Tgl,effectivealternative (W/m2K) (W/m2K)

1 1.85 0.69 0.58 1.65 0.30 0.13 2 1.14 0.59 0.44 1.08 0.23 0.10 3 1.14 0.35 0.30 1.07 0.28 0.08 4 1.11 0.23 0.22 1.04 0.22 0.06 5 1.14 0.58 0.44 1.08 0.47 0.12 6 1.14 0.35 0.30 0.92 0.19 0.08 7 1.14 0.35 0.30 1.14 0.20 0.17

Table 5.12 Solar factor and solar transmittance for the fi xed horizontal external louvres.

Month g-mean g-mean g-mean Tmean Tmean Tmean sunshade window system sunshade window system (%) (%) (%) (%) (%) (%)

1 84.6 34.4 29.1 84.4 29.4 24.8 2 74.4 34.1 25.4 74.0 29.4 21.7 3 58.5 33.9 19.8 57.9 29.2 16.9 4 44.2 33.6 14.9 43.5 29.1 12.7 5 45.2 33.1 15.0 44.7 29.0 13.0 6 49.0 33.0 16.2 48.7 28.9 14.1 7 51.1 32.9 16.8 51.0 28.9 14.7 8 40.5 33.2 13.4 40.2 29.1 11.7 9 48.9 33.4 16.3 48.5 29.1 14.1 10 68.3 33.9 23.1 68.1 29.3 19.9 11 80.5 34.4 27.7 80.2 29.3 23.5 12 87.1 34.5 30.0 86.8 29.4 25.5

A detailed description of the frame construction is given in Appendix G.


155

5.2.2 Description of 100% glazed buildingA view of the 100% glazed building is shown in Figure 5.22.

Figure 5.22 100% glazed alternative.

The U values and g values for the windows are the same as for the 60% alternative, since the same ratio between glazing and frame area was as-sumed. The façade construction of the 100% glazed building with internal or intermediate shading devices is shown in Figure 5.23.


156

1880 mm

600 mm

50 mm

60 mm

60 mm

50 mm

450 mm

580 mm

1130 mm

openable

openable openable

1200 mm

2400 mm

1150 mm

300 mm

50 mm

Figure 5.23 100% glazed façade construction (with intermediate or internal shading devices).

The façade construction of the 100% glazed building with fi xed external horizontal louvres is shown in Figure 5.24.


157

2000 mm

600 mm

50 mm

50 mm

450 mm

700 mm

1130 mm

openable

1200 mm

2400 mm

1150 mm

300 mm

50 mm

Figure 5.24 100% glazed facade with external fi xed horizontal louvres.

5.3 Description of double skin glazed alternatives

5.3.1 WIS 3.0 simulations

5.3.1.1 Geometry of the “standard” double façade boxTwo double skin façade construction types were assumed: a multi storey type and a box window type. Since the software used for the current simula-tions (WIS 3) is a two dimensional computing tool the cavity width does not infl uence the output of the simulations. For all the cases, however, the width was assumed to be 3.5 m. The cavity height in the case of a multi storey high façade was assumed to be 10 m, 20 m, and 30 m, while for the box window it was assumed 3.5 m. The depth varies from 0.2 m up to 1.6 m. No frame was considered for the WIS 3 simulations.

5.3.1.2 Geometry of the airfl ow windowFor the airfl ow window cases a box window façade was assumed with a total height and width of 3.5 m. The mechanically ventilated cavity is


158

0.3 m deep. In the airfl ow window cases air is supplied (11.4 l/sec) to the room and exhausted through the cavity to the heat exchanger for heat recovery purposes.

5.3.1.3 Description of the openings

The openings for the multi storey high double façade are assumed to be horizontal at the top and the bottom of the cavity. At the top of the façade dampers are applied (as shown in Figure 5.25), in order to control the openable area all year round. Since the dampers occupy some area, when the cavity is fully opened the actual ventilated area (for even cavity depths) is the 87% of the area if no dampers were applied. When the cavity depth is uneven (e.g. 0.3 m, 0.5 m, etc) the actual opening area is the same as if the cavity depth was 0.1 m smaller.

Figure 5.25 Dampers for the multi storey high façade.

For the box window the height and width of the cavity were assumed to be 3.5 m. The cavity depth varied (depending on the case) from 0.2 m up to 1.6 m. The openings of the façade in this case are vertical. However, since WIS 3 can not handle vertical openings, equivalent horizontal openings were assumed.


159

5.3.2 IDA ICE 3.0 Input (zone level)

5.3.2.1 Offi ce description (IDA ICE 3.0 - zone level)In order to carry out the parametric study on a zone level, a typical single offi ce room was considered. The offi ce was considered to be on the second fl oor of the building. The occupancy, the equipment and the other internal loads (with their schedules) are described in detail in Appendix D. All the installations were considered to be the same as in the reference building.

5.3.2.2 Geometry of the box window

A box window façade was assumed, covering both the window and wall area of total height of 3.5 m and width of 2.4 m. The depth of the cavity was assumed to be 0.8 m for the “standard” double façade and 0.3 m for the airfl ow window case.

At the upper and lower parts of the cavity, dampers were assumed (same as those described in the WIS 3 input). When the cavity is fully open the actual opening area with the dampers is the 87%. When the cavity is closed a very small opening of 0.01 m2 was assumed, since the cavity is not completely sealed. A discharge coeffi cient of 0.65 for the top and 0.55 for the bottom opening was assumed (Kalyanova et. al, 2006).Thus, the equivalent leakage area (ELA) is calculated from:

area freeCAELA d ××=

where

A: is the area of the opening, with the dampers considered (m2)Cd: is the discharge coeffi cientfree area: actual opening area (with the dampers considered)

The Equivalent Leakage Area (ELA) for each zone is given in Table 5.13, while more detailed cavity characteristics are presented in Appendix J.


160

Table 5.13 Equivalent Leakage Area (ELA) for double façade opennings of different zones.

Zone type ELA (m2) – bottom ELA (m2) – upper leak leak

Corner offi ces (short façade) 1.84 2.18Corner offi ces (long façade) 1.58 1.87Double offi ce rooms (long façade) 1.84 2.18Double offi ce rooms (short façade) 1.58 1.87Single offi ce rooms 1.06 1.25Corridors 0.70 0.83

5.3.2.3 Geometry of the multi storey high façadeA multi storey high façade (0.8m deep) was assumed covering the whole building façade. In order to simulate correctly the air temperature strati-fi cation along the 24 m of the cavity, fi ve box window cavities (fl oors 1, 2, 3, 4 and 5) were linked (the ground fl oor was assumed with a single skin façade). Since the airfl ow model in IDA ICE 3.0 is a one node one, the cavity air temperatures calculated are the average air temperature of each level.

The dampers were placed at the top opening of the façade. Grills inside the cavity and between the different fl oors (used for maintenance purposes) were also assumed. The actual top opening (inserted in IDA ICE 3.0) was calculated as before; 87% for fully open and 0.01 m2 for fully closed cav-ity. The equivalent leakage area (ELA) for the upper and lower openings is calculated as before and presented i detail in Appendix J.

5.3.2.4 Properties of the inner and outer skinIn order to insert the properties of the glazing in IDA ICE 3.0 the ther-mal transmittance and the total and direct solar transmittance had to be calculated. These calculations were carried out in WIS 3 for normal angle of incidence. The inner and outer skin properties are presented in Tables 5.13 and 5.14 and described further in section 4.1.3.


161

Table 5.14 Inner and outer skin properties for the “standard” double façade alternatives.

DSF Case U value g value Tsol Outer skin Inner skin Outer skin Inner skin Outer skin Inner skin

A 5.79 2.90 0.789 0.746 0.742 0.685D 5.79 2.74 0.532 0.746 0.389 0.685E 5.79 1.46 0.532 0.655 0.389 0.565F 5.79 1.31 0.789 0.215 0.742 0.179

Table 5.15 Inner and outer skin properties for the airfl ow window alterna-tives

AW Case U value g value Tsol Outer skin Inner skin Outer skin Inner skin Outer skin Inner skin

A 2.87 5.92 0.688 0.846 0.624 0.82D 2.71 5.92 0.419 0.846 0.338 0.82E 1.45 5.92 0.369 0.846 0.299 0.82F 1.30 5.92 0.213 0.846 0.169 0.82G 1.31 5.66 0.214 0.738 0.176 0.677

The properties of the panes used are presented in more detail in Appendix H. The properties of the “standard” double façade and airfl ow window modes (when no ventilation occurs in the cavity) are described in Appendix I. The frame to window ratio was assumed to be 0.28 for the inner and 0.18 for the outer skin. Uframe was assumed to be 1.6 W/m2K.

5.3.2.5 Shading devicesTwo types of shading device were considered (white and blue venetian blind with a slat angle of 45°), in order to study their impact on the energy use and thermal comfort; their properties are described further in Appendix K. The thermal and solar transmittance (g and Tsol) have been calculated with the Parasol software. Since the shading is used mostly during summer, an average multiplier value (for the U, g and Tsol values) was assumed. This average includes a fi ve month period, from May to September. The multiplier is inserted in IDA ICE 3.0 as the inner shading coeffi cient of the outer skin (outer pane for the “typical” double façade cases and intermediate one for the airfl ow window mode). Eight orienta-tions were taken into account for the calculations: north, northwest, west,


162

southwest south, southeast, east and northeast as shown in Figures 5.26 and 5.27. The multipliers of the total and direct solar transmittance are also presented in Appendix K.

0.58

0.6

0.62

0.64

0.66

0.68

0.7

0.72

0.74

0.76

0.78

0.8

0.82

0.84

0.86

0.88

0.9

north northeast east southeast south southwest west northwest

DF A,F White blinds DF A,F Blue blinds DF D,E White blinds DF D,E Blue blindsAW A White blinds AW A Blue blinds AW D White blinds AW D Blue blindsAW E White blinds AW E Blue blinds AW F,G White blinds AW F,G Blue blinds

Figure 5.26 Multipliers for the gshading as used in the IDA ICE 3.0 simula-tions.

The multipliers for direct solar transmittance are almost the same when dark (blue) shading devices are applied. For white venetian blinds, however, the multipliers follow the same trend as before: the shading effect is higher for outer panes with larger Tsol.


163

0.08

0.1

0.12

0.14

0.16

0.18

0.2

0.22

0.24

0.26

0.28

0.3

north northeast east southeast south southwest west northwest

DF A,F White blinds DF A,F Blue blinds DF D,E White blinds DF D,E Blue blindsAW A White blinds AW A Blue blinds AW D White blinds AW D Blue blindsAW E White blinds AW E Blue blinds AW F,G White blinds AW F,G Blue blinds

Figure 5.27 Multipliers for the Tsol shading.

5.4 Assumptions made during the calculations

Condensation issues were not considered during this study. For the double façade mode the double pane (thermal barrier) is placed as inner skin. In reality, during winter, when the openings are closed the increased air temperature inside the cavity decrease the risk of condensation, while in the mechanically ventilated cases the cold outdoor air could be an issue.

For naturally ventilated cavities the stack effect was considered to be the main driving force all year round. During summer, the main driving force is the thermal buoyancy, while during winter is the wind effect (Saelens 2002). However the naturally ventilated double facades are most often closed during the winter, in order to increase the air temperature inside the cavity providing higher surface temperatures of the inner skin. Moreover, simulations carried out during the IEA Task 34 (Testing and Validation of Building Energy Simulation Tools) show that the impact of wind effect does not infl uence much the energy use for heating and cooling the zone behind the double façade or the surface temperatures of the inner layer.

Another assumption made, was that no heat absorption by the grills (in between the fl oors of a multi storey high facade) was considered. In the cases in which shading devices are applied this mistake is decreased since


164

the grills are shaded (mostly when the shading devices are placed close to the external pane). In the cases, however, when no shading devices applied (mostly during winter), the absorption by the grills result in an increase of the air temperature inside the cavity.

Results and discussion

165

6 Results and discussion

In this chapter the results of the simulations are presented and discussed in three parts:

• reference building• single skin glazed alternatives• double skin glazed alternatives

In the fi rst Section (results from the reference building alternatives) the parameters varied are the building orientation, the control set points and the plan type. The output of the simulations concerns the energy use, the mean air temperatures, the directed operative temperatures and the thermal comfort indices (PMV and PPD). The results are studied and discussed on both a building and a zone level.

In the second Section the glazing and shading device type are varied for building alternatives of 60% and 100% window to external wall area ratios. The simulations were carried out for both open plan and cell type offi ce buildings for different control set points. The studied parameters were the same as in the reference building and the simulations were carried out on both a building and a zone level.

The third Section is divided into three parts: (a) simulations of double skin façades on a component level for steady state boundary conditions, (b) all year round simulations on a zone level and (c) all year round simulations on a building level with integrated double skin façades. The main aim of the fi rst part (pilot study) was to study the possibilities and limitations of different double skin façade constructions, while reducing the amount of alternatives to be further simulated. The second part focuses on the proper integration of double skin façades regarding energy use and thermal comfort issues. Several alternatives were simulated varying the façade mode (naturally, mechanically and hybrid ventilated double façades and airfl ow windows) and the glazing and shading device type used. In the third part double skin façade alternatives were selectively simulated, in order to provide output values for purposes of comparison (especially with the single skin façade glazed building alternatives).


166

6.1 Reference building

6.1.1 Energy use As stated in Subsection 4.1.1, 18 building alternatives were generated with 30% window to external wall area ratio (2 plan types, 3 orientations and 3 control set points). The energy use for each alternative is presented in Appendix L, while the results are discussed below.

6.1.1.1 Impact of fl oor plan typeFor the reference building (identical long and short façades), the build-ing’s orientation does not really infl uence its energy use, neither for the cell nor the open plan type (as described in Subsection 6.1.1.2). Thus, in order to decrease the number of comparisons, an average energy use was assumed for the 3 building orientations. Due to the higher internal loads of the open plan, the energy use for heating is slightly lower than for the cell type (see Figure 6.1); the open plan has a higher occupant density and different set points and power for lights used (6 W/m2 for the corridors of the cell type and 12 W/m2 for all the open plan space), as described in Subsection 5.1. For the same reason, the cooling demand of the build-ing and the server rooms is higher for the open plan (the energy demand for the operation and cooling of the server rooms is proportional to the number of occupants (Jensen, 2003)). As described in Subsection 5.1.7, for the normal set points (cell type) the lights were switched on, in order to ensure a minimum level of 500 lux at the workplace, while for the open plan the lights were always switched on (since IDA ICE 3.0 can not handle any set point for artifi cial lighting in the non rectangular zones of the open plan type). Additionally, power of 12 W/m2 was assumed for the working spaces (offi ces and meeting rooms for the cell type and the whole space of the open plan), while power of 6 W/m2 was assumed for the corridor of the cell type. As a result, the energy use for lighting the open plan is 5 kWh/m2a higher; most of it due to the different installed power, since the impact of lighting control set points is limited at least for the reference (30% window to external wall area ratio) case. Although the occupancy level of the open plan is higher and one could assume that the energy use for the equipment would be higher, the larger number of meeting rooms of the open plan reduces the number of printers and faxes (see Subsection 5.1.9), bringing the internal loads (equipment) of the two layouts to almost the same level.


167

123

5108

22

1411

52

127

613

6

211917

45

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

Spaceheating

Cooling Lighting Equipment Pumps,fans

Serverrooms

Coolingserverrooms

Total

Ene

rgy

use

(kW

h/m

²a)

SS-30%-Cell-average-normal SS-30%-Open-average-normal

Figure 6.1 Impact of plan type on the energy use of the reference building (normal control set points).

In Figure 6.2, the difference in energy use for heating, cooling and lighting is compared between the two plan types for the three control set points. The energy use for heating the cell type offi ce building (with strict set points) is 5.6 kWh/m2a (10%) higher than for the open plan one. The increase in energy use for the normal and poor control is 7.4 kWh/m2a (14%) and 8.9 kWh/m2a (19%) respectively. The positive value in Figure 6.2 shows higher energy use for the cell plan. As expected, in all the cases the energy use for heating is higher for the cell type due to the lower internal gains. On the contrary, the cooling demand of the open plan type is much higher (42%, 56%, and 48%). It can be concluded that the higher internal loads of the open plan result in more “useful” heat that can be stored in the cases with wider temperature variation (such as in poor set points); for the same reason the effect is the opposite for the cooling demand.

At this point the difference in the ventilation strategy of the two plan types should be noted. As described in Subsection 5.1.8, the air supplied to the offi ces (cell type) passes through the doors (or leaks) to the corridor from where it is extracted, while for the open plan type the air is supplied and exhausted from the same zone. Since in the cell type offi ce building there are no cooling beams installed in the corridor, the air temperature in the corridor at times rises above the upper control set point limit. This may cause an increase in the air temperature in the corridors, but since there are no occupants (used as sensors) placed in that zone, no discomfort is considered. On the other hand, for the open plan type building, the


168

whole fl oor area is considered as working area and thus cooled according to the upper set point temperature limit.

The corridor of the cell type plan is almost 44% of the total fl oor area and the rated input per unit (light) is 6 W/m2 instead of 12 W/m2 (open plan and working spaces of cell plan). For the strict set points the lights are assumed switched on during the working hours for both the cell and open plan types. However, the increased demand for proper lighting of the working area results in a 32% increase in energy use. For the normal and poor set points this difference increases, up to 35% and 36% respectively, due to the light control set points (500 lux and 300 lux) applied for the offi ces and meeting rooms of the cell plan.

Finally, there is a 24% higher energy demand for the operation and cooling of the server rooms (regardless of the set point) for the open plan, due to the increased number of occupants (the assumed energy use is 175 kWh/am2 per occupant for operating the server rooms and 87.5 kWh/am2 per occupant for cooling the server rooms (Jensen, 2003)).

The higher internal gains of the open plan type result in 6% higher total energy use for the strict set point and in 3% increase for the normal one. For the poor set point the cell plan uses slightly less energy as shown in Figure 6.2. As expected, the stricter the set points, the higher the impact of the plan type on the total energy use.

5,6

-4,6

-7,6

7,4

-6,2

-5,0

-4,0-3,1

-5,1

-8,3

0,5

8,9

-10

-9

-8

-7

-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

7

8

9

10

Energy use for heating Energy use for cooling Energy use for lighting Total energy use

Ene

rgy

diff

eren

ce b

etw

een

cell

and

open

pla

n (k

Wh/

m²a

)

Strict control set points Normal control set points Poor control set points

Figure 6.2 Impact of plan type on the energy use of the reference building.


169

6.1.1.2 Impact of orientationAs pointed out above, the orientation has a very small (negligible) impact on the energy use for the reference building. The main reason is that the two long and two short facades are identical (including the shading devices used). If the façades were different, there would of course have been an impact of orientation.

6.1.1.3 Impact of control set pointsFor the cell type offi ce building the energy use for heating decreases by 7% for the normal and by 16% for the poor temperature set points (compared with the strict one) as shown in Figure 6.3. For the open plan type the heating demand decreases by 11% and 24%. Although the lower temperature limit is the same for the strict and normal control set points (22ºC), the strict set points reduce the capacity for storing heat (thermal mass), increasing the heating demand.

For the normal control set points the maximum permissible air tem-perature is 24.5°C, while for the strict and poor set points it is 23°C and 26°C respectively. The cooling demand for the normal (compared with the strict) set point is 45% lower for the cell plan and 40% lower for the open plan type. The decrease for the poor type is 65% and 64% respectively.

7

45

2

10

16

65

3

17

11

40

0

12

24

64

0

21

0

5

10

15

20

25

30

35

40

45

50

55

60

65

70

Space heating Cooling Lighting Total

Ene

rgy

diff

eren

ce (

%)

Strict - normal (cell plan) Strict - poor (cell plan)Strict - normal (open plan) Strict - poor (open plan)

Figure 6.3 Impact of control set points on the energy use of the reference build-ing.


170

The minimum limit of 500 and 300 lux applied for the normal and poor set points results in an almost negligible decrease in energy use for light-ing, 2% and 3% for the two control set points for the cell type plan. The impact of different set points on the energy use for lighting is calculated for the whole cell type building, although the set points are applied only to the working space. Thus, out of 14.7 kWh/am2 (total energy use for the strict control) 10.5 kWh/am2 (72%) are used in the working areas and 4.2 kWh/m2a (28%) in the common spaces. For the normal set points the energy decrease (0.3 kWh/am2) refers only to the working areas, so the energy use is reduced for this control to 10.2 kWh/am2, and to 10 kWh/am2 for the poor set points. For the open plan type, the lights were assumed switched on during the working hours for all the three types of control set points, thus the energy use remains the same. It can be con-cluded that the three different lighting strategies (for the reference building alternatives) result in small differences in energy use for lighting.

Finally, as can be expected, the total energy use is lower for alternatives with less strict temperature control set points. The total energy use of the cell type reference building decreases by 10% for the normal and by 17% for the poor set point, compared with the strict control set points. For the open plan type, the decrease is 12% and 21% respectively.

6.1.2 Indoor climate on a building levelThe mean air temperatures and thermal comfort indices are studied, in order to evaluate the indoor climate of the different building alternatives. In general, monthly average values, as presented below, are good indica-tors of the indoor climate building performance but do not provide any information about the variation over the year in individual zones. Thus, in this section monthly average air temperatures and thermal comfort indices were studied on a building level, while more detailed discussion focusing on specifi c zones is carried out in Subsection 6.1.3.

6.1.2.1 Weighted average mean air temperaturesAs can be expected, control set points have a larger infl uence on average mean air temperatures of the working area of the reference building, compared with the orientation and plan type. As shown in Figure 6.4, the mean air temperature difference between the cell and open plan type of the reference building is very small for strict set points and it increases as the permissible air temperature variation (normal and poor set points)


171

increases. As expected, the open plan type is warmer due to the higher internal loads.

22,0

22,5

23,0

23,5

24,0

24,5

25,0

25,5

26,0

26,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e m

ean

air

tem

pera

ture

(°C

)

SS-30%-Cell-average-strict SS-30%-Open-average-strict SS-30%-Cell-average-normal

SS-30%-Open-average-normal SS-30%-Cell-average-poor SS-30%-Open-average-poor

Figure 6.4 Weighted average mean air temperatures for the working area of the reference building.

For the normal control set point the maximum permissible air temperature (24.5°C) results in higher air temperature differences during February (0.6 °C) while the poor set point (26°C) results in higher differences dur-ing April (1.4°C).

The impact of orientation on the (weighted average) mean air tem-peratures of the building is almost negligible (see Figure 6.5), due to the identical short and long façades and the moderate window areas; further studies, however, are carried out, in order to study the impact of orienta-tion on specifi c zones.


172

22,0

22,5

23,0

23,5

24,0

24,5

25,0

25,5

26,0

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e m

ean

air

tem

pera

ture

(°C

)

SS-30%-Cell-NS-poor SS-30%-Cell-NS45-poor SS-30%-Cell-EW-poor

Figure 6.5 Impact of the orientation on the weighted average temperature (cell type, poor control set points).

Although the diagrams above are a good indicator of the average tem-perature variation during the year, they do not provide any information concerning the number of hours with a certain air temperature. Therefore, the number of hours between certain (weighted) average mean air tem-peratures for the working space of the reference building (for the cell and open plan and for the three control set points) is presented below.

The open plan offi ce building (normal control set point) is warmer than the cell type one as shown in Figure 6.6. The number of hours close to the upper permissible temperature limit is as high as 45% for the cell and 70% for the open plan type, increasing the risk of potential overheating.


173

050

100150200250300350400450500550600650700750800850900

21.6

- 2

1.7

21.8

- 2

1.9

22.0

- 2

2.1

22.2

- 2

2.3

22.4

- 2

2.5

22.6

- 2

2.7

22.8

- 2

2.9

23.0

- 2

3.1

23.2

- 2

3.3

23.4

- 2

3.5

23.6

- 2

3.7

23.8

- 2

3.9

24.0

- 2

4.1

24.2

- 2

4.3

24.4

- 2

4.5

24.6

- 2

4.7

24.8

- 2

4.9

Weighted average mean air temperature (°C)

Num

ber

of w

orki

ng h

ours


Figure 6.6 Number of working hours in a year with certain average weighted air temperatures for the reference building (normal set points).

The number of hours between certain air temperatures for the three control set points of the cell type offi ce plan is shown in the Figure 6.7. As can be noticed the temperature variation increases for poorer set points. For the poor set points only 18% of the working hours are close to the upper limit for the cell and 44% for the open plan type.

050

100150200250300350400450500550600650700750800850900950

21.1

- 2

1.2

21.3

- 2

1.4

21.5

- 2

1.6

21.7

- 2

1.8

21.9

- 2

2.0

22.1

- 2

2.2

22.3

- 2

2.4

22.5

- 2

2.6

22.7

- 2

2.8

22.9

- 2

3.0

23.1

- 2

3.2

23.3

- 2

3.4

23.5

- 2

3.6

23.7

- 2

3.8

23.9

- 2

4.0

24.1

- 2

4.2

24.3

- 2

4.4

24.5

- 2

4.6

24.7

- 2

4.8

24.9

- 2

5.0

25.1

- 2

5.2

25.3

- 2

5.5

25.5

- 2

5.6

25.7

- 2

5.8

25.9

- 2

6.0

26.1

- 2

6.2

Weighted average mean air temperature (°C)

Num

ber

of w

orki

ng h

ours

SS-30%-Cell-average-strict SS-30%-Cell-average-normal SS-30%-Cell-average-poor

Figure 6.7 Number of working hours in a year with a certain average weighted air temperatures for the reference building (normal set points, cell plan).


174

6.1.2.2 Perception of thermal comfortThe PMV and PPD values were calculated differently for the cell and open plan type offi ce buildings. Due to the non rectangular geometry of the zones (of the open plan type), the simulations were simplifi ed (energy model of IDA ICE 3.0). Therefore, an average radiant temperature of the external walls was assumed, diminishing the importance of occupants’ posi-tion, when the operative temperature is calculated. This can be considered an acceptable assumption, since for the open plan type the distance of the occupants the surrounding walls is larger. For the simulations of the cell type building the position of the occupants is important, since the climate model used (in IDA ICE 3.0) takes into account the radiant temperature of each surface (glazed area and opaque wall) of the façade.

• Impact of offi ce layout on the perception of thermal comfort

For the strict control set points the weighted average PMV varies form -0.58 to -0.05 for the cell and from -0.51 to 0 for the open plan type as shown in Figure 6.8. As expected, due to the higher internal loads the open plan is slightly warmer than the cell type. However, the average difference in the weighted average PMV values for the two plan types is very small (less than 0.1) throughout the year.

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e PM

V

SS-30%-Cell-average-strict SS-30%-Open-average-strict

Figure 6.8 Weighted average PMV for strict set points of the reference build-ing.


175

The number of working hours for certain average PPD values for strict set points of the reference building is shown in Figure 6.9. According to ISO Standard 7730 (1984) PPD values of 10% (or less) for 90% of the (working) time can ensure a very good thermal environment. For the cell type an average PPD of 10% or less corresponds to 66% of the working hours, while for the open plan the same PPD corresponds to 78%. For the cell type a maximum average PPD of 15% corresponds to 93% and for the open plan to 95% of the working hours. PMV values closer to zero (open plan type) mean lower PPD values and improved thermal conditions.

050

100150200250300350400450500550600650700750800850900

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

Weighted average PPD

Num

ber

of w

orki

ng h

ours


Figure 6.9 Number of working hours for certain average PPD for strict set points of the reference building.

For the normal control set points, the weighted average PMV of the open plan varies between -0.4 and +0.4, while for the cell type it varies between -0.55 and +0.3, as shown in Figure 6.10. The difference in the weighted average PMV values between the two plan types is slightly higher during the winter. The tendency can be explained by the lower occupancy during the summer months.

For the normal control set points the lower average PPD values of the open plan imply a better indoor thermal environment, as shown in Figure 6.11. For the cell type a maximum average PPD of 10% corresponds to 73% of the working hours, while for open plan the same PPD corresponds to 82%. For the cell type an average PPD of 15% (or less) corresponds to 93% and for the open plan to 96% of the working hours. The yearly weighted average PMV difference is 0.13.


176

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e PM

V


Figure 6.10 Weighted average PMV for normal set points of the reference build-ing.

050

100150200250300350400450500550600650700750800

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%


Num

ber

of w

orki

ng h

ours


Figure 6.11 Number of working hours for certain average PPD for normal set points of the reference building.

The weighted average PMV values for the poor control set points vary between -0.6 and +0.72 for the open plan and between -0.77 and +0.6 for the cell type as shown in Figure 6.12. The difference in the PMV values between the two interior types increases even more during the winter. The


177

yearly weighted average PMV difference is 0.23, which is higher than for the strict and normal set points. This can be explained by the wider air temperature variation allowed by the poor set points.

For the poor control set points, the weighted average PMV of the open plan results in slightly lower PPD values as shown in Figure 6.13. For the cell type an average PPD of 10% corresponds to 39% of the working hours, while for open plan the same PPD corresponds to 43%. For both the cell and open plan types an average of 15% PPD corresponds to 63% of the working hours.

-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,10,00,10,20,30,40,50,60,70,8

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e PM

V

SS-30%-Cell-average-poor SS-30%-Open-average-poor

Figure 6.12 Weighted average PMV for poor set points of the reference build-ing.


178

0

50

100

150

200

250

300

350

400

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%

30-3

2%

32-3

4%

34-3

6%


Num

ber

of w

orki

ng h

ours


Figure 6.13 Number of working hours for certain average PPD for poor set points of the reference building.

• Impact of control set points on the perception of thermal com-fort

The impact of the different control set points on the perception of the thermal environment is presented in Figures 6.14 and 6.15 (cell type). From January until May and from the middle of September to the end of December the normal set points result in a more neutral thermal environ-ment, since the weighted average PMV values are closer to zero. From the end of May until the beginning of September, however, the strict set points ensure an indoor air temperature that improves the perception of comfort for the occupants. Both controls are very close to the recommendations of ISO Standard 7730 (1984), since the PMV hardly exceeds the limits of ±0.5. For the poor control set points, however, the PMV varies from -0.8 to +0.6, exceeding the comfort levels. Lower cooling and higher heating air temperature set points (as in normal or strict set points) are preferable for providing an acceptable thermal environment.

For the strict set points a maximum average PPD of 10% corresponds to 66% of the working hours, while for the normal and poor ones the same PPD corresponds to 73% and 40% respectively. The same values for maximum average PPD of 15% correspond to 93%, 93% and 63% of the working hours for the three set points.


179

-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,10,00,10,20,30,40,50,60,70,8

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e PM

V


Figure 6.14 Weighted average PMV for the cell type plan of the reference build-ing.

0

50

100

150

200

250

300

350

400

450

500

550

600

650

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%

30-3

2%

32-3

4%

34-3

6%


Num

ber

of w

orki

ng h

ours


Figure 6.15 Number of working hours for certain average PPD for cell plan (reference building).

The impact of strict, normal and poor control set points on the perception of thermal comfort for the open plan type is presented in Figures 6.16 and 6.17. As already stated, due to the higher internal loads in the open plan the weighted average PMV values slightly increase. From January until the beginning of May and from the middle of September to the


180

end of December the normal set points provide a more neutral thermal environment. From May till the beginning of September the PMV values of the normal set points increase, causing slight discomfort. The weighted average PMV of the normal control set points varies between ±0.4, while those for the strict set points vary between -0.5 and 0. For the poor con-trol the variation is from -0.6 to +0.75, exceeding the comfort levels (ISO Standard 7730, 1984).

For the strict set points a weighted average PPD of 10% corresponds to 78% of the working hours while for the normal and poor set points the same average corresponds to 82% and 43% respectively. The same values for a weighted average PPD of 15% (or less) correspond to 95%, 96% and 63% of the working hours.

-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,10,00,10,20,30,40,50,60,70,8

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e PM

V

SS-30%-Open-average-strict SS-30%-Open-average-normal SS-30%-Open-average-poor

Figure 6.16 Weighted average PMV for the open type plan of the reference build-ing.


181

050

100150200250300350400450500550600650700750800850900

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%

30-3

2%

32-3

4%


Num

ber

of h

ours

SS-30%-Open-average-strict SS-30%-Open-average-normal SS-30%-Open-average-poor

Figure 6.17 Number of working hours for certain average PPD for open plan (reference building).

From the above it can be concluded that a narrow variation in PMV val-ues can be achieved by strict temperature set points, which will however result in an increased energy use. The selection of correct upper and lower temperature limits is essential for the provision of an improved thermal environment, in order to avoid creating too cold or too warm thermal conditions. In order to ensure that the optimal upper and lower tempera-ture limits are selected, parameters such as internal loads, thermal mass, etc, should be considered. The stricter set points require more attention to these limits.

6.1.3 Indoor climate on a zone levelThe mean air temperature, the directed operative temperature and thermal comfort indices on a zone level for the reference building are studied. North, east, south and west oriented zones were considered for the three control set points. The zones studied are all the occupied ones of the cell type offi ce building (i.e. double and single offi ce rooms, meeting rooms and corner offi ce rooms). The mean air temperatures were studied, in order to investigate their variation within the limits defi ned by the set points. The number of hours with certain directed operative temperatures works as a “link” between the mean air temperatures and the way the occupants perceive their thermal environment. Finally the monthly average weighted


182

PMV values describe whether the occupants feel “warm” or “cold” during the year, while the average PPD values are a quantitive way of evaluating the quality of thermal environment.

6.1.3.1 Mean air temperatures and potential overheating problem

The window to external wall area ratio of the reference building is 30% as described in Subsection 5.1.1. However, the meeting rooms are the only zones that have a higher glazing area than the rest of the zones (65% win-dow to external wall area ratio, while this is 30% is for the other zones). In order to study the impact of orientation on the mean air temperatures, the meeting rooms were selected. As shown in Figure 6.18 there is a peak for the number of working hours with air temperatures close to the up-per permissible limit, which is very little dependent on the orientation. The results are similar for zones with 30% window to external wall area ratio.

0

50

100

150

200

250

300

350

400

450

500

21.7

- 2

1.8

21.9

- 2

2.0

22.1

- 2

2.2

22.3

- 2

2.4

22.5

- 2

2.6

22.7

- 2

2.8

22.9

- 2

3.0

23.1

- 2

3.2

23.3

- 2

3.4

23.5

- 2

3.6

23.7

- 2

3.8

23.9

- 2

4.0

24.1

- 2

4.2

24.3

- 2

4.4

24.5

- 2

4.6

Mean air temperature (°C)

Num

ber

of w

orki

ng h

ours

Meeting room-North Meeting room-East Meeting room-South Meeting room-West

Figure 6.18 Working hours with certain mean air temperatures for the meting room (reference building, cell type plan, and normal control set points), which has 65% window to external wall area ratio.

Due to the large peak number of working hours with air temperatures close to the upper permissible mean air temperature limit, the potential overheating problem was studied. This could have been studied by calculat-ing the cooling load for each zone; since, however, the output (regarding energy use) from IDA ICE 3.0 is available only on a building level and


183

simulation of individual zones would have been far too time consuming, an alternative way had to be found. Thus, the number of hours close to the upper permissible temperature limit (hours with cooling demand, in order to keep the air temperature within the permissible limits) was calculated instead. This indicator is not precise but in a comparative study it can show the risk of overheating problems for different orientations, glazed areas, control set points and internal loads for each zone type (Figure 6.19).

As expected, the south orientated zones are slightly warmer. Due to the relatively high internal gains and the low upper air temperature limit of 23ºC (strict set points), the single and double offi ces have similar number of hours with cooling demand. Mainly due to the lower internal gains (lower occupant density), the corner offi ce reaches the temperature limit of 23ºC for fewer hours as shown in Figure 6.19.

The different occupancy of the meeting rooms (occupied only 4 instead of 8 hours per day) makes the comparison of this zone type with the rest useless. However, useful conclusions can be drawn, when the impact of orientation on a zone with larger glazing area (such as the meeting rooms with window area to room volume ratio of 0.67 m2/m3) and the rest of the zones (e.g. single and double offi ce room with window area to room volume ratio of 0.258 m2/m3 and 0255 m2/m3 respectively) are compared. As shown in Figure 6.19, the peak of the curve in the south oriented façade of the meeting room is slightly larger than for the other zones.

The high internal gains of the single and double offi ce rooms, combined with the low upper temperature limit, result in a large number of hours with cooling demand, decreasing the effect of the orientation (for the single room, the difference between the north and south oriented single offi ce is 3% and for the double offi ce 1%). The lower internal gains of the corner room and the larger glass area of the meeting rooms, however, result in a somewhat larger difference in the number of hours when cool-ing might be needed. Regardless of the zone type, the south orientation is the warmest, and the north orientation is the one with the lowest cooling demand, as expected.


184

175317691751 1742

1724177817371719

656751

637632

12571352

12351156

0

200

400

600

800

1000

1200

1400

1600

1800

2000

North East South West

Num

ber

of w

orki

ng h

ours

bet

wee

n 23

.9 a

nd 2

3.1°

C

Double room Single room Meeting room Corner room

Figure 6.19 Potential overheating problem of the reference building for the strict control set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

When normal set points (with upper air temperature limit 24.5ºC) are applied, the number of hours with potential cooling demand drops dra-matically, regardless of the type of zone (as shown in Figure 6.20). When Figures 6.19 and 6.20 are compared it can be seen that for strict set points the curves for the single and double offi ces are fairly similar, while for the normal ones, the curve of the single offi ce zone is somewhat lower than the double one. This means that narrow temperature variations with lower upper temperature limits diminish the impact of internal loads on the cool-ing demand (mostly in zones that are already quite densely occupied).

Another interesting conclusion that can be drawn when these two graphs are compared is that strict set points reduce the impact of orientation on the cooling mainly in densely occupied zones or in zones with larger glazed areas, such as meeting rooms (since in both cases the temperature could easily rise up to 23ºC).


185

1075

930

1132

1036

756

932978

880

433

557

457392

648692680

511

0

100

200

300

400

500

600

700

800

900

1000

1100

1200


Num

ber

of w

orki

ng h

ours

bet

wee

n 24

.4 a

nd 2

4.6°

C


Figure 6.20 Potential overheating problem of the reference building for the normal control set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

For the poor set points the number of hours when cooling is used drops dramatically, mainly due to the higher upper temperature limit (26ºC), as shown in Figure 6.21. The increased glass area of the meeting room, however, keeps the values relatively high, compared with the other zone types. This means that the impact of poor set points on cooling demand is smaller for zones with larger glass areas (regardless of the internal loads of the zone). The curve type of the offi ces has changed for this set point, since the number of hours with air temperatures up to 26ºC in the south facing offi ces is lower than in the east facing ones (due to the occupants’ schedule and the upper temperature limit).


186

492

266

467446

267

470 423

427

292

386

316

274244

230277

114

0

50

100

150

200

250

300

350

400

450

500

550


Num

ber

of w

orki

ng h

ours

bet

wee

n 25

.9°C

and

26.

1°C


Figure 6.21. Potential overheating problem of the reference building for the poor control set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

6.1.3.2 Directed operative temperaturesThe directed operative temperatures for different zones (double, corner offi ce and meeting room), control set points and orientations are studied for the different zones of the reference building, in order to fi nd their im-pact on the occupants’ comfort. The occupants are placed 1 m from the façade in all the zones. The corner offi ces have two external walls, one with window and one opaque. The one denoted by capital letters (see Figure 6.22) refers to the orientation where the window is placed; the other one refers to the opaque wall.

For the strict control set points (22ºC-23ºC) the directed operative temperature varies from 21.5ºC to 25ºC for the double and corner offi ces and from 20ºC to 28ºC for the meeting rooms as shown in Figure 6.22 (south orientation). As expected, the variation in the directed operative temperature in the meeting room (due to the larger window area) is larger than in the rest of the zones. It can be noticed that although the orientation has limited impact on the mean air temperatures, the directed operative temperatures vary considerably. This can be explained by the nature of the set points used (based on mean air and not operative temperatures) and also by the fact that solar gains vary considerably with the orientation.


187

0

25

50

75

100

125

150

175

200

225

…-2

0.1

20.2

- 2

0.3

20.5

- 2

0.6

20.8

- 2

0.9

21.1

- 2

1.2

21.4

- 2

1.5

21.7

- 2

1.8

22.0

- 2

2.1

22.3

- 2

2.4

22.6

- 2

2.7

22.9

- 2

3.0

23.2

- 2

3.3

23.5

- 2

3.6

23.8

- 2

3.9

24.1

- 2

4.2

24.4

- 2

4.5

24.7

- 2

4.8

25.0

- 2

5.1

25.3

- 2

5.4

25.6

- 2

5.7

25.9

- 2

6.0

26.2

- 2

6.3

26.5

- 2

6.6

26.8

- 2

6.9

27.1

- 2

7.2

27.4

- 2

7.5

27.7

- 2

7.8

Operative temperature (°C)

Num

ber

of w

orki

ng h

ours

Double-South Meeting room-South Corner SOUTH-West

Figure 6.22 Number of working hours between certain directed operative tem-peratures for south oriented zones for the reference building (strict control set points). The lower case letters refer to the opaque wall. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

In Figure 6.23 a diagram of the number of hours with certain directed operative temperatures is presented for meeting rooms with strict control set points for the north and south orientations. The results for the east and west facing meeting rooms are quite similar and they vary somewhere in between the north and south oriented alternatives (maximum temperature 26.5ºC and 27ºC respectively). The north facing meeting room is cooler (maximum directed operative temperature 25.4ºC), while the directed operative temperature of that facing south reaches 28ºC. The tendencies for the other zones are similar, but somewhat lower due to the smaller glazing areas used. The reason for selecting the meeting rooms for this comparison is that due to the larger glazed area (65% instead of 30% for the rest of the zones) the tendencies are more evident.

Directed oprerative temperature (°C)


188

0

5

10

15

20

25

30

35

40

45

…-2

0.1

20.2

- 2

0.3

20.5

- 2

0.6

20.8

- 2

0.9

21.1

- 2

1.2

21.4

- 2

1.5

21.7

- 2

1.8

22.0

- 2

2.1

22.3

- 2

2.4

22.6

- 2

2.7

22.9

- 2

3.0

23.2

- 2

3.3

23.5

- 2

3.6

23.8

- 2

3.9

24.1

- 2

4.2

24.4

- 2

4.5

24.7

- 2

4.8

25.0

- 2

5.1

25.3

- 2

5.4

25.6

- 2

5.7

25.9

- 2

6.0

26.2

- 2

6.3

26.5

- 2

6.6

26.8

- 2

6.9

27.1

- 2

7.2

27.4

- 2

7.5

27.7

- 2

7.8


Num

ber

of w

orki

ng h

ours

Meeting room-North Meeting room-South

Figure 6.23 Number of working hours between certain directed operative tem-peratures for the 65% (reference) glazed meeting room (strict control set points). The meeting room has 65% window to external wall area ratio.

In order to study the impact of temperature set points on the directed operative temperatures, the north and south single offi ces are compared, as shown in Figure 6.24. Since the lower permissible air temperature limit for both cases is the same (22°C), the minimum directed operative temperature is the same. For the north orientation the maximum directed operative temperature reaches 24°C for the strict set points and 25.4°C for the normal ones. For the south orientation the temperature reaches 25°C and 26.4°C respectively. For both set points the directed operative temperature is 1°C higher than the permissible air temperature limit for the north orientation and 2°C higher for the south orientation.



189

0

25

50

75

100

125

150

175

200

225

250

21.7

- 2

1.8

21.9

- 2

2.0

22.1

- 2

2.2

22.3

- 2

2.4

22.5

- 2

2.6

22.7

- 2

2.8

22.9

- 2

3.0

23.1

- 2

3.2

23.3

- 2

3.4

23.5

- 2

3.6

23.7

- 2

3.8

23.9

- 2

4.0

24.1

- 2

4.2

24.3

- 2

4.4

24.5

- 2

4.6

24.7

- 2

4.8

24.9

- 2

5.0

25.1

- 2

5.2

25.3

- 2

5.4

25.5

- 2

5.6

25.7

- 2

5.8

25.9

- 2

6.0

26.1

- 2

6.2


Num

ber

of w

orki

ng h

ours

Strict-North Strict-South Normal-North Normal-South

Figure 6.24 Impact of control set points on directed operative temperature in north and south oriented single offi ces of the reference building. The offi ces have 30% window to external wall area ratio.

A comparison of the number of working hours with certain directed operative temperatures (Figure 6.25) is carried out for the north facing single offi ces and the meeting rooms (strict and normal control). For the meeting rooms, the large impact of the glass area on the directed opera-tive temperatures is evident. For the single offi ces the difference (between maximum permissible air and directed operative temperature) is 1°C (regardless of the set point), while for the meeting rooms it is 2.5°C for the strict and 3°C for the normal set points. The impact of the glass area on the directed temperature is larger for the south oriented zones.



190

0

25

50

75

100

125

150

175

200

225

250

…-2

0.1

20.2

- 2

0.3

20.5

- 2

0.6

20.8

- 2

0.9

21.1

- 2

1.2

21.4

- 2

1.5

21.7

- 2

1.8

22.0

- 2

2.1

22.3

- 2

2.4

22.6

- 2

2.7

22.9

- 2

3.0

23.2

- 2

3.3

23.5

- 2

3.6

23.8

- 2

3.9

24.1

- 2

4.2

24.4

- 2

4.5

24.7

- 2

4.8

25.0

- 2

5.1

25.3

- 2

5.4

25.6

- 2

5.7

25.9

- 2

6.0

26.2

- 2

6.3

26.5

- 2

6.6

26.8

- 2

6.9

27.1

- 2

7.2

27.4

- 2

7.5

27.7

- 2

7.8


Num

ber

of w

orki

ng h

ours

Meeting room-Strict Meeting room-Normal Single office-Strict Single office-Normal

Figure 6.25 Impact of control set points on directed operative temperature in single offi ces and meeting rooms for the reference building (north orientation, 3rd alternative). The meeting room has 65% window to external wall area ratio and the offi ces 30%.

6.1.3.3 Perception of thermal comfortThe thermal comfort indices (PMV and PPD) for zones with different orientations and set points are studied in this Section. As stated above, the occupants were placed 1m from the external wall (window).

For strict temperature set points the PMV and PPD values are quite similar for the simulated alternatives. The monthly average PMV for the south oriented zones is always below 0 for the offi ces, while for the meeting rooms (higher glazing area) it is positive from early May to late September (as presented in Figure 6.26). Clearly, the corner offi ce room is the one with the lower PMV values, due to the lower internal gains and higher external wall area to room volume ratio.



191

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

Double office Single office Meeting room Corner office

Figure 6.26 Monthly average PMV for south oriented zones of the reference building with strict set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

For the north oriented zones, the meeting room is slightly colder than the single and double offi ces during winter (while for the south oriented ones, it was warmer almost throughout the year) as shown in Figure 6.27. The monthly average PMV values for the offi ces do not change much, when the north and south oriented façades are compared, due to the strict set points and the small (compared at least with the meeting rooms) glazed area; the difference in the PMV values is larger, however, for the meeting rooms.


192

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.27 Monthly average PMV for north oriented zones of the reference build-ing with normal set point. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

Regardless of the orientation it appears that the strict temperature set point (22ºC - 23ºC) is quite low for the cell type reference building, since the PMV values exceed the recommended limit of -0.5 (ISO Standard 7730, 1984) during winter, while during summer the monthly average PMV values do not exceed zero. The lower PMV values during July (compared with June and August) are due to the lower internal gains (75% of the occupants are working during June and August and 50% during July) and not to lower outdoor temperatures.

In Figure 6.28 a diagram of the number of hours with certain PPD for north facing zones is presented. The single and double offi ces perform similarly, while the corner offi ce results in higher PPD values due to the lower internal loads and the larger external wall area. The PPD values of the meeting room have a larger spread due to the larger glazing area. It has to be noted that in this case the total amount of working hours is smaller, since the meeting rooms are occupied only 50% of the time.


193

0

50

100

150

200

250

300

350

400

450

500

550

600

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

PPD

Num

ber

of w

orki

ng h

ours


Figure 6.28 Number of working hours with certain PPD for north facing zones of the reference building with strict set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

The impact of orientation on the PPD is limited for the strict set points due to the narrow permissible air temperature variation (as with PMV values). In Figure 6.29 the percentage of working hours with PPD of 10% and 15% (or lower) is presented. Regardless of the orientation, the corner offi ce has the lowest PPD values. The single and double offi ces perform quite well, while the south oriented meeting room has similar values for PPD of 10% (or less).

Taking into account the low PMV values discussed earlier and the PPD values presented below, it is obvious that the low upper temperature limit (e.g.23°C) combined with the narrow permissible temperature variations of the strict set points results in a cold thermal environment all year round (PMV values between -0.7 and 0). This is evident for the corner zones, in which the low internal gains (and thus lower air temperatures) reduce drastically the number of working hours with PPD values of 10 and 15%. The larger glazing area of the south oriented meeting room, however, increases the operative temperatures and results in a warmer thermal en-vironment, more acceptable to the occupants (due to the non-optimal strict set point).


194

6366

6965

6772

67

92 93 939595 959696

66

81

89

8279

63

72

6360

5154

5753

75 7782

77

0

10

20

30

40

50

60

70

80

90

100

North East South West North East South West

PPD of 10% PPD of 15%

% o

f wor

king

hou

rs


Figure 6.29 % of working hours with PPD of 10% and 15% (or less) for zones with strict control set points (reference building). The meeting room has 65% window to external wall area ratio and the offi ces 30%.

For the normal control set points the monthly average PMV values of the south oriented zones are shown in Figure 6.30. The meeting room appears to be warmer (compared with the single and double offi ces) between the middle of March and the middle of October. The monthly average PMV values in this case vary between -0.65 and +0.53, while for the same case (but for strict set points) they vary between -0.65 and +0.15. For the other zones the monthly average PMV values exceed zero (while for the strict set points they are negative all year round, see Figure 6.26) from May to September. During the winter months the PMV for the normal set points is similar to that for the strict set points, since the lower permissible tem-perature limit of 22ºC is kept the same.


195

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

0,6

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.30 Monthly average PMV for south oriented zones of the reference build-ing with normal set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

For the north oriented zones the monthly PMV values are presented in Figure 6.31. The meeting room is once more warmer than the single and double offi ces between April and September and all year round compared with the corner offi ce. The PMV in this case varies between -0.7 and +0.5, while for the same case (but for strict set points) the PMV is between -0.7 and +0.1.


196

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.31 Monthly average PMV for north oriented zones of the reference build-ing with normal set points. The meeting room has 65% window to external wall area ratio and the offi ces 30%.

The large glazing area of the meeting rooms results in a wide variation in the monthly average PMV values. However, it is only the south oriented meeting room that slightly exceeds the recommended PMV limit of +0.5 (ISO Standard 7730, 1984). The effect of the normal set points on the rest of the zones is similar (but on a lower scale).

The number of working hours with PPD of 10% and 15% (or less) is presented in Figure 6.32. When Figures 6.29 and 6.32 are compared, it can be noticed that the number of hours with PPD of 15% or less is similar for normal and strict set points, regardless of the zone type and the orienta-tion. The number of hours with PPD of 10% or less, however, is larger for the normal control cases mostly for the offi ces (single, double and corner with 30% window area) for all the orientations i.e. the thermal comfort is better. This shows that the choice of a narrow variation in permisible air temperature for zones with small window to external wall area ratios requires very careful selection of the higher and lower limits; in this case the upper air temperature limit for the strict set points was selected lower than it should have been, resulting in higher discomfort of the occupants (due to the low PMV values). Thus, a narrow air temperature set point is not necessarily recommended, since it is quite diffi cult to know precisely a priori the parameters that determine the selection of upper and lower air temperature limits. A comparison of the number of working hours with PPD lower than 10% for zones with large window to external wall area


197

ratios (Figures 6.29 and 6.32), however, shows that strict set points are essential, in order to lower the PPD values (since the difference between the directed operative temperatures and the air temperatures increases regardless of the zone type and the orientation). In this case the lower upper air temperature limit ensures the low PPD values.

7577

8076

93 94 94

7174

7975

95

4642

49 48

78 78 80 80

5357

6055

75 77

83

77

95 959796

0

10

20

30

40

50

60

70

80

90

100



% o

f wor

king

hou

rs


Figure 6.32 Number of working hours with PPD of 10% and 15% (or less) for zones with normal control set points (reference building). The meeting room has 65% window to external wall area ratio and the offi ces 30%.

For the poor control set points the monthly average PMV values exceed the recommended values for an acceptable indoor thermal environment (ISO Standard 7730, 1984), as shown in Figure 6.33. The variation in monthly average PMV values for the south oriented zones is between -0.9 and +0.9, with a smaller variation for the cell offi ce rooms. The results for the north, east and west oriented zones are similar.


198

-0,9-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,10,00,10,20,30,40,50,60,70,80,9

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.33 Monthly average PMV for south oriented offi ces and meeting room with poor set points (reference building). The meeting room has 65% window to external wall area ratio and the offi ces 30%.

The unsatisfactory thermal environment resulting from the poor set points can be proved by the low number of working hours with PPD values lower than 10% and 15%. Mostly for the meeting rooms, it is for no more than 27% of the working hours that PPD values are lower than 10% (the PPD values of 15% do not exceed 43%), as shown in Figure 6.34. For this reason the PMV and PPD are not studied further in this section.


199

45

39

48

42

6966

7167

39

34

43

36

6360

6762

22 22

2724

3735

4340

32

27

3431

49 49

5550

0

10

20

30

40

50

60

70

80

90

100



% o

f wor

king

hou

rs


Figure 6.34 Number of working hours with PPD of 10% and 15% (or less) for zones with poor control set points (reference building). The meeting room has 65% window to external wall area ratio and the offi ces 30%.

6.2 Single skin glazed alternatives (60% and 100% window to external wall area ratios)

6.2.1 Energy use As stated in Subsection 4.1.2, 42 building alternatives were generated with 60% window to external wall area ratio (7 windows and shading device types, 2 plan types and 3 control set points) and another 42 for alternatives with 100% window to external wall area ratio. A detailed description of the window alternatives used was given in Subsection 4.1.2, while their properties were presented in Subsection 5.2.1.2. The energy use for the highly glazed alternatives is presented in Appendix L, while the results are discussed below.

In the fi rst 100% glazed alternative (with clear panes) the high thermal transmittance of the windows resulted in insuffi cient heating capacity. The opposite problem (insuffi cient cooling capacity) is noticed in the fi fth al-ternative (mostly in the cell type) due to the high total solar transmittance (and also geffective) of the glazing. These 9 extreme cases were simulated


200

once more with increased heating and cooling capacity, in order to deter-mine the energy use for keeping the air temperature in the working areas within the set point limits. However, since the original reference building was designed with a certain heating and cooling capacity, the parametric study is carried out considering the original alternatives. In Appendix L the energy use for alternatives with increased heating and cooling capacity is presented.

6.2.1.1 Impact of fl oor plan type and orientationFor the 60% and 100% glazed alternatives the impact of plan type on energy use for heating and cooling is similar to that for the reference build-ing (higher heating demand for the cell type and higher cooling demand for the open type) as described in Subsection 6.1.1.1. The energy use for lighting depends on the visual transmittance of the window system (with and without the use of the shading devices). As mentioned in Subsection 5.2.1.2, however, the daylight availability in IDA ICE 3.0 depends on the direct solar transmittance, since Tvis is not considered. The tendency, regarding the energy for lighting the different glazing alternatives, is quite correct, since in almost all the simulated cases, windows with lower g values also have lower Tvis values (see Table 5.11), resulting in increased energy demand for artifi cial lighting.

In Figure 6.35 the difference in heating and cooling demand between the cell and open plan types is presented (the absolute values are given in Appendix L). Once more it can be noticed that the wider temperature set points allow more “useful” heat to be stored; thus, the effect on the cooling demand is the opposite.


201

-8,0-7,0-6,0-5,0-4,0-3,0-2,0-1,00,01,02,03,04,05,06,07,08,09,0

10,011,0

heating cooling heating cooling heating cooling

Strict Normal Poor

Ene

rgy

diff

eren

ce (

kWh/

m2 a)

Alternative 1 Alternative 2 Alternative 3 Alternative 4 Alternative 5 Alternative 6 Alternative 7

Figure 6.35 Impact (difference between cell and open types) of plan type on the heating and cooling demand of 60% glazed alternatives with dif-ferent window types.

When the total energy demand difference between the cell and open plan type alternatives is compared (Figure 6.36), it can be noticed that the impact of plan type on the energy use regardless of the window type used is increased for strict control set points (60% glazed alternatives).

-7,0

-6,0

-5,0

-4,0

-3,0

-2,0

-1,0

0,0

1,0

2,0

3,0

4,0

5,0

6,0

Strict Normal Poor

Ene

rgy

diff

eren

ce (

kWh/

m2 a)

Alternative 1 Alternative 2 Alternative 3 Alternative 4 Alternative 5 Alternative 6 Alternative 7

Figure 6.36 Impact (difference between cell and open types) of plan type on the total energy use of 60% glazed alternatives with different window types.


202

Regarding the impact of orientation, the difference in energy use for the north south and east-west oriented 100% glazed alternatives (3) increases by up to 3% (from 0.1 of the reference building). Due to the small im-pact of the orientation on the total energy use of the building (because of the identical long and short façades), no further study of the infl uence of orientation on the total energy use has been made on a building level.

6.2.1.2 Impact of windows and shading devices for the 60% and 100% glazed alternatives

An energy use comparison of the 60% and 100% single skin glazed alterna-tives with different windows and shading devices applied is presented in this Subsection. For the comparison, only cell type alternatives with normal control set points are described. The results are similar for the other set points and the open plan. The absolute energy use values of alternatives with strict and poor set points are presented in Appendix L, as well as the open plan type cases. A brief description of the window alternatives used is given below.

The fi rst (60% and 100%) glazed alternatives have a triple clear pane window of Uwindow=1.97 W/m2K (as described in Subsection 5.2.1.2). When its thermal transmittance is reduced to 1.27 W/m2K in the second alternative (and the total solar transmittance from 0.69 to 0.58), the total energy use decreases by 12% for the 60% and by 14% for the 100% glazed alternative as shown in Figure 6.37).

For the second (60% glazed) alternative the energy use for heating drops by 31%, while for the 100% glazed alternative it is slightly higher (33%). The reduced thermal transmittance of the window results in a 20% increase in cooling demand for the 60% and in a 17% increase for the 100% glazed alternatives. Although the total solar transmittance slightly drops in the second case, the much lower thermal transmittance insulates the building in such a way that it doesn’t allow the heat to “escape” dur-ing summer, resulting in increased cooling demand. The energy use for lighting (with a set point of 500 lux at the desktop) increases by 4% for the 60% and by 5% for the 100% glazed building alternatives, due to the reduced (in the second case) direct solar transmittance (Tsol) of the glazing system (see Table 5.11), since, as already stated in Subsection 5.2.1.2, the daylight availability in IDA ICE 3.0 is calculated by the direct solar and not the visual transmittance.


203

7250

92

59

20

24

30

37

13,4

13,8

12,9

13,5

45

45

45

45

0

20

40

60

80

100

120

140

160

180

200

normal (1) normal (2) normal (1) normal (2)

60% glazed alternatives 100% glazed alternatives

Ene

rgy

use

(kW

h/m

2 a)

Space heating Cooling Lighting Rest

Figure 6.37 Comparison of energy use for the 1st and 2nd glazed alternatives (cell type).

In the third alternative the triple glazing is replaced by double glazing with a lower total solar transmittance (g=0.35 instead of g=0.58). The intermediate venetian blinds (geffective=0.225) are also replaced by internal ones (geffective=0.276). The total energy use drops only by 1% for the 60% and by 4% for the 100% glazed alternative (Figure 6.38).

For the third alternative the energy use for heating increases by 6% for the 60% and by 10% for the 100% glazed alternative due to the lower total solar transmittance of the glazing (the U value was not changed). For the same reason cooling demand drops by 26% for the 60% and by 25% for the 100% glazed building alternative. It has to be noted that in the third alternative the blinds are internal, while for the second they are intermediate resulting in a lower geffective for the second case (0.23 vs. 0.28 for the third case), which explains the rather small drop in cooling demand. The energy use for lighting the 60% glazed offi ces increases (by 3% for the 60% and by 4% for the 100% glazed building alternative) in the third alternative due to the lower direct solar transmittance (see Table 5.11).


204

50 54 59 65

24 18

37 2713,8 14,2

13,514,0

45 45

4545

0

20

40

60

80

100

120

140

160



Ene

rgy

use

(kW

h/m

2 a)


Figure 6.38 Comparison of energy use for the 2nd and 3rd glazed alternatives (cell type).

In the fourth alternative the double glazing is replaced by one with a lower total solar transmittance (g=0.23 instead of g=0.35). The total energy use drops only by 2% for the 60% and by 3% for the 100% glazed building alternative as shown in Figure 6.39.

The heating demand for the fourth 60% and 100% glazed building alternatives increases by 3%. The cooling demand drops by 26% for the 60% glazed alternative and by 27% for the 100% glazed one. The decrease in cooling demand in this case is larger due to the lower geffective (0.22 instead of 0.28), since the position of the blinds was kept the same in the third and fourth alternatives. The energy use for lighting the 60% glazed offi ces slightly increases (by 1% for the 60% and by 2% for the 100% glazed building alternative) in the fourth alternative, due to the lower direct solar transmittance (see Table 5.11).


205

54 5565 68

18 13

27 1914,2 14,4

14,014,2

45 45

4545

0

20

40

60

80

100

120

140

160



Ene

rgy

use

(kW

h/m

2 a)


Figure 6.39 Comparison of energy use for the 3rd and 4th glazed alternatives (cell type).

In the fi fth alternative the double glazing is replaced by one with a solar factor higher than in the third alternative (g=0.58 instead of g=0.35). The total energy use increases by 9% for the 60% and by 12 % for the 100% glazed building alternative Figure 6.40.

The heating demand drops by 10% for the 60% and by 12% for the 100% glazed building alternative (case (5) compared with case (3)), while the cooling demand increases dramatically (by 99% for the 60% and by 102% for the 100% glazed building alternative). The energy use for lighting the 60% glazed offi ces drops (by 3% for the 60% and by 5% for the 100% glazed building alternative), in the fi fth alternative due to the higher direct solar transmittance (see Table 5.11).


206

54 4865 58

18 36

2754

14,2

13,814,0

13,445

4545

45

0

20

40

60

80

100

120

140

160

180



Ene

rgy

use

(kW

h/m

2 a)



A comparison between the second and fi fth glazed building alternatives is carried out, in order to study the impact of shading device position on energy use (since U and g values are kept the same). In the second alterna-tive the blinds are intermediate, while in the fi fth they are internal. When the venetian blinds are placed in between the panes, the solar factor of the system (window and shading device) is much lower than when they are internal (geffective=0.225 instead of geffective=0.469). The total energy use for the 60% glazed buildings increases by 8% for the fi fth 60% glazed building alternative and by 10% for the 100% glazed building, as shown in Figure 6.41.

Since the shading devices are not often used during the heating period, the higher geffective of the fi fth alternative has a limited impact on the heating demand. The energy use for heating the 60% glazed buildings is almost the same for the second and fi fth alternatives, while for the 100% glazing alternatives the heating demand drops by 3%. During the cooling periods, however, the cooling demand of the fi fth alternative increases dramatically since the venetian blinds are used more often. The increase is 47% for the 60% and 45% for the 100% glazed building alternative. The energy use for lighting the 60% and 100% glazed offi ces is almost the same, due to the similar direct solar transmittance (Tsol) of the two alternatives (see Table 5.11).


207

50 4859 58

2436

375413,8

13,8

13,5

13,445

45

45

45

0

20

40

60

80

100

120

140

160

180



Ene

rgy

use

(kW

h/m

2 a)


Figure 6.41 Comparison of energy use for the 2nd and 5th glazed alternatives (cell type).

In the sixth alternative internal screens were applied. When this case is compared with the third one (same glazing system but with internal blinds which gives a geffective of 0.19 instead of 0.28), there is a slight decrease in cooling demand and there is not much effect on the heating demand (Figure 6.42).

54 5465 66

18 16

27 2414,2 14,2

14,0 13,9

45 45

45 45

0

20

40

60

80

100

120

140

160



Ene

rgy

use

(kW

h/m

2 a)




208

In the seventh alternative the internal blinds (of the third alternative) are replaced by fi xed external louvres (with lower effective total solar transmit-tance). The heating demand increases by 9% for the 60% and by 11% for the 100% glazed building alternative (when compared with the third one) as shown in Figure 6.43. This can be explained by the fact that the shading in the seventh case is applied all year round and consequently geffective=g=0.19. The shading in the third alternative, however, is ap-plied 100% when the incident light inside the glass exceeds 100W/m2. The cooling demand drops by 60% for the seventh 60% glazed building alternative and by 66% for the 100% glazed one. The total energy use for the 60% glazed buildings decreases by 4% for the seventh 60% and by 7% for the 100% glazed building alternative. The energy use for lighting is slightly higher in the seventh alternative, since the shading devices are applied throughout the year (by 2% and 3% for the 60% and 100% glazed cases). The seventh alternative results in the lowest total energy use, due to the lower g value of the system (since the shading is applied all year round and thus the total solar transmittance is equal to the effective one). Another parameter that has to be further noticed, is that although the shading for the seventh case is always applied, the total solar transmittance varies considerably during the year due to the monthly average g and Tsol values inserted in IDA ICE 3.0 (e.g. 0.13 during May and 0.28 during January). This permits valuable heat gains during the winter (resulting in a small increase in heating demand), while having the opposite effect during summer (resulting in a large drop in cooling demand).

54 59 65 72

18 7

27 914,214,4

14,0

14,3

4545

45

45

0

20

40

60

80

100

120

140

160



Ene

rgy

use

(kW

h/m

2 a)




209

A cross comparison of the 60% and 100% glazed alternatives with dif-ferent windows and shading devices applied, shows that the difference in total energy use (compared with the reference building) is reduced, when the thermal and total solar transmittance decrease. As expected, lower gglazing values result in lower cooling and slightly higher heating demand (e.g. alternatives 2 and 3). The value of geffective (total solar transmit-tance when shading is applied) is crucial for the cooling demand. When the alternatives 2 and 5 (same glazing but different position of shading results in different geffective) are compared, the cooling demand differs considerably (higher cooling when the shading is applied indoors). The low window thermal transmittance results in lower heating demand (as expected), which can be seen, when the fi rst and the remaining glazing alternatives are compared.

6.2.2 Indoor climate on a building level

6.2.2.1 Weighted average mean air temperaturesIn order to carry out parametric studies (on a building level) for study-ing the impact of window and shading device type on average air tem-peratures, 100% glazed alternatives with normal control set points were considered.

In the case of the reference building (30% window to external wall area ratio) the mean air temperature greatly depends on the set point applied in each case (as discussed in Subsection 6.1.2.1). When the glazing area in-creases, however, (60% and 100% glazed alternatives) the potential number of hours outside the permissible air temperature limits may increase, due to the insuffi cient heating and/or cooling capacity. For example in the fi rst alternative (high thermal transmittance of the window system) and the fi fth case (double pane with low thermal transmittance and relatively high geffective value with internally placed blinds) the air temperature exceeds the limits set by the control set points. Although in most of the cases the number of working hours exceeding the temperature limits is lower than 5%, for the fi rst 100% glazed alternative the temperatures are lower for almost 10% of the working hours, while the opposite problem (high air temperatures) occurs in the fi fth alternative, due to the increased g and geffective values of the window. In Figure 6.44 the weighted average mean air temperatures of the fi rst and fi fth alternatives are compared. The temperature difference between these two cases is obvious, with the fi fth alternative exceeding the upper permissible air temperature limit during the summer months. The insuffi cient heating and cooling capacity are examined in greater detail on the zone level. The energy use for these two


210

cases (both cell and open plan) is recalculated (and presented in Appendix L) with increased heating and cooling capacity, in order to calculate the energy demand needed for each alternative. For the rest of the parametric study the original alternatives were used in order to study the impact of window type on the perception of thermal comfort.

21,0

21,5

22,0

22,5

23,0

23,5

24,0

24,5

25,0

25,5

26,0

26,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d av

erag

e m

ean

air

tem

pera

ture

(°C

)

Strict (1) Normal (1) Poor (1) Strict (5) Normal (5) Poor (5)

Figure 6.44 Weighted average air temperatures for the 1st and 5th 100% glazed alternatives.

6.2.2.2 Impact of window and shading type on the perception of thermal comfort for the 60% and 100% glazed alternatives

Since the poor control set points result in high PPD values (not accept-able according to ISO standard 7730, 1984), the perception of thermal comfort indices are studied only for the strict and normal set points of the two plan types. A comparison of the weighted monthly average PMV values and the number of hours with certain PPD values for the 60% and 100% alternatives follows, in order to study the infl uence of the façade elements on the perception of the thermal environment. A table with the number of hours with PPD values lower than 10% and 15% for the 3 control set points for the cell and open plan type 30%, 60% and 100% glazed buildings are presented in Appendix M.

When the thermal transmittance of the fi rst 60% glazed alternative is reduced (alternative 2 compared with alternative 1), the weighted monthly average PMV values improve during the heating season (for the strict set points, during January the weighted average PMV of the fi rst alternative


211

is -0.67, while for the second one it is -0.55; for the normal set points the weighted average PMV of the fi rst alternative is -0.67, while for the second one it is -0.51), as shown in Figure 6.45. During the summer months, however, the weighted average PMV values for the strict set points are around 0 and for the normal set points they are between 0.3 and 0.4.

-0,9

-0,8

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d m

onth

ly a

vera

ge P

MV

Strict set point (1) Strict set point (2) Normal set point (1) Normal set point (2)

Figure 6.45 Weighted monthly average PMV for the 1st and 2nd 60% glazed alternatives (strict and normal control set points).

For the 100% glazed alternatives the PMV values are reduced even more during the heating season (for the strict set points, during January the weighted average PMV of the fi rst alternative is -0.82, while for the second one it is -0.62; for the normal set points the weighted average PMV of the fi rst alternative is -0.81 while for the second one it is -0.59), as shown in Figure 6.46. During the summer months, however, the weighted monthly average PMV values for the strict set points are around 0.1 and for the normal set points they are between 0.4 and 0.5.


212

-0,9

-0,8

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d m

onth

ly a

vera

ge P

MV


Figure 6.46 Weighted monthly average PMV for the 1st and 2nd 100% glazed alternatives (strict and normal control set points).

The increase in glazed area results in a greater difference in the weighted average PMV values between the alternatives 1 and 2 mostly during the heating periods (for the strict set points of the 60% glazed alternative the difference is 0.12 during January, while for the 100% glazed one it is 0.2; for the normal set point the difference is 0.16 and 0.22 respectively). This can be partly explained by air temperature difference between the two alternatives but mainly it is caused by the difference in the radiant temperature (which could be expected from the difference in U value). The directed operative temperature is used, in order to study this effect. Since the difference in the weighted average directed operative temperatures between the fi rst and second alternative is larger than the weighted average air temperatures (Figure 6.47), it is evident that the impact of the radiant temperature is greater (the negative values show that the 1st alternative is colder than the 2nd).


213

-1,3-1,2-1,1-1,0-0,9-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,10,00,1

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Tem

pera

ture

diff

eren

ce (

°C)

Directed Operative Temp. Strict 1-2 Directed Operative Temp. Normal 1-2

Air Temp. Strict 1-2 Air Temp. Normal 1-2

Figure 6.47 Difference between weighted average directed operative temperature and air temperature for the 1st and 2nd 100% glazed alterna-tives.

A diagram with the number of hours with certain (weighted average) di-rected operative temperatures is shown in Figure 6.48. The lower thermal transmittance of the window of the second alternative increases the number of hours with directed operative temperatures closer to the maximum permissible temperature limit for each set point.


214

050

100150200250300350400450500550600650700750800

21.0

- 2

1.1

21.2

- 2

1.3

21.4

- 2

1.5

21.6

- 2

1.7

21.8

- 2

1.9

22.0

- 2

2.1

22.2

- 2

2.3

22.4

- 2

2.5

22.6

- 2

2.7

22.8

- 2

2.9

23.0

- 2

3.1

23.2

- 2

3.3

23.4

- 2

3.5

23.6

- 2

3.7

23.8

- 2

3.9

24.0

- 2

4.1

24.2

- 2

4.3

24.4

- 2

4.5

24.6

- 2

4.7

24.8

- 2

4.9

Directed operative temperature (°C)

Num

ber

of h

ours

Strict (1) Normal (1) Strict (2) Normal (2)

Figure 6.48 Number of hours with certain (weighted average) directed operative temperatures for 1st and 2nd 100% glazed alternatives.

For the 60% glazed buildings with strict control set points the (weighted average) PPD values are lower than 10% during 72% of the working hours for the second alternative, and during 61% of the working hours for the fi rst alternative (Figure 6.49). For the normal set points the number of hours decreases to 70% and 57% respectively. The number of hours with (weighted average) PPD values lower than 15% increases by up to 94% for the second and by 83% for the fi rst alternative (for the strict control set points). For the normal set point the number of hours is 94% and 82% for the two alternatives.


215

050

100150200250300350400450500550600650700750800

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%

30-3

2%

32-3

4%


Num

ber

of h

ours

Strict (1) Strict (2) Normal (1) Normal (2)

Figure 6.49 Weighted average PPD for the 1st and 2nd 60% glazed alterna-tives.

For the 100% glazed building, the (weighted average) PPD values are lower than 10% during 68% of the working hours for the second alternative, and during 46% of the working hours for the fi rst alternative for the cell type (Figure 6.50). For the normal set point the number of hours decreases to 55% and 31% respectively. The number of hours with (weighted aver-age) PPD values lower than 15% increases by up to 87% for the second and by 65% for the fi rst alternative (for the strict control set points). For the normal set point the number of hours is 84% and 60% for the two alternatives.


216

0

50

100

150200

250

300

350

400

450

500

550600

650

700

750

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%

30-3

2%

32-3

4%

34-3

6%

36-3

8%

38-4

0%


Num

ber

of h

ours


Figure 6.50 Weighted average PPD for the 1st and 2nd 100% glazed alterna-tives.

When the triple window is replaced by a double one (with a lower total solar transmittance) and the blinds are placed internally (third alternative), the weighted monthly average PMV values of the 100% glazed alternatives decrease (more for the normal than for the strict set points, as shown in Figure 6.51). Since the thermal transmittance of the glazing was kept the same, the difference in the PMV values is small during the winter months. The results for the 60% glazed buildings are similar.


217

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d m

onth

ly a

vera

ge P

MV


Figure 6.51 Weighted average PMV for the 2nd and 3rd 100% glazed alterna-tives.

For the 60% glazed building with strict set points, the (weighted average) PPD values are lower than 10% during 72% of the working hours for the second and during 68% of the working hours for the third alternative (Figure 6.52). For the normal set points the number of hours decreases to 70% and 70% respectively. The number of hours with (weighted average) PPD values lower than 15% increases by up to 94% for the second and by 93% for the third alternative (for the strict control set points). For the normal set points the number of hours is 94% and 93% for the two alternatives.


218

050

100150200250300350400450500550600650700750800

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%


Num

ber

of h

ours


Figure 6.52 Weighted average PPD for the 2nd 3rd 60% glazed alternatives.

For the strict control set points of the 100% glazed buildings, the (weighted average) PPD values are lower than 10% during 68% of the working hours for the second and during 62% of the working hours for the third alternative (Figure 6.53). For the normal set points the number of hours decreases to 55% and 57% respectively. The number of hours with (weighted average) PPD values lower than 15% increases by up to 87% for the second and by 84% for the third alternative (for the strict control set points). For the normal set point the number of hours is 84% and 82% for the two alternatives.


219

050

100150200250300350400450500550600650700750

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%

30-3

2%

32-3

4%

Wighted average PPD

Num

ber

of h

ours


Figure 6.53 Weighted average PPD for the 2nd and 3rd 100% glazed alterna-tives.

A further decrease and increase in the total solar transmittance of the fourth and fi fth alternatives respectively, results in lower and higher weighted average PMV values as shown in Figure 6.54. The weighted average PMV values of the fi fth alternative vary between the recommended limits ±0.5 (which correspond to lower than 15% of weighted average PPD values), but on the other hand they exceed the limits of ±0.3 (which correspond to lower than 10% of PPD values) for a longer time.


220

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d m

onth

ly a

vera

ge P

MV

Strict set point (3) Strict set point (4) Strict set point (5)

Normal set point (3) Normal set point (4) Normal set point (5)

Figure 6.54 Weighted average PMV for the 3rd, 4th and 5th 60% glazed alter-natives (strict and normal control set points).

For the 60% glazed building, the (weighted average) PPD values are lower than 10% during 66% of the working hours for the fourth and during 71% of the working hours for the fi fth alternative for strict set points (Figure 6.55). For the normal set points the number of hours increases for the fourth to 71% and drops for the fi fth to 65%. The fi fth alterna-tive gives lower PPD values for the strict set points, since the (weighted average) PMV is between ±0.3 between the beginning of April and the end of October, while for the fourth one the PMV varies between these limits between the end of April and the middle of October. For the normal set points, however, the PPD values are lower for the fourth alternative, since the high total solar transmittance (of the fi fth alternative) results in higher than +0.3 PMV values from the middle of May until the end of August, while for the fourth alternative they are lower than +0.3 during these months. The number of hours with PPD values lower than 15% increases by up to 92% for the fourth and by 93% for the fi fth alternative (for the strict control set points). For the normal set point the number of hours is 93% and 90% for the two alternatives. The number of hours with PPD values lower than 15% is lower in the fourth alternative mostly due to the higher directed operative temperatures during January, February and December (PMV higher than -0.5).

For the strict control set points of the 100% glazed buildings, the PPD values are lower than 10% during 60% of the working hours for the fourth, and during 64% of the working hours for the fi fth alternative for the cell


221

type (Figure 6.56). For the normal set points the number of hours decreases to 59% and 50% respectively. The number of hours with PPD values lower than 15% increases by up to 83% for the fourth and by 84% for the fi fth alternative (for the strict control set points). For the normal set point the number of hours is 83% and 78% for the two alternatives.

050

100150200250300350400450500550600650700750

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%


Num

ber

of h

ours


Figure 6.55 Weighted average PPD for the 4th and 5th 60% glazed alterna-tives.

0

50

100

150

200

250

300

350

400

450

500

550

600

650

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%


Num

ber

of h

ours


Figure 6.56 Weighted average PPD for the 4th and 5th 100% glazed alterna-tives.


222

The sixth and seventh alternatives have the same glazing as the third alter-native. Instead of internal blinds, the sixth alternative has internal screens and the seventh one fi xed external louvres. The effective g value of both alternatives is lower than in the third alternative, as described in Subsec-tion 5.2.1.2. The main difference between the alternative with the fi xed external louvres and the other two is that in the fi rst the shading devices are always applied, while in the others they are drawn for a set point of 100 W/m2 on the surface of the glass. As shown in Figure 6.57, the weighted monthly average PMV for the strict set points of the seventh alternative is quite low. However, the 6th alternative does not really differ from the 3rd one. For this reason, the sixth alternative is be studied further.

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Wei

ghte

d m

onth

ly a

vera

ge P

MV

Strict set point (3) Strict set point (6) Strict set point (7)

Normal set point (3) Normal set point (4) Normal set point (7)

Figure 6.57 Weighted average PMV for the 3rd, 6th and 7th 100% glazed alternatives (strict and normal control set points).

For the seventh 60% glazed alternative, the PPD values are lower than 10% during 68% of the working hours for the strict and during 71% for the normal set point. The number of hours with weighted average PPD values lower than 15%, increases by up to 93% for the strict and by 92% for the normal set point. For the strict control set points of the 100% glazed buildings, the PPD values are lower than 10% during 57% of the working hours for the strict and during 60% for the normal set point (Figure 6.58). The number of hours with PPD values lower than 15%, increases by up to 82% for the strict and by 83% for the normal set points respectively.


223

0

50

100

150

200

250

300

350

400

450

500

550

600

650

700

0-2%

2-4%

4-6%

6-8%

8-10

%

10-1

2%

12-1

4%

14-1

6%

16-1

8%

18-2

0%

20-2

2%

22-2

4%

24-2

6%

26-2

8%

28-3

0%


Num

ber

of h

ours

Strict (60%) Strict (100%) Normal (60%) Normal (100%)

Figure 6.58 Weighted average PPD for the 7th 60% and 100% glazed alterna-tives.

The fi rst glazed alternative (triple clear glazing) results in both high and low PMV values due to the high thermal and total solar transmittance. When the thermal transmittance of the windows (second alternative) increases, the heating demand drops and the (minimum) PMV values during the winter increase. The (maximum) PMV values during the summer months are almost the same, since the solar factor values were kept the same. When the solar factor of the glazing in the third alternative is reduced the PMV values drop during the summer, resulting in a lower number of hours with PPD up to 10% (for a PPD of 15% the number of hours for the 2nd and 3rd alternatives is almost the same). A further decrease in the solar factor of the fourth alternative brings similar results as before, increasing the number of dissatisfi ed occupants. When the effective total solar transmittance increases (as in the second alternative but with inner venetian blinds) overheating problem occurs. For the strict set points the fi fth alternative appears to provide a better thermal environment, while for the normal set points the temperature increases causing discomfort problems. The alternative with internal screens (sixth) provides thermal environment of quality similar to that in the third alternative. Finally, when the fi xed horizontal external louvres are applied (seventh alternative) the upper limit of 23°C appears to be very low, increasing the PPD values. For the normal control however the PMV increases, giving PPD values similar to those in the third alternative.


224

For strict set points the best performing alternative is the 2nd (since the high total solar transmittance “corrects” the mistaken, low upper air temperature limit of 23°C). For normal set points the alternatives 2, 3, 4, 6 and 7 perform similarly due to the low U, g and geffective values (PPD lower than 15% for 90% of the working hours). The 7th alternative, however, gives a larger number of hours with PPD values lower than 10% (which correspond to PMV values between ±0.3).

In general, the upper permissible air temperature of 23°C appears to be fairly low, since the monthly average PMV during summer hardly reaches the neutral conditions (= 0). The air temperature set points are crucial for the provision of improved thermal environment since (a) the air temperature variation infl uences the PMV variation and (b) the cor-rect selection of upper and lower permissible air temperature limits can place the PMV closer to the neutral condition (0) axis (in such a way that the PPD values will minimize). The large glazing area has a similar effect to the “less strict” set points since the large variation in the radiant temperatures will affect the operative temperatures, increasing the PMV variation. In order to minimize this effect, the use of “stricter” set points (with correct selection of upper and lower temperature limit) in highly glazed buildings is preferable.

6.2.3 Indoor climate on a zone levelDirected operative temperatures and thermal comfort indices on a zone level are studied in this section. The orientation of the zones is north, east, south and west for the strict, normal and poor control set points. The studied zones are double and single offi ce rooms, meeting rooms and corner offi ce rooms of the cell type 100% glazed offi ce building. The third building alternative (typical construction for a glazed facade) was chosen as a 100% glazed reference case for the comparisons on a zone level, in order to study the impact of glazing on each orientation. The potential overheating problem for zones with increased glazing area (i.e. meeting rooms with 65% of window to external wall area ratio) is discussed in Subsection 6.1.3.

6.2.3.1 Directed operative temperaturesZones with different set points and orientations were studied (third 100% glazed alternative), in order to investigate the impact of increased glazed area on the directed operative temperatures.

The number of working hours with certain directed operative tempera-tures of the 100% glazed meeting rooms is presented in Figure 6.59 (strict


225

set points). The orientations selected are the north and south (the east and west vary in between). The minimum directed operative temperatures are almost 21ºC for all orientations; the maximum values, however, differ by 3ºC (24.5ºC for the north and 27.5ºC for the south facing meeting room). For the normal control set points the directed operative temperature varies from 21ºC to 31ºC for the south oriented meeting room with 9% of the working time being higher than 27.5ºC. For the north oriented meeting room the directed operative temperature varies from 21ºC to 26ºC.

05

10152025303540455055606570

21.2

- 2

1.3

21.5

- 2

1.6

21.8

- 2

1.9

22.1

- 2

2.2

22.4

- 2

2.5

22.7

- 2

2.8

23.0

- 2

3.1

23.3

- 2

3.4

23.6

- 2

3.7

23.9

- 2

4.0

24.2

- 2

4.3

24.5

- 2

4.6

24.8

- 2

4.9

25.1

- 2

5.2

25.4

- 2

5.5

25.7

- 2

5.8

26.0

- 2

6.1

26.3

- 2

6.4

26.6

- 2

6.7

26.9

- 2

7.0

27.2

- 2

7.3

27.5

- 2

7.7

27.8

- 2

7.9


Num

ber

of w

orki

ng h

ours

Meeting room-North Meeting room-South

Figure 6.59 Number of working hours between certain directed operative tem-peratures for the 100% glazed meeting room (strict control set points, 3rd alternative).

The impact of window to fl oor area ratio on the directed operative tem-peratures is presented in Figure 6.60 for a south oriented double and corner offi ce zone with strict set points. For the double offi ce with one external wall the directed operative temperature starts at 21ºC and it ex-ceeds 27ºC only 1% of the time. For the corner offi ce, however (with the external wall area twice than in the double offi ce) the directed operative temperature is lower than 21ºC for 25% and higher that 27ºC for 8% of the working time. For the normal control set points, the directed operative temperatures vary even more.


226

0

10

20

30

40

50

60

70

80

90

100

…-2

0.1

20.2

- 2

0.3

20.5

- 2

0.6

20.8

- 2

0.9

21.1

- 2

1.2

21.4

- 2

1.5

21.7

- 2

1.8

22.0

- 2

2.1

22.3

- 2

2.4

22.6

- 2

2.7

22.9

- 2

3.0

23.2

- 2

3.3

23.5

- 2

3.6

23.8

- 2

3.9

24.1

- 2

4.2

24.4

- 2

4.5

24.7

- 2

4.8

25.0

- 2

5.1

25.3

- 2

5.4

25.6

- 2

5.7

25.9

- 2

6.0

26.2

- 2

6.3

26.5

- 2

6.6

26.8

- 2

6.9

27.1

- 2

7.2

27.4

- 2

7.5

27.7

- 2

7.8


Num

ber

of w

orki

ng h

ours

Double-South Corner South-West

Figure 6.60 Number of hours between certain directed operative temperatures for the 100% glazed south oriented zones (strict control set points, 3rd alternative).

From the above Figures it is evident that, in order to keep the directed operative temperatures at reasonable levels, strict set points should be ap-plied in zones with fully glazed external walls (see number of hours above 24.5°C in Figure 6.59). For corner offi ce zones (with increased external wall area to room volume ratio) not even strict set points can ensure directed operative temperatures within accepted levels (see number of hours above 28°C in Figure 6.60).

6.2.3.2 Perception of thermal comfortIn this part the glazing and type of shading device were also varied (while for the study of the directed operative temperature only the third case was considered), in order to investigate the thermal comfort indices of different zones.

Beginning with the third widow alternative (considered as reference for the 100% glazed building cases) with strict set points, the monthly average PMV value varies for different zone types, as shown in Figure 6.61. The wider monthly PMV variations occur for the corner offi ces due to the increased external wall area to room volume ratio. The values for the southwest oriented corner offi ce vary from -1 to 0.4, while for the


227

rest they vary from -0.6 to 0.2. For the rest of the glazing alternatives the tendency is similar.

-1,0

-0,9

-0,8

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

Corner (southwest) Double (southwest) Single (southwest) Meeting room (southeast)

Figure 6.61 Monthly average PMV for zones of the third 100% glazed building with strict set points.

Since the more extreme monthly average PMV values occur for the corner offi ce rooms, mainly this zone type was considered for further comparison. In the fi rst 100% glazed case, with three clear panes and intermediate venetian blinds, the high thermal transmittance of the windows resulted in insuffi cient heating capacity. Thus, the case was simulated once more with increased heating and cooling capacity, in order to determine the energy use for keeping the air temperature in the working areas within the permissible set point limits. A comparison of the double and corner offi ce zones with normal and increased heating capacity is presented in Figure 6.62. The lack of suffi cient heat capacity results in differences in PMV values only for the corner offi ces, since the values for the double offi ce are completely identical. During the winter months the difference reaches 0.8 (monthly average PMV of -1.7 during December for the alternative with real heat capacity), while during summer the values are the same, as expected.


228

-1,8

-1,6

-1,4

-1,2

-1,0

-0,8

-0,6

-0,4

-0,2

0,0

0,2

0,4

0,6

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

Corner (real h.c.) Double (real h.c.) Corner (incresed h.c.) Double (incresed h.c.)

Figure 6.62 Monthly average PMV for the corner and double offi ce of the fi rst glazing case with real and increased heating capacity (100% glazed building, strict set points, southwest oriented zones).

The opposite problem (insuffi cient cooling capacity) is noticed in the fi fth alternative due to the high solar factor (and also high geffective) of the window. The corner and double offi ces (case with real and increased cooling capacity) are compared in Figure 6.63. Once more the insuffi cient cooling capacity is considered only for the corner zones since the monthly average PMV values for the double offi ce are identical. For the corner zone, however, the monthly average PMV reaches 1.1 with real and 0.6 with increased cooling capacity. There is also a small difference in PMV values during spring and autumn, since cooling is needed occasionally due to the strict control.


229

-1,0

-0,8

-0,6

-0,4

-0,2

0,0

0,2

0,4

0,6

0,8

1,0

1,2

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

Corner (real c.c.) Double (real c.c.) Corner (incresed c.c.) Double (incresed c.c.)

Figure 6.63 Monthly average PMV for the corner and double offi ce of the fi fth glazing case with real and increased cooling capacity (100% glazed building, strict set points, southwest oriented zones).

In order to study the impact of window and shading device type on the perception of thermal comfort, the highly glazed corner offi ces with southwest facing alternatives (1-7) were compared (Figure 6.64). When the heating capacity is increased, the fi rst alternative with the triple clear pane gives quite high monthly average PMV values during the summer due to the high thermal transmittance of the windows (otherwise the PMV values are very low). The second alternative is the second warmest due to the high solar factor of the glazing; g=0.584 and the lower (com-pared with the fi rst alternative) thermal transmittance. The only alterna-tive warmer than the second one is the fi fth. In this alternative the same glazing was used while the intermediate blinds were replaced by internal ones, increasing the effective solar factor and thus the directed operative temperatures and the PMV values (as in the fi rst alternative, the case with increased cooling capacity was chosen; otherwise the PMV values would have been much higher as shown in Figure 6.62). The monthly average PMV values of the fourth alternative are much lower than those of the third one due to the lower solar factor of the glazing (g=0.28 instead of g=0.35). The sixth alternative, with the same total solar transmittance as the third one, is slightly warmer due to the internal screens (lower geffective than the internal blinds). Finally, the seventh alternative with the fi xed external louvres is colder throughout the year due to the fi xed external louvres (low geffective all year round).


230

-1,0

-0,8

-0,6

-0,4

-0,2

0,0

0,2

0,4

0,6

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

1st alt. 2nd alt. 3rd alt. 4th alt. 5th alt. 6th alt. 7th alt.

Figure 6.64 Monthly average PMV for 100% glazed southwest corner offi ces (strict set points).

The impact of set points on the perception of thermal comfort (number of hours with PPD lower than 10% and 15%) for different zones is pre-sented in detail in Appendix M. In general the number of hours with PPD lower than 15% is somewhat lower when normal set points are applied as shown in Figure 6.65. A larger difference between the two set points can be noticed in the “warmer” alternatives (alternatives with higher PMV values), since the higher permissible temperature limits combined with the higher g and geffective values cause higher discomfort.

The highest percentage of working hours with PPD lower than 15% for normal set points occurs for alternative 7 (with low geffective), while alternatives 2, 4 and 6 have similar values. All these alternatives have low geffective and U values. For the strict set points the result is slightly different and the percentages are all higher. The poorest alternatives are the ones with high U and geffective values.


231

74

9087 87

82

88 86

70

8582

85

75

8587

0

10

20

30

40

50

60

70

80

90

100


% n

umbe

r of

hou

rs w

ith

PPD

low

er th

an 1

5%

strict normal

Figure 6.65 Percentage of working hours with PPD lower than 15% for the southwest corner offi ce zones.

When the percentage of working hours with PPD values lower than 10% is examined (for the corner offi ces with strict and normal set points), the positive effect of narrow air temperature variation is evident (Figure 6.66). The percentage difference in this case increases which shows that “quality wise” the strict set points are essential for zones with large glazing areas, such as corner offi ces (since PPD lower than 10% corresponds to PMV values between ±0.3, while PPD lower than 15% corresponds to PMV values between ±0.5).


232

37

49

4246

34

46 44

24

3230

33

25

31

42

0

10

20

30

40

50

60

70

80

90

100


% n

umbe

r of

hou

rs w

ith

PPD

low

er th

an 1

0%

strict normal

Figure 6.66 Percentage of working hours with PPD lower than 10% for the southwest oriented corner offi ce zones.

A similar comparison of the percentage of working hours with PPD lower than 10% for strict and normal set points for the double offi ces (Figure 6.67) shows that the thermal environment improves drastically due to the lower ratio of external wall area to room volume. Since the difference (in % of working hours with PPD lower than 10%) between strict and normal set points decreases, it is also evident that the importance of nar-rower permissible air temperature variation (such as in strict set points) increases as the glazing area increases.


233

57

71

65 6462

6662

47

59 5962

53

61

68

0

10

20

30

40

50

60

70

80

90

100


% n

umbe

r of

hou

rs w

ith

PPD

low

er th

an 1

0%

strict normal

Figure 6.67 Percentage of working hours with PPD lower than 10% for the southwest oriented double offi ce zones.

6.3 Double skin façades

6.3.1 Simulations on a component level (pilot study using WIS 3)

Before the results obtained by the pilot study are analyzed, it would be useful to briefl y describe the construction types on which the WIS 3 simu-lations are based. The simulations were carried out for a box window and a multi storey façade as shown in Figure 6.68. For the “standard” double façade mode during summer the air is extracted through the cavity to the outside either naturally (A-B-C) or mechanically (A-B-D-C); during winter the air remains in the cavity for increased thermal insulation, or after preheated in the cavity it is used as inlet supply air for the AHU (A-B-D-E) respectively. For the airfl ow window cases the indoor air enters the cavity and is driven (through the cavity) to the AHU (E-B-D); during the heating season heat is recovered while during summer the air is just conducted to the outside.


234

E B

D

AHU

C

A

E

A

D

AHU B

C

B

B

Figure 6.68 Multi storey and box window constructions.

6.3.1.1 Pre study: reducing the number of “standard” double façade alternatives

Seven glazing alternatives were initially considered for the “standard” dou-ble façade and airfl ow window modes, as described in Subsection 4.1.3.1. A pre study, however, was carried out in order to reduce the number of glazing alternatives as described in this section. The main parameter studied, in order to achieve this is the temperature of different layers at the horizontal and vertical centres of the cavity of the double skin façade.

For this study a box window 3.5 m high and 800 mm deep was as-sumed. Cases with opened and closed cavity opening were considered; for the cases in which the cavity is opened no dampers are applied (100% opened cavity). The air inside the cavity is naturally ventilated and the only driving force assumed is the thermal buoyancy (wind effects are neglected). The air enters the cavity from the outdoors and leaves to the outdoors. The shading devices (when applied) are assumed at the horizontal centre of the ventilated cavity.

The simulations were carried out (see also Table 4.1 in Subsection 4.1.3.1) for a:

• typical summer day (open cavity - with and without shading de-vices)

• extreme summer day (open cavity - with and without shading devices)

• winter day (open – closed cavity - without shading devices)


235

The properties of the “standard” double façade glazing systems are de-scribed in Table 6.1 and in more detail (e.g. visual transmittance) in Appendix I.

Table 6.1 Properties of the “standard” double façade glazing systems as calculated by WIS 3.

DSF

Cas

e

Ext

erna

l pa

ne

Gap

(80

0mm

)

Inte

rmed

iate

pa

ne

Gap

(12

mm

)

Inte

rnal

pan

e

U v

alue

gla

zing

sy

stem

W/m

²K,

clos

ed c

avit

y

U v

alue

inn

er

skin

W/m

²K

g va

lue

Tso

l

A Clear pane 8mm

Ventilated cavity

Clear pane 4mm Air

Clear pane 4mm 1.93 2.87 0.627 0.53

B Clear pane 8mm

Ventilated cavity

Clear pane 4mm Argon

Low E Coated 4mm

1.15 1.46 0.551 0.447

C Clear pane 8mm

Ventilated cavity

Optigreen (solar control tinted) 6mm

Argon Clear pane

4mm 1.85 2.73 0.516 0.326

D

Optigreen (solar

control tinted) 8mm

Ventilated cavity


Clear pane 4mm 1.85 2.74 0.404 0.279

E

Optigreen (solar

control tinted) 8mm

Ventilated cavity


Low E Coated 4mm

1.15 1.46 0.354 0.264

F Clear pane 8mm

Ventilated cavity

Solar control + lowE (soft coated) 6mm

Argon Clear pane

4mm 1.04 1.31 0.301 0.15

G

Solar control +lowE (hard

coated) 8mm

Ventilated cavity


Low E Coated 4mm

1.14 1.46 0.443 0.335

The temperatures at the vertical and horizontal centres of different layers, for an opened cavity without shading during a typical summer day, were calculated as shown in Figure 6.69. In these cases it has to be noted that, since the cavity is ventilated, the thermal transmittance of the system (for each alternative) is the thermal transmittance of the inner skin (see Table 6.1). In order to compare the simulated alternatives, the fi rst case (A) with 3 clear panes is considered as a reference case. When the low E pane replaces the inner clear one (case B), a slight increase in tempera-ture (1°C, from 28°C to 29°C) can be noticed at the inner pane, due to a substantial decrease in the thermal transmittance of the inner skin. In the third case (C) the intermediate solar control pane results in an increase


236

of almost 8°C (at the intermediate pane) and almost 3°C (to 31°C) at the inner one (when compared with the reference case, due to the increased indirect solar transmittance). When the solar control (body tinted) pane is placed as the outer one (case D and E), the increase in temperature (at this pane) is approximately 8°C. When these two cases are compared, it is evident that the low g values of the outer skin reduce the effect of low thermal transmittance (see U values of the inner skin for the cases D and E), resulting in slightly lower inner pane temperatures (compared with the inner pane temperatures of the cases A and B with clear outer pane). This can be explained as being due to the larger proportion of solar radiation that is absorbed at the outer skin. In the sixth case (F) only one advanced intermediate pane (with solar control and low E coating) was applied, resulting in an inner pane temperature 4°C higher than the indoor air temperature (and similar ones as the reference case). Both cases C and F have solar control intermediate panes. When these two cases are compared, however, the temperature of the intermediate pane is higher for the case F due to the low E coating (lower thermal transmittance). The position (facing to the inside) of the low E coating for the case F is crucial for the inner pane temperature. The additional insulation that the low E coating provides allows the inner pane to maintain lower temperatures (similar to the reference case). Finally, when a solar control and low E coated pane is placed as external pane, the temperature of the external pane increases as expected. The low E coating at the inner pane has similar effect. Due to the well ventilated cavity the cases (G) and (E) perform in a similar way.

Thus the lowest inner surface temperature was achieved with alternative D, because of a low g-value, not the lowest, and a high U value (allowing the heat to be transmitted from indoors to the cavity). However, the inner pane temperature of the case E is similar due to the same low total solar transmittance of the outer skin; the impact of lower thermal transmittance, on the other hand, is limited.


237

20

22

24

26

28

30

32

34

36

38

40

42

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ven

tila

ted

cavi

ty

bord

er

Inte

rmed

iate

pan

e (c

entr

e)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)

A1 B1 C1 D1 E1 F1 G1

Figure 6.69 Calculated temperatures for different layers at the vertical and horizontal centres of a naturally ventilated double skin façade for a typical summer day, opening depth: 800mm – no shading devices.

When shading devices are applied inside the cavity, part of the transmitted radiation is absorbed increasing the temperature inside the cavity. Accord-ing to the existing literature (Poirazis, 2006), suffi cient heat extraction can ensure low air and inner surface temperatures. The infl uence of the air temperature on the inner pane’s temperature is studied later in this Subsec-tion. In Figure 6.70 the temperatures of the different layers are presented for the different glazing alternatives during a typical summer day, when the movable solar shading in the cavity is applied. When the cases with and without shading devices are compared, an increase in the cavity air temperature is obvious. As shown in Figure 6.70 the inner layer’s tempera-tures are only slightly higher than the indoor air temperatures due to the decreased geffective value of the system and the heat extraction through the cavity. The tendency for the different glazing alternatives is similar to the case without the shading devices, but at a smaller scale. The results for an extreme summer day are similar (yet more signifi cant).


238

20

22

24

26

28

30

32

34

36

38

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ext

erna

l sub

cavi

ty (

cent

re)

bord

er

Shad

ing

devi

ce (

cent

re)

bord

er

Inte

rnal

sub

cavi

ty (

cent

re)

bord

er

Inte

redi

ate

pane

(ce

ntre

)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)

A1s B1s C1s D1s E1s F1s G1s

Figure 6.70 Calculated temperatures for different layers at the vertical and hori-zontal centres of a double façade for a typical summer day, opening depth: 800mm – shading devices.

During winter, the cavity of the “standard” double façade mode is closed, in order to reduce the system’s thermal transmittance and hence the heating demand. At other times (usually in mechanical ventilated cavities) the air is preheated before reaching the Air Handing Unit (AHU). Furthermore, in the existing literature (Poirazis, 2006) buildings with fully opened cavities during all the year are also mentioned. In this case the inner skin is crucial for the performance of the double façade.

For a ventilated cavity without shading devices, as shown in Figure 6.71, the cases with low E inner pane (B, E, F and G) are the most appropriate in terms of surface temperatures, as the U value of the inner skin is the lowest. The cases with clear inner and intermediate panes are the ones with the lowest inner surface temperatures, as expected, since they have the highest U values.


239

02468

1012141618202224

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ven

tila

ted

cavi

ty

bord

er

Inte

rmed

iate

pan

e (c

entr

e)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)

A3 B3 C3 D3 E3 F3 G3

Figure 6.71 Calculated temperatures for different layers at the vertical and hori-zontal centres of a double façade for a winter day, opening depth: 800mm – no shading devices.

When the cavity is closed, the temperatures at the vertical and horizontal centres of each layer for cavities with closed openings are shown in Figure 6.72. The inner surface temperature of the reference case is slightly above 20°C, while the air temperature inside the cavity (centre) is approximately 10°C. The inner surface temperature of the second case increases due to the low E coating (lower U value). The air temperatures, however, are lower than in the reference case, since the insulation of the low E coating lowers the heat fl ow from indoors to the cavity. The solar control body tinted intermediate pane of the third alternative (C) results in high air temperatures in the cavity and inner surface temperatures slightly above 23°C. The curve of the sixth alternative (F) with the intermediate solar control and low E coating is very similar. The main reason for that is the high absorption of the body tinted pane. Since the temperatures of the inner and intermediate panes are very similar, the insulation that the low E coating provides seems not to be needed; however, in cases with lower solar gains, the increased thermal insulation that the low E pane (case F) provides can lead to reduced thermal losses. When the solar control body tinted pane is placed as external pane and the inner and intermediate panes are clear ones (case D), the air temperatures are slightly below 16°C and the inner surface temperatures slightly above 20°C. When a low E coating is placed at the inner pane (case E) the air temperature decreases and the inner surface temperature increases (similar effect with cases A and B).


240

Finally, for the case (G) due to the solar control and low E external pane the air temperature is 11°C, and due to the inner low E coated pane the inner surface temperature is around 23°C. The two low E coated panes (external and inner ones) of the latter case result in a low temperature inside the cavity but a suffi cient one at the inner pane.

The highest inner surface temperature is obtained for cases C and F, due to the intermediate solar control pane which leads to higher indirect solar transmittance. Lower inner pane temperatures occur in the cases A and D with high thermal transmittance values.

02468

101214161820222426

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ven

tila

ted

cavi

ty

bord

er

Inte

rmed

iate

pan

e (c

entr

e)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)

A3 B3 C3 D3 E3 F3 G3

Figure 6.72 Calculated temperatures for different layers at the vertical and hori-zontal centres of a double façade for a winter day, opening depth: 0 mm – no shading devices.

During night the best performing alternatives are the ones with the lowest U value, since the g value does not have any impact on the inner pane temperatures. The inner pane temperatures during night are of no interest (the building is occupied during day), so no further study was carried out on a component level.

For the “standard” double façade the alternatives chosen were:

• Case A: Selected as a reference case. The fi rst case has the highest thermal transmittance (1.93 W/m2K) (when the cavity is closed) and the highest total solar transmittance (0.627).

• Case D: This case has the solar control body tinted pane as an exterior layer and clear intermediate and inner panes. The thermal transmittance of this alternative is 1.85 W/m2K (closed cavity) and the total solar transmittance is 0.404. This case was chosen as a fi rst improvement of the case A by adding a solar control pane.


241

• Case E: In this case both a solar control body tinted (external) and a low E pane (intermediate) was applied. In this way the temperatures at the inner layer are kept within reasonable levels. The thermal transmittance for this alternative is 1.15 W/m2K (closed cavity) and the total solar transmittance 0.354. This is a second improvement by adding a low E intermediate pane to increase the insulation.

• Case F: This alternative has the lowest thermal (1.04 W/m2K with closed cavity) and total solar transmittance (0.301). Only the intermediate pane is an advanced one (low E and solar control coat-ings), while the inner and external panes are clear. This alternative has the same coatings as the case E but their position changes; as a result, the intermediate advanced pane results in increased indirect transmittance.

6.3.1.2 Parametric study: infl uence of cavity geometry on system performance

In this case a multi storey standard double façade cavity (with different height and depth) was chosen, in order to study the airfl ows and the vertical air temperature profi le along the ventilated cavity. The parameters varied were the cavity height and depth, the size of openings and the position of shading device inside the cavity.

The cavity depth varied from 200 to 1600 mm. In the cases with fully open cavity it could be expected that the depth of the openings is equal to the depth of the cavity. However, since the dampers occupy space (see Subsection 5.3.1.3) the opening depth to cavity depth percentage is pre-sented in Table 6.2.

Table 6.2 Cavity and opening depth for the multi storey façade (even depths).

Cavity depth (mm) Opening depth (mm) % of opened area

200 173 87 400 346 87 400 346 87 600 519 87 800 692 87 1000 865 87 1200 1038 87 1400 1211 87 1600 1384 87


242

Infl uence of cavity height and depth on the airfl ows The height of the façade infl uences the airfl ow inside the naturally ven-tilated cavity and consequently the air temperatures. Initially the fi rst double façade case with three clear panes (case A) was considered, in order to study the airfl ow rate for 10 m, 20 m and 30 m high cavities of different depths.

The air velocity inside the cavity (10 m, 20 m and 30 m high) is shown in Figure 6.73. As expected, the higher cavities result in higher air velocities (for the same cavity depth), since the stack effect is stronger.

0,0

0,1

0,2

0,3

0,4

0,5

0,6

0,7

0,8

0,9

1,0

1,1

1,2

1,3

1,4

1,5

1,6

0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6

Cavity depth

Air

vel

ocit

y (m

/s p

er m

cav

ity

wid

th)

10m 20m 30m

Figure 6.73 Calculated air velocity for 10, 20 and 30m high multistorey facades (case A, cavity depths 0.2-1.6 m, typical summer day).

The airfl ow rate (l/s per metre cavity width) for the 10 m, 20 m and 30 m high cavities is shown in Figure 6.74. As expected, the wider cavities (with larger openings) provide larger airfl ows (absolute numbers). A comparison of the airfl ow per meter of cavity height, however, shows that shorter cavi-ties can ventilate more effi ciently, since as shown in Figure 6.75 the airfl ow rate per metre of cavity height is larger. However, the total air fl ow will increase with the height of the cavity.


243

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6

Cavity depth (m)

Air

flow

(l/

s pe

r m

cav

ity

wid

th)

10m 20m 30m

Figure 6.74 Calculated air fl ows as a function of cavity depth for 10, 20 and 30m high multistorey façades (case A).

0

5

10

15

20

25

30

35

40

45

50

55

60

0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6Cavity depth (m)

Air

flow

(l/

s pe

r m

cav

ity

heig

ht a

nd w

idth

)

10m 20m 30m

Figure 6.75 Calculated air fl ows per metre of cavity height as a function of cavity depth for 10, 20 and 30m high multistorey facades (case A).

The diagram in Figure 6.76 shows the differences in airfl ow rates between cavities 10 and 20 m high and 20 and 30 m high façades.


244

0

1

2

3

4

5

6

7

8

9

10

0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6

Cavity depth (m)

Air

flow

diff

eren

ce (

l/s/

m c

av. h

eigh

t and

wid

th)

10-20m 20-30m

Figure 6.76 Calculated airfl ow differences (per m of cavity height) between cavi-ties 10 and 20 m high and 20 and 30 m high.

It can be noticed that:

• The airfl ow difference is higher in the fi rst case than in the second one independently of the cavity depth (since the fi rst curve is higher than the second one). This means that the more the cavity height increases the smaller is the infl uence of the height on the airfl ow.

• The airfl ow difference increases when the cavity depth increases (since the deeper the façade, the larger the difference). This means that the infl uence of cavity depth is smaller in higher cavities.

Infl uence of cavity height and depth on the temperature profi le along the cavityThe air temperature profi le inside a 10 m high cavity (case A, typical summer, no shading devices) is presented in Figure 6.77. The cavity is assumed fully opened (87%). As expected, the wider cavities provide lower air temperatures at the top of the cavity due to the larger airfl ows. The air temperature differences decrease as the cavity depth increases. In this case, as noticed in Figure 6.77, when the cavity gets wider than 1 m, the temperature differences along the cavity become very small.


245

20,0

20,5

21,0

21,5

22,0

22,5

23,0

23,5

24,0

24,5

25,0

0 1 2 3 4 5 6 7 8 9 10Cavity height (m)

Air

tem

pera

ture

(°C

)

0.2m 0.4m 0.6m 0.8m 1m 1.2m 1.4m 1.6m

Figure 6.77 Calculated air temperature as a function of height along a 10m high cavity for different cavity depths (case A, typical summer, no shading devices).

The results for a 20 and a 30 m high cavity are similar as shown in Figure 6.78.

20

21

22

23

24

25

26

27

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Cavity height (m)

Air

tem

pera

ture

(°C

)

0.2m 0.4m 0.6m 0.8m 1m 1.2m 1.4m 1.6m

Figure 6.78 Calculated temperature in the cavity as a function of height along a 30m high cavity for different cavity depths (case A1, typical summer, no shading devices).


246

When Figures 6.77 and 6.78 are compared, it appears that, independently of the cavity depth, the air temperature at the 10th metre of a 10 m high cavity is higher than that at the 10th metre of a 30 m high one. This can be explained by the stronger stack effect that takes place in the taller cav-ity. In terms of temperature profi le the cavity height is not very important for fi nding the optimal cavity depth, since similar results can be obtained regardless of the cavity height. Optimal cavity depth can be considered the minimum depth after which the air temperature decrease drops, ensuring suffi cient heat extraction. However, this may vary for different weather conditions and glazing combinations.

Another interesting result is the non linear curve type of the 0.2 m deep cavity (for 10 m high cavity) and the curves of the 0.2 and 0.4 m deep cavities (for the 30 m high cavities). Due to the lower airfl ows, the narrower the cavity (of a certain height), the more intense is the air temperature increase at lower heights. The tendency as the cavity height increases is similar.

Larger airfl ows result in a lower rise in air temperature in the cavity. The air in the cavity is heated by energy losses from the inside and solar radiation from the outside, absorbed in the glass panes surrounding the gap. If the fl ow is low, the gap air temperature approaches the average temperature of the surrounding panes at a lower height than if the airfl ow is high. The temperature rise will then be curved as seen in Figure 6.78.

Infl uence of percentage of opened area on the airfl ows and the temperature profi le along the cavityDue to the construction type of dampers, when they are fully opened, the % of opened area is 87% for even opening size (0.2 m, 0.4 m, 06 m, etc) but it varies for odd ones. In Table 6.3 the opening area for different cavity depths (0.5, 0.7 and 0.9 m) is presented.

Table 6.3 Cavity and opening depth for the multistorey façade (even and odd depths).

Cavity depth (mm) Opening depth (mm) % of opened area

400 346 87 500 346 69 600 519 87 700 519 74 800 692 87 900 692 77


247

The airfl ow difference between the 0.4 and 0.5 m (20m high) cavity is almost negligible as it is for the 0.6 and 0.7 and for the 0.8 and 0.9 m deep ones (cavities with the same opening size). The small differences are caused by the higher fl ow resistance of the narrower cavities. Due to the larger cavity depth, however, the air velocity is higher at the even cavities as shown in Figure 6.79. The larger differences in the air velocities in narrower cavities can be explained by the higher fl ow resistance of the narrower cavities.

0,000,050,100,150,200,250,300,350,400,450,500,550,600,650,700,750,800,850,900,951,00

0,4 0,5 0,6 0,7 0,8 0,9

Cavity depthair velocity (m/sec per m width) airflow (m3/sec per m width)

Figure 6.79 Calculated cavity air velocity and airfl ow for different cavity depths for a 20 m high multistorey facade (case A, cavity depths 0.4-0.9 m, typical summer day, no shading).

The higher air velocity in the cavities with even depth results in lower air temperatures, since the air can be extracted easier. The temperature profi le along a 20 m high cavity is shown in Figure 6.80. The temperature difference between the cavities with even and odd depths decreases as the depth increases.


248

20,0

20,5

21,0

21,5

22,0

22,5

23,0

23,5

24,0

24,5

25,0

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Cavity height (m)

Air

tem

pera

ture

(°C

)

0.4m 0.5m 0.6m 0.7m 0.8m 0.9m

Figure 6.80 Calculated air temperature as a function of height for different cavity depths for a 20 m high multistorey facade (case A, no shading).

It can be concluded that:

• The opening size is very crucial for the size of the air fl ows. Airfl ows in cavities with the same opening size are almost the same (inde-pendently of the cavity depth). The air temperatures however tend to increase more in narrower cavities due to the lower air velocity.

Infl uence of glazing on cavity depthIn this section the infl uence of glazing type on the airfl ow and tempera-ture profi le along a 20 m high cavity are studied. For a typical summer day the cavity depths studied were 0.6, 0.8, 1 and 1.2 m, while for an extreme summer day the cavity depths considered were 0.6, 0.8, 1, 1.2, 1.4 and 1.6 m.

The infl uence of glazing type on the optimal depth (minimum depth for effi cient heat extraction) is presented in Figure 6.81. In this case only the temperatures at the exit were considered for an extreme summer day. As expected, due to the lower absorptivity, the case with the three clear panes results in the lowest temperatures at the exit of the cavity. The cases with the solar external pane (with clear or low E coated internal pane) are somewhat warmer and the one with the highest temperatures is the one with the intermediate advanced coating (low E and solar control).


249

From the Figure below, it can be concluded that the solar control pane increases the need for wider cavities, in order to achieve effi cient heat extraction (air temperature difference for 0.6 m is higher than the one for 1.6 m); this effect is stronger, when the solar control pane is placed as intermediate one (case F instead of outer pane in cases D and E), since a larger part of the absorbed heat is transmitted into the cavity.

32

33

34

35

36

37

38

39

40

0,6 0,8 1,0 1,2 1,4 1,6

Cavity depth (m)

Air

tem

pera

ture

at e

xit (

°C)

case A case D case E case F

Figure 6.81 Temperatures at the exit of a 20 m high cavity during an extreme summer day (cases A, D, E, F, no shading).

When shading devices are applied the vertical air temperature profi le is calculated in both the external (1) and internal (2) sub cavity. The external sub cavity is the one between the external pane and the shading devices and the internal one is that between the shading devices and the intermediate pane. At this point the shading devices are assumed to be placed in the middle of the ventilated cavity. For the case with the three clear panes (case A) the air temperatures in the two sub cavities are very similar as shown in Figure 6.82. The results are also very similar for a typical summer day. With shading the cavity air temperature is higher, but less solar energy enters the room behind the double skin façade.


250

30

31

32

33

34

35

36

37

38

39

40

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Cavity height (m)

Air

tem

pera

ture

(°C

)

ext 0.6m int 0.6m ext 0.8m int 0.8m

ext 1m int 1m ext 1.2m int 1.2m

Figure 6.82 Air temperature profi le along the internal and external sub cavity for different cavity depths for an extreme summer day (case A, shad-ing).

The vertical air temperature profi le was calculated also for the cases D, E and F. In order to reduce the number of simulations, only a 0.8 deep cavity was assumed. The difference in the air temperature between the internal and external sub cavity is minimum for the case with the three clear panes (case A), as shown in Figure 6.83. The difference in the case with the advanced intermediate pane (solar control and low E) is slightly larger, since the external pane is a clear one. Finally, the sub cavities in the cases with the solar control outer pane (cases D and E) show the largest air temperature difference.

The solar control outer pane performs best during a summer day, since the temperatures at the external sub cavity are high and at the inner one low (when shading is used). During winter, however, shading is off and the solar control outer pane gives rise to high temperatures inside the cavity (as wanted).


251

30

31

32

33

34

35

36

37

38

39

40

41

42

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

Cavity height (m)

Air

tem

pera

ture

(°C

)

case A ext case A int case D ext case D intcase E ext case E int case F ext case F int

Figure 6.83 Air temperature profi le along the internal and external sub cavity for the cases A, D, E and F for an extreme summer day (depth 0.8m, shading).

Infl uence of the position of shading device on the air temperaturesIn the existing literature (Poirazis, 2006) it is mentioned that the position of shading devices inside the cavity can have a relatively large impact on the energy and on the thermal performance of double skin façades. In order to investigate the impact of the position of shading device on the inner layer temperatures, a 20 m high and 0.8 m deep cavity was selected and the cases A, D, E and F were simulated for an extreme summer day.

In Figure 6.84 the air temperature (at the vertical centre) of the inter-nal sub cavity for different depths, and the inner layer temperatures for the cases A, D, E and F, are presented. The main aim of this comparison was to investigate the infl uence of shading device position on the thermal comfort. As shown in Figure 6.84, the variation in air temperature for the internal sub cavity ranges from 5°C (case D) to 9°C (case F) depending on the depth. However, the variation in the inner pane’s surface temperature is much smaller for all glazing alternatives. From the above it can be con-cluded that in the naturally ventilated cavities (with double pane glazing unit as inner skin) the position of shading devices is not important for the thermal comfort during the summer unless the inner skin is openable (box window cases) for natural ventilation purposes.


252

20

22

24

26

28

30

32

34

36

38

40

42

case A case D case E case F case A case D case E case F

air temperatures (°C) inner surface temperatures (°C)

Tem

pera

ture

(°C

)

0.1m 0.2m 0.3m 0.4m 0.5m 0.6m 0.7m

Figure 6.84 Air temperature (at the vertical centre) of the internal sub cavity and inner surface temperatures for different depths of the inner sub cavity for the cases A, D, E and F (extreme summer day).

6.3.1.3 Performance of the glazing alternatives

Double façade mode

Summer function (naturally ventilated cavity)

The temperature was calculated at the vertical and horizontal centres of each layer of a box window façade 800 mm deep and 3.5 m high. The calculations were carried out for an extreme summer day. For the cases that shading devices are not applied, all alternatives except the one with the three clear panes (case A) perform similarly in terms of the inner pane surface temperature (32°C for the cases D, E and F and 36°C for case A, with indoor air temperature of 25°C). For the cases D and E a large part of the absorption takes place at the outer pane, while for the case F the inter-mediate solar control and low E coated pane results in a dramatic increase in temperature (of the intermediate pane), as shown in Figure 6.85.


253

20222426283032343638404244464850525456586062

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ven

tila

ted

cavi

ty

bord

er

Inte

rmed

iate

pan

e (c

entr

e)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)

A D E F

Figure 6.85 Calculated temperatures for different layers at the vertical and horizontal centres of a double façade for an extreme summer day, naturally ventilated cavity-no shading devices.

When shading is applied, the surface temperature of the inner pane decreases as expected, since part of the solar radiation is absorbed by the venetian blinds. As shown in Figure 6.86 for the cases that no low E was applied (cases A and D) (i.e. high U values of the inner skin) the inner pane temperature is slightly higher (approximately 2°C) than in the cases with inner and intermediate one (cases E and F).


254

20222426283032343638404244464850525456586062

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ext

erna

l sub

cavi

ty (

cent

re)

bord

er

Shad

ing

devi

ce (

cent

re)

bord

er

Inte

rnal

sub

cavi

ty (

cent

re)

bord

er

Inte

redi

ate

pane

(ce

ntre

)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)

A D E F

Figure 6.86 Calculated temperatures for different layers at the vertical and horizontal centres of a double façade for an extreme summer day, naturally ventilated cavity-shading devices.

Winter function (closed cavity)

For the winter case a box window cavity was selected, in order to study the air temperatures inside the cavity and the temperatures at the inner layer. In Figure 6.87 temperature at the horizontal and vertical centres of the layers for the cases A, D, E and F is presented assuming that no shading devices were applied. The cavity was assumed completely closed. At the inner layer the surface temperatures vary from 20°C to 24°C for the different cases, while the air temperatures inside the cavity vary from 10°C to 17°C. The case F (with the advanced low E and solar control intermediate pane) has the highest air and surface temperatures. The higher inner pane tempera-tures (when the alternatives E and F are compared) can be explained by (a) the lower U value and (b) the higher indirect solar transmittance of the F case. As shown in Table 6.1 the g values for the alternatives E and F are 0.35 and 0.3 respectively. The indirect transmittance is however 0.09 and 0.15 (g, Tsol) for the two cases. This means that even if the total solar transmittance is lower in the F case, the indirect transmittance is higher resulting in higher temperatures of the inner pane. Thus, the position of the solar control pane in a double façade system can be important for the inner pane temperatures.


255

02468

101214161820222426

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ven

tila

ted

cavi

ty

bord

er

Inte

rmed

iate

pan

e (c

entr

e)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)


Figure 6.87 Calculated temperatures for different layers at the vertical and hori-zontal centres of a double façade for a winter day, closed cavity- no shading devices.

When shading devices are applied (Figure 6.88), the variation in the air temperature inside the cavity is between 20°C and 25°C, while the inner surface temperatures vary from 20°C to 22°C. The increase in air tempera-ture inside the cavity is caused by the absorption of the shading devices.


256

02468

101214161820222426

Out

door

air

tem

pera

ture

Out

door

sur

face

tem

pera

ture

Ext

erna

l pan

e (c

entr

e)

bord

er

Ext

erna

l sub

cavi

ty (

cent

re)

bord

er

Shad

ing

devi

ce (

cent

re)

bord

er

Inte

rnal

sub

cavi

ty (

cent

re)

bord

er

Inte

redi

ate

pane

(ce

ntre

)

bord

er

win

dow

air

gap

(cen

tre)

bord

er

Inte

rnal

pan

e (c

entr

e)

Indo

or s

urfa

ce te

mpe

ratu

re

Indo

or a

ir te

mpe

ratu

re

Tem

pera

ture

(°C

)


Figure 6.88 Calculated temperatures for different layers at the vertical and horizontal centres of a double façade for a winter day, closed cavity- shading devices.

A parametric study with different positions of shading devices inside the closed cavity was carried out for winter conditions for the cases A, D and F; as expected, the shading device position does not infl uence the tem-peratures at the inner surface.

Summer – winter function (mechanically ventilated cavity)

The main purpose of the mechanically ventilated DSF mode is to preheat the air inside the cavity during the heating season and to distribute it through the AHU into the zones as supply air (if needed it can be further heated in the AHU, in order to meet the required supply temperature). The increase in air temperature should lead to energy savings during the heating season. These simulations will be carried out in IDA on a building level, in order to reduce the energy use all year round. However, simulations are carried out in WIS 3, in order to study the increase in temperature with different glazing alternatives with and without shading devices.

In order to calculate the air temperatures at the outlet of the cavity, a 20 m high multistorey façade was chosen and the simulations were carried out for a typical winter day (in order to study the warmest case); the airfl ows needed for ventilation of the offi ce zones are shown in Table 6.4. The two airfl ow rates picked for the simulations are the minimum (typical single offi ce) and the maximum (meeting room for 12 persons).


257

Table 6.4 Airfl ow rates for mechanically ventilated cavities (double façade mode). The airfl ow rates are chosen according to the ventilation rates of the different zones.

Zone type Total Ventilation Ventilation ventilation per m of per m of rate (l/s) width for the width for the box window 20m high cavity (l/s/m cavity (l/s/m width) width)

Typical single offi ce (1 occupant) 10 4.2 24Double offi ce (2 occupants) 20 5.6 32Corner offi ce (1 occupant) 15 4.2 24Meeting room (6 occupants) 21 5.8 33Meeting room (8 occupants) 28 5.2 30Meeting room (12 occupants) 42 6 34

For the winter case the air temperatures at the exit for the different glaz-ing alternatives are presented in Figure 6.89. In the cases with shading devices (placed at the middle of the cavity) an average temperature of the two sub cavities was assumed. As expected, in all the cases there is a small decrease in the air temperature at the exit, when the airfl ow rate increases from 24 l/s/per m cavity width to 34 l/s/per m cavity width. The outlet air temperature of the case A (thee clear panes) is around 9°C when shading devices are not applied and around 16.5°C with shading devices. When a solar control external pane is applied (case D) the air temperature at the exit increases (around 11°C) for the cases without shading devices due to the higher absorption of the external pane. For the same reason the air temperature difference for the cases with and without shading drops (by approximately 4.5°C) compared with the case A. The results are similar when a low E internal pane is applied (case E). The lower air temperatures in case E (compared with the case D) can be explained by the lower heat transmission from indoors to the cavity. When the cases D and E with shading applied are compared, it can be noticed that the air temperatures at the exit slightly decrease. This can be explained by the larger absorption of the external pane. Finally, in the case with the advanced intermedi-ate pane (low E + solar control pane) the air temperature increase is the highest both in the cases with (approximately 18 to 17°C) and without (approximately 15°C) shading devices. In this case, the air temperature increases due to the solar control pane treatment, while the low E coating reduces the heat transmission to or from the indoors. Since shading devices


258

are usually not applied during winter, the preferred cases for the winter conditions should be the cases F and D. On the other hand it is likely that there will be energy losses through the windows due to the higher U values then for E and F. This energy aspect is analyzed and discussed on a zone level using IDA ICE 3.0 (Subsection 6.3.2).

9,3 9,0

11,511,0

10,49,7

15,314,7

17,216,4

16,015,3 15,3

14,6

18,217,5

0123456789

1011121314151617181920

24 l/s/mwidth

34 l/s/mwidth

24 l/s/mwidth

34 l/s/mwidth

24 l/s/mwidth

34 l/s/mwidth

24 l/s/mwidth

34 l/s/mwidth

A3 D3 E3 F3

Air

tem

pera

ture

at e

xit (

°C)

no shading with shading

Figure 6.89 Calculated air temperatures at the exit of the mechanically ventilated cavity during a typical winter day (800mm deep cavity).

In order to calculate the inner pane’s surface temperatures, a box window façade (3.5 m high and 0.8 m deep) was considered. The surface tem-perature of the inner layer was calculated for both winter and (extreme) summer conditions. For the winter case the maximum airfl ow (6 l/sec per m width) was assumed, while for the summer case the minimum one was considered (4.2 l/sec per m width). The main reason for these assumptions was to study the worst possible cases causing thermal discomfort problems. Both winter cases (with and without shading devices) were simulated as shown in Figure 6.90.

A D E F


259

20,121,5

19,820,9

21,9 22,022,723,1

0123456789

10111213141516171819202122232425

no shading shading

Winter

Inne

r su

rfac

e te

mpe

ratu

re (°

C)


Figure 6.90 Calculated inner surface temperatures of mechanically ventilated box window facades (winter/extreme summer, with/without shading devices).

During a typical winter day the inner pane surface temperatures are quite close to the indoor air temperatures (23°C), mostly for the alternatives E and F, with low U values. For the other two alternatives the surface temperatures are close to 20°C, when shading is not applied. In the cases A and D the inner pane temperature is somewhat higher, when shading is applied, due to the clear inner pane. In the cases E (inner low E pane) and F (intermediate low E pane), however, the inner surface temperature remains almost the same for the cases with and without shading devices. During an extreme summer day the difference between the inner surface temperature and the indoor air temperature is 10 to 12°C for the cases without the shading devices and 9 to 14°C for the cases in which shading is applied.

Simulations were carried out with naturally ventilated box window cases, in order to study the infl uence of the reduced airfl ows (mechanical ventilated cases) on the inner surface temperatures for an extreme summer day. For the naturally ventilated cavities the airfl ow rate is much larger (than in the assumed mechanically ventilated ones) resulting in lower inner surface temperatures as shown in Figure 6.91. The airfl ow rates for the naturally ventilated cases are presented in Table 6.5 (the air fl ow rate for the mechanically ventilated cavities is 4.2 l/sec per m of width of the façade). Regardless of whether shading is applied or not, a drop in the surface temperature can be noticed for the naturally ventilated cases due


260

to the highest airfl ow rates inside the cavity. In the mechanically ventilated cases A and D, when shading is applied, the inner surface temperature slightly increases, while for the cases E and F it decreases. The absorbed radiation in the blinds results in higher air temperatures inside the cav-ity. In the cases with intermediate and inner clear panes (A and D) the air temperature increase has a considerable effect on the inner surface temperatures (resulting in higher inner surface temperatures when shad-ing is applied), while for the cases E and F the infl uence is much smaller (resulting in lower inner surface temperatures when shading is applied). In the naturally ventilated cases, however, the increase in air temperature inside the cavity when shading is applied results in a more intense stack effect and thus higher airfl ows. The inner surface temperatures in all cases drop when shading is applied (most often cases during extreme summer days).

02468

1012141618202224262830323436384042


Inne

r su

rfac

e te

mpe

ratu

re (

°C)

no shading mechanical vent. no shading natural vent.shading mechanical vent. shading natural vent.

Figure 6.91 Calculated inner surface temperatures for box window facades (ex-treme summer day).

Table 6.5 Airfl ow rates for naturally ventilated box window facades.

case A case D case E case F (l/sec per m (l/sec per m (l/sec per m (l/sec per m of width) of width) of width) of width)

no shading 167 242 248 274shading 255 267 248 284


261

Finally, a comparison between a naturally and a mechanically ventilated cavity during an extreme summer day shows that (especially when shad-ing is applied) the suffi cient heat extraction can reduce the inner surface temperatures. The drop in temperature can reach 9°C in the cases with clear inner and intermediate panes and 6°C in the cases with a low E coated pane. Moreover it is obvious that in order to ensure low inner surface temperatures during extreme summer days, suffi cient airfl ow in the cavity should be provided. If the temperature does not drop enough with a naturally ventilated cavity, then the possibility of fan supported ventilation (Poirazis, 2006) should be examined.

Airfl ow window modeFor the airfl ow window cases the exhaust air (of the offi ce space) enters the cavity and through the cavity is driven to the AHU. Thus, the temperature of the inlet air (to the cavity) is constant (i.e. 23°C for a winter day, 24°C for a typical summer and 25°C for an extreme summer day, same as the indoor air temperatures ). Only box window façades were studied for the airfl ow window mode, focusing on (a) the (inner surface) temperatures of the inner pane, for thermal comfort purposes and (b) the air temperatures at the exit from the cavity, for heat recovery purposes. The different glazing alternatives used are described in more detail in Subsection 4.1.3.1.

A brief parametric study was initially carried out, regarding the impact of cavity depth on the inner surface temperatures and the air temperatures at the vertical and horizontal centres of the cavity. The calculations were made for a typical winter and an extreme summer day for cases without shading devices. The study showed that neither the air nor the surface temperatures are infl uenced by the depth of the mechanically ventilated cavity. For this reason a 200 mm deep cavity was selected for the simula-tions. The properties of the airfl ow window glazing systems are described in Table 6.6 and in more detail (e.g. visual transmittance) in Appendix I.


262

Table 6.6 Properties of the airfl ow window glazing systems as calculated by WIS 3.

AW

Cas

e

Ext

erna

l pan

e

Gap

(12

mm

)

Inte

rmed

iate

pa

ne

Gap

(20

0m

m)

Inte

rnal

pan

e

U v

alue

gla

zing

sy

stem

W/m

²K,

clos

ed c

avit

y

U v

alue

inne

r sk

in W

/m²K

gva

lue

Tso

l

A Clear pane 8mm

Air Clear pane

4mm Ventilated

Clear pane 4mm

1.93 5.92 0.627 0.53

B Clear pane 8mm

Argon Low E 4mm

Ventilated Clear pane

4mm 1.15 5.92 0.561 0.447

C Clear pane 8mm

Argon Optigreen

6mm Ventilated

Clear pane 4mm

1.83 5.92 0.529 0.326

D Optigreen 8mm

Argon Clear pane

4mm Ventilated

Clear pane 4mm

1.85 5.92 0.404 0.297

E Optigreen 8mm

Argon Low E 4mm

Ventilated Clear pane

4mm 1.15 5.92 0.354 0.264

F Solar

control+lowE 8mm

Argon Clear pane

8mm Ventilated

Clear pane 4mm

1.04 5.92 0.195 0.157

G Solar

control+lowE 8mm

Argon Clear pane

4mm Ventilated Low E 6mm 0.824 5.66 0.191 0.149

Inner surface temperatures of airfl ow window façades

In order to carry out the parametric studies with the different glazing alternatives a typical winter and an extreme summer day were considered (where the results are more evident). Cases with and without shading were considered for the extreme summer conditions, in order to study the po-tential overheating problem, while the airfl ows were selected for the worst case in terms of heat gain or extraction (6 l/sec per m width of façade for winter and 4.2 l/sec per m width of façade for summer conditions).

During a typical winter day the inner surface temperature of all the simulated alternatives varies from 21°C to 24°C, as shown in Figure 6.92 (the inlet air temperature is the same as the indoor air temperature, 23°C). The highest inner surface temperature is noticed in the third (C) case with an intermediate solar control pane and clear inner and outer panes (its performance regarding energy use is studied all year round on a zone level in Subsection 6.3.2 using IDA ICE 3.0 software) . The lowest temperature is noticed in the fourth (D) case in which the solar control pane was placed as an outer pane and the inner and intermediate panes are clear. For an extrene summer day the highest surface temperatures are still noticed in the third case, while the lowest are noticed in the seventh (G) case with an advanced (low E + solar control) outer pane, a clear intermediate pane and


263

a low E inner pane (case with the lowest U and g values). The inner surface temperature variation of the simulated alternatives is from 32°C to 47°C, without shading, and from 33°C to 50°C when shading is applied.

When three clear panes are applied (case A) the inner surface tempera-ture is almost 22°C, while for the extreme summer case the temperature rises up to 37°C when shading is not applied and up to 48°C for the case with shading. The surface temperatures for the second (B) case in which the clear intermediate pane was replaced by a low E coated one are slightly higher.

By replacing the intermediate clear pane with a solar control one (case C) the surface temperature slightly rises even more during a typical winter day. For an extreme summer day, however, the highly absorbing intermediate pane results in increased air temperatures inside the cavity, and consequently (due to the relatively high thermal transmittance of the inner clear pane) high inner surface temperatures. When shading is applied the inner surface temperature increase is lower than in the cases A and B, since the solar control intermediate pane reduces the effect of shading devices.

When a body tinted solar control pane is placed as an outer one (cases D and E) the surface temperatures drop to 21°C in the case D and to 22°C in the case E, for the typical winter day. During an extreme summer day (without shading devices) the inner surface temperatures are 38°C for the case D (clear intermediate pane) and 36°C for the case E (low E coated intermediate pane). A larger drop in inner surface temperatures (compared with the cases A, B and C) can, however, be noticed when shading devices are applied during an extreme summer day. In this case the temperatures rise to 42°C and 41°C respectively (i.e. much lower than the alternatives studied before), since the absorption takes place at the outer pane (same effect as case C but this time at the outer pane). When the cases D and E are compared it is obvious that the low E coated pane (with lower U value) placed as intermediate pane (case E) replacing the clear one (case D) results in a slightly higher (1°C) surface temperature during winter, while during the summer this slight temperature increase brings the opposite results. The difference between the two alternatives is however quite small.

In the last two cases an advanced solar coated (low E + solar control) pane is applied. In the case F the intermediate and inner panes are clear ones, while in the case G the inner pane is replaced by a hard coated low E one. In these two cases the inner surface temperature during the winter is around 22°C (slightly higher for the case G), while for an extreme summer day the temperature rise is the lowest (compared with all previous cases) as it does not exceed 32°C (no shading applied). When shading is applied,


264

the increase in the inner surface temperature is very small especially in the case with the inner low E coated pane (case G).

02468

10121416182022242628303234363840424446485052

typical wintersurface temperature (no

shading)

extreme summer extreme summersurface temperature (shading)

Tem

pera

ture

(°C

)

A B C D E F G

Figure 6.92 Inner surface temperatures for an airfl ow box window (with and without shading devices).

Outlet air temperatures of airfl ow window façades

The outlet air temperatures are calculated only for the winter case, since in this case the air is brought to the AHU for heat recovery. In order to study the infl uence of glazing combination on the outlet air temperature a 200 mm deep box cavity was selected with airfl ow of 6 l/sec per m width of façade. The airfl ow window alternatives are studied both with and without shading device.

When shading is not applied, the highest outlet air temperature is no-ticed when the solar control pane is placed as intermediate pane (case C) as shown in Figure 6.93. The lowest temperatures are noticed for the case A, in which three clear panes are applied (case with the highest thermal transmittance). When shading is applied the cases with clear outer pane are the ones with the highest air temperatures at the exit of the cavity (cases A, B and C).

In more detail, when shading is not applied the air temperature at the exit for the fi rst case (A) is 19°C (4°C of temperature drop), while, when shading is applied, the air temperature rises to 27.5°C. When a low E coated pane is applied as intermediate pane (case B) resulting in a lower U value, the air temperature increases to 23°C and 28.5°C respectively. The advanced low E and solar control intermediate pane (case C) results


265

in a higher air temperature when shading is not applied (24°C) but in a lower one (26°) when shading is applied, due to the higher absorption at this pane. When a solar control pane is applied as outer pane (case D), the air temperature at the exit is 19°C, without shading, and 21.5°C with shading. When a low E pane is applied as intermediate one (case E), replacing the clear pane of the case D, the air temperature at the exit increases to 22°C and 24.5°C respectively. The temperatures are somewhat lower when the advanced coated pane (low E + solar control) is used as the outer one (cases F ad G).

0

2

4

6

8

10

12

14

16

18

20

22

24

26

28

30

A1 B1 C1 D1 E1 F1 G1

Air

tem

pera

ture

at t

he e

xit (

°C)

no shading shading

Figure 6.93 Air temperature at exit for different airfl ow window cases (winter day).

In general, solar control panes result in smaller air temperature differences between the cases with and without shading devices, since by absorbing a larger amount of radiation (than clear panes) they reduce the effect of shading. The concept of heat recovery is studied more extensively in the IDA ICE 3.0 simulations and conclusions are further discussed below.

6.3.2 Parametric studies on a zone level (IDA ICE 3.0)As described in Subsection 4.1.3.2, a parametric study of different double skin façade alternatives was carried out, in order to evaluate their perform-ance and better understand the infl uence of design parameters on energy use and thermal comfort. A typical cell offi ce (for one person) zone with normal control set points was selected for this study. Due to the large amount of output data, the results are presented selectively.

A B C D E F G


266

6.3.2.1 “Standard” double façade mode (naturally ventilated cavity)

For the naturally ventilated “standard” double façade mode the alterna-tives selected for the IDA ICE 3.0 simulations were A, D, E and F (see Subsection 6.3.1.1). Two types of shading devices were used (see Subsec-tion 5.3.2.6) in the cavity with a depth of 0.8 m (see Subsection 5.3.2.3).The cavity depth was selected after consideration of minimum depth to ensure effi cient heat extraction from the cavity as discussed in Subsection 6.3.1.2.

For the naturally ventilated cavities nine different damper control set points were considered. For the fi rst case the dampers were assumed always open. For the second case (set point of 10°C) the dampers were completely closed for cavity air temperature below 8°C and completely open above 12°C; between 8 and 12°C the dampers open linearly. The dampers for the other set points (12, 14, 16, 18, 20 and 22°C) are similar. In the last case the openings were considered always closed.

The output parameters studied for the naturally ventilated cavities are related to energy use and indoor climate issues. In more detail, the study focuses on the infl uence of design parameters (such as glazing and shading device type and the opening control of dampers) on energy use and inner pane temperatures. Furthermore their impact on operative temperature and comfort indices is presented.

Energy useThe design parameters studied are the façade orientation and the type of glazing and shading device. Each of these cases was simulated for the different damper control set points mentioned above and the alternatives with minimum total energy use were selected as optimal ones.

The glazing type is crucial for energy savings. As shown in Figure 6.94, the alternative F (solar control + low E coating intermediate pane) performs best regardless of the façade orientation. The alternatives A and D have increased heating demand (due to the high U value of the system). The lower g value of the alternative D (solar control outer pane), results in a lower cooling and a higher heating demand. In total there is a slightly higher total energy demand for the north and east but a lower one for the south and west oriented façades. The alternative E has a lower heating demand due to the low E inner coating (lower thermal transmittance); the lower thermal transmittance of the inner skin of the alternative E (1.46 instead of 2.74 W/m2K, see Table 6.1) does not have much infl uence on the cooling demand. The alternative F with the intermediate solar control and low E coating performs better during both the heating and cooling seasons,


267

due to the lower U and g values. The performance of the cases with blue (instead of white) venetian blinds is similar; the impact of shading devices is studied further in Subsection 6.3.2.6.

91314209131316912129 18111017

89

101

13813

4

89

97

134

132

91

104

143137

92

108

14914

1

0102030405060708090

100110120130140150160

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

north east south west

Hea

ting

and

coo

ling

dem

and

(kW

h/m

2 a)

cooling heating

Figure 6.94 Energy use (heating and cooling) for a cell offi ce for cases with white venetian blinds (“standard” double façade mode, naturally ventilated cavity with optimal temperature damper control).

Indoor ClimateIn order to study the thermal performance of the double façade alternatives, the inner pane temperatures and the monthly average PMV values were studied. Although similar output was obtained by WIS 3.0 simulations, this time all year round simulations were carried out, providing a more realistic view of the double façade thermal performance as to Scandinavian climatic conditions. For the simulations, cavities with optimal (as to energy use) temperature damper control were chosen.

Inner pane temperatures

For studying the inner pane temperatures a south oriented façade was selected, with optimal (as to energy use) damper control. The glazing type has a considerable infl uence on the inner pane temperatures of the ven-tilated cavity as shown in Figure 6.95, which was also shown by the WIS 3.0 simulations. The best choice for glazing is alternative F, with a low U and g value of the inner skin. The inner pane temperature distribution is larger for the alternatives (A and D) without low E pane (higher U value) and as expected somewhat colder than the ones with low E panes.


268

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Inner pane temperature (°C)

Num

ber

of h

ours

DSF A DSF D DSF E DSF F

Figure 6.95 Number of hours (in a year) for certain inner pane temperatures for different glazing alternatives for a cell offi ce (south orientation, ventilated cavity with optimal temperature damper control, white venetian blinds).

Monthly average PMV

The Predicted Mean Vote (monthly average PMV values during the working hours) was studied in this section. North, east, south and west oriented single offi ces with ventilated façades were simulated. All four glazing alternatives were selected with white venetian blinds.

Due to the lower U value (when the dampers are closed) the alterna-tives E and F perform better than the A and D during the winter months (north oriented façades, see Figure 6.96). The higher indirect transmit-tance and slightly lower U values of the case F further improve the PMV values during winter (when compared with the case E), as explained in Subsection 6.1.3.1. During summer on the other hand the warm air is extracted through the ventilated cavity (cooling the intermediate pane), so the performance of alternative F is similar to that with the case E (see Figures 6.85 and 6.86). In the PMV comparison, however, it can be seen that the PMV values are lower in the case F. This can be explained by damper control set points, in combination with the fact that the cavity air temperatures tend to be higher in the case F. The dampers in this case are opened for more hours during the summer months resulting in lower U values (1.31 instead of 1.15 W/m2K of case E) and lower inner pane temperatures. The alternative with the outer solar control pane (D) also performs well during the summer but it has pretty low PMV values during


269

winter, due to a high U value. The lower PMV values (during summer) of the alternative D compared with those of alternative E during summer can be explained with the higher transmission losses from the zone to the cavity (due to the lack of the inner low E pane). Finally, the alternative A performs poorly during both winter and summer due to the high U and g values.

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.96 Monthly average PMV values for the north oriented naturally ven-tilated cases for a cell offi ce (white venetian blinds).

The monthly average PMV values during summer of the south oriented alternatives are somewhat lower (better being closer to zero) than those of the north oriented ones, as shown in Figure 6.97, due to the increased use of shading devices. Since the shading (due to the daylight control) is used more in the south oriented façade (e.g. 31% for the south and 1% for the north oriented alternative F during the summer months), the system g value drops, lowering the inner pane temperatures and consequently the monthly average PMV values.

Another conclusion that can be drawn by observing the curves in Figure 6.97 is that the PMV values of the south oriented A and F alternatives (with clear outer pane) tend to decrease even more (compared with D and E that have outer solar control pane). Since the shading is used more for the cases with clear outer pane (due to the daylight control), the g values of the south oriented cases A and F are lower during the summer months. Furthermore, the more often the shading is applied, the warmer the cavity gets and the greater is the number of hours with opened dampers, result-ing in lower inner pane temperatures and lower monthly average PMV values for the south oriented cases.


270

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.97 Monthly average PMV values for the south oriented naturally ven-tilated cases for a cell offi ce (white venetian blinds).

6.3.2.2. “Standard” double façade mode (mechanically ventilated cavity)

For the mechanically ventilated “standard” double façade mode the four glazing alternatives selected were the same as for the naturally ventilated one. For this mode the air enters the AHU in three ways: (a) from outdoors, (b) from the double skin façade cavity or (c) from both. All year round, the outlet air from the double skin façade cavity is warmer than the outdoor air. During the heating season the use of this air as supply air may have a positive effect on energy demand, while during summer the outdoor air should be preferred, in order to avoid increasing the cooling demand. In order to mix the outdoor air and cavity air properly, so as to meet the supply air temperature (into the zones), a mixing box was designed in IDA ICE 3.0. The parameters studied for the mechanically ventilated cavities are similar to those for the naturally ventilated ones.

Energy useThe design parameters studied are the façade orientation and the type of glazing. As for the naturally ventilated cases the glazing type is crucial for the energy savings. As shown in Figure 6.98, the DSF F (solar control + low E coated intermediate pane) performs best regardless of the façade orientation, while due to the high U value the alternatives A and D have increased heating demand. The total energy demand for the north and


271

east is slightly higher, while for the south and west oriented façades it is lower. The heating demand of the DSF E case is lower due to the low E inner coating. When the alternatives E and F are compared it is obvious that the position of the solar control pane is as crucial for the mechanically ventilated cases, as it was for the naturally ventilated ones. The intermediate solar control + low E pane results in a lower heating demand due to reduced transmission losses through the outer skin, while the lower g value of the alternative F results in a lower cooling demand during summer.

19

11 11 9

26

17 16 12

28 20 18 13

32

21 19 1380

93

12912

4

78

89

12411

9

82

97

13412

8

86

102

14213

3

0102030405060708090

100110120130140150160

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F


Hea

ting

and

coo

ling

dem

and

(kW

h/m

2 a)

cooling heating

Figure 6.98 Energy use (heating and cooling) for cases with white venetian blinds for a cell offi ce (standard double façade mode, mechanically ventilated cavity).

Indoor ClimateThe parameters studied in this section were the inner pane temperatures and the monthly average PMV values.

Inner pane temperatures

As for the naturally ventilated cavity a south oriented façade was selected for studying the inner pane temperatures. The glazing type considerably infl uences the inner pane temperatures of the ventilated cavity as shown in Figure 6.99 (cases with white venetian blinds). The low U and g values of the alternative F result once more in narrower temperature variation of the inner pane; the distribution is larger for the alternatives A and D without low E pane (higher U value) and as expected somewhat colder. When the inner pane temperatures for the south oriented naturally (Figure 6.89) and mechanically (Figure 6.99) ventilated cavities are compared it


272

is obvious that the latter ones are much higher. This can be explained by the insuffi cient heat extraction due to the low airfl ow rates.

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

Temperature (°C)

Num

ber

of h

ours


Figure 6.99 Number of hours (in year) for certain inner pane temperatures for different glazing alternatives for a cell offi ce (south orientation, white venetian blinds, mechanically ventilated cavity).

Monthly average PMV

The monthly average PMV is presented for the north and south oriented double skin alternatives. As expected, for the north oriented façade the DSF F appears to be closer to the zero axis (due to the low U and g values) both during winter and summer (as shown in Figure 6.100. The low E coating of the DSF E results in higher PMV values (compared with the DSF D), while the solar control outer pane results in similar PMV values during summer. The clear outer pane of the DSF A results in slightly higher PMV values (than DSF D) all year round.


273

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.100 Monthly average PMV values for the north oriented mechanically ventilated cases for a cell offi ce (white venetian blinds).

The trend for the monthly average PMV values for the south oriented façade is similar as shown in Figure 6.101. As expected, when the naturally and mechanically ventilated cases are compared, the PMV values are some-what higher during summer due to the lower air fl ow rates (mainly during the summer months) inside the cavity. Higher airfl ow rates can partly solve this problem but they will increase the energy demand for fans.

The lack of effi cient heat extraction through the cavity has also an-other effect on the PMV values of the south oriented façades. While for the naturally ventilated cases the larger amount of absorbed radiation during the summer months, combined with the effi cient heat extraction, resulted in lower PMV values, for the mechanically ventilated cases the insuffi cient ventilation rates lead to increased air temperatures inside the cavity, which infl uences the inner pane temperatures (as shown in Figure 6.99) and increases the PMV values.


274

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.101 Monthly average PMV values for the south oriented mechanically ventilated cases for a cell offi ce (white venetian blinds).

6.3.2.3 “Standard” double façade mode (hybrid ventilated cavity)

For the hybrid ventilated “standard” double façade mode the same four glazing alternatives (as for the naturally and mechanically ventilated ones) were simulated in IDA ICE 3.0. For this mode the cavity is mechanically ventilated all year round, using the same mixing box as the one described in Subsection 6.3.2.2. In the hybrid ventilated case, however, the dampers open when the cavity air temperature exceeds a certain limit (for more effi cient heat extraction purposes); this temperature limit was selected in such a way that the total (heating and cooling demand) is minimized. The main aim of the “hybrid” mode is to combine effi cient heat recovery (as in the mechanically ventilated cases) and reduced cooling demand due to the better ventilated cavity (as in the naturally ventilated cases).

Energy useThe design parameters studied are the façade orientation and the type of glazing. As shown in Figure 6.102, the DSF F (solar control+low E coat-ing intermediate pane) performs best once more regardless of the façade orientation , while due to the high U value the alternatives A and D have increased heating demand. The total energy demand for the north and east is slightly higher, while for the south and west oriented façades it is


275

lower. The heating demand of the DSF E case is lower due to the low E inner coating.

17 10 10 9 17 11 13 10 15 13 13 10

20 14 14 1182

98

13413

0

82

94

129

127

86

101

140

133

89

107

145

137

0102030405060708090

100110120130140150160

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F


Hea

ting

and

coo

ling

dem

and

(kW

h/m

2 a)

cooling heating

Figure 6.102 Energy use (heating and cooling) for cases with white venetian blinds for a cell offi ce (standard double façade mode- hybrid ventilated cavity).

Indoor ClimateThe parameters studied in this section were the monthly average PMV values for the north and south oriented façades.

Monthly average PMV

For the north oriented façades the alternative A (clear panes) has the high-est monthly average PMV values during the summer months due to the high g values. The alternative with the inner low E and outer solar control pane (E) is next, while the alternatives D and F appear to perform best during the cooling period. As for the naturally ventilated cases, the inner low E pane prevents the transmission losses from the offi ce space to the cavity resulting in slightly warmer conditions during the cooling period. During the heating season the alternatives with the lowest U values (E and F) perform best, while the intermediate solar control + low E pane (case F) results in more hours with open cavity, due to the increased air cavity temperatures and thus lower monthly average PMV values.


276

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.103 Monthly average PMV values for the north oriented hybrid ventilated cases for a cell offi ce (white venetian blinds).

The south oriented alternatives follow a similar trend. As for the natu-rally ventilated cases the increased use of shading devices combined with effi cient heat extraction (caused partly by the more intense stack effect during the summer) results in lower monthly average PMV values during the summer months for the south oriented façades; the effect is larger for the cases with clear outer pane (A and F).


277

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.104 Monthly average PMV values for the south oriented hybrid ventilated cases for a cell offi ce (white venetian blinds).

6.3.2.4 Airfl ow window modeIn the airfl ow window cases the air from the offi ce zone is exhausted through the cavity to the heat exchanger for heat recovery purposes. The supply air to the zone is 11.5 l/s (equal to the airfl ow rate inside the cav-ity). More information concerning the geometry of the airfl ow window cases can be found in Subsection 5.3.1.2, while a description of the type and position of shading devices can be found in Subsection 5.3.2.6. The glazing properties of the studied alternatives are described in Appendices H and I.

Energy useIn Figure 6.105 the energy use of the airfl ow window alternatives is pre-sented. In general, the north facing alternatives tend to perform slightly better than the rest mainly due to the lower cooling demand. As expected, AW A (case with three clear panes) has the highest total (heating and cool-ing) demand, due to the high U and g values (1.93 W/m2K and 0.627, as shown in Appendix I). The outer solar control pane (AW D) results in a lower cooling and in a slightly higher heating demand (due to the lower g value of the outer skin). For the south and west orientation the decrease in total demand reaches 11%. When a low E intermediate pane replaces the clear one (AW E), decreasing the U value of the system (1.15 W/m2K), the heating demand drops drastically (approximately 33% regardless of


278

orientation, compared with AW D), since the air temperature drop inside the cavity is smaller due to the better insulation of the outer skin. The cooling demand, however, increases (although the g value is lower than in the two previous cases), since during the cooling period the low E inter-mediate pane does not allow the heat to transmit from the cavity to the outside. When the advanced low E + solar control pane is placed as the outer one (AW F), the cooling demand drops, while the heating demand increases due to the lower g value of the outer skin (compared with AW E); the decrease in total demand is more noticeable in the south, west and east orientations due to the larger cooling demand. Finally, when the clear inner pane (of the E case) is replaced by a low E pane the total demand drops further. The cooling demand in this case slightly increases (although the solar transmittance slightly drops) due to the improved insulation of the inner skin (less heat transmission from indoors to the cavity). AW G results in the lowest energy use for heating and cooling, due to the low thermal transmittance (both outer and inner skin).

26

13 17 10 11

45

26 33

18 19

59

35

46

25 26

60

34

44

23 2470

83

82

122

116

65

78

75

115

108

73

88

86

127

121

77

93

91

13512

7

0102030405060708090

100110120130140150160170180

AW

A

AW

D

AW

E

AW

F

AW

G

AW

A

AW

D

AW

E

AW

F

AW

G

AW

A

AW

D

AW

E

AW

F

AW

G

AW

A

AW

D

AW

E

AW

F

AW

G


Hea

ting

and

coo

ling

dem

and

(kW

h/m

2 a)

cooling sum heating sum

Figure 6.105 Energy use (heating and cooling) for cases with white venetian blinds for a cell offi ce (AW cases).

Indoor Climate

Inner pane temperaturesThe inner pane temperatures have been studied in this section for the different glazing alternatives. For all the cases white venetian blinds were assumed (when applied). As shown in Figure 6.106 the inner pane temperatures of the north facing alternatives hardly exceed 28°C. For the


279

cases without low E pane (AW A and AW D) the temperature of the inner pane is somewhat lower due to the high thermal transmittance. The inner pane temperature of the alternatives E and F (with U of 1.15 and 1.04 W/m2K and g values of 0.354 and 0.195) increases, when low E panes are applied. The temperature variation is smaller in the case F due to the lower g value, while the slightly lower U value does not affect the lower inner pane temperatures. When the inner clear pane is replaced with a low E pane (case G), a slight decrease in the U value results in higher inner pane temperatures.

0

100

200

300

400

500

600

700

800

900

1000

1100

1200

1300

14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Inner pane temperature (°C)

Num

ber

of h

ours

AW A AW D AW E AW F AW G

Figure 6.106 Inner pane temperatures for north facing AW alternatives for a cell offi ce (white venetian blinds).

The inner pane temperatures, however, increase dramatically in the south, west and east orientations, as shown in Figure 6.107. When no solar control pane is applied (AW A), the number of hours with temperatures above 30°C is 860 for the south, 650 for the west and 400 for the east oriented façade. As expected, the solar control outer pane (AW D) causes a decrease in the inner pane temperatures, especially for the south facing zone. When a low E intermediate pane replaces the clear one (AW E), the inner pane temperature slightly increases, although the solar transmittance of AW E is lower than that of AW D (0.354 and 0.404 respectively). The increased thermal insulation of the intermediate pane in combination with the low airfl ow rates results in higher air temperatures inside the cavity and thus higher inner pane temperatures. Finally, the number of hours with inner pane temperatures higher than 30°C is much lower for the last two AW alternatives (AW F and AW G), mainly due to the lower total solar


280

transmittance. The low E inner pane of AW G results in somewhat lower inner pane temperatures.

406

645

229

535

426

275

627

465

48

107142

2465

116

868

050

100150200250300350400450500550600650700750800850900

East South WestNum

ber

of h

ours

wit

h in

ner

pane

tem

pera

ture

abo

ve 3

0°C


Figure 6.107 Number of hours with inner pane temperatures above 30°C for a cell offi ce (white venetian blinds).

Monthly average PMV

The monthly average PMV values for the north oriented facades are pre-sented in Figure 6.108. As expected, the cases without low E pane (AW A and AW D) have the lowest PMV values during the winter months, due to high U values. During spring and autumn, however, the PMV values of AW A increase due to the high total solar transmittance. AW D (case with outer solar control pane) is a somewhat cooler option during summer compared with AW A. When the low E pane is placed as the intermediate one (AW E) the PMV improves during the heating season but not during summer (compared with AW D), since less heat can be transmitted to the outdoors through the outer skin, increasing the air temperature and consequently the inner pane temperature. Finally, the monthly average PMV values are improved by AW F and AW G with the solar control+low E outer pane. The PMV during winter improves further, when the inner low E pane (AW G) is applied.


281

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.108 Monthly average PMV values for the north oriented AW cases for a cell offi ce (white venetian blinds).

For the south oriented façade the monthly average PMV values are shown in Figure 6.109. The tendencies remain the same as in the north oriented façade but in this case the PMV values are somewhat higher during the whole year. The alternatives performing best are the ones with the solar control + low E outer pane (AW F and AW G) for both the north and south oriented façades.

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.109 Monthly average PMV values for the south oriented AW cases for a cell offi ce (white venetian blinds).


282

6.3.2.5 Impact of the “ventilated façade” conceptIn order to estimate the energy savings and the indoor climate improve-ments achieved by different ventilated facade concepts, a comparison between “standard” double façades, airfl ow window cases and a non-venti-lated façades was carried out at a zone level. The “standard” double façade mode (naturally, mechanically and hybrid) ventilated glazing alternatives were compared with the same ones in which no ventilation occurs. Ac-cording to personal communication with Dr. Bengt Hellström (Division of Energy and Building Design, Lund University) the difference in thermal transmittance values between a closed double skin façade (with cavity depth of 0.8 m) and a single skin one with the same glazing (but cavity depth of 0.012 m) is 0.1 W/m2K. This difference is also what IDA ICE 3.0 results in. So consideration of the all year round closed double façade as a single skin one is reasonable. Similar comparisons have been carried out for the airfl ow window cases. This comparison can be considered as a comparison between (triple glazed) single skin and double skin alternatives with intermediate venetian blinds. In order to reduce the size of output, south and north oriented façades were selected for the comparisons.

Energy useWhen the cooling demand of the different north and south oriented alter-natives is compared, it is obvious that the naturally and hybrid ventilated double façades use less energy than the mechanically ventilated and the closed double skin facades (see Figure 6.110). As expected, the effect of ventilated cavities is larger for the south oriented facades, with the alterna-tive F giving the lowest energy use. When the naturally and hybrid venti-lated façades are compared with the (all year round) closed ones, a larger drop in cooling demand can be observed in the case A (with the higher total solar and thermal transmittance; 52% lower cooling demand in the naturally ventilated case compared with the one with a closed cavity, for the south oriented zone). This can be explained by the fact that the heat extraction is more important due to the higher total solar transmittance of the outer skin (which results in increased air temperatures in the cav-ity) and the high thermal transmittance of the inner skin (increased heat transmission from the cavity to indoors). The outer solar control pane (case D) reduces the effect of ventilation, decreasing the cooling demand by 46% (naturally ventilated south oriented zone compared with the case with closed dampers). When the inner low E inner pane replaces the clear one (case E) the drop is 40%, while for the case with the advanced intermediate pane (case F) the drop is 42% (slightly higher drop due to


283

the higher indirect solar transmittance, when the cavity is closed). The results for the hybrid ventilated cases are similar.

02468

10121416182022242628303234

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

north south

Coo

ling

dem

and

(kW

h/m

2 a)

closed naturally mechanically hybrid

Figure 6.110 Cooling demand for façades with natural, mechanical, hybrid and no ventilation for the north and south oriented zones for a cell offi ce (normal set points, white venetian blinds).

A comparison between a closed cavity and cavities with natural, mechani-cal and hybrid ventilation (Figure 6.111) shows that the closed and the mechanically ventilated cavities are the ones performing slightly better during winter. It should be stated, however, that for the naturally venti-lated cavity the dampers are not completely closed (sealed), allowing a low airfl ow inside the cavity. This assumption was made, since it is similar to the real case. The hybrid ventilated case has also slightly higher energy use for heating due to the damper control used (depending on the cavity air temperature); as a result, in some of the cases where the dampers are open to allow natural ventilation, heating is still needed. A more sophisticated damper control was assumed on a building level.


284

0102030405060708090

100110120130140150

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

north south

Hea

ting

dem

and

(kW

h/m

2 a)


Figure 6.111 Heating demand for façades with natural, mechanical, hybrid and no ventilation for the north and south oriented zones for a cell offi ce (normal set points, white venetian blinds).

As shown in Figure 6.112 the impact of ventilated façade on the total energy demand is limited. For the north facing alternatives the closed and mechanically ventilated cases perform best regardless of the glazing case, since the heating demand is the dominating factor; increased heating demand due to small openings, when the dampers are assumed closed can lead to slightly higher total energy use. For the south oriented cases the hybrid ventilated cavities perform slightly better than the rest (except in the case F in which the mechanically ventilated case is slightly better). The savings for the naturally ventilated case can reach 4% (south oriented, case A), for the mechanically ventilated case 4% (south oriented, case F) and for the hybrid case 7% (south oriented, case A).


285

0102030405060708090

100110120130140150160170

DSF

A

DSF

D

DSF

E

DSF

F

DSF

A

DSF

D

DSF

E

DSF

F

north south

Tot

al d

eman

d (k

Wh/

m2 a)

naturally mechanically hybrid closed

Figure 6.112 Total (heating and cooling) demand for cavities with natural, me-chanical, hybrid and no ventilation for the north and south oriented zones for a cell offi ce (normal set points, white venetian blinds).

Regarding the airfl ow window cases, a comparison between these and the closed double façades was carried out (Figure 6.113). The U and g values of the system are kept the same, while the geffective is slightly different due to the position of shading devices. Regardless of the orientation, the cooling demand for the north facing airfl ow window cases increases, while the heating demand drops. Due to the reduced heating demand, the total demand is lower for the airfl ow window cases.


286

26 20 13 11 17 11 10 9

127

140

135 14

8

91 108

93 92

0102030405060708090

100110120130140150160170

aw df (closed) aw df (closed) aw df (closed) aw df (closed)


Tot

al d

eman

d (k

Wh/

m2 a)

cooling heating

Figure 6.113 Total demand (heating and cooling) for the double façade and air-fl ow window north oriented zones for a cell offi ce with normal set points.

When the south oriented zones are compared, however, the cooling de-mand increases dramatically due to the small airfl ow rates and the increased inlet air temperatures (Figure 6.114). Another reason for the increased cooling demand is the position of shading devices. The shading devices for the airfl ow window cases are placed between the intermediate and inner panes (while for the double façade cases they were placed between the outer and intermediate panes), increasing the amount of transmitted heat through the inner skin; the effect of shading devices is more impor-tant for the south oriented zones, since in this case they are used more. When the heating demand is examined, there is a drop regardless of the glazing case. The total demand is in most of the cases increased for the south facing façades.


287

59

27 35

20

46

18 25

13

108

128

115

132

75

96 78

86

0102030405060708090

100110120130140150160170

aw df (closed) aw df (closed) aw df (closed) aw df (closed)


Tot

al d

eman

d (k

Wh/

m2 a)

cooling heating

Figure 6.114 Total demand (heating and cooling) for the double façade and air-fl ow window south oriented zones for a cell offi ce with normal set points.

Indoor climateFor the indoor climate comparisons, monthly average PMV for the south oriented façades were considered, while PPD values were examined on a building level.

For the case with the three clear panes the naturally and hybrid venti-lated cavities perform similarly resulting in lower monthly average PMV values as presented in Figure 6.115. For these cases the PMV values vary from -0.6 to 0.26.


288

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.115 Monthly average PMV values for the south oriented alternatives for a cell offi ce (case A, normal set points, white venetian blinds).

The trend for the glazing alternatives D and E is similar (but lower), in which the monthly average PMV values vary from -0.6 to 0.2 and from -0.55 to 0.2 respectively (Figure 6.116). For the case F (with solar control and low E intermediate pane), however, a difference can be noticed in the naturally and hybrid ventilated cases. This difference can be explained by the increased number of hours with open cavity for the naturally ventilated cases. Since the temperature set point for the dampers (hybrid ventilate case) is set, in order to decrease the total energy demand, it was decided to have a much higher set point (in order to use the preheated air for heat recovery purposes for more hours).


289

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V


Figure 6.116 Monthly average PMV values for the south oriented alternatives for a cell offi ce (case F, normal set points, white venetian blinds).

The airfl ow window cases tend to be warmer as discussed previously (Sub-section 6.3.2.4). The monthly average PMV values were compared with the closed and naturally ventilated double façades, in order to investigate the impact of façade mode on thermal comfort. As shown in Figure 6.117 for the south oriented case A the monthly average PMV values are higher for the airfl ow window than for the double façade alternatives during the whole year. The tendency for the cases D, E and F is similar. The impact of the increased PMV values on thermal comfort is examined quantatively on a building level.


290

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

closed df naturally ventilated df aw

Figure 6.117 Monthly average PMV values for the south oriented (closed and naturally ventilated) double façade and airfl ow window alternatives for a cell offi ce (case A, normal set points, white venetian blinds).

The tendency for the north facing alternatives is similar. The difference in PMV values between the double façade and airfl ow window alternatives is smaller (than for the south oriented ones) as shown in Figure 6.118. Since the shading devices are used less than in the south facing cases, the increased heat transmission (due to the application of shading devices) through the inner skin is limited during the summer months, resulting in smaller differences in PMV values during the whole year. The differences are reduced for the other glazing alternatives D, E and F.


291

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

closed df naturally ventilated df aw

Figure 6.118 Monthly average PMV values for the north oriented (closed and naturally ventilated) double façade and airfl ow window alternatives for a cell offi ce (case A, normal set points, white venetian blinds).

6.3.2.6. Impact of shading device typeTwo types of shading devices were used for the parametric studies. Their properties are given in Appendix K and also described in Subsection 5.3.2.6. The impact of shading type is larger for the closed and mechani-cally ventilated cases, due to the low airfl ow rates (of the latter one), which result in high air temperatures inside the cavity (Figure 6.119). On the other hand, the impact of shading device type on the cooling demand for the naturally and hybrid ventilated cases is limited due to the effi cient heat extraction from the cavity. For the case with three clear panes the increase is substantial (up to 56% in cases with closed or mechanically ventilated cavity) due to the high total solar transmittance of the outer skin (large amount of radiation is absorbed by the shading) and the high thermal transmittance of the inner skin (large proportion of the created heat is transmitted to the indoors). The solar control outer pane (case D) reduces the effect of shading devices, while the number of hours when shading is used drops (since the daylight sensor is at the inner skin). For the mechanically ventilated case there is a slight drop in cooling demand, since even the low ventilation rate is enough in this case to extract the increased heat created by the high absorbing shading device. The impact of shading on cooling demand is very small for the case with solar control outer and low E inner pane, while fi nally due to the clear outer pane the


292

closed and mechanically ventilated case F has increased cooling demand (due to the low total solar transmittance of the outer pane, the shading effect is more intense).

-2-10123456789

1011121314151617181920


Incr

ease

in c

oolin

g de

man

d (k

Wh/

m2 a)


Figure 6.119 Increase in cooling demand when the blue shading devices replace the white ones for a cell offi ce (normal set points, south oriented zones).

The impact of shading type on heating demand is almost negligible. For the cases A and F (high total solar transmittance of the outer skin) the shading there has a positive effect for the closed façade, while for the mechanically ventilated cases the effect is negative.

In general, darker shading devices (with higher absorptance) have a negative effect on the total energy demand for the different double façade alternatives as shown in Figure 6.120. The increase for the case A is larger due to the increased cooling demand. For the cases with effi cient heat extraction (naturally and hybrid ventilated cavities) the effect is very small.


293

-2-10123456789

1011121314151617


Incr

ease

in to

tal d

eman

d (k

Wh/

m2 a)


Figure 6.120 Increase in total demand when the blue shading devices replace the white ones for a cell offi ce (normal set points, south oriented zones).

A comparison between white and blue venetian blinds regarding energy use has been carried out for air fl ow windows as well. For the north oriented facades the difference in energy use is very small, (since the venetian blinds are applied for fewer hours). For the south oriented façade, however, there is an increase in total demand in the cases with the darker shading devices due to the increased cooling demand, as shown in Figure 6.121. The heat-ing demand, however, slightly increases due to the higher absorbtivity of the blinds, which results in an increase in air temperatures in the cavity. The differences in the cases with high U and g values are larger.


294

59

84

35 43 46 51

25 25 26 26

108

106

115 11

4

75

75

78 78

65 65

0102030405060708090

100110120130140150160170180190200

white blue white blue white blue white blue white blue


Tot

al d

eman

d (k

Wh/

m2 a)

cooling sum heating sum

Figure 6.121 Energy use (heating & cooling) for air fl ow window cases with white and blue venetian blinds for a cell offi ce (normal set points, south orientation).

The impact of shading device type on thermal comfort is also studied for the ventilated cavities. Since the mechanically ventilated and closed cavities perform similarly, and the shading device type does not really infl uence the inner pane temperatures of the naturally and hybrid ventilated façade (since the damper control is dependent on the cavity air temperature), only the south oriented mechanically ventilated case was examined. Further PPD comparisons were carried out on a building level. The larger differences in monthly average PMV values are noticed in the south oriented case A (with the three clear panes), as shown in Figure 6.122. The difference in PMV values is still rather small; thus, no further investigation of the impact of shading device on thermal comfort was carried out.


295

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

DSF A white DSF A blue

Figure 6.122 Monthly average PMV values for the mechanically ventilated, south oriented case A, for a cell offi ce.

For the airfl ow window cases the darker shading devices have a nega-tive effect on thermal comfort, since they increase the monthly average PMV values when applied (during summer months). This increase can be noticed mostly for the south oriented zones, in which the shading is applied more often. For this reason, the impact of shading devices was not studied further.

6.3.3 Parametric studies on a building level (IDA ICE 3.0)

As described in Subsection 4.1.3.3, buildings with different fully glazed double skin façades were studied. The parameters varied for this study were the façade type (multi storey and box window), mode (“standard” double façade and airfl ow window) and ventilation strategy (naturally, mechanically and hybrid ventilated cavities).

The two best performing “standard” fully glazed double skin façade alternatives were selected for further study on a building level (cases E and F). Both cases have low U and g values due to low E and solar control panes (outer and intermediate for case E and advanced intermediate for F). When these two cases are compared with regard to optical and thermal properties it can be noted that:


296

• the thermal transmittance is slightly lower in case F than in E (1.04 instead of 1.14 W/m2K)

• the total solar transmittance is somewhat lower in case F than in E (0.3 instead of 0.354), while the difference for the direct solar transmittance is higher (0.151 instead of 0.264)

• the light transmittance of case F is lower than for case E (0.416 instead of 0.551)

The best performing alternative, with regard to energy use and thermal comfort, on a zone level was F. However, due to the low light transmittance (according to personal communication with the WSP consultant Peter Pertola low light transmittance is below 0.5) and the high temperatures noticed at the intermediate advanced pane (which can lead to possible cracks) the case E was also selected as an alternative solution.

For the airfl ow window cases the glazing alternatives chosen for simula-tions on a building level are E (low E inner, clear intermediate and solar control outer pane), F (clear inner and intermediate and solar control + low E outer pane) and G (low E inner, clear intermediate and solar control + low E outer pane). The best performing alternative according to the simulations on a zone level is G, with F and E next. The light transmit-tance of the cases F and G is low (0.357 and 0.344 respectively), so case E was also selected (with a light transmittance of 0.551). The case G, on the other hand, has very low thermal transmittance and a comparison with the “standard” double façade and single skin alternatives would be quite “unfair”; that is why case F (with similar thermal and optical proper-ties to those of the double façade F) was selected. Finally, the case G was simulated on a building level, in order to investigate the importance of inner low E pane on energy use.

The thermal and optical properties of the different glazing alternatives can be found in more detail in Appendix I.

As stated in Subsection 5.3.2.2, the ground fl oor of the building has a single skin façade (for both the “standard” double façade and airfl ow window alternatives). The glazing assumed for the ground fl oor has the same properties as the reference single skin glazing alternative (third case, see Figure 4.2). For the rest of the fl oors a double skin façade was assumed with glazing as stated above. For the “standard” double façade mode white shading devices are selected for the southwest and southeast orientations (in order to reduce the increased cooling demand) and blue for the northeast and northwest (in order to avoid increasing the heating demand); for the airfl ow window cases only white venetian blinds were selected. The cav-ity depth of the “standard” double façade mode on a building level was assumed 0.8 m and for the airfl ow windows 0.3 m.


297

A brief description of the naturally, mechanically and hybrid venti-lated double façade and the airfl ow window building alternatives is given below.

Naturally ventilated, box window “standard” double façade: The air in this case is inserted from the lower opening of the cavity (when the façade is ventilated) and is extracted through the upper opening of the box window construction. The dampers were set to open, when the indoor (offi ce zone) mean air temperature reaches 23°C (while for the simulations on the zone level the dampers were functioning depending on the cavity air temperatures as described in Subsection 6.3.2.1).

Naturally ventilated, multi storey “standard” double façade: The air in this case is inserted from the lower opening of the cavity of the second fl oor (when the façade is ventilated) and is extracted through the upper opening of the top fl oor. The dampers were set to open when the indoor (offi ce zone) mean air temperature of the zones of the last fl oors reach 23°C.

Mechanically ventilated “standard” double façade: The outdoor air enters the cavity at a controlled rate and ends up in the AHU. A mixing box that mixes the outdoor and cavity air, in order to bring air tempera-tures as close as possible to the supply air temperature is installed in the AHU. If further heating or cooling is needed, then this is provided by the heating and cooling coils. The cavity air is used as supply air (when needed), while the exhaust air from the zone is driven to the AHU for heat recovery purposes.

Hybrid ventilated “standard” double façade: The hybrid ventilated case is a “combination” of the natural and mechanical cases. A temperature control opens the cavity dampers allowing natural ventilation, when the indoor air temperature exceeds 23°C (naturally ventilated case). When the indoor air temperature is below 23°C, the ventilation inside the cavity is the same as in the mechanically ventilated “standard” double façade, with a mixing box optimizing the mixing of outdoor and cavity air for minimizing the energy demand.

Airfl ow windows: For the airfl ow windows the exhaust air from the offi ce zones enters through a bottom opening and passes through the cavity to the AHU for heat extraction purposes. The airfl ow rates in the cavity for each zone are the same as the exhaust rates, as stated in Appendix F.

6.3.3.1 Energy useA comparison regarding the total energy use of the two best performing glazing alternatives is carried out in this section. The main purpose of this comparison is to investigate the impact of ventilated façades for already


298

well performing glazing alternatives on a building level. It has to be noted that for this simulations the same glazing and ventilation modes were applied for all the orientations and no optimization for each orientation was considered. In practice the optimal façade type for each orientation should be considered. However, a comparison of the energy performance between single skin (with intermediate venetian blinds) and double skin alternatives can still be carried out.

In Figure 6.123 a comparison for case E (solar control outer, low E intermediate and clear inner pane) is carried out. The larger drop in heat-ing demand (compared with the case with closed cavity) can be noticed for the airfl ow window and the mechanically ventilated “standard” double façade cavity, while a smaller drop can be seen in the hybrid ventilated case. As expected, the heating demand for the naturally ventilated cavity is higher, due to the realistic assumption that closed dampers are leaky (higher thermal transmittance). The results are opposite regarding the cooling demand. The naturally and hybrid ventilated façades perform best while the mechanically ventilated façade is slightly higher. Finally, the airfl ow window case has the highest cooling demand due to the low airfl ow rates and the increased inlet air temperatures (during the spring and autumn months) and the low thermal transmittance of the inner skin. The decrease in total energy use is larger for the hybrid ventilated case (138 instead of 145 kWh/m2a noticed for the case with closed cavity), while for the airfl ow window case the total energy use increases to 148 kWh/m2a.

When the multi storey and box window naturally ventilated “standard” double façades are compared an increase of 12 kWh/m2a of heating and a drop of 5 kWh/m2a of cooling demand can be seen. This can be explained by the temperature set points of the dampers. In order to avoid overheat-ing of the cavity air, the dampers open when the indoor air temperature of the offi ce of the last fl oor reaches 23°C. The cavity air temperature infl uences the inner pane temperature and the indoor air one. Since the cavity air temperature is higher at the top fl oors than in the fi rst ones, if the damper set points are connected with the top fl oors, more hours they will be opened resulting higher heating and lower cooling demand; op-posite would have been the effect if the damper set points were connected with the fi rst fl oors.


299

67 6981

64 66 63

18 12

7

16 1225

14,414,4

14,414,4 14,4

14,4

45 4545

45 45

45

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

closed naturally (boxwindow)

naturally(multi storey)

mechanically hybrid airflowwindow

Ene

rgy

use

(kW

h/m

2 a)


Figure 6.123 Total energy use for the glazing alternative E (closed, naturally, mechanically and hybrid ventilated “standard” double façade and airfl ow window, normal set points).

Similar comparisons are carried out for case F (clear inner and outer pane and solar control + low E intermediate pane). As shown in Figure 6.124 the decrease in cooling demand of the naturally ventilated cavity (when compared with the closed one) is 8 kWh/m2a. In case F the intermedi-ate pane tends to overheat resulting in increased heat transmission to the inside; when natural ventilation occurs the temperature of the intermedi-ate pane drops resulting in much lower cooling demand. For the same reason, a slight increase in heating demand can be noticed. As in case E, the mechanically ventilated case (when compared with the naturally ven-tilated one) results in increased cooling and decreased heating demand. The hybrid case is the one performing best (136 instead of 143 kWh/m2a for the case with closed cavity), with the heating demand slightly higher than in the mechanically ventilated case and with the cooling demand the same as in the naturally ventilated case. Finally, the airfl ow window case performs worst in terms of total energy use.


300

66 69 63 66 64

18 1015 10 15

14,614,6 14,6 14,6

14,6

4545 45 45

45

0

10

20

30

40

50

60

70

80

90

100

110

120

130

140

150

closed naturally mechanically hybrid airflow window

Ene

rgy

use

(kW

h/m

2 a)


Figure 6.124 Total energy use for the glazing alternative F (closed, naturally, mechanically and hybrid ventilated “standard” double façade and airfl ow window, normal set points).

From the Figures above it can be concluded that the decrease in total energy use is rather small, when double skin facades are used instead of single skin ones. However, the impact of ventilated façades on a zone level (Subsection 6.3.2.5) can give a better understanding as to which are the energy saving possibilities, since the study was focused only on heating and cooling demand (for which the savings occur) and was carried out for different oriented zones to point out the energy saving possibilities of each mode for each orientation. Although optimal double skin façade integration should defi nitely include consideration of the orientation, the comparison carried out above can give an understanding of the energy saving possibilities mostly for the hybrid ventilated case which can perform differently for the north, south, west and east oriented facades.

6.3.3.2 Thermal comfortWhen the monthly average PMV values of the different modes are com-pared for the glazing alternative E, it can be noticed that the cases with closed and mechanically ventilated cavities perform similarly (with the mechanically ventilated case giving slightly lower PMV values than the case with closed cavity), while the naturally and hybrid ventilated ones perform


301

the same (giving slightly lower PMV values); fi nally, the airfl ow window cases give the highest values during the whole year (Figure 6.125).

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

closed naturally ventilated mechanically ventilatedhybrid ventilated airflow windows

Figure 6.125 Monthly average PMV for the glazing alternative E (closed, naturally, mechanically and hybrid ventilated “standard” double façade and airfl ow window, normal set points).

Regarding the number of hours (%) with PPD values lower than 10% and 15%, the double façade alternatives perform similarly to the case with closed cavity (62% and 85%) while for the airfl ow window cases there is a slight increase (67% and 90%) due to the higher (and closer to the 0 axis) PMV values during the winter months.

For case F with the advanced (low E + solar control) intermediate pane the monthly PMV values slightly drop during the summer months due to the lower total solar transmittance of the glazing unit. The percentage of working hours with PPD values lower than 10% is 64% for the closed and mechanically ventilated cavities, while for the naturally and hybrid ventilated ones it is 63%. For the airfl ow window the percentage increases to 72%. Weighted average PPD values lower than 15% are noticed for 86% of working hours for the double façade and for 92% for the airfl ow window alternatives.


302

6.4 Comparison of single and double skin façade building alternatives

6.4.1 Impact of glazing size on energy use and thermal comfort of single skin buildings

In order to study the impact of glass area on the energy use, 30%, 60% and 100% glazed alternatives with triple clear glazing (as in the reference building) were generated (in reality windows used for highly glazed alter-natives have lower thermal transmittance). A cross comparison diagram of energy use of the 30%, 60% and 100% glazed alternatives (cell type) with strict and normal set points is presented in Figure 6.126. The increase in the total energy use for the 60% glazed building is 23% regardless of the set point (compared with the reference building).

The increase for the 100% glazed alternatives is 45% for the strict and 47% for the normal set points. Both the heating and cooling demand in-crease in the highly glazed building alternatives as shown in Figure 6.126. However, the increase in cooling demand of the 100% glazed building, which reaches 112% for the strict and 177% for the normal set points, is substantial.

One of the main arguments for using increased glazed areas in buildings is the provision of better indoor environment due to daylight. However, the increased window area does not necessarily lead to a reduction in en-ergy use for lighting the building properly. To make use of daylight more effi ciently, attention has to be paid to how the daylight is controlled and brought into the building. Traditional control of solar shading and lighting was applied to the cases studied in this report.


303

5676

95

5272

92

20

31

43

11

20

30

14,7

14,7

14,7

14,4

13,4

12,9

45

45

45

45

45

45

0102030405060708090

100110120130140150160170180190200

30% 60% 100% 30% 60% 100%

strict set points normal set points

Ene

rgy

use

(kW

h/m

2 a)


Figure 6.126 Impact of glazing size on energy use (triple clear pane with interme-diate shading devices, strict and normal set points).

As already stated, the perception of thermal comfort depends on the per-missible mean air temperatures, but also on the surface temperatures that surround the occupant. Thus, the size of the window is crucial for the perception of thermal comfort as seen in Figure 6.127. The considered glazed alternative with the triple clear glazing tends to give both high and low PMV values due to the high thermal and total solar transmittance. For the reference building alternative the monthly average PMV varies from -0.55 to 0.3, while for the 60% and 100% glazed alternatives the monthly average PMV varies from -0.68 to 0.36 and from -0.84 to 0.43 respectively.


304

-0,9

-0,8

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

30% 60% 100%

Figure 6.127 Monthly average PMV for the 30%, 60% and 100% glazed alterna-tives with triple clear pane and intermediate shading devices with normal set points.

6.4.2 Impact of glazing type on energy use and thermal comfort

In order to study briefl y the impact of the windows and shading devices on the energy use, the seven 100% single skin glazed alternatives with normal set points (cell type plan) are compared (Figure 6.128). For the best performing case the total energy use of the glazed alternative is only 20 % higher than for the reference building, (122 kWh/m²a with three clear panes and normal set points).

A decrease in the thermal transmittance of the window (alternatives 2-7) results in a reduction in the energy use for heating and a smaller increase in cooling demand (comparison of alternatives 1 and 2). The alternatives (2nd and 5th) with high glazing solar factor values (0.584) have also a slightly lower heating demand (compared with the 3rd one with g=0.354), while the one (4th) with lower g values (0.27) has a slightly higher heating demand. The effect on cooling demand is the opposite; the 4th alternative uses less energy for cooling than the 3rd and 5th.

Another parameter studied was the position of shading devices on the energy use. Intermediate blinds result in lower geffective values and thus lower energy use for cooling. When the 2nd and 5th alternatives (same window and shading device properties) are compared, it is obvious that the cooling demand increases dramatically (37%), when the blinds are


305

placed intervally (5th alternative). The heating demand is almost the same (slightly higher in the 2nd alternative), since the blinds were used mostly during the warm periods. When fi xed external louvres are applied (7th alternative), the cooling demand reduces dramatically while the heating demand increases.

Different types of internal shading (blinds in the 3rd and screens in the 6th) with similar properties do not much infl uence the energy use. The type of glazing (solar factor values) infl uences the energy use for lighting in the building. For a set point of 500 lux at the work place the energy use increases up to 14.3 kWh/m2a, when fi xed external louvres are applied (seventh alternative), from 12.9 kWh/m2a (fi rst alternative). Often, the low g and geffective values lead to increased lighting demand.

92

59 65 6858 66 72

30

37 27 1954

24 9

12,9

13,5 14,014,2

13,4

13,914,3

45

45 4545

45

4545

0102030405060708090

100110120130140150160170180190

1st alt 2nd alt 3rd alt 4th alt 5th alt 6th alt 7th alt

Ene

rgy

use

(kW

h/m

2 a)


Figure 6.128 Impact of glazing type and shading device position on energy use (100% glazed alternatives, normal control set points).

The fi rst glazed alternative (triple clear glazing) tends to give both high (during summer) and low (during winter) monthly average PMV values due to the high thermal and total solar transmittance (Figure 6.129). Although the g value of the windows slightly increases in the second al-ternative, the PMV values are still slightly higher due to the much lower thermal transmittance; the effect during winter is similar. When the total solar transmittance of the glazing in the third alternative is reduced the PMV values drop during the summer, resulting in lower PPD values. A further decrease in the total solar transmittance of the fourth alternative brings similar results as before, lowering the PMV values during the cooling period. When geffective increases (the fi fth case is the same as the second


306

one but the shading is placed internally instead of between the panes), the PMV values increase dramatically. The alternative with internal screens (sixth) has PMV values similar to the third one. Finally, when the fi xed horizontal external louvres are applied (seventh alternative) the PMV is lower all year round.

-0,9-0,8-0,7-0,6-0,5-0,4-0,3-0,2-0,10,00,10,20,30,40,50,60,7

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

1st alt 2nd alt 3rd alt 4th alt 5th alt 6th alt 7th alt

Figure 6.129 Impact of glazing type and shading device position on monthly average PMV values (100% glazed alternatives, normal control set points).

6.4.3 Comparison of buildings with single and double skin façades

In Figure 6.130 single and double skin alternatives with the same thermal and optical properties are compared (thermal transmittance of glazing 1.14 kWh/m2a and solar transmittance of glazing 0.354). In the fi rst case internal shading was applied resulting in a high geffective value. When fi xed horizontal external louvres (second case) replace the internal venetian blinds (fi rst case) the heating demand increases (7 kWh/m2a), while the cooling demand drops drastically (18 kWh/m2a) resulting in lower energy use. Intermediate placed shading devices (third case) result in lower heating (due to the increased solar gains during the heating season) and increased cooling demand (due to the increased indirect solar transmittance and to the daylight solar control of the shading that allows shading not to be used, when cooling is needed). A hybrid ventilated double façade (fourth case) with venetian blinds results in a slight decrease (when compared with the third case) in both heating (due to the reduced transmission losses


307

from the cavity through the outer skin) and cooling demand (due to the heat extraction from the cavity when open). Finally, the airfl ow window (fi fth case) has the lowest heating but increased cooling demand (due to the low airfl ow rates and the increased inlet to the cavity air temperature) that leads to high total energy use.

65 72 67 66 63

27 9 18 12 25

14,0

14,3 14,414,4

14,4

45

45 4545

45

0102030405060708090

100110120130140150160

single skininternal shading

single skinexternal shading

single skinintermediate

shading

double skinhybrid ventilated

airflow window

Ene

rgy

use

(kW

h/m

2 a)


Figure 6.130 Energy use for single and double skin alternatives with similar ther-mal and optical properties of the glazing (100% glazed alternatives, normal control set points).

When the monthly average PMV values (Figure 6.131) of these alternatives are compared, it can be noted that the case with internal shading devices has low values during winter and high values during summer. For the case with external shading the PMV values drop during summer (values closer to the 0 axis), but also during spring and autumn resulting in higher PPD values. The monthly average PMV values for the single skin case with intermediate venetian blinds are placed in between the previous two cases, while for the hybrid ventilated cases they are slightly lower mainly during the spring and summer. Finally, the airfl ow window case results in similar (with the case with internal shading) values during summer but somewhat increased values during winter, improving the thermal comfort.


308

-0,7

-0,6

-0,5

-0,4

-0,3

-0,2

-0,1

0,0

0,1

0,2

0,3

0,4

0,5

Janu

ary

Febr

uary

Mar

ch

Apr

il

May

June

July

Aug

ust

Sept

embe

r

Oct

ober

Nov

embe

r

Dec

embe

r

Mon

thly

ave

rage

PM

V

single skin internal shading single skin external shading single skin intermediate shadingdouble skin hybrid ventilated airflow window

Figure 6.131 Monthly average PMV values for single and double skin alternatives with similar thermal and optical properties of the glazing (100% glazed alternatives, normal control set points).

6.4.4 Comparison of best performing alternativesFinally, a comparison of the “best performing” alternatives as to energy use and thermal comfort is carried out (Figure 6.132). An improved version of the reference (30% glazed) single skin building was simulated. The Uglazing and Uframe were assumed 1.14 W/m2K and 1.6 W/m2K, while the g value was set to 0.354 (for the initial reference case with triple clear pane the values were 1.85 W/m2K, 2.31 W/m2K and 0.69 W/m2K respectively). The low thermal transmittance and smaller area of widows result in a very low (compared with the rest of the cases) heating and cooling demand and the lowest total demand (113 kWh/m2a). The next two alternatives have the same glazing type but larger window to external wall area ratio (60% and 100% respectively). For buildings with larger glazed area the heating, cooling and thus total energy use increases by 18 kWh/m2a for the 60% and by 38 kWh/m2a for the 100% glazed alternative (from 113 to 131 and 151 kWh/m2a). On considering the fully glazed case for comparison, it is seen that fi xed externally placed shading devices reduce the total energy use by 10 kWh/m2a (to 141 kWh/m2a) due to a drastic decrease in the cooling demand. A hybrid ventilated double façade (with the same glazing type as in the two previous cases) can result in a further drop of 3 kWh/m2a in total energy use compared with the case with fi xed


309

external louvres; for a more “fair” (considering the concept of clean, highly glazed façade) comparison, the total demand decreases by 13 kWh/m2a compared with the case with internal shading devices (total energy use of 138 kWh/m2a). Finally, an airfl ow window case with even lower thermal and optical transmittance was considered (U=0.824, g=0.191, Tvis=0.344; inner low E coated, and advanced solar control + low E outer pane); this case results in even lower energy use (132 kWh/m2a), but the low visual transmittance can often be a drawback. Energy wise, the poorer perform-ance of highly glazed offi ce buildings is inevitable. Careful selection of shading and glazing, however, can decrease the difference, resulting in reasonable solutions.

4454

65 72 6656

10

18

27 912

1614,7

14

14

14 14,415

45

45

45

45 4545

0102030405060708090

100110120130140150160

improvedreferencebuilding

60% (3rd)glazed singleskin building

100% (3rd)glazed singleskin building

100% (7th)glazed singleskin building

100% (E)glazed doubleskin hybridventilated

100% (G)glazed airflow

window

Ene

rgy

use

(kW

h/m

2 a)


Figure 6.132 Comparison of the energy use for the “best performing” alternatives, normal set points.

Regarding the quality of indoor environment it is evident that smaller glazing areas provide a more stable environment with fewer dissatis-fi ed occupants. In more detail, PPD values lower than 10% and 15% are achieved for 81% and 97% of the working hours for the improved reference building, while for the alternatives with larger glazed area (and the same window type) these values drop to 70% and 93% (for the 60% glazed building) and to 57% and 82% (for the 100% glazed building). The improvement is rather small when external shading is applied, since PPD values lower than 10% and 15% occur for 60% and 83% of the working hours; the improvement with a hybrid ventilated double façade cavity (63% and 85% respectively) is similarly small. Finally, due to the increase in the already low PMV values during winter, the airfl ow window


310

case provides clearly improved thermal environment resulting in the same PPD values as the improved reference alternative (PPD values lower than 10% and 15% are achieved for 81% and 97% of the working hours).

Conclusions

311

7 Conclusions

The energy effi ciency and thermal performance of highly glazed offi ce buildings are often questioned. However, nowadays glazed buildings are increasingly being built around the world, because (a) there is a growing tendency among architects to use large areas of glass in the façade, often with the aim of contributing to a better view of the outside and access to daylight, (b) users often like the idea of increased glass area, relating it to a better view of the outside and a more pleasant indoor environment and (c) many companies prefer the distinctive image of themselves (e.g. transparency or openness) that a glazed offi ce building can provide.

Achieving good building performance, when using fully glazed façades is a great challenge. The energy effi ciency and provision of an acceptable indoor climate are issues that should be considered. This thesis compares the performance (as to energy use and indoor climate issues) of an offi ce building with different façade alternatives, both conventional and highly glazed ones. Optimizing the energy and indoor climate performance of highly glazed buildings by achieving proper construction and integration of single and double skin façade systems is an important and challenging goal.

This chapter is divided into three main sections:

• Energy use and thermal comfort for highly glazed offi ce buildings located in Scandinavia. Within this section the main conclusions from the simulations are described. The aspects discussed are the impact of plan type, orientation, control set points, glazing area and type and façade (single and double skin) on energy use and thermal comfort.

• Methods used to determine the energy and indoor climate perform-ance of single and double skin façade glazed offi ce buildings.

• Improving the energy effi ciency and thermal comfort of offi ce build-ings with highly glazed façades. Suggestions are given for further studies that can be carried out regarding the components and their effi cient integration. The possibilities of double skin façade systems are further discussed.


312

7.1 Energy use and thermal comfort for highly glazed offi ce buildings located in Scandinavia

7.1.1 Plan type The plan type of offi ce buildings plays a considerable role in building performance regarding energy use and indoor environment issues. There are two clearly distinguishable plan types: the open plan and cell offi ce. The use of the offi ce space, ventilation rates, occupant density and the quantity of offi ce equipment are some of the main parameters that differ between these two types.

Due to the higher internal loads, mainly caused by higher occupancy and lower ventilation rates, an open plan offi ce building is likely to have higher cooling and lower heating demand. In this study the increase in cooling demand for an open plan compared to a cell offi ce reference build-ing (30% window to external wall area ratio) was 57% (6 kWh/m²a) for the normal set points for the indoor air temperature, while the decrease in heating demand was 14% (7 kWh/m²a). The energy use for lighting the open plan offi ce space increases (5 kWh/m²a), since all the space is considered to be working area. The tendencies for the highly glazed build-ing alternatives are similar.

The risk of discomfort increases in cell offi ce zones, compared with zones of the open plan type, due to the larger external wall to occupied offi ce fl oor area ratios (the occupants of the open plan are more evenly distributed within the whole fl oor space). In fully glazed building alter-natives the risk of comfort problems is higher; especially in cases when windows with high thermal, total solar and effective solar transmittance were applied. The variation in operative temperatures increased, leading to an unacceptable thermal environment. The increased fl oor to external wall area ratio of the open plan type, combined with the position of the occupants within the working area (in the open plan some occupants are distributed more evenly in the working space, decreasing the impact of radiant temperature on the perceived thermal comfort), results in a more stable thermal environment.

7.1.2 Control set pointsThe control set points for the indoor air temperature are crucial for the energy and the thermal performance of a given building alternative. The

Conclusions

313

three mean air temperature set points considered within this thesis were the strict (22-23°C), the normal (22-24.5°C) and the poor (21-26°C).

In general, the energy use increases when strict set points are applied. For the studied typical cell type reference building with 30% window to external wall area ratio (with 3 clear panes) the difference in energy use for heating between the strict – normal and strict - poor set points respectively is 7% (4 kWh/m²a) and 16% (9 kWh/m²a) . This difference tends to decrease as the glazing area increases (e.g. 5% and 7% are the corresponding values for a 100% glazed alternative); the difference for cooling demand is similar. When windows with lower thermal and total solar transmittance are used, the difference in energy use between the different set points drops.

Despite a common perception that strict control set points often lead to an improved thermal environment, this study has shown that the key to choosing proper set points is the consideration of design parameters such as plan type, internal loads and glazing area and type. In general, the choice of set point requires the determination of (a) the permissible air temperature variation and (b) the minimum and maximum permissible air temperature. Two of the main parameters that highly infl uence the PMV values are the indoor air and operative temperatures. For a given position of an occupant inside the offi ce space, the operative temperature depends on the surrounding surface temperatures. In order to achieve an improved thermal environment (assuming that the indoor air is well mixed), the optimal indoor air temperature should take into consideration the surface temperatures of, mainly, the external walls.

In general, a wide variation in indoor air temperatures results in PMV curves with larger difference between minimum and maximum PMV, while the maximum and minimum air temperature limits infl uence the absolute PMV levels. Since, however, the PMV values are also infl uenced by the radiative temperatures of the surrounding surfaces and the occupant’s position, it is obvious that the indoor air temperature is not the only factor that determines the quality of thermal environment.

As mentioned before, the two parameters that should be determined when deciding the set points are (a) the permissible air temperature vari-ation and (b) the minimum and maximum air temperatures. Three of the most important input design parameters to be considered for this selection are (a) the internal loads (e.g. the open plan requires lower maximum air temperature limits due to the increased occupancy), (b) the window size (since a larger variation in surface temperatures results in a larger variation in PMV values) and (c) the glazing and shading device type used (since the different thermal and optical properties have a large impact on the radiative temperatures).


314

For the reference building alternatives (30% window to external wall area ratio) strict set points may lead to comfort problems with excessive energy use, when the maximum and minimum temperature limits are not chosen properly (e.g. the strict set points with maximum air temperature limit of 23°C resulted in negative PMV values all year round, while the increased PMV values of the normal set points with maximum air temperature limit of 24.5°C resulted in similar PPD values with reduced cooling demand). Thus, for building alternatives with lower window to external wall area ratios, the PMV values can still vary within the accept-able PMV limits of ±0.5 (ISO Standard 7730), when set points with wider variation are allowed. For highly glazed facades narrow variations may have to be chosen.

When the glazing area increases, windows with low thermal transmit-tance can ensure inner pane temperatures closer to the indoor air ones. For the cases when low inner pane temperatures occur, increasing the minimum temperature limit can only improve the situation to some extent, since cold drafts can still occur, lowering the quality of thermal environment. During summer, attention should be paid to the selection of the maximum permissible air temperature, in order to avoid excessive cooling of the offi ce space. Windows with low total solar transmittance and/or shading devices that result in low geffective values (i.e. externally placed louvres or naturally ventilated double façades with venetian blinds) partly lower the inner pane temperatures. The maximum permissible air temperature should be decided according to the PMV values given for a certain window alternative. For cases with internally placed shading devices the maximum air temperature of 23°C was considered suffi cient, while for cases with lower g and geffective values the air temperature limit of 24.5°C was preferred. Another parameter that should be considered in highly glazed buildings is the larger inner pane temperature variation that occurs during the year. Thus, stricter set points (narrower air tem-perature variation) should be preferred in general, requiring more careful consideration of the input parameters, leading to optimal selection of maximum and minimum limits that will ensure acceptable PMV values throughout the year.

The selection of optimal temperature set points that ensure an accept-able thermal environment (avoiding at the same time excessive heating or cooling) is a complicated task. The selection of strict set points does not always guarantee improved indoor thermal environment, since if the input design parameters (e.g. occupancy, internal loads, window type and size, etc) are not considered thoroughly and the maximum and minimum permissible temperatures are not selected correctly, both discomfort and excessive heating or cooling may occur. Highly glazed buildings are more likely to lead to comfort problems and the selection of proper temperature

Conclusions

315

set points is more diffi cult. However, in any case the set points should be selected individually for each building and in cases with increased glazing areas further analysis should be carried out.

7.1.3 OrientationThe orientation of a building may have an effect on the energy use; how-ever, if the design is the same for the opposite sides of the façades the effect is negligible. Strict temperature control set points (22-23°C) decrease the impact of orientation on energy use.

Although the orientation of a zone has limited impact on the mean air temperatures, the directed operative temperatures may vary considerably. Wider permissible air temperature variations increase impact of the ori-entation in each zone (wider variation in operative temperature), mostly for zones with highly glazed areas.

It is evident that, in order to achieve an acceptable indoor thermal environment, proper set points should be selected after considering zones with different orientations. For buildings with larger window to external wall area ratios the need for careful set point selection increases, since in these cases the impact of orientation on the perception of thermal comfort increases.

7.1.4 Glazing areaAs expected, the total use of energy increases with the glazing area. As-suming a triple clear glazed window, the increase in energy use of a stud-ied typical 60% single skin glazed building, compared with the studied typical reference building, is 28 kWh/m²a (23%) for the normal indoor temperature control set points. The increase for the studied typical 100% single skin glazed alternatives is 57 kWh/m²a (47%). This difference in energy use would of course decrease if glazed facades with lower thermal and total solar transmittance (including solar shading, which results in lower geffective values) were chosen.

Increased window area does not necessarily mean a substantial decrease in the electricity for lighting. When the 30% and 100% alternatives are compared the decrease in energy does not exceed 1.3 kWh/m²a for the normal (with a set point of 500 lux at desktop) and 1.9 kWh/m²a for the poor set points (with a set point of 300 lux at desktop) and is even smaller for glazing with lower visual transmittance. Thus, the energy savings for lighting highly glazed buildings are often very small. The access to day-light might be higher with increased glazing area, but the quality of visual comfort could be compromised due to glare problems.


316

The larger the window to external wall area ratio, the larger the comfort problems are during the whole year, especially when the comparison is based on triple-glazed (clear glass) windows. During the winter the PMV drops from -0.55 (reference building) to -0.82 (100% glazed alternative) and during the summer it increases from 0.3 (reference building) to 0.45 (100% glazed alternative). This corresponds to a number of hours with PPD values lower than 10% of 73% for the reference building, 57% for the 60% glazed alternative and 31% for the 100% glazed alternative (with normal control set points). Thus, the need of glazing with lower thermal and total solar transmittance is evident, especially for highly glazed alternatives.

7.1.5 Glazing and solar shading typeThe window properties (U, g and geffective) have, as expected, a remarkable impact on energy use, which is also a main conclusion from this study. However, one of the main aims of the study was also to quantify the impact for Scandinavian climatic conditions. A description of the energy use, as simulated for the cases with normal control set points for the indoor air temperature, is given below.

Large glazing areas with three clear panes (with high U and g values) result in a high energy use for heating and cooling. When the three clear panes were replaced by glazing with lower U values (from 1.85 to 1.14 W/m2K), the total energy use for the studied fully glazed building with normal control set points dropped from 180 to 155 kWh/m²a mainly due to reduced heating. By decreasing the total solar transmittance of the glazing (0.35 instead of 0.58), the energy use for cooling dropped (from 37 to 27 kWh/m²a), while the energy use for heating increased (from 59 to 65 kWh/m²a); a further decrease in the total solar transmittance had a similar effect.

The position of shading, however, had a positive effect, with the in-termediate placed shading decreasing the cooling demand, while having almost no effect on the heating demand. Simulations have shown that for intermediate blinds the cooling demand (of a 100% glazed alternative with normal control set points) is 37 kWh/m²a, while when the blinds are internally placed, it increases to 54 kWh/m²a (both cases have Uglaz-

ing of 1.14 and gglazing of 0.58). Finally, externally placed shading devices decreased substantially the cooling demand compared with intermediate shadings. The cooling demand then dropped from 27 to 9 kWh/m²a, although the heating demand increased by 7 kWh/m²a, for alternatives with Uglazing of 1.14 and gglazing of 0.35. This case resulted in the lowest total energy use (141 kWh/m²a).

Conclusions

317

For 100% glazed alternatives (a) a decrease in thermal transmittance of the windows can lead to considerable energy savings for heating, (b) low total solar transmittance of the glazing unit results in lower cooling and higher heating demand and (c) low effective total solar transmittance by using intermediate or external shading devices can achieve a lower cooling demand, while it doesn’t have much effect on the energy use for heating.

As to indoor climate, highly glazed single skin alternatives give a wide variation in PMV values resulting in a poorer thermal environment. In general, narrower PMV variation can be achieved by applying windows with low thermal transmittance and low g and geffective values. However, the PMV values during winter still exceed the PMV comfort limit of -0.5 due to the low radiant temperatures. Fixed, external shading devices lower the PMV values during summer months but also during spring and autumn.

7.1.6 Double skin façadesIn order to reduce the energy use and improve the indoor thermal environ-ment of highly glazed buildings, double skin façades can be implemented. The double skin façade is a system consisting of two glass skins (single or double) placed in such a way that air can fl ow through the intermediate cavity. In principle the main purpose of the double skin façades, as to energy use and thermal comfort, is to allow the useful solar gains into the building when shading is not needed, and to extract, through the ventilated cavity, the heat absorbed by the shading to lower the cooling demand of the building. The distance between the skins usually varies from 0.2 m up to 2 m. For protection and heat extraction reasons the solar shading devices are placed inside the cavity. The ventilation of the cavity can be natural, fan supported or mechanical; the origin and destination of the air can also vary depending on the location, the use and HVAC strategy of the building.

The two modes considered in this study were the “typical double façade” and the “airfl ow window” mode. In the “typical double façade” air always enters the cavity from outdoors. The cavity can be (a) closed during the heating season for increased thermal insulation and opened during the cooling season for heat extraction purposes (naturally ventilated cavity), (b) used for preheating the air supplied in the AHU (as supply air) dur-ing the heating season and extracted during cooling periods (mechani-cally ventilated case) or (c) opened during the cooling periods, as in the naturally ventilated cases and used for preheating the air supplied in the AHU, as in the mechanically ventilated cases (hybrid ventilated cavity).


318

For the “airfl ow window” mode, the air always enters the cavity from the indoor offi ce space (exhaust air) all year round. The aim is to improve the inner glass temperatures for extreme winter and extreme summer condi-tions. During the heating season the cavity air ends up in the AHU for heat recovery purposes, while during the cooling season the air is extracted to the outside.

A main requirement to be fulfi lled when ventilated cavities of double skin façades are designed is effi cient heat extraction during the summer months. For naturally ventilated cavities the key parameters for achieving this are the characteristic height of the cavity (height to depth ratio) and the inlet and outlet opening size of the dampers. In general, the area occupied by the dampers when the cavity is fully open should be as small as possible, since the opening size is crucial for the size of the air fl ows. Cavities with the same opening size result in almost the same airfl ows independently of the cavity depth. Another parameter that has to be considered is the type and position of panes, since they have a considerable impact on the minimum cavity depth that ensures effi cient heat extraction.

During the parametric studies on a component level it was noted that solar control outer and especially intermediate panes require wider cavi-ties, due to the increased cavity air temperatures that result in high inner pane temperatures. Finally, for naturally ventilated cases, the position of shading device inside the cavity has limited impact on the thermal com-fort; the lower the thermal and indirect solar transmittances of the inner skin, the smaller is the impact. However, if the operable cavity is used for natural ventilation of the offi ce space, then the shading position should be considered.

The ventilation rate is crucial for the inner pane temperatures. Low airfl ow rates in the ventilated cavity can increase the risk of overheating the air and infl uence the inner pane temperatures. Especially in cases when no low E coated pane is applied at the inner skin, suffi cient heat extraction reduces the risk of thermal discomfort. This problem is more evident in mechanically ventilated cavities with lower airfl ow rates. In airfl ow window cases, in which the air enters the cavity from the offi ce zone, the overheating risk of the cavity air increases; especially when the indoor air temperature is lower than the outdoor air temperature, which is typical for Scandinavia. In that case a low E coated pane could reduce the heat transmission through the inner skin, keeping the cooling demand at reasonable levels.

Offi ce zones with all year round closed, naturally, mechanically or hybrid ventilated cavities perform well, mainly due to a decreased heating demand, when equipped with a window with an advanced intermediate pane (solar control and low E coated). Reducing the heating demand is essential for offi ce buildings located in Nordic countries. The position

Conclusions

319

of the advanced pane in this case is essential for its performance, since the increased indirect solar transmittance results in reduced transmission losses though the outer skin when the cavity is closed. This is true at least compared with an alternative with an outer solar control and low E intermediate pane. On the other hand, during the summer months, the suffi cient ventilation results in effi cient heat extraction and low cooling demand. The performance is better for a north oriented offi ce zone, since the heating demand is normally larger for a zone facing north. The effect is similar on thermal comfort; PMV values are higher in the winter and slightly lower during summer due to the lower total solar transmittance. Similar but less intense is the trend for the mechanically and hybrid ven-tilated cavity, with an advanced (low E + solar control) intermediate pane performing best.

A single skin case with intermediate blinds compared with the naturally and hybrid ventilated double skin case shows that the savings in cooling demand of the south oriented zones are important, while for the north they are very small. Larger are the savings in the glazing alternatives with high thermal and total solar transmittance values. The heating demand tends to increase, however, due to (a) leaky dampers in the case when the cavity is considered closed and (b) not optimal damper control set points (dependence of the dampers’ temperature set points on the cavity air temperature). When the total energy use is compared, the single skin façades perform better than the double for the north oriented zones for the two reasons mentioned above. This can be explained by the fact that the simulations were carried out for Sweden where the heating demand is a dominating factor. However, for the south oriented zones a decrease in energy use can be noted with the double façade cases, with the hybrid ventilated one performing best. In general, hybrid ventilated cases tend to perform best as to energy use (regardless of the orientation), while airfl ow windows provide a better indoor thermal environment, mostly with north oriented façades, due to the higher PMV values during winter.

The energy savings, which can be achieved for a highly glazed offi ce building by using double skin façades instead of single skin façades, are rather small. A typical double façade hybrid building (best performing) has a total energy use of 137 kWh/m2a vs. 144 kWh/m2a for a single skin façade building (same glazing cases with low U and geffective values, non-ventilated cavity and intermediate shading). As regards thermal comfort, the double façade alternatives perform similarly to the single skin ones (PPD values), while the airfl ow window cases perform slightly better due to the higher (and closer to the 0 axis) PMV values during the winter months.

Airfl ow windows with the same thermal and optical properties, on the other hand, result in high energy use (but still lower than the inter-


320

nally placed shading devices) due to increased cooling demand. However, when the inner clear pane is replaced with a low E hard coated pane, the performance increases dramatically. In airfl ow window cases with low E inner pane the quality of the thermal environment can improve drasti-cally, reaching the comfort levels of a 30% glazed building with improved window thermal and optical properties.

7.1.7 Best performing alternativesOften, highly glazed buildings perform poorly when compared to a tra-ditional building with e.g. 30% glazed facade (reference case) with regard to energy use and thermal comfort. As to energy use, windows with low thermal and total solar transmittance are essential for improving the build-ing’s performance, especially since the energy use may vary substantially in highly glazed buildings, depending upon design. The position of shading devices also plays a major role. Fixed external louvres decrease the cooling and increase the heating demand. The effect is opposite for the internally placed venetian blinds resulting in the highest total energy use due to the drastic increase in the cooling demand. Glazed façades with intermedi-ate venetian blinds can perform slightly poorer than those with externally placed ones due to the larger cooling demand. However, if a double skin façade is used and the cavity with the shading device is ventilated, the total demand can drop further.

The difference in total energy use between a reference building (30% glazed alternative) and a highly glazed hybrid ventilated double skin façade, could for the studied cases be reduced by 25 kWh/m2a, (113 kWh/m2a vs. 138 kWh/m2a), if the glazing properties are kept the same. When exter-nal louvres are placed in a single skin alternative (100% glazed) with the same window properties, the building performs similarly (increase in total demand by 3 kWh/m2a). The same alternative but with internal venetian blinds instead results in a further increase by 10 kWh/m2a giving a total energy use of 151 kWh/m2a. An airfl ow window alternative with further improved glazing properties results in a total demand of 132 kWh/m2a; only 19 kWh/m2a higher than in the reference case.

As regards the quality of indoor environment it is evident that smaller glazing areas provide a more stable environment with fewer dissatisfi ed occupants. If the glazing is selected carefully (i.e. low U values), the percentage of dissatisfi ed occupants will still increase mainly due to the low inner surface temperatures during winter. The reference building can provide a thermal environment, in which there will be fewer than 15% of dissatisfi ed occupants for 97% of the working hours. This value drops for the alternative with the hybrid ventilated double façade to 85% of the

Conclusions

321

working hours. However, if the glazing is further improved it is possible that the quality of thermal environment can reach the level of the reference case by using an airfl ow window.

7.2 Methods for determining energy and indoor climate performance

Energy and indoor climate simulations of an offi ce building should pref-erably be carried out already at an early design stage and then be refi ned during the actual design. This will ensure improved energy and indoor climate performance of the building. In order to achieve energy effi cient buildings and improved thermal environment, it is essential that the steps described below should be followed:

• Validation of the calculation methods used should be the fi rst step in a successful energy effi cient building design. Understanding of the possibilities and limitations of the computing tools and the physical models used, enables accurate predictions. Validation tests that have been carried out in the past can help the user recognize the strengths and weaknesses of each tool and use the right tool for the right purpose or even combine different ones when needed.

• Simulations on a component level (such as those carried out for the double skin façade alternatives) can provide the necessary background to the possibilities and limitations of the system used. These simulations can often be very detailed and require a lot of time and effort, in order to meet a certain level of accuracy. In practice, however, deciding the acceptable level of accuracy is not an easy task, since this depends on the scale and complexity of the project, the experience of the user and the available time.

• Performance and quality specifi cations have to be fulfi lled. Once the validation of the calculation methods has been completed and a deeper understanding of the integrated components has been achieved, it is essential that the client and the design team, includ-ing the engineer responsible for energy and thermal simulations, prioritize the goals and decide to what extent achieving low energy and improved thermal environment can and should infl uence the building design. In practice, many mistakes may originate from the lack of communication and clear goals of the design team.

• Parametric studies on a zone level: A clear view of the parameters that can be varied, in order to lower the energy use and improve the thermal performance is the fi rst step for selecting the simulated


322

alternatives. In general, parametric studies on a zone level can pro-vide useful information regarding the impact of different design parameters on the building performance, especially regarding com-fort. The reduced simulation time can allow the user to investigate a larger number of alternatives, but can not predict absolute values for the whole building performance.

• Simulations on a building level: Once the best performing alter-natives on a zone level are selected, the whole building should be modelled. However, if the implimented building model is too simplifi ed (e.g. assuming the whole fl oor as one zone or calculating the average cavity air temperature of a multi storey façade), this can lead to wrong values and overestimation of the building perform-ance. Simulations on a building level can predict and improve the building performance, as long as input parameters (e.g. internal loads, thermal mass, control set points, ventilation rates, etc) are correctly estimated.

7.2.1 Lessons learnt from the simulation workA challenge in this thesis has been to correctly simulate buildings with double skin façades, since the modelling of especially naturally ventilated double skin façades is a complicated task. On a component level the software tool WIS 3 was used. WIS 3 can provide a wide range of output values but on the other hand the simulations are carried out for steady-state boundary conditions, which do not allow the user to investigate the façade performance throughout the year. The determination of input parameters, such as cavity geometry, position of shading devices inside the cavity etc, allows the user to carry out an extensive parametric study, in order to understand the infl uence of the key parameters on the façade performance.

Energy and indoor climate simulations on a zone and building level were carried out using IDA ICE 3.0. The dynamic building energy simu-lation tool IDA ICE 3.0 allows energy and indoor climate simulations of complicated buildings with integrated double skin façades during the whole year. The input options regarding a double skin façade e.g. cavity geometry, shading position, are not so many and certain values (such as discharge coeffi cients for the openings, etc) have to be calculated by the user before being input in the software. At the advanced level of IDA ICE 3.0 the user has a great variety of options. Some of these options, used within this study, were (a) different modelled ventilation modes (such as the naturally, mechanically and hybrid ventilated double facades and airfl ow windows), (b) different damper controls, e.g. dampers opening

Conclusions

323

according to the cavity air temperature for the zone simulations, while on a building level they were dependent on the indoor air temperature and (c) multi storey façades simulated by connecting cavities at different fl oors. In general IDA ICE 3.0 allows a wide variety of confi gurations, since the user can build his/her own case. The main drawback, however, is that the user has to be quite experienced, since most of the models have to be developed at the advanced level.

Correct modelling of offi ce buildings can be a complicated task. The large number of offi ce zones can be a drawback for the speed of simula-tion, often limiting the number of parametric studies. On the other hand, simplifying the building model too much can lead to wrong assumptions which diminish the quality of output results and mislead the design team regarding the energy and thermal building performance.

Further improvement of the advanced building simulation tools, which includes easier interfaces, reduced simulation time, but also more options for the user to create individual building components and insert them into the building model, are essential for broader use. In order to correctly use building simulation software, documentation of the physical and math-ematical models included should be available to the user. In this way, the possibilities and limitations would be clear when the user attempts to do his/her own modelling.

7.3 Improving the energy and indoor climate performance of highly glazed buildings: general recommendations and further studies

Low energy building design would be a relatively easy task, if it were always prioritized as one of the main performance requirements to be achieved. However, in practice the building “concept” is often initially created lead-ing to a building design that does not always take into account energy and thermal performance. An example of this is the increasing number of highly glazed offi ce buildings built all around the world during the last decades. Despite their poorer energy and thermal performance, most architects (following the trend of highly glazed areas), companies (which want to create a distinctive image for themselves) and users (who relate highly glazed façades to a more pleasant environment) prefer this building type.

Before anything else, it has to be stated that the low energy building design is directly connected with the location of the building, since the


324

climatic conditions are essential to its performance; especially for buildings with large glazing areas, general recommendations should be avoided. Thus, careful and individual design is an indisputable requirement, in order to achieve improved building performance.

For Swedish climatic conditions, windows with low thermal transmit-tance are essential. This is especially true for highly glazed buildings, in order to improve the building’s performance as to energy use and thermal comfort during the winter months. The results have shown that in cases with windows with low g values, the increase in heating demand is una-voidable; this is not the case when windows with relatively high g but low geffective (when shading is used) values are applied. Similar results were obtained for the perception of thermal comfort, since the cold pane surfaces result in negative PMV values during winter. Hybrid ventilated double façades, however, can reduce the heating demand and improve the quality of thermal environment. Airfl ow windows have a positive effect on thermal comfort mainly during winter due to the increased PMV values. The results show that airfl ow windows may perform better in north facing façades. Air fl ow windows with (hard coated) low E inner pane result in a radical improvement of the indoor climate, since the PPD values are the same as for a 30% glazed building with windows of low U and g values.

On the other hand, glazing with low total solar transmittance values reduces the cooling demand and decreases the risk of overheating. Fixed external louvres have defi nitely a positive effect on cooling, but since they are applied all year round they have a negative effect during winter, increas-ing the heating demand. Internally placed venetian blinds have the opposite effect, but the radical increase in cooling demand and radiant temperatures leads to poor all-year performance. Intermediate placed movable shading improves the building’s performance, since they result in higher g values when the shading is not in use and solar heat gain is needed, but also due to low geffective values that result in acceptable performance during sum-mer. When the cavity in which the shading is placed in the double façade is ventilated, a further improvement can be noticed.

When integrated properly, double skin façades result in improved (compared with single skin facades) energy and thermal performance of the building, mainly when used on the south façade. However, since the cooling demand is rather limited for Scandinavian climatic conditions, their impact is also limited. Hybrid ventilated façades perform better than naturally or mechanically ventilated ones, since they effi ciently combine heat recovery during winter and effi cient heat extraction by natural ven-tilation during summer.

Optimal damper control set points and suffi ciently sealed cavities when dampers are closed are essential to the system’s performance. The usually higher construction cost of double skin façades, on the other hand, makes

Conclusions

325

their application in buildings located in Scandinavia questionable. Further investigation of their possibilities for buildings located in warmer climates, however, should be carried out, since their impact on the building per-formance may be promising.

Individual building design that takes into consideration the type of façade including the size and type of glazing and the position of shading devices, the temperature set points, the building occupancy and plan type, can defi nitely lead to improved building performance. A holistic view that takes into account the interaction of the different building components is essential for low energy building design. If this is established, even in highly glazed cases, the building performance can reach reasonable levels as to energy use and indoor climate.

Further studies are suggested in the following fi elds:

• impact of building shape on the energy and thermal performance of highly glazed buildings

• detailed CFD calculations on double skin facades; infl uence of geometrical characteristics on airfl ows

• optimization of DSF constructions and modes for different climates to study the potential of energy savings due to the use of double skin façades

• optimization of damper control set points for double skin façades (cavity air, indoor air, operative temperatures, etc)

• further development of dynamic building energy simulation tools with integrated double façade

• further validation of tools for double skin façade cases with venetian blinds


326

Summary

327

8 Summary

8.1 IntroductionHighly glazed offi ce buildings are considered to be airy, light and trans-parent with more access to daylight than traditional buildings, but their energy effi ciency is often questioned. However, nowadays glazed buildings are increasingly being built around the world, because (a) there is a growing tendency among architects to use large areas of glass in the façade, often with the aim of contributing to a better view of the outside and access to daylight, (b) users often like the idea of increased glass area, relating it to a better view of the outside and a more pleasant indoor environ-ment and (c) many companies prefer the distinctive image of themselves (e.g. transparency or openness) that a glazed offi ce building can provide. Achieving low energy use and improved thermal environment when using a fully glazed façade can be a great challenge; energy effi ciency and the provision of an acceptable indoor climate are essential for improved build-ing performance. Other parameters to be taken into consideration during the decision and design process are visual comfort, building aesthetics, sociological and psychological determinants (such as visual and acoustical privacy), life cycle cost, etc. By prioritizing at an early stage the goals to be achieved, the design team can improve the building performance and fulfi ll the design requirements.

8.2 BackgroundToday, there is insuffi cient knowledge of the function, energy use and indoor environment of offi ce buildings with highly glazed facades for Scandinavian conditions. Therefore, a project was initiated to gain knowledge of the possibilities and limitations of glazed offi ce buildings in Scandinavian climates. This means further development of calculation methods and analysis tools, improvement of analysis methodology, calcu-lation of life-cycle costs (LCC), compilation of advice and guidelines for the construction of glazed offi ces, and strengthening and improving the


328

competence regarding sustainable buildings in Sweden. This thesis, which is a part of the “Glazed Offi ce Buildings” project, aims to (a) determine how the energy and indoor climate performance can be analysed, (b) clarify and quantify how highly glazed façades affect the energy use and thermal comfort and (c) determine how the design can be inproved with regard to energy effi ciency and thermal comfort. Through extensive parametric studies the impact of design parameters on building performance has been studied. Optimizing energy and indoor climate performance of single skin highly glazed buildings was the fi rst goal to be achieved. The proper integration of double skin façade systems was also investigated with the aim of further improvements.

8.3 MethodsA virtual reference building was initially created, which was considered to be representative of Swedish offi ce buildings built in the late nineties with regard to design, energy and indoor climate performance. The design of the building was determined by researchers from the Division of Energy and Building Design, architects and engineers from WSP and Skanska. First, detailed performance specifi cations for energy and indoor climate were established and then typical constructions were determined for the reference building. System descriptions and drawings were prepared. The building was approved by a reference group. Finally a validation of the simulated performance of the reference building showed that the perform-ance specifi cations were fulfi lled.

The reference building is a 6 storey building of rectangular shape (21 m high, 66 m long and 15.4 m wide; the room height is 2.7 m) and the distance between fl oors is 3.5 m. Two plan types were assumed for the simulations; cell and open plan. In order to reduce the simulation time in IDA ICE 3.0, but still be able to analyse the indoor climate for individual rooms, the number of zones was reduced to 11 per fl oor for the cell type and to 7 per fl oor for the open plan type. Input parameters such as occupancy, HVAC strategy, internal loads, etc, were decided after thorough consideration and long discussions with experts of the working group; these parameters are described in great detail in the report. Three different control set point intervals for the indoor temperature were cho-sen for the simulations of the building alternatives. The normal control set point (22 - 24.5°C) is considered the standard (reference) case, since the lower and upper temperature limits are common practice in modern Swedish offi ces. However, the two other control set points (21 - 26°C (poor) and 22 - 23°C (strict)) can provide useful information concerning

Summary

329

the variation in energy use with indoor temperature and the perception of thermal comfort.

For this building, a parametric study of energy use and indoor climate was carried out. The building construction, the HVAC system, the occu-pancy, etc were modelled in great detail. Parameters such as the building’s orientation, plan type, control set points, façade elements (window type and area, shading devices, etc) and façade type (single and double skin) were varied during the simulations, while others such as building shape, occupants’ activity and schedules, etc were kept the same. A sensitivity analysis based on the simulated alternatives was carried out regarding the occupants’ comfort and the energy use for operating the building.

The window to external wall area ratio of the simulated single skin building alternatives varied from 30% (reference case), to 60% and 100%. For the 30% glazed alternatives a triple-glazed (clear glass) window with a venetian blind between two of the panes was assumed, while the build-ing’s construction was kept the same. The simulations were carried out for three orientations, three control set points (strict, normal and poor) and two plan types (cell and open). For the 60% and 100% single glazed alternatives, 7 different (commercially available) window constructions were studied. The rest of the building construction was kept the same as for the 30% glazed alternative. Each case was then simulated for both cell and open plan type and for strict, normal and poor control set points. In total 84 (60% and 100%) glazed alternatives were simulated.

The simulation tools used were (a) IDA ICE 3.0, a dynamic building energy simulation tool, and (b) WIS 3, a calculation tool developed to calculate the thermal and solar characteristics of window systems and com-ponents. Validation tests have shown that both programs give reasonable results and are applicable to detailed simulations. In order to analyse the large amount of output data from the IDA ICE 3.0, simulations a post processor in MS Excel was developed.

The performance parameters examined on a building level in this report are (a) heating demand, cooling demand and electricity for light-ing, pumps and fans, etc, (b) weighted (average) air temperatures for the working area, (c) number of hours between certain (weighted) average air temperatures for the working space, (d) weighted average PMV and (e) number of working hours for certain average PPD. On a double skin façade component level the main parameters calculated were the temperatures of different layers at the vertical and horizontal centre and the airfl ow rates and temperatures along the ventilated cavity.


330

8.4 Discussion and conclusions

8.4.1 Glazing areaIn general, the total use of energy increases with increased glazing area. Assuming a triple clear glazed window the energy use of the studied typical 60% single skin glazed building is 23% higher for the strict and normal indoor temperature control set points, compared with the studied typical reference building. The increase for the 100% single skin glazed alterna-tives is 45 % and 47 % respectively. This difference in energy use would of course decrease if glazed facades with lower thermal and total solar trans-mittance (including solar shading, which results in lower geffective values) were chosen. Furthermore, increased window area does not necessarily mean a substantial decrease in the electricity for lighting. According to the simulations the maximum savings did not exceed 2 kWh/m²a. Additionally, the larger the window to external wall area ratio, the larger are the comfort problems during the whole year, especially when triple-glazed (clear glass) windows are used. Thus, the need of glazing with lower thermal and total solar transmittance is evident, especially for highly glazed alternatives.

8.4.2 Glazing and solar shading typeThe window properties (U, g and geffective values) have a remarkable impact on energy use, as concluded from the study and as already expected. For the Swedish climatic conditions, windows with low thermal transmittance are essential, in order to improve the building’s performance during the winter months regarding energy use and thermal comfort. The results have shown that in cases with windows with low g values, the increase in heating demand is unavoidable. Similar results were obtained for the perception of thermal comfort, since the cold pane surfaces result in negative PMV values during winter. On the other hand, glazing with low total solar transmittance values reduces the cooling demand and decreases the risk of overheating. Fixed external louvres have defi nitely a positive effect on cooling, but since they are applied all year round they increase the heating demand during winter. Movable shading and effi cient heat extraction (as in the case of exterior shading) can be achieved with the use of venetian blinds in double skin façades. Internally placed venetian blinds have the opposite effect. The radical increase in cooling demand and radiant tem-peratures leads to poor year- round performance. Intermediate moveable shading improves performance, since they result in higher g values when

Summary

331

the shading is not in use and solar gain is needed, but also due to low gef-

fective values that provide acceptable performance during summer.

8.4.3 Double skin facadesIn order to reduce energy use and improve the indoor thermal environ-ment, double skin façades were introduced. The double skin façade is a system consisting of two glass skins (single or double) placed in such a way that air can fl ow through the intermediate cavity. In principle the main purpose of the double skin façades (as to energy use and thermal comfort) is to allow the useful solar gains into the building when shading is not applied, and to extract through the ventilated cavity the heat absorbed by the shading to lower the cooling demand of the buildings. The distance between the skins usually varies from 0.2 m up to 2 m. For protection and heat extraction reasons the solar shading devices are placed inside the cavity. The ventilation of the cavity can be natural, fan supported or mechanical; the origin and destination of the air can also vary depending on the location, the use and HVAC strategy of the building.

A main requirement when ventilated cavities of double skin façades are designed, is effi cient heat extraction during the summer months. For naturally ventilated cavities the key parameters for achieving that are the characteristic height of the cavity (height to depth ratio) and the inlet and outlet opening sizes of the dampers. Another parameter that has to be considered is the type and position of panes, since they have a considerable impact on the minimum cavity depth that ensures effi cient heat extrac-tion. For naturally ventilated cases, the position of shading device inside the cavity has limited impact on the thermal comfort, unless the openable cavity is used for natural ventilation of the offi ce space.

The energy savings achieved for a glazed offi ce building by using double skin façades instead of single skin façades are rather small. When the double façade hybrid case (best performing as to energy use) is compared with single skin façade buildings (same glazing cases with low U- and geffective values, no ventilated cavity and intermediate shading), the total energy use decreases by 7 kWh/m2a (137 instead of 144 kWh/m2a). Regarding thermal comfort, the double façade alternatives perform similarly to the single skin ones (PPD values), while the airfl ow window cases perform slightly better due to the higher (and closer to the 0 axis) PMV values during the winter months. Airfl ow windows with the same thermal and optical properties, on the other hand, result in high energy use due to increased cooling demand, but still lower than for the case with internally placed shading devices. However, when the inner clear pane is replaced with a low E hard coated pane, the performance improves dramatically.


332

As to thermal comfort the airfl ow windows perform better than the rest of the cases due to the increased PMV values during winter months. In cases with low E inner pane the quality of the thermal environment can improve drastically, reaching the comfort levels of a 30% glazed building with improved window thermal and optical properties.

8.4.4 Other parameters that infl uence the building performance

The plan type of offi ce buildings plays a considerable role in the build-ing performance regarding energy use and indoor environment issues. The two clearly distinguishable plan types studied in this thesis were the open plan and cell offi ce. The use of the offi ce space, ventilation rates, occupant density and the quantity of offi ce equipment are some of the main parameters that differ between these two types. This results in (a) often higher cooling and lower heating demand of the open plan offi ce building due to higher internal loads, mainly caused by higher occupancy and lower ventilation rates and (b) the risk of discomfort increases in cell offi ce zones, compared with zones of the open plan type, due to the larger external wall to offi ce fl oor area ratios (the occupants of the open plan are more evenly distributed within the whole fl oor space). In fully glazed building alternatives the risk of comfort problems is higher, especially in cases when windows with high thermal, total solar and effective (i.e. shading is applied) solar transmittance were applied.

The control set points for the indoor air temperature are crucial for the energy and the thermal performance of a given building. Despite a com-mon perception that strict control set points lead to an improved thermal environment, this study has shown that the key to choosing proper set points is the consideration of design parameters such as plan type, internal loads and glazing area and type. In general, the choice of set point requires the determination of (a) the permissible air temperature variation and (b) the minimum and maximum permissible air temperature limits. The selection of strict set points does not always guarantee an improved indoor thermal environment, since if the input design parameters (e.g. occupancy, internal loads, window type and size, etc) are not considered thoroughly and the maximum and minimum permissible temperature limits are not selected correctly, both discomfort and excessive heating or cooling may occur. Highly glazed buildings are more likely to lead to comfort problems and the selection of proper temperature set points is more diffi cult. In any case, the set points should be selected individually for each building and in cases with increased glazing areas further analysis should be carried out.

Summary

333

The orientation of a building may have an effect on the energy use; however, if the design is the same for the opposite sides of the façades the effect is negligible. The orientation of a zone has limited impact on the mean air temperatures, but the directed operative temperatures may vary considerably. The wider the permissible air temperature variation, the bigger is the impact of the orientation in each zone (wider variation in operative temperature) mostly for zones with highly glazed areas. This could provide useful information for the choice of the permissible air temperature variation of certain zone types, in order to keep the operative temperature within acceptable limits.

8.4.5 Best performing alternativesOften, highly glazed buildings perform poorly when compared to a tra-ditional building with e.g. 30% glazed facade (reference case) with regard to energy use and thermal comfort. As to energy use, the use of windows with low thermal and total solar transmittance is essential for improving the building’s performance, since especially in highly glazed buildings the energy use may vary substantially, depending upon design. The position of shading devices also plays a major role; fi xed external louvres decrease the cooling and increase the heating demand. The effect is opposite for internally placed venetian blinds resulting in the highest total energy use due to the drastic increase in the cooling demand. Glazed façades with intermediate placed venetian blinds can perform slightly poorer than externally placed shadings due to the larger cooling demand, unless the cavity where the shading devices are placed is ventilated. In this case the total demand can drop further.

The difference in total energy use between the reference building (30% glazed alternative) and the best performing highly glazed building (hybrid ventilated double façade), could be limited to only 25 kWh/m2a in cases when the glazing properties were kept the same. When external louvres are placed in a single skin alternative with the same window properties, the building performs similarly. The same alternative but with internal venetian blinds instead results in a further increase by 10 kWh/m2a (total energy use of 151 kWh/m2a). An airfl ow window alternative with further improved glazing properties results in a total demand of 132 kWh/m2a; only 19 kWh/m2a higher than in the reference case.

Regarding the quality of the indoor environment, it is evident that smaller glazing areas provide a more stable environment, with fewer dis-satisfi ed occupants. If the glazing is selected carefully, the percentage of dissatisfi ed occupants will still increase mainly due to the low inner surface temperatures during winter. For the reference case with the same window


334

properties as described above, the reference building can provide a thermal environment, in which there will be fewer than 15% dissatisfi ed occupants for 97% of the working hours. This value drops for the alternative with the hybrid ventilated double façade to 85% of the working hours. However, if the glazing is further improved, it is possible that the quality of thermal environment may reach the level of the reference case.

8.4.6 Determination of energy and indoor climate performance

In principle, energy and indoor climate simulations have to be carried out already at an early design stage and then be refi ned during the actual design. This will ensure improved energy and indoor climate performance of the building. In order to achieve energy effi cient building design and improved thermal environment it is essential to (a) validate the calcula-tion methods, (b) carry out simulations on a component level in order to gain the necessary background to the possibilities and limitations of the system, (c) prioritize the performance and quality requirements to be fulfi lled and (d) carry out simulations on a zone and on a building level. Parametric studies on a zone level can provide useful information regarding the impact of different design parameters on the building performance, while parametric studies on a building level predict absolute values of the building.

8.5 General recommendations for improvements of highly glazed offi ce buildings

In general, low energy building design would be a relatively easy task if it were always prioritized as one of the main performance requirements to be achieved. However, in practice the building “concept” is often initially created leading to a building design that does not always take into account energy and thermal performance. An example of this is the increasing number of highly glazed offi ce buildings built all around the world during the last decades. Within this study, however, an effort has been made to quantify the performance of highly glazed offi ce buildings, as to energy use and indoor climate issues, for Scandinavian climatic conditions and suggest improving solutions. It is always required that the low energy building design is directly related to the building location, since the cli-

Summary

335

matic conditions are essential to its performance. Especially for buildings with large glazing areas, general recommendations should be avoided. For instance a double façade performing well in a south European country could be unacceptable for Scandinavia, and vice versa. Thus, careful and individual design is an indisputable requirement, in order to achieve im-proved building performance.

For Swedish climatic conditions during the winter months, windows with low thermal transmittance are essential, especially for highly glazed buildings, in order to improve the building’s performance as to energy use and thermal comfort. Glazing with low total solar transmittance values reduces the cooling demand and decreases the risk of overheating. The best solution is external movable solar shading to arrive at low g-values during periods with warm weather, as in winter low g values are not needed and the solar gains can contribute to space heating. However, if the g value of the glazing is too low, the light transmittance might be insuffi cient. The external movable solar shading can also be located in the cavity of a double skin façade, in order to be protected. When g values and U values are chosen for a façade, the area and orientation of the glazing have to be taken into account. The total solar gains and total thermal losses of a zone are important.

When integrated properly, double skin façades result in improved energy and thermal performance of the building mainly when applied on the south façade, but since cooling is not the main issue for Scandinavian climatic conditions, their impact is limited.

Individual building design that takes into consideration the type of façade including the size and type of glazing, the position of shading de-vices, the temperature set points, the building occupancy and plan type, can defi nitely lead to improved building performance. If this is established, even in highly glazed cases, the building performance can reach reason-able levels as to energy use and indoor climate. However, a building with low energy demand cannot be achieved with a highly glazed building in a Scandinavian climate.


336

References

337

References

Achermann, M., & Zweifel, G. (2003). RADTEST — Radiant Heating and Cooling Test Cases. Universtiy of Applied Sciences of Central Switzerland, Lucerne, Switzerland.

Action Plan for Energy Effi ciency: Realising the Potential, COMMISSION OF THE EUROPEAN COMMUNITIES, Brussels, 19.10.2006

Architect's Council of Europe, Brussels, Energy Research Group, Univer-sity College Dublin, Suomen Arkkiteillitto, Helsinki and SOFTECH, Turin (1999). A Green Vitruvious. ISBN: 1-873936-94-X

Adalberth, K. (2000). Energy Use and Environmental Impact of New Resi-dential Buildings. Department of building Physics, Lund University, Lund, Sweden. Report TVBH-1012.

Andersen, I. (2000). A Multi-Criteria Decision-Making Method for Solar Building Design. PhD thesis for the degree of Doctor Ingeniør at the Nowegian University of Science and Technology, Faculty of Architec-ture, Planning and Fine Arts, Department of Building Technology, Norway.

ASHRAE (American Society of Heating, Refrigerating and Air Conditio-ning Engineers). (2001). Handbook of Fundamentals. Atlanta, USA

ASHRAE (American Society of Heating, Refrigerating and Air Conditio-ning Engineers). (2005). Handbook of Fundamentals. Atlanta, USA

ASHRAE Standard 55 (1992). Thermal Environmental Conditions for Human Occupancy

BLAST software (1998). USA CERL and the University of Illinois. Web address: http://www.cecer.army.mil/

BRIEFING: Directive 2002/91/EC of the European Parliament and of the Council of 16 December 2002 on the Energy Performance of Buildings, 7 November, 2006


338

Bring, A., Sahlin, P., & Vuolle, M. (1999). Models for Building Indoor Climate and Energy Simulation. Report of IEA, Task 22, Subtask B: Building Energy Analysis Tools Model Documentation

BSim 4 software (2004). Danish Building Research Institute Web address: http://www.sbi.dk

Bülow-Hübe, H., (2007) Daylight in Offi ce Buildings. Division of Energy and Building Design, Department of Architecture and Built Environ-ment, Lund Institute of Technology, Lund University, Sweden.

Button, D., & Pye, B. (1993). Glass in building; A guide to modern ar-chitectural glass performance. Pilkington Glass Ltd. with Butterworth Architecture. ISBN 0750605901

CAPSOL software. Physibel, Belgium. Web address: http://www.physibel.be/

Champagne, C. (2002). Computational Fluid Dynamics and Double Skin Facades. Assignment for the Architectural Engineering Computer Labs, Pennsylvania State University, USA Web address: http://www.arche.psu.edu/courses/ae597J/Champage-Hw1.pdf

Carlson, P.-O. (2003). Glazed facades – double skin facades – requirements and methods. Arkus, Byggförlaget skrift nr 42, Stockholm (in Swe-dish).

Carlsson, P.O., 2005. Construction with glass. Swedish glass trade union, Stockholm, Sweden (in Swedish).

Chen, Q. (1997). Computational fl uid dynamics for HVAC: successes and failures. ASHRAE Transactions, 103(1), 178-187.

Chrisoffersen, J. (1995). Daylight utilization in offi ce buildings. PhD thesis, Danish Building Research Institute, Report 258.

Claessens, J., & DeHerte, A. Active Solar Heating and Photovoltaics. Solar Energy in European Offi ce Buildings. Energy Research Group, School of Architecture, University College of Dublin, Ireland. Web address: http://erg.ucd.ie/mid_career/mid_career.html

Compagno, A. (2002). Intelligent Glass Facades (5th revised and updated edition). Berlin: Birkhäuser

Crawley, D. B., Hand, J. W, Kummert, M., & Griffi th, B. T. (2005). Contrasting the Capabilities of Building Energy Performance Simulation Programs. Proceedings of Building Simulation IBPSA 2005, 15-18 August 2005, Montreal, Quebec, Canada.

References

339

Derob-LTH software (2007). Lund Institute of Technology, Lund Uni-versity. Web address: http://www.derob.se

Di Maio, F., & van Paassen, A.H.C. (2000). Second skin façade simula-tion with Simulink code. International symposium air conditioning in high rise buildings (Shanghai), s.n.,s.l.,ISBN 2-913149-07-3, cat c, Projectcode: 06A-V

Di Maio, F., & van Paassen, A.H.C. (2001). Modelling the air infi ltrations in the second skin facade. Proceedings of the 4th international confe-rence on indoor air quality, ventilation and energy conservation in buildings, October 2-5 (pp. 873-880). Hong Kong: City University of Hong Kong. (TUD)

Ding, W., Hasemi, Y. & Yamada, T. (2004). Natural Ventilation of a Double-Skin Façade with Solar Chimney. Energy and Buildings Vol. 37, No. 4, 411-418, 2005.

DIRECTIVE 2002/91/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 16 December 2002 on the energy performance of buildings Offi cial Journal of the European Commu-nities, 2003

Djunaedy, E., Hensen, J.L.M., & Loomans, M.G.L.C. (2002). Towards a Strategy for Airfl ow Simulation in Building Design Center for Building & Systems. TNO - TU/e. Technische Universiteit Eindhoven, the Netherlands. Web address: http://sts.bwk.tue.nl/erdj/papers/room-vent2002.pdf

DOE-2.1E software. Lawrence Berkeley National Laboratory. Web address: http://simulationresearch.lbl.gov/

ECOTECT software. c/o Centre for Research in the Built Environment. Web address: http://www.squ1.com

Energiboken (1995), Swedish Council for Building Research (in Swedish) Offi ce Buildings.

Energy plus software (v. 1.2.2). U S Department of Energy. Web address: http://www.energyplus.gov

Erhorn, H., & Stoffel, J. (eds.) (1996). ADELINE 2.0, Advanced dayligh-ting and electric lighting integrated new environment, Documentation of the software package ADELINE, version 2.0 (9 volumes), IEA Solar Heating and Cooling Program, Task 12: Solar Building Energy Ana-lysis Tools.


340

ESp-rsoftware (v. 10.1). Department of Mechanical Engineering, Univer-sity of Strathclyde. Web address: http://www.esru.strath.ac.uk/

Faist, A. P. (1998). Double Skin Walls. Institut de technique du batiment. Department d’ Architecture. École Polytechnique Fédéral de Lausanne (EPFL), Switzerland

Fanger, P. O. (1970). Thermal Comfort: Analysis and Applications in Environmental Engineering: Danish Technical Press, Copenhagen, Denmark

Flamant, G., Heijmans N., Guiot E., Gratia, E. & Bruyer,e I. (2004). Ventilated Double Facades. Determination of the energy performances of ventilated double facades by the use of simulation integrating the control aspects – Modelling aspects and assessment of the applicability of several simulation software. Final report, Belgian Building Research Institute, Dept. of Building Physics, Indoor Climate and Building Services, December, 2004.

Fuller, S. K., & Petersen, S. R. (1996). Life Cycle Costing Manual. Buil-ding and Fire Research Laboratory, Offi ce of Applied Economics, Gaithersburg, MD 20899

Gertis, K. (1999). Sind neuere Fassadenentwickelungen bauphysikalisch sinnvoll? Teil 2: Glas-Doppelfassaden (GDF) (in German), Bauphysik, vol. 21, pp. 54-66.

Gratia, E. and De Herde, A. (2004). Natural Ventilation in a Double-Skin Façade. Energy and Buildings 36(2004) 137-146.

Hanssen, O. S. (2000). Evaluation of association between indoor air climate wellbeing and productivity, Healthy Buildings, Helsinki, Finland.

Hendriksen, O. J., Sørensen, H., Svensson, A., & Aaqvist, P. Double Skin Facades – Fashion or a Step towards Sustainable Buildings

Hensen, J.L.M. (2002). Integrated Building (and) airfl ow Simulation: an overview. Proceedings form the Ninth International Conference on Computing in Civil and Building Engineering, Taipei, Taiwan. Web address: http://www.bwk.tue.nl/fago/hensen/publications/02_iccc-be_airfl ow.pdf

Hensen, J.L.M., Bartak, M., & Drkal, F. (2002). Modelling and simulation of double-skin facade systems. ASHRAE Transactions, vol. 108:2,

IDA software (v. 3.0). Equa. Web address: http://www.equa.se/

References

341

ISO Standard 7730, (1984). Moderate thermal environments – determination of the PMV and PPD indices and specifi cations of conditions for thermal comfort. International Organization for Standardization. Geneva.

ISO Standard 15099, (2003). Thermal performance of windows, doors and shading devices – Detailed calculations. International Organization for Standardization. Geneva.

Jacobs, P., & Henderson, H. (2002). State-of-the-Art Whole Building, Building Envelope and HVAC Component and System Simulation and Design Tools. Prepared for the Air-Conditioning and Refrgeration Technology Institute, Arlington, VA

Jaroš, M., Charvát, P., Švorèík, P., & Gorný, R. Possibilities of CFD Simu-lation of Solar Heated Spaces. Brno University of Technology, Faculty of Mechanical Engineering. Institute of Power Engineering, Dept. of Thermodynamics and Environmental Engng, Brno, Czech Republic. Web address: http://dt.fme.vutbr.cz/~cfd/solar/paper3.htm

Jaros, M., Gorny, R., Katolicky, J., & Sedlak, J. (2002). Numerical and Experimental Investigation of the Conditions in the Double Solar Energy Façade. Proceedings of Indoor Air 2002.

Jensen, S. Ø., Poulsen, C. S., & Rasmussen, L. (2003). Cooling of Server rooms (Køling i serverrum). Center for Energi i bygninger and Center for Køle-og Varmepumpeteknik, Teknologisk Institut, Denmark, (in Danish).

Judkoff R., & Neymark, J. (2001), International Energy Agency Building Energy Simulation Test (BESTEST) and Diagnostic Method. IEA SHC Task 12 Building Energy Analysis and Design Tools for Solar Appli-cations. NREL Report TP -472-6231, Golden: NREL.

Kalyanova, O., & Heiselberg, P. K. (2005). Comparative Test Case Specifi ca-tion – Test Cases DSF100_2 and DSF400_3. Report for IEA ECBCS Annex 43/SHC Task 34 Validation of Building Energy Simulation Tools. Aalborg University, Department of Civil Engineering

Kalyanova, O., & Heiselberg, P. K. (2006). Empirical Test Case Specifi cation : Test Cases DSF200_3 and DSF200_4. IEA ECBCS Annex43/SHC Task 34 : Validation of Building Energy Simulation Tools. Aalborg University, Department of Civil Engineering

Kleibrink, (2002). The changing world of offi ce work. Detail 2002/9.

Lee, E., Selkowitz, S., Bazjanac, V., Inkarojrit, V., & Kohler, C. (2002). High-Performance Commercial Building Façades. Building Technologies


342

Program, Environmental Energy Technologies Division, Ernest Or-lando Lawrence Berkeley National Laboratory (LBNL), University of California, Berkeley, USA (LBNL – 50502) Web address: http://gaia.lbl.gov/hpbf/documents/LBNL50502.pdf

Li, Shang-Shiou. (2001). A Protocol to Determine the Performance of South Facing Double Glass Façade System. Msc Thesis in Architecture Sub-mitted to the Faculty of the Virginia Polytechnic Institute and State University, USA. Web address: http://scholar.lib.vt.edu/theses/avai-lable/etd-04212001- 152253/unrestricted/

Limb, M. (1992). Air Infi ltration and Ventilation Glossary; Technical Note AIVC. Air Infi ltration and Ventilation Centre. University of Warwick Science Park, Barclays Venture Centre, Coventry, Great Britain. ISBN 0946075581.

Liddament, M. W. (1996). A guide to Energy Effi cient Ventilation. Air Infi ltration and Ventilation Centre (AIVC).

Manz, H., & Frank, Th. (2005). Thermal Simulation of Buildings with Double-Skin Facades. Energy and Buildings, 37 (2005) 1114-1121.

Nathanson, T. (1995). Indoor Air Quality in Offi ce Buildings: A Technical Guide. Report of the Federal-Provincial Advisory Committee on En-vironmental and Occupational Health, Minister of National Health and Welfare, Canada. ISBN 0-662-23846-X

Nilson, A., Uppström, R., & Hjalmarsson, C. (1996). Rendering the use of energy effi cient in offi ce buildings – a gain not only for the environment! Swedish council for building research, T10:1996, Stockholm, Sweden (in Swedish).

Nilsson, P.E., (2003). Achieving the Desired Indoor Climate; Energy Effi ci-ency Aspects of System Design. Printed by Narayana Press, Denmark. ISBN 91-44-03235-8

Oesterle, E., Lieb, R-D., Lutz, M., & Heusler, W. (2001). Double Skin Facades – Integrated Planning. Prestel Verlag: Munich, Germany

Park, C.-S., Augenbroe, G., Sadegh, N., Thitisawat, M., & Messadi, T. (2003). Occupant Responsive Optimal Control of Smart Facade Systems. Proceedings of Building Simulation'03, August, 2003, Netherlands.

Pedersen, S. (2001). Low Energy Building Design. Lecture note, BYG, DTU, U-018, ISSN 1396-4046, Danish Technical University (DTU), Lyngby, Denmark

References

343

Peterson, F., (1987). Heating Systems. Department of Heating and Ventila-tion. Royal Institute of Technology, Stockholm, Sweden

Peterson, F., (1991). Climate Calculations. Department of Heating and Ventilation. Royal Institute of Technology, Stockholm, Sweden

Pilkington Glass Catalogue, (2004)

Poirazis, H. (2004). Double skin facades for offi ce buildings; a literature review. Division of Energy and Building Design, Department of Construction and Architecture, Lund Institute of Technology, Lund University, Sweden. ISBN 91- 85147 – 02 -8

Poirazis, H. (2005a). Introductory report to the Glazed Offi ce Building Project. Division of Energy and Building Design, Department of Architecture and Built Environment, Lund Institute of Technology, Lund University, Sweden. Report EBD-R--05/10

Poirazis, H. (2005b). Single Skin Glazed Offi ce Buildings; Energy Use and Indoor Climate Simulations. Division of Energy and Building Design, Department of Architecture and Built Environment, Lund Institute of Technology, Lund University, Sweden. Report EBD-T--05/4

PowerDomus software. Thermal Systems Laboratory, Curitiba, Brazil.

Poirazis H. (2006). Double Skin Façades; a literature review. Report for the IEA SHC Task 34 ECBCS Annex 43.

Public Technology Inc. and US Green Building Council, (1996). Sustai-nable Building Technical Manual; Green Building Design, Construc-tion and Operations. Sponsored by U.S. Department of Energy, U.S. Environmental Protection Agency

REPAB, (2003). Årskostnader Kontor 2003 (Yearly costs Offi ce buildings). REPAB AB, Mölndahl (in Swedish).

Saelens, D., & Hens, H. (2001). Experimental evaluation of naturally ventilated active envelopes, International Journal of Thermal Envelopes and Building Science, vol. 25, nr. 2, pp. 101-127.

Saelens, D., Carmeliet, J., & Hens, H. (2001). Modelling of Air and Heat Transport in Active Envelopes, Proceedings of ICBEST 2001, Interna-tional Conference on Building Envelope Systems and Technologies, Ottawa, Canada, pp. 243-247.


344

Saelens, D. (2002). Energy Performance Assessments of Single Storey Multiple-Skin Facades. PhD thesis, Laboratory for Building Physics, Department of Civil Engineering, Catholic University of Leuven, Belgium.

Saelens, D., Carmeliet, J., & Hens, H. (2003). Energy performance as-sessment of multiple skin facades. International Journal of HVAC&R Research 9 (2): 167-186. Web address: http://www.bwk.kuleuven.ac.be/bwf/pdf_artikels/I_J_HVACR_DS_2003.pdf

Saelens, D., Blocken, B., Roels, S., Hens, H. (2005). Optimization of the energy performance of the multiple skin facades. Proceedings of the Ninth International IBPSA Conference, MOndreal, Canada

SCB, (2001). Energy statistics for non-residential premises. Statistics Sweden, Statistiska meddelande EN 16 SM 0203 (in Swedish).

Sjodin L., (2007). LCC Calculations of Glazed Offi ce Buildings. WSP Ma-nagement Internal Report, Malmö, Sweden (in Swedish)

Stec, W., & van Paassen, A.H.C. Integration of the Double Skin Façade with the buildings. Energy in Built Environment, Energy Technology, TU Delft, Mekelweg 2, 2628 CD, Delft, The Netherlands.

Stec, W., and van Paassen, A.H.C. (2003). Defi ning the Performance of the Double Skin Facade with the Use of the Simulation Model. Proceedings of Building Simulation '03, 1243-1250, Netherlands.

Svensson, A. (2000). Glazed facades – a question of image or? Journal Byggkontakt May 2/2000 (in Swedish).

Svensson, A., & Åqvist, P. (2001). Double skin glazed facades – Image or at step on the road to a sustainable society? Arkus skrift nr 37, Stockholm (in Swedish)

TAS software (v.907). EDSL Ltd. Web address: http://ourworld.compu-serve.com/homepages/edsl

Travesi, J. G., Maxwell, K., Klaassen, M. & Holtz. (2001). Empirical Validation of Iowa Energy Resource Station Building Energy Analysis Simulation Models, A report of Task 22, Subtask A Building Energy Analysis Tools Project A.1 Empirical Validation, June 2001.

TRISCO software. Physibel, Belgium. Web address: http://www.physibel.be/

TRNSYS software (v16.037). University of Wisconsin, USA Web address: http://sel.me.wisc.edu/trnsys

References

345

U.S. Army Corps of Engineers, (1997). Design Guide for Interiors. DG 1110-3-122

van Dijk, D. H.A.L. & Henk Oversloot, H.P. (2003). WIS, the Eu-ropean Tool to Calculate Thermal and Solar Properties of Windows and Window Components. Proceedings of Building Simulation '03, 259-266, Netherlands.Web address: http://www.ibpsa.org/PDFs/BS03%20Papers/BS03_0259_266.pdf

van Paassen, A.H.C., & Stec, W. (2001). Controlled Double Facades and HVAC. 7th world congress Clima 2000/Napoli 2001, September 15-18 (CD-ROM). Indoor environment technology: towards a global approach (Napels), REHVA, Brussels, 2001, p. 1-15

VVS, (2000). Classifi cation of indoor climate systems. Guidelines and spe-cifi cations. R1, Swedish HVAC association (in Swedish).

Wall, M., & Bülow-Hübe, H. (2001). Solar protection in buildings. (Report TABK- 01/3060). Lund: Lund University, Dept of Construction and Architecture.

Wall, M., & Bülow-Hübe, H. (2001). Solar Protection in Buildings. Part 2: 2000 – 2002 (2003). (Report EBD-R—03/1). Lund, Sweden: Energy and Building Design, Dept of Construction and Architecture, Lund Institute of Technology, Lund University.

Wall, M., & Kvist, H. (2003), SunSpace- a user friendly design tool to pre-dict temperatures and energy demands for different types of glazed spaces. Report EBD-R--03/2, the department of Energy and Building Design, Lund Institute of Technology, Lund University.

Wilkins, C., & Honsi, M. H. (2000). Heat Gain from Offi ce Equipment. ASHRAE Journal.

Wouters, P. (1999-2000). Quality in Relation to Indoor Climate and Energy Effi ciency. An analysis of trends, achievements and remaining challenges. PhD thesis in Université Catholique de Louvain (UCL), Faculté des Sciences Appliquées, Département Architecture et Construction (ARCH), Belgium.

Wyon, D. (1986). The effects of indoor climate on productivity and per-formance. Simplifi ed summary of results: ‘Indoor climate, accidents, human effi ciency and comfort’, Revised version of a summary prepared by the author for Scandinavian Indoor Climate Year, published in Swedish in VVS & Energi, 59-65.


346

Wyon, D. (1987). The importance of our indoor climate. Fläkt Review No. 71, Indoor climate systems, Fläkt AB, Stockholm, Sweden, ISSN 0015-3400.

Wyon, P. D. (2000a). Enhancing Productivity While Reducing Energy Use in Buildings. Proceedings of the Conference held at The Ronald Reagan Building and International Trade Center Washington, D.C

Wyon, D. (2000b). Individual control at each workplace: the means and the potential benefi ts In: Clements-Croome, D. (editor). Creating the productive workplace. E&FN Spon. London and New York.

Appendix A

347

Appendix A

Germany


348

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

D

üsse

ldor

f ci

ty g

ate

(Düs

seld

orfe

r St

adtt

or)

BB

RI,

(200

2),

Oes

terle

et a

l.,

(200

1), L

ee e

t al

., (2

002)

, C

ompa

gno,

(2

002)

Cor

ridor

fa

çade

Th

e fa

çade

was

des

igne

d to

na

tura

lly v

entil

ate

the

room

s w

ith o

utsi

de a

ir du

ring

long

pe

riods

of t

he y

ear.

The

first

ye

ars o

f ope

ratio

n sh

ow th

at th

e bu

ildin

g ca

n be

nat

ural

ly

vent

ilate

d fo

r rou

ghly

70-

75%

of

the

year

.

Two

corr

idor

wid

ths a

re

enco

unte

red

in th

e bu

ildin

g (9

0 cm

and

140

cm

). Th

e en

tire

build

ing

is e

nclo

sed

in a

gla

ss

skin

so th

at a

56-

met

er-h

igh

atriu

m sp

ace

is c

reat

ed a

t the

ce

ntre

.

The

oute

r lay

er c

onsi

sts o

f a 1

2 m

m sa

fety

gla

ss a

nd th

e in

ner i

s a

low

–E

glaz

ing

with

a w

oode

n fr

ame.

The

sola

r blin

ds a

re

situ

ated

nea

r the

out

er g

lazi

ng

laye

r.

No

com

plet

e ai

r co

nditi

onin

g of

the

offic

e ro

om w

as

inst

alle

d. T

he o

ffice

ro

oms a

re e

quip

ped

with

chi

lled

ceili

ng.

AR

AG

200

0 T

ower

O

este

rle e

t al.,

(2

001)

, C

ompa

gno,

(2

002)

Shaf

t-box

sy

stem

Fr

ee w

indo

w v

entil

atio

n is

po

ssib

le fo

r 50-

60%

of t

he

year

. Dur

ing

win

ter,

the

air-

extra

ct sh

aft i

s als

o de

sign

ed to

be

clo

sed

if re

quire

d

The

faça

de is

roug

hly

70 c

m

deep

. Eac

h of

the

box

win

dow

s ha

s its

ow

n 15

cm

hig

h ai

r-in

take

ope

ning

in th

e fo

rm o

f a

clos

able

flap

The

inne

r faç

ade

laye

r is

cons

truct

ed w

ith c

onve

ntio

nal

verti

cally

piv

otin

g al

umin

um

case

men

ts w

ith lo

w-E

gla

zing

. Lo

uvre

d bl

inds

wer

e in

stal

led

in

the

oute

r thi

rd o

f the

faça

de.

Dur

ing

perio

ds o

f ex

trem

e w

eath

er

cond

ition

s, a

high

le

vel o

f the

rmal

co

mfo

rt ca

n be

at

tain

ed w

ith

mec

hani

cal

vent

ilatio

n.

Hea

dqua

rter

s of

C

omm

erzb

ank

Com

pagn

o,

(200

2)

Mul

ti st

orey

hi

gh

Two

varia

tions

on

the

prin

cipl

e of

the

“buf

fer z

one”

for n

atur

al

vent

ilatio

n of

the

offic

es w

ere

used

: as a

dou

ble

skin

faça

de

and

as a

win

ter g

arde

n.

Thre

e st

orey

seal

ed o

uter

skin

, a

cont

inuo

us c

avity

and

an

inne

r fa

çade

with

ope

rabl

e w

indo

ws.

The

oute

r ski

n co

nsis

ts o

f 8 m

m

toug

hene

d gl

ass.

Air

louv

res

wer

e pr

ovid

ed a

t the

low

er a

nd

uppe

r end

s of t

he c

avity

.

Non

e

Eur

othe

um

Lee

et a

l.,

(200

2)

Box

win

dow

Fr

esh

air i

s sup

plie

d th

roug

h 75

-mm

dia

met

er h

oles

in th

e ve

rtica

l met

al fi

ns o

n ea

ch si

de

of th

e gl

azin

g un

it. W

arm

air

is

extra

cted

thro

ugh

an e

xter

ior

open

ing

at th

e ce

iling

leve

l.

The

faça

de g

rid is

135

0 m

m

wid

e 33

50 m

m ta

ll an

d 34

0mm

de

ep. E

ach

unit,

whi

ch is

pre

-fa

bric

ated

off

site

, con

sist

s of a

6-

grid

span

, one

-sto

rey

tall.

The

inte

rnal

skin

con

sist

s of

ther

mal

ly-b

roke

n al

umin

um

fram

es a

nd d

oubl

e-pa

ne,

man

ually

-ope

rate

d, ti

lt-an

d-tu

rn

win

dow

s. Th

e ex

tern

al sk

in

cons

ists

of s

ingl

e-pa

ne, f

ixed

gl

azin

g. P

ower

-ope

rate

d bl

inds

ar

e ap

plie

d.

Onl

y pa

rt of

the

build

ing

is d

esig

ned

with

a d

oubl

e-sk

in

faça

de, w

hich

pr

ovid

es n

atur

al

vent

ilatio

n fo

r mos

t of

the

year

.

Appendix A

349

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

D

ebis

he

adqu

arte

rs

Lee

et a

l.,

(200

2), C

resp

o,

Oes

terle

et a

l. (2

001)

Cor

ridor

fa

çade

Th

e ex

tern

al sk

in is

ope

nabl

e.

Ope

ning

the

exte

rnal

skin

to a

gr

eate

r deg

ree

has a

pos

itive

in

fluen

ce o

n th

e ve

ntila

tion,

si

nce

it he

lps t

o re

mov

e th

e he

at

in th

e in

term

edia

te sp

ace.

Th

e us

ers c

an o

pen

the

inte

rior

win

dow

s for

nat

ural

ven

tilat

ion.

A

mec

hani

cal v

entil

atio

n pl

an

was

inst

alle

d to

pro

vide

par

tial

air c

ondi

tioni

ng fo

r tho

se

perio

ds in

win

ter a

nd su

mm

er

whe

n ex

trem

e w

eath

er

cond

ition

s pre

vail.

Wal

kway

gril

ls o

ccur

at e

very

flo

or w

ithin

the

70-c

m w

ide

inte

rstit

ial s

pace

and

are

co

vere

d w

ith g

lass

to p

reve

nt

verti

cal s

mok

e sp

read

bet

wee

n flo

ors.

The

inne

r ski

n co

nsis

ts o

f a

doub

le lo

w-E

insu

latin

g gl

azin

g in

alu

min

um fr

ames

. The

ex

terio

r ski

n co

nsis

ts o

f 12m

m

thic

k la

min

ated

gla

ss lo

uvre

s. Sl

idin

g lo

uvre

blin

ds w

ere

inst

alle

d in

fron

t of t

he in

ner

faça

de.

Nig

ht-ti

me

cool

ing

is a

utom

ated

. The

m

ain

obje

ctiv

e of

th

e cl

ient

s and

the

plan

ner w

as to

cr

eate

an

envi

ronm

enta

lly

sust

aina

ble

and

user

-fr

iend

ly b

uild

ing.

To

achi

eve

thes

e go

als,

larg

e sc

ale

inve

stig

atio

ns a

nd

rese

arch

wor

k w

ere

unde

rtake

n.

(GSW

) H

eadq

uart

ers

Lee

et a

l.,

(200

2)

Mul

ti st

orey

hi

gh

Cro

ss v

entil

atio

n; th

e do

uble

fa

çade

ope

rate

s as a

ther

mal

sh

aft.

Dur

ing

the

heat

ing

seas

on, t

he a

ir ca

vity

act

s as a

th

erm

al b

uffe

r whe

n al

l op

erab

le w

indo

ws a

re c

lose

d.

War

m a

ir is

retu

rned

to th

e ce

ntra

l pla

nt v

ia ri

sers

for h

eat

reco

very

.

The

cavi

ty is

0.9

m d

eep.

In

terio

r dou

ble

pane

win

dow

s th

at a

re o

pera

ted

both

man

ually

an

d au

tom

atic

ally

and

a se

aled

10

-mm

ext

erio

r gla

zing

laye

r. W

ide,

ver

tical

, per

fora

ted

alum

inum

louv

res a

re lo

cate

d in

th

e ca

vity

and

are

aut

omat

ical

ly

depl

oyed

and

man

ually

ad

just

able

.

Rad

iant

hea

ting

and

cool

ing

are

prov

ided

to

the

build

ing.

V

ario

us b

uild

ing

syst

ems s

uch

as

light

ing

and

diff

user

s are

in

tegr

ated

.

Hal

ense

estr

aße

Lee

et a

l.,

(200

2)

Cor

ridor

fa

çade

D

urin

g th

e su

mm

er, t

he c

avity

is

mec

hani

cally

ven

tilat

ed. A

t ni

ght,

inte

rnal

hea

t gai

ns a

re

rem

oved

with

mec

hani

cal

vent

ilatio

n. D

urin

g th

e w

inte

r, so

lar g

ains

pre

war

m th

e ai

r in

the

inte

rstit

ial s

pace

.

The

cavi

ty is

0.8

5 m

dee

p.

The

12-m

m si

ngle

-pan

e ex

tern

al sk

in o

f thi

s dou

ble-

skin

fa

çade

is c

ompl

etel

y se

aled

w

hile

the

inte

rnal

skin

con

sist

s of

slid

ing

doub

le p

ane

glas

s do

ors.

A b

lind

was

inst

alle

d w

ithin

the

1-st

orey

hig

h in

ters

titia

l spa

ce.

Onl

y th

e to

p w

est-

faci

ng se

ven

stor

ies

of th

is te

n st

orey

bu

ildin

g ar

e de

sign

ed w

ith a

do

uble

skin

faça

de.

The

doub

le-s

kin

faça

de re

duce

s noi

se

from

the

adja

cent

hi

ghw

ay to

war

ds th

e w

est.


350

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

G

alle

ries

L

afay

ette

C

ompa

gno,

(2

002)

Cor

ridor

fa

çade

Th

e in

let a

nd o

utle

t ven

ts a

re

plac

ed a

t eac

h flo

or. T

he

open

ings

rem

ain

perm

anen

tly

open

. The

faça

de e

nabl

es

natu

ral v

entil

atio

n of

the

offic

es

for m

ost o

f the

yea

r. If

the

outs

ide

tem

pera

ture

is to

o lo

w

or to

o hi

gh, a

mec

hani

cal

vent

ilatio

n sy

stem

is s

witc

hed

on.

The

cavi

ty is

0.2

m d

eep.

Th

e 29

mm

thic

k in

sula

ting

glas

s uni

t with

an

8 m

m g

lass

on

the

outs

ide

and

a 6

mm

low

–E

coa

ted

glas

s on

the

insi

de,

has a

cav

ity fi

lled

with

arg

on.

Perf

orat

ed lo

uvre

s of s

tain

less

st

eel a

re fi

tted

as so

lar c

ontro

l.

The

unus

ually

de

sign

ed D

oubl

e Sk

in F

açad

e is

in

tend

ed to

serv

e as

an

info

rmat

ion

carr

ier a

nd a

ct a

s an

optic

al a

ttrac

tion.

It

also

serv

es a

s pr

otec

tion

agai

nst

the

exte

rnal

noi

se.

Pots

dam

er

Plat

z 1

Oes

terle

et a

l.,

(200

1)

Box

win

dow

Th

e ve

ntila

tion

of th

e ca

vity

an

d th

e in

tern

al ro

oms i

s ef

fect

ed v

ia a

6 c

m h

igh

gap

bene

ath

the

oute

r piv

otin

g ca

sem

ent.

The

cavi

ty is

roug

hly

0.22

m

deep

. Th

e in

tern

al w

indo

w is

a lo

w-E

gl

azin

g in

an

oak

fram

e an

d w

ith a

lum

inum

cov

er st

rips

exte

rnal

ly. A

louv

red

blin

d is

in

stal

led,

the

loca

tion

of w

hich

w

as o

ptim

ized

in re

spec

t of i

ts

rear

ven

tilat

ion

by d

esig

ning

the

uppe

r lou

vres

to b

e fix

ed a

t a

flatte

r ang

le, s

o th

at th

ey re

mai

n pe

rman

ently

ope

n, e

ven

whe

n th

e bl

ind

is lo

wer

ed.

The

com

bina

tion

of

win

dow

ven

tilat

ion

with

add

ition

al

mec

hani

cal s

uppo

rt un

der e

xtre

me

wea

ther

con

ditio

ns

allo

ws a

ver

y hi

gh

degr

ee o

f the

rmal

co

mfo

rt to

be

achi

eved

.

Deu

tsch

er

Rin

g V

erw

altu

n-gs

gebä

ude

Lee

et a

l.,

(200

2)

Cor

ridor

type

Th

e to

p of

the

four

-sto

rey

faça

de h

as a

rain

proo

f ope

ning

w

ith o

verla

ppin

g gl

ass p

anes

th

at a

llow

air

exch

ange

. The

ca

vity

is n

atur

ally

ven

tilat

ed fo

r he

at e

xtra

ctio

n

Som

e of

the

inte

rior w

indo

ws

are

oper

able

to a

llow

for

clea

ning

with

in th

e in

ters

titia

l sp

ace.

Wal

kway

gril

ls o

ccur

at

ever

y flo

or w

ithin

this

in

ters

titia

l spa

ce.

The

exte

rior s

kin

is to

ughe

ned,

so

lar c

ontro

l, si

ngle

pan

e. T

he

inte

rior s

kin

cons

ists

of a

low

-E

coat

ed, d

oubl

e-gl

azed

, win

dow

. B

linds

are

pos

ition

ed in

terio

r to

the

inte

rnal

gla

ss w

indo

ws

Non

e

Prin

t Med

ia

Aca

dem

y B

ohre

n an

d B

oake

, (20

01)

Box

win

dow

C

ross

ven

tilat

ion.

A c

ross

ve

ntila

tion

cont

rol s

yste

m

mod

erat

es th

e bu

ffer

spac

e be

twee

n th

e ou

ter a

nd in

ner

glaz

ing.

The

cavi

ty is

roug

hly

0.46

m

deep

. Si

ngle

gla

ss e

xter

ior p

ane

and

a se

aled

dou

ble

glas

s pan

e on

the

inne

r sid

e. T

he sh

adin

g sy

stem

is

a m

echa

nica

l alu

min

ium

bl

ind

syst

em th

at c

ontro

ls so

lar

heat

gai

n.

The

build

ing'

s ce

ntra

l sys

tem

co

ntro

ls th

e ra

te o

f ai

r flo

w in

to th

e ca

vity

spac

e.

Appendix A

351

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

R

WE

AG

H

eadq

uart

ers

Lee

et a

l.,

(200

2), K

ragh

, (2

000)

, Col

lins,

(200

0), S

pace

m

odul

ator

, A

rons

, (20

00)

Box

win

dow

Th

e ex

tra a

ir ca

vity

act

s as a

th

erm

al b

uffe

r, de

crea

sing

the

rate

of h

eat l

oss b

etw

een

outs

ide

and

insi

de. F

resh

air

is

supp

lied

thro

ugh

the

open

ing

at

the

botto

m a

nd w

arm

air

is

exha

uste

d th

roug

h th

e op

enin

g at

the

top

of th

e fa

çade

. Dur

ing

extre

me

cold

con

ditio

ns, t

he

win

dow

s are

clo

sed.

War

m a

ir is

retu

rned

to th

e ce

ntra

l pla

nt

via

riser

s for

hea

t rec

over

y in

th

e w

inte

r. Th

e ty

pe o

f ve

ntila

tion

is d

iago

nal.

Ope

nabl

e in

terio

r ful

l-hei

ght,

doub

le-p

ane

glas

s doo

rs th

at

can

be o

pene

d 13

.5 c

m w

ide

by

the

occu

pant

s (an

d w

ider

for

mai

nten

ance

). Th

e ca

vity

is

0.5m

dee

p, 3

.6m

hig

h an

d 2m

w

ide.

The

exte

rior l

ayer

is a

10-

mm

ex

tra-w

hite

gla

ss. T

he in

terio

r la

yer c

onsi

sts o

f ful

l-hei

ght,

doub

le-p

ane

glas

s doo

rs. A

n an

ti-gl

are

scre

en is

pos

ition

ed

on th

e in

terio

r. R

etra

ctab

le

vene

tian

blin

ds a

re p

ositi

oned

ju

st o

utsi

de th

e fa

ce o

f the

sl

idin

g gl

ass d

oors

with

in th

e ca

vity

. Day

light

, dire

ct so

lar

and

glar

e ca

n be

con

trolle

d w

ith b

linds

and

an

inte

rior a

nti-

glar

e sc

reen

.

The

desi

gn o

f the

R

WE

faça

de sy

stem

w

as in

fluen

ced

by

the

clie

nts’

des

ire

for o

ptim

um u

se o

f da

ylig

ht, n

atur

al

vent

ilatio

n, a

nd

sola

r pro

tect

ion.

Vic

tori

a L

ife

Insu

ranc

e B

uild

ings

Le

e et

al.,

(2

002)

, C

ompa

gno,

(2

002)

Mul

ti st

orey

Fr

esh

air i

s sup

plie

d at

the

botto

m le

vel a

nd is

ext

ract

ed a

t 21

m h

eigh

t thr

ough

pow

er-

oper

ated

ven

ts.

The

cavi

ty is

0.8

m d

eep.

Th

e ex

tern

al sk

in c

onsi

sts o

f 15

mm

lam

inat

ed so

lar c

ontro

l gl

azin

g; th

e in

tern

al sk

in

cons

ists

of s

olar

con

trol

glaz

ing.

Alu

min

um 5

0mm

dee

p lo

uvre

s are

inte

grat

ed in

to th

e ca

vity

.

The

mai

n ad

vant

age

of th

e do

uble

-ski

n fa

çade

syst

em is

the

impr

ovem

ent i

n th

erm

al c

omfo

rt.

Vic

tori

a E

nsem

ble

Oes

terle

et a

l.,

(200

1)

Mul

ti st

orey

Th

e fa

çade

is u

sed

excl

usiv

ely

as a

mea

ns o

f reg

ulat

ing

ther

mal

insu

latio

n fo

r diff

eren

t w

eath

er c

ondi

tions

. The

inta

ke

of a

ir is

at t

he fo

ot o

f the

bu

ildin

g an

d ex

tract

ed a

t roo

f le

vel.

Ope

nabl

e da

mpe

rs. A

cen

tral

cont

rol s

yste

m k

eeps

the

flaps

cl

osed

whe

n ex

tern

al

tem

pera

ture

s are

low

, ens

urin

g m

axim

um th

erm

al in

sula

tion.

W

hen

exte

rnal

tem

pera

ture

s ris

e, th

e fla

ps a

re o

pene

d to

al

low

the

vent

ilatio

n of

the

cavi

ty p

reve

ntin

g ov

erhe

atin

g.

Con

tinuo

us st

rips o

f fla

ps w

ere

inst

alle

d ar

ound

the

entir

e bu

ildin

g at

the

foot

and

the

top

of th

e fa

çade

to c

ontro

l te

mpe

ratu

res.

Non

e

Gla

dbac

her

Ban

k O

este

rle e

t al.,

(2

001)

Shaf

t box

A

ir-in

take

and

ext

ract

ope

ning

s en

sure

a sa

tisfa

ctor

y su

pply

of

exte

rnal

air

for t

he ro

oms w

hen

the

inne

r faç

ade

is o

pen.

A

sing

le la

yer o

f ref

lect

ing,

su

n-sc

reen

gla

zing

was

inse

rted

in th

e ou

ter s

kin.

Adj

usta

ble

shad

ing

in th

e ca

vity

.

The

Dou

ble

Skin

Fa

çade

was

a p

art

of a

refu

rbis

hmen

t pr

ojec

t.


352

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

D

B C

argo

B

uild

ing

Oes

terle

et a

l.,

(200

1)

Com

bina

tion

of b

ox-

win

dow

and

co

rrid

or fa

çade

ty

pes

Nat

ural

ven

tilat

ion.

The

co

nstru

ctio

n of

a d

oubl

e-sk

in

faça

de m

ade

win

dow

ve

ntila

tion

poss

ible

, the

reby

ov

erco

min

g th

e pr

oble

m o

f a

non-

open

able

faça

de w

ith

inev

itabl

e ai

r-co

nditi

onin

g of

th

e ad

join

ing

room

s. A

par

tial

air-

cond

ition

ing

syst

em w

as

inst

alle

d, p

rovi

ding

a 2

.2 fo

ld

hour

ly a

ir ch

ange

(ach

).

The

cavi

ty is

0.2

3m d

eep.

A

lum

inum

louv

red

blin

ds

Taki

ng in

to a

ccou

nt

the

savi

ngs m

ade

in

the

air-

cond

ition

ing,

th

e si

mpl

e fo

rm o

f co

nstru

ctio

n an

d th

e hi

gh d

egre

e of

pr

efab

ricat

ion

of th

e fa

çade

resu

lted

in

an e

cono

mic

al

solu

tion.

Bus

ines

s Pr

omot

ion

Cen

tre

and

the

Tec

hnol

ogy

Cen

tre

Com

pagn

o,

(200

2)

Box

win

dow

A

ir is

inje

cted

at s

light

ly h

ighe

r th

an a

mbi

ent p

ress

ure

into

the

low

er p

art o

f the

cav

ity a

nd th

e w

arm

ing

effe

ct re

sults

in a

na

tura

l sta

ck e

ffec

t.

The

faça

de c

onsi

sts o

f 1.5

0 ×

3.30

m to

ughe

ned

12 m

m th

ick

pane

s sus

pend

ed in

ver

tical

al

umin

um m

ullio

ns. T

he in

ner

faça

de sk

in c

onsi

sts o

f sto

rey

high

Out

side

is a

6 m

m fl

oat g

lass

, in

side

is a

n 8

mm

lam

inat

ed

glas

s with

low

-E a

nd th

e ca

vity

be

twee

n is

fille

d w

ith a

rgon

ga

s. Pe

rfor

ated

, com

pute

r-co

ntro

lled

alum

inum

louv

res

are

inco

rpor

ated

into

the

cavi

ty

Sinc

e th

e bu

ildin

g w

ent i

nto

oper

atio

n,

over

heat

ing

prob

lem

s hav

e be

en

repo

rted

in th

e to

p flo

ors.

Appendix A

353

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

Sa

nom

atal

o U

uttu

, (20

01)

Mul

ti st

orey

Th

e ca

vity

is c

lose

d an

d ca

n be

ven

ted

by m

otor

-op

erat

ed v

ents

at t

he to

p an

d bo

ttom

, whi

ch a

re

cont

rolle

d by

ther

mos

tats

.

The

dept

h of

the

cavi

ty is

0.

7m T

he in

ner e

nvel

ope

cons

ists

of t

hree

gla

ss

laye

rs, w

hile

the

oute

r co

nsis

ts o

f tw

o.

inne

r gla

ss: t

ough

ened

and

lam

inat

ed 6

m

m,

mid

dle

glas

s: to

ughe

ned

4 m

m a

nd

oute

r gla

ss: t

ough

ened

and

sola

r con

trol 6

m

m. T

he sp

ace:

arg

on a

nd k

rypt

on g

as.

The

oute

r env

elop

e: to

ughe

ned

and

lam

inat

ed 6

+6 m

m p

anes

. B

linds

exi

st in

side

the

inne

r env

elop

e.

A m

aint

enan

ce

gond

ola

fixed

ont

o th

e gi

rder

s of t

he

roof

ena

bles

out

side

m

aint

enan

ce.

SysO

pen

Tow

er

Uut

tu, (

2001

)

Box

win

dow

N

o in

form

atio

n gi

ven

The

dept

h of

the

cavi

ty is

0.

55m

Th

e in

ner e

nvel

ope

cons

ists

of 2

k=2k

4-18

, 26

mm

thic

k gl

ass a

nd th

e ou

ter

enve

lope

con

sist

s of

1k=

1k8

tem

pere

d, 8

mm

thic

k gl

ass.

Aut

omat

ic so

lar b

linds

are

pla

ced

insi

de

the

cavi

ty.

Non

e

Mar

tela

U

uttu

, (20

01)

Mul

ti st

orey

Ea

ch fl

oor h

as tw

o se

rvic

e do

ors t

o th

e ca

vity

. V

entil

ator

s are

inst

alle

d at

th

e co

rner

s of t

he c

avity

ar

ea in

ord

er to

mov

e w

arm

ai

r thr

ough

the

corn

ers.

The

dept

h of

the

cavi

ty is

0.

7m

Inne

r hea

t ins

ulat

ing

glas

s, 4

mm

+ 4

mm

la

min

ated

du

e to

the

rail

requ

irem

ents

O

uter

12

mm

tem

pere

d gl

ass

The

doub

le-s

kin

faça

de is

tota

lly

sepa

rate

d fr

om th

e m

ain

fram

e of

the

offic

e bu

ildin

g.

Itäm

eren

tori

U

uttu

, (20

01)

Mul

ti st

orey

Th

e w

indo

ws o

f the

inne

r en

velo

pe a

re fi

xed.

H

owev

er v

entil

atio

n do

ors

open

to th

e in

term

edia

te

spac

e. T

he in

term

edia

te

spac

e ha

s gra

vita

tiona

l ve

ntila

tion.

The

dept

h of

the

cavi

ty is

0.

925m

Th

e ou

ter g

laze

d sk

in c

onsi

sts o

f 6-8

mm

to

ughe

ned

glas

s. Th

e ci

rcul

ar p

art o

f the

bu

ildin

g ha

s lam

inat

ed g

lass

. Mot

oriz

ed

sola

r sha

ding

blin

ds a

re p

lace

d ou

tsid

e th

e in

ner e

nvel

ope’

s win

dow

s.

Non

e

Nok

ia

Ruo

hola

hti

Uut

tu, (

2001

)

Mul

ti st

orey

N

o in

form

atio

n gi

ven

No

info

rmat

ion

give

n In

ner:

doub

le in

sula

ting

glas

s O

uter

: 6 m

m te

mpe

red

glas

s. Th

e to

p of

th

e ca

vity

has

an

adju

stab

le lo

uvre

, whi

le

the

botto

m is

ope

n.

The

doub

le sk

in

faça

de h

elps

to

rest

rict t

he e

xces

sive

am

ount

of s

olar

hea

t an

d tra

ffic

noi

se.

Finland


354

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

So

nera

U

uttu

, (20

01)

Mul

ti st

orey

Th

e ca

vity

form

ed is

ope

n at

bot

tom

and

top.

N

o in

form

atio

n gi

ven

Inne

r ski

n: g

reen

6 m

m g

lass

(out

er),

sele

ctiv

e 4

mm

Eko

plus

- gla

ss (i

nner

) ar

gon

gas 1

5 m

m (i

n be

twee

n)

Out

er e

nvel

ope:

4+4

mm

lam

inat

ed g

lass

(in

bet

wee

n 0.

76 m

m o

pal s

heet

) el

emen

ts. O

ne o

f the

gla

sses

is c

lear

and

th

e ot

her o

ne is

gre

y w

ith a

silk

scre

en-

prin

ted

patte

rn.

Non

e

Hig

h T

ech

Cen

tre

Uut

tu, (

2001

)

Cor

ridor

faça

de

No

info

rmat

ion

give

n Th

e de

pth

of th

e ca

vity

is

0.34

2m a

nd is

not

ac

cess

ible

.

The

inne

r env

elop

e co

nsis

ts o

f tw

o di

ffere

nt k

inds

of w

indo

ws.

The

low

er

win

dow

s con

sist

of:

- flo

at g

lass

6 m

m

- arg

on g

as 1

8 m

m

- cle

ar se

lect

ive

float

gla

ss 6

mm

an

d th

e up

per w

indo

ws c

onsi

st o

f: - t

empe

red

float

gla

ss 6

mm

- a

rgon

gas

18

mm

- c

lear

sele

ctiv

e flo

at g

lass

4 m

m

The

oute

r gla

ss sk

in c

onsi

sts o

f 10

mm

te

mpe

red

glas

s.

The

inne

r env

elop

e’s

win

dow

s can

be

open

ed to

per

form

cl

eani

ng in

side

the

cavi

ty. O

utsi

de

clea

ning

is

perf

orm

ed fr

om a

ho

ist.

Rad

iolin

ja

Uut

tu, (

2001

) M

ulti

stor

ey

The

air i

n th

e ca

vity

can

be

use

d fo

r hea

ting

and

cool

ing

purp

oses

.

The

dept

h of

the

cavi

ty is

0.

65m

. In

ner s

kin:

6 m

m se

lect

ive

glas

s (in

ner)

, 4

mm

floa

t gla

ss (m

iddl

e) a

nd 6

mm

te

mpe

red

glas

s (ou

ter)

Th

e ou

ter e

nvel

ope:

12

mm

thic

k te

mpe

red

glas

s. M

otor

ized

sola

r sha

ding

bl

inds

are

pla

ced

in th

e ca

vity

.

Non

e

Nok

ia K

2 U

uttu

, (20

01)

Mul

ti st

orey

Th

e ca

vity

is o

pen

from

th

e bo

ttom

and

eac

h flo

or

has v

ents

, whi

ch c

an b

e op

ened

.

The

dept

h of

the

cavi

ty is

0.

6m.

Inne

r env

elop

e: d

oubl

e in

sula

ting

glas

s O

uter

env

elop

e: 6

mm

thic

k te

mpe

red

glas

s. So

lar b

linds

are

inst

alle

d in

the

cavi

ty.

Non

e

Appendix A

355

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

Is

o O

men

a m

all

Uut

tu, (

2001

)

Mul

ti st

orey

Th

e ca

vity

is n

ot o

pen

at th

e to

p. T

he c

avity

is c

lose

d at

its

side

s. Th

e bo

ttom

of t

he

cavi

ty is

clo

sed

with

a

lam

inat

ed g

lass

for s

ound

in

sula

tion

purp

oses

.

The

dept

h of

the

cavi

ty is

1m

. Th

e in

ner e

nvel

ope’

s gla

ss p

anes

are

floa

t gl

ass.

The

oute

r env

elop

e ha

s 8 m

m th

ick

tem

pere

d gl

ass.

Two

of th

e fa

çade

s in

clud

e a

doub

le-

skin

faça

de. O

ne o

f th

em is

inst

alle

d to

re

duce

the

traff

ic

nois

e.

Kon

e B

uild

ing

Uut

tu, (

2001

) M

ulti

stor

ey

The

cavi

ty is

ope

n fr

om th

e bo

ttom

and

eac

h flo

or h

as

vent

s, w

hich

can

be

open

ed.

The

dept

h of

the

cavi

ty is

0.

58m

. In

ner e

nvel

ope:

insu

latin

g gl

ass

Out

er e

nvel

ope:

8 m

m te

mpe

red

clea

r gl

ass p

anes

with

a si

lk sc

reen

pat

tern

.

Non

e

Nok

ia

Kei

lala

hti

Uut

tu, (

2001

)

Mul

ti st

orey

Th

e ca

vity

is o

pen

from

the

botto

m a

nd e

ach

floor

has

ve

nts,

whi

ch c

an b

e op

ened

.

The

dept

h of

the

cavi

ty is

0.

69m

. Th

e in

ner e

nvel

ope:

2k6

-12

sele

ctiv

e gl

ass,

argo

n ga

s in

betw

een

The

oute

r env

elop

e: 6

mm

tem

pere

d cl

ear

glas

s. So

lar b

linds

are

pla

ced

outs

ide

the

inne

r env

elop

e to

rest

rict t

he e

xces

sive

am

ount

of s

olar

hea

t. A

t the

upp

er e

nd o

f th

e in

term

edia

te sp

ace

mot

oriz

ed lo

uvre

s ar

e pl

aced

.

Non

e

Kor

ona

Uut

tu, (

2001

) M

ulti

stor

ey

In w

inte

r fre

sh a

ir is

take

n fr

om th

e so

uthe

rn si

de o

f th

e bu

ildin

g an

d us

ed in

the

HV

AC

syst

em. I

n su

mm

er

the

fres

h ai

r is t

aken

from

th

e no

rther

n si

de.

Part

of th

e ca

vity

is a

bout

2

met

es d

eep

and

anot

her

partl

y w

ider

to fo

rm

win

ter g

arde

ns.

The

win

dow

s in

the

inne

r env

elop

e ha

ve a

se

lect

ive

2k in

sula

ting

glas

s whe

re th

e ou

ter g

lass

is K

Gla

ss a

nd th

e in

ner g

lass

is

cle

ar 4

-6m

m la

min

ated

gla

ss. T

he o

uter

en

velo

pe c

onsi

sts o

f cle

ar, 6

mm

thic

k flo

at g

lass

and

par

tly a

lso

sele

ctiv

e gl

ass.

The

cylin

dric

al fo

rm

of th

e bu

ildin

g ha

s an

ene

rgy

savi

ng

effe

ct; t

he a

rea

of

the

enve

lope

is sm

all

com

pare

d to

the

volu

me

of th

e bu

ildin

g.

JOT

A

utom

atio

n G

roup

U

uttu

, (20

01)

Mul

ti st

orey

The

dept

h of

the

cavi

ty is

1m

. In

ner e

nvel

ope

(trip

le):

6 m

m a

ntis

un,

gree

n gl

ass (

oute

r), 4

mm

cle

ar g

lass

(m

iddl

e), 4

mm

cle

ar g

lass

(inn

er)

Out

er e

nvel

ope:

10

mm

tem

pere

d, g

reen

su

n pr

otec

tive

glas

s pan

es.


356

Bui

ldin

g Fa

çade

type

V

entil

atio

n st

rate

gy

Con

stru

ctio

n Pa

nes a

nd sh

adin

g de

vice

s C

omm

ents

K

ista

Sci

ence

T

ower

, Kis

ta

Cor

ridor

fa

çade

D

iago

nal v

entil

atio

n.

Pref

abric

ated

one

-sto

rey

high

alu

min

um

cons

truct

ion

Non

-ope

nabl

e w

indo

ws.

Out

er sk

in: 8

/10

mm

H, n

on-c

olou

red

Inne

r ski

n: d

oubl

e-pa

ne se

aled

gla

zing

un

its w

ith L

owE

glas

s, no

n-co

lour

ed,

1.35

m o

n ce

ntre

V

enet

ian

blin

ds w

ill b

e in

stal

led

on

the

north

side

Two

of th

e th

ree

faca

des

(tria

ngul

ar fl

oor

cons

truct

ion:

pla

n) a

re

doub

le sk

in fa

cade

s, th

e th

ird (t

o th

e no

rth) i

s a

sing

le sk

in fa

cade

. N

OK

IA

Hou

se, K

ista

M

ulti

Stor

ey

One

-sto

rey

high

ca

vity

div

ided

into

fiv

e sl

its

cons

truct

ion

The

dept

h of

the

cavi

ty is

0.

7m w

ith g

angw

ays o

n ea

ch fl

oor.

Out

er sk

in: 1

0 m

m

Inne

r ski

n: D

oubl

e pa

ne se

aled

gla

zing

un

it w

ith o

uter

pan

e of

soft

coat

ed L

owE

glas

s, 12

mm

arg

on g

as a

nd in

ner p

ane

of

300/

30 c

lear

gla

ss fo

r the

wal

l bel

ow th

e w

indo

w. M

otor

ized

ven

etia

n bl

inds

, co

ntro

lled

by a

pyr

anom

eter

.

Rad

iato

rs a

nd a

ctiv

e co

olin

g be

ams.

Dis

trict

he

atin

g.

Arl

anda

, Pir

F,

Sig

tuna

M

ulti

Stor

ey

Mot

oriz

ed e

xhau

st

open

ing

at r

oof l

evel

. 9.

5 m

long

Non

-ope

nabl

e w

indo

ws.

O

uter

ski

n: 6

mm

floa

t gla

ss. L

ower

big

pa

nes:

12

mm

H D

iam

ant S

ecur

it (i

ron

free

). I

nner

ski

n: 6

mm

Pla

nith

erm

Fut

ur,

20m

m a

rgon

, 6 m

m c

lear

floa

t. Lo

wer

big

pa

nes:

8m

m P

lani

ther

m F

utur

, 16

mm

ar

gon,

8.7

6 m

m C

ontr

aspl

it. V

enet

ian

blin

ds

Non

e

AB

B B

usin

ess

Cen

ter,

So

llent

una

Mul

ti St

orey

A

ir en

ters

at t

he

botto

m th

roug

h th

e gr

atin

g an

d le

aves

at

the

top

thro

ugh

mot

oriz

ed

cont

rolle

d da

mpe

rs

The

dept

h of

the

cavi

ty is

0.

8m w

ith g

angw

ays o

n ea

ch fl

oor.

Non

-ope

nabl

e w

indo

ws.

Alu

min

um fr

ame

cons

truct

ion

with

8 m

m

H si

ngle

pan

es in

the

oute

r ski

n an

d Lo

wE

glas

s in

the

inne

r ski

n (d

oubl

e pa

ne

seal

ed g

lazi

ng u

nit)

with

Arg

on fi

lling

. V

enet

ian

blin

ds a

nd w

ith g

ratin

g ga

ngw

ay o

n ea

ch fl

oor.

Inne

r cur

tain

s ha

ve b

een

adde

d fo

r day

light

con

trol.

Cal

cula

tions

with

dou

ble

skin

faça

de re

sulte

d in

lo

wer

coo

ling

dem

and

than

si

ngle

skin

faca

de. S

ound

pr

oofin

g ag

ains

t the

m

otor

way

E4.

Gla

shus

Ett

M

ulti

stor

ey

At t

he to

p an

d at

the

botto

m th

ere

are

auto

mat

ical

ly

cont

rolle

d da

mpe

rs

for c

ontro

lling

the

airf

low

in th

e ca

vity

.

On-

site

ere

cted

stee

l co

nstru

ctio

n. N

on-

open

able

win

dow

s.

Out

er sk

in: 2

×8 m

m P

lani

bel T

op N

. Se

aled

gla

zing

uni

ts w

ith a

rgon

. Sam

e pa

rts a

re la

min

ated

and

har

dene

d. U

-va

lue=

1,1,

and

for t

he fa

çade

< 1

,3.

Inne

r ski

n: 8

mm

sing

le-p

ane

hard

ened

gl

ass.

Day

light

redi

rect

ion

with

au

tom

atic

ally

con

trolle

d ve

netia

n bl

inds

man

ually

co

ntro

lled

for e

ach

faça

de

and

floor

.

Sweden

Appendix B

357

Appendix BPerformance specifi cations for the reference building

The project team within the project “Glazed Offi ce Buildings” developed performance specifi cations for the reference building, a typical offi ce building of the nineties. The performance specifi cations were approved by a reference group.

Energy useFor the entire building:

Energy use, kWh/m2a Reference

District heating 80Electricity for pumps, fans etc. 20Electricity for lighting, PCs etc. 50Electricity for cooling 30Total use of electricity 100 Total use of energy 180

m2 refers to non-residential/premises fl oor area (LOA see Swedish standard SS 021053 Area and volume of buildings)

Energy use, kW/m3s Reference

Ventilation 2.5

Indoor environmentThe performance specifi cations below are valid for the whole building


358

Air tightness

Description Reference

Air tightness <1.8 litre/(m²s) at 50 Pa difference pressure

Performance specifi cations for roomsThe performance specifi cations below are valid within the occupied zone for every offi ce room. Work places are in a cell-type or open-plan or combination offi ce.

S summer conditions where clothing of 0,6 clo is assumedW winter conditions where clothing of 1 clo is assumed

Description Reference Comments

Sound: from HVAC LAeq< 35 dB, LAmax < 35 dB, LCeq < 50 dB; The performance

from outside LAeq< 30 dB; specifi cations are

airborne sound insulation to other rooms >44 dB applicable during (>40 dB for walls with door), to corridors > 30 dB offi ce hours.

reverberation time for open-plan < 0.4 s for cell-type offi ce room < 0.5 s

Light: should be possible to shade against direct sun light. daylight factor min 1% and max 5%

Lighting: Cell-type: 300 -500 lux, < 12 W/m2 electricity Corridors: > 100 lux, < 6 W/m2 electricity Open-plan: 300 - 600 lux, < 12 W/m2

colour rendering index (Ra-index) > 80

luminance distribution working material : nearest surrounding : surrounding areas 10 : 3 : 1

Luminance within the normal visual fi eld < 1000 cd/m2

Outside the normal visual fi eld < 2000 cd/m2

Internal gains: from offi ce equipment: < 20 W/m2

from PC + monitors: < 125 W i.e. < 5 W/m2

from offi ce copier: < 400 W i.e. < 1 W/m2

from laser printer: 50 W i.e. < 2 W/m2

from persons: 4 W/m2 (100 W/person)

Ventilation: > 0,35 l/sm2 + > 7 l/sperson, air change effi ciency > 40%

IAQ: < 1000 ppm CO2

Thermal comfort: air temperature minimumS 22°C and maximumS 24-26°C

air velocityW < 0.15 m/s

radiant temperature asymmetryW from vertical surfaces < 10 K

vertical air temperature differenceW between 1,1 and 0.1 m < 3 K

Appendix C

359

Appendix CArchitectural drawings

Figure C.1 Drawings of the cell type offi ce plan.


360

Figure C.2 Drawings of the open type offi ce plan.

Appendix C

361

Figure C.3 Cross section of the reference building.


362

Figure C.4 Gable elevation of the reference building.

Appendix C

363

Figure C.5 Front elevation of the reference building.


364

Appendix D

365

Appendix DGeometry of the cell and open plan thermal zones

Cell Type: Geometry of the offi ces - thermal zones

Corner offi ce roomsThermal zone type A1(1), A9(1), B1(4), B4(4), B6(4), B9(4), C1(1), C4(1), C6(1), C9(1). The offi ce area is 15.1 m2 (including half inter-nal wall a and half b). The numbers in the parenthesis show how many identical thermal zones are in the building. Typical corner offi ce room is shown in Figure D.1. The total area of the all the corner offi ce rooms is 331.8 m2.

Figure D.1 Typical corner offi ce.

3.6 m

4.19 m

a

b


366

Typical double offi ce roomsThermal zone type A2(8), A6(5), B2(28), B8(28), C2(7), C8(7). The of-fi ce area is 15.3 m2 (including whole internal wall b and the half a). The numbers in the parenthesis show how many identical thermal zones are in the building. Typical double offi ce room is shown in Figure D.2. The total area of the typical double offi ce rooms is 1272.8 m2.

Figure D.2 Typical double offi ce room.

Typical single offi ce roomsThermal zone type A3(4), A7(12), B3(56), B7(56), C3(14), C7(14). The offi ce area is 10.05 m2 (including whole internal wall b and half a). The numbers in the parenthesis show how many identical thermal zones are in the building. Typical single offi ce room is shown in Figure D.3. The total area of the typical single offi ce rooms is 1568 m2.

a

b 4.19 m

3.66 m

Appendix D

367

Figure D.3 Typical single offi ce room.

Typical meeting roomsThermal zone type A10(1), B5(4), B10(4), C5(1), C10(1). The offi ce area is 12.96 m2 (including whole internal wall b and half a). The numbers in the parenthesis show how many identical thermal zones are in the build-ing. Typical meeting room is shown in Figure D.4. The total area of the typical meeting rooms is 142.56 m2.

Figure D.4 Typical meeting room.

a

b

a

b

4.19 m

2.4 m

3.6 m

3.6 m


368

Meeting room (12 persons)Thermal zone type A4(3). The offi ce area is 25.14 m2 (including half internal wall b and whole a). The numbers in the parenthesis show how many identical thermal zones are in the building. A 12 persons meeting room is shown in Figure D.5. The total area of the meeting rooms (12p) is 75.42 m2.

Figure D.5 Meeting room for 12 persons.

Meeting room (8 persons)Thermal zone type A8(1). The offi ce area is 20.2 m2 (including half in-ternal wall b and whole a). The numbers in the parenthesis show how many identical thermal zones are in the building. An 8 persons meeting room is shown in Figure D.6. The total area of the meeting room (8p) is 20.2 m2.

a

b

6.0 m

4.19 m

Appendix D

369

Figure D.6 8 persons meeting room.

Storage room (1st fl oor)Thermal zone type A5(1). The room area is 56.2 m2 (including half b). The numbers in the parenthesis show how many identical thermal zones are in the building. . The storage room is shown in Figure D.7. The total area of the storage room is 56.2 m2.

a

b

4.19 m

4.82 m


370

Figure D.7 Storage room.

b 15.6 m

3.6 m

Appendix D

371

Table D.1 Properties of the cell type offi ces.

C

orne

r of

fice

room

D

oubl

e of

fice

room

Si

ngle

off

ice

room

Ty

pica

l m

eetin

g ro

om

Mee

ting

room

(1

2 pe

rson

s)

Mee

ting

room

(8

per

sons

) St

orag

e ro

om

Wat

er ra

diat

or (P

max

=

1000

W/u

nit)

1 un

it 1

unit

1 un

it 1

unit

2 un

its

2 un

its

4 un

its

HVAC

Coo

ling

Bea

ms (

num

ber o

f un

its, d

esig

n ai

r flo

w)

1 un

it d.

a.f.=

15

l/s

1 un

it

d.a.

f.= 1

5 l/s

1

unit

d.a

.f.=

10 l/

s

2 un

its

d.a.

f.= 2

1 l/s

(e

ach)

3 un

its

d.a.

f.= 2

8 l/s

(e

ach)

2 un

its

d.a.

f.= 2

8 l/s

(e

ach)

-

Exte

rnal

wal

l with

win

dow

s Lo

ng e

xter

nal

faca

de

Long

ex

tern

al

faca

de

Long

ext

erna

l fa

cade

Sh

ort e

xter

nal

faca

de

Long

ext

erna

l fa

cade

Lo

ng e

xter

nal

faca

de

-

Construct.

Exte

rnal

wal

l with

out

win

dow

s Sh

ort e

xter

nal

faca

de

Shor

t ext

. fa

cade

-

- -

- -

Num

ber o

f occ

upan

ts

1 1

1 6

12

8 -

Occ

upan

t’s sc

hedu

le

Sche

dule

for

offic

es

Sche

dule

for

offic

es

Sche

dule

for

offic

es

Sche

dule

for

mee

ting

room

s Sc

hedu

le fo

r m

eetin

g ro

oms

Sche

dule

for

mee

ting

room

s -

Clo

thin

g (s

: dur

ing

sum

mer

, w: d

urin

g w

inte

r)

1 cl

o (w

), 0.

6 cl

o (s

) 1

clo

(w),

0.6

clo

(s)

1 cl

o (w

), 0.

6 cl

o (s

) 1

clo

(w),

0.6

clo

(s)

1 cl

o (w

), 0.

6 cl

o (s

) 1

clo

(w),

0.6

clo

(s)

-

Occupants

Act

ivity

leve

l (s

ittin

g, re

adin

g)

1 m

et

1 m

et

1 m

et

1 m

et

1 m

et

1 m

et

-

Num

ber o

f uni

ts

PC: 1

25W

, PR

: prin

ter :

30W

, Fa

x : 3

0W

1 PC

, 1 p

rinte

r 1

fax

2 PC

s

1 PC

-

- -

-

Sche

dule

Eq

uipm

ent’s

sc

hedu

le

Equi

pmen

t’s

sche

dule

Eq

uipm

ent’s

sc

hedu

le

- -

- -

Equipment

Ave

rage

em

itted

hea

t per

uni

t 61

.67W

12

5W

125W

-

- -

- N

umbe

r of u

nits

1

1 1

1 1

1 1

Rat

ed in

put p

er u

nit (

12W

/m2 )

175

W

175

W

113

W

148

W

291

W

233

W

233

W

Lum

inou

s effi

cacy

* 41

.67

lm/W

41

.67

lm/W

41

.67

lm/W

41

.67

lm/W

41

.67

lm/W

41

.67

lm/W

41

.67

lm/W

Lights

Con

vect

ive

frac

tion

0.3

0.3

0.3

0.3

0.3

0.3

0.3

Uni

ts d

esks

(d),

chai

rs (c

), bo

oksh

elve

s (b)

1

(d),

(c),

(b)

2 (d

), (c

), (b

) 1

(d),

(c),

(b)

1 (d

), 6

(c)

1 (d

), 12

(c)

1 (d

), 8

(c)

shel

ves

Con

stru

ctio

n D

efau

lt fu

rnitu

re

Def

ault

furn

iture

D

efau

lt fu

rnitu

re

Def

ault

furn

iture

D

efau

lt fu

rnitu

re

Def

ault

furn

iture

D

efau

lt fu

rnitu

re

Furniture

Are

a 10

m2

10 m

2 7

m2

15 m

2 25

m2

20 m

2 50

m2


372

Corridor for the ground fl oorSince the corridors and the common spaces are (almost) internal thermal zones, they are not so interesting for the energy and thermal comfort simulations. Thus, they were considered as one thermal zone as shown below.

Thermal zone type A11(1). The corridor area (including the reception and 1 meeting room is) 470 m2. The numbers in the parenthesis show how many identical thermal zones are in the building. The total area of the corridor of the ground fl oor is 470 m2 (Figure D.8).

Figure D.8. Corridor of the ground fl oor.

Corridors for the 1st-5th fl oorThermal zone type B11(4), C11. The corridor area (including the recep-tion and 1 meeting room is) 444.8 m2. The numbers in the parenthesis show how many identical thermal zones are in the building. The total area of the corridors is 2224 m2 (Figure D.9).

Figure D.9 Corridor of the 1st-5th fl oor.

Appendix D

373

Table D.2. Properties of the cell type corridors.

Open Plan Type: Geometry of the offi ces - thermal zones

Typical corner zonesThermal zone type A1(1), B1(4), B4(4), C1(1), C4(1). The zone area is 258.7 m2 (including half internal wall a and half b). The numbers in the parenthesis show how many identical thermal zones are in the building. Typical corner zone is shown in Figure D.10. The total area of the corner offi ce rooms is 2846 m2.

Corridor (ground floor) Corridor (1st-5th floor) Water radiator (Pmax = 1000W/unit) 2 units 4 units

HVAC Cooling Beams (number of units, design air flow) - -

External wall with windows Short external facade Short external facade Construct. External wall without windows Long external facade Long external facade Number of occupants - 0 Occupant’s schedule - - Clothing (s: during summer, w: during winter)

- - Occupants

Activity level (sitting, reading) - -

Number of units Copy machines: 500W PR: printer : 50W, Fax : 30W

4 Printers, 4 Copy machines, 2 Faxes

4 Printers, 4 Copy machines, 2 Faxes

Schedule Equipment’s schedule Equipment’s schedule Equipment

Average emitted heat per unit 226W 226W Number of units 1 1 Rated input per unit (6W/m2) 2796 W 2460 W Luminous efficacy* 41.67 lm/W 41.67 lm/W Lights

Convective fraction 0.3 0.3 Units desks (d), chairs (c), bookshelves (b) Chairs, etc Chairs, etc

Construction Default furniture Default furniture Furniture Area 100 m2 80 m2


374

Figure D.10 Typical corner zone.

Intermediate zonesThermal zone type A8(1), B(8), C8(1). The zone area is 430 m2 (includ-ing half internal wall a and half b). The numbers in the parenthesis show how many identical thermal zones are in the building. Typical corner zone is shown in Figure D.11. The total area of the intermediate zones is 2577 m2.

Figure D.11 Typical intermediate zone.

a

b

Appendix D

375

Reduced corner zoneThermal zone type A4 (1). The zone area is 203 m2 (including half inter-nal wall a and half b). The numbers in the parenthesis show how many identical thermal zones are in the building. Typical corner zone is shown in Figure D.12. The total area of the reduced corner zone is 203 m2.

Figure D.12 Reduced corner zone.

Meeting rooms and Storage roomThese two zones are completely identical with the meeting room for 8 persons (cell type) and the storage room (cell type) correspondently.


376

Table D.3 Properties of the open plan type zones.

Typical corner zone

Intermediate zones

Reduced corner zone

Meeting room (8 persons)

Storage room

Water radiator (Pmax = 5000W/unit) 2 units 2 units 2 units 4 units 2 units

HV

AC

Cooling Beams (number of units, design air flow)

1 unit d.a.f.= 131 l/s

1 unit d.a.f.= 202 l/s

2 units d.a.f.= 28 l/s (each)

- 3 units d.a.f.= 28 l/s (each)

External walls (a) Long external facade

Long external facade

Long external facade -

Long external facade

Con

stru

ct.

External walls (b) Short external facade - - - -

Number of occupants 16 24 8 - 12

Occupant’s schedule Schedule for offices

Schedule for offices

Schedule for meeting rooms

- Schedule for meeting rooms

Clothing (s: during summer, w: during winter)

1 clo (w), 0.6 clo (s)

1 clo (w), 0.6 clo (s)

1 clo (w), 0.6 clo (s) - 1 clo (w),

0.6 clo (s) Occ

upan

ts

Activity level (sitting, reading) 1 met 1 met 1 met - 1 met

Number of units PC: 125W, PR: printer : 30 or 50W, fax : 30W, copy machines: 500W

16 PCs 4 PR (30W), 4 faxes

24 PCs 4 PR (50W), 4 c.m.

- - -

Schedule Equipment’s schedule

Equipment’s schedule - - - Eq

uipm

ent

Average emitted heat per unit 93.4 W 159.3W - - -

Number of units 1 1 1 1 1 Rated input per unit (12W/m2) 3104 W 5148 W 233 W 233 W 291 W

Luminous efficacy* 41.67 lm/W 41.67 lm/W 41.67 lm/W 41.67 lm/W 41.67 lm/W Ligh

ts

Convective fraction 0.3 0.3 0.3 0.3 0.3 Units desks (d), chairs (c), bookshelves (b), Common furniture (cf)

16 (d), (c), (b), (cf)

24 (d), (c), (b), (cf) 1 (d), 8 (c) shelves 1 (d), 12 (c)

Construction Default furniture

Default furniture

Default furniture

Default furniture

Default furniture

Furn

iture

Area 160 m2 260 m2 20 m2 50 m2 25 m2

Appendix E

377

Appendix E Corrected theoretical U-values (including thermal bridges)

The thermal bridges caused by the steel columns and the wooden studs were calculated using the two-dimensional software Heat 2 (version 6).

Due to the fact that when defi ning a thermal zone in IDA ICE 3.0 one has to insert the loss factor for thermal bridges (W/°C) for each thermal zone type (depending on the geometry of the zone) we decided to calculate the U-value of the external wall including the thermal bridges (using Heat 2) and then design an equivalent new wall in IDA ICE 3.0 assuming that the loss factor = 0. The 4 steps followed are:

1. We defi ned wall structure without the steel columns and the ma-terial properties in IDA ICE 3.0 (heat conductivity, density and specifi c heat). IDA ICE 3.0 calculated the U-value of the external wall (=0.2201Wm-2K-1) as show in Figure E.1.

Figure E.1 U-value (without the columns) calculated by IDA ICE 3.0.


378

2. We built the same model in Heat 2 and calculated the U-value. Since the minimum material properties and thickness (minimum thickness 10mm) were not exactly the same, we tuned the model in order to get the same U-value with the one calculated by IDA ICE 3.0 (step 1) as shown in Figure E.2.

Figure E.2 U-value (without columns) calculated by Heat 2.

The construction was:

• 120mm facing bricks• 40mm air gap• 10mmgypsum• 150mm insulation• 10mm gypsum

The properties of the materials were the same with the IDA ICE 3.0 model

The boundary conditions assumed were:

• Inside temperature: 1°C• Outside temperature: 0°C (temperature difference 1°C in order to

get the Wm-2K-1)• Heat fl ow from both sides q=0 Wm-1, i.e. sideways in the wall

The U value calculated was 0.2027 Wm-2K-1. The calculations were steady state.

Appendix E

379

3. We added the columns in the Heat 2 model as shown in Figure E.3. The distance between the centres of the steel columns is 2.4m, between the centres of studs 0.6m and between centres of the fl oors 3.5m.

Figure E.3 External wall construction (with columns).

Using the same assumptions as in step 2, we simulated the new model (Figure E.4). The new U-value calculated was 0.27 Wm-2K-1.

Figure E.4 Simulation with the steel columns (Heat 2).

4. We assumed an equivalent external wall (with the same U-value) in IDA ICE 3.0 by reducing the insulation from 0.137m to 0.1125m as shown in Figure E.5.


380

Figure E.5 Final model of the external wall in IDA ICE 3.0 (reduced thermal insulation).

Appendix F

381

Appendix FAHU- Ventilation rates

AHU for Cell typeFor the cell type air the air exchange with the outdoor environment takes place in 4 different ways:

o Supply air to the offi ces and meeting rooms during the working hours (mechanical ventilation)

o Infi ltration to all the zones always (natural ventilation)o Exhaust air from the corridors during the working hours (mechani-

cal ventilation)o Exfi ltration from all the zones always (natural ventilation)

In all the offi ces and the total ingoing air is due to mechanically supplied air and the infi ltration and the total outgoing air is due to exfi ltration. The supplied air to the offi ces is extracted though leakages to the corridor and exhausted from there through the AHU. The infi ltration and exfi ltration are assumed to be 0.1ach for all the zones. In the meeting rooms all the air supplied is also exhausted. The reason for that is that if the supplied air (VAV CO2 control) was extracted to the corridors the total exhaust air would be unknown. This means that any mistake in the estimation of the VAV supplied air would destroy the balance of the total supply – exhaust air infl uencing the AHU effi ciency. However, the VAV CO2 control applied in the meeting rooms caused 2 problems for the simulations:

o Infi ltration (increase the mechanical ventilation) could not be added since the rooms were supplied with the necessary air so we had to add this infi ltration to the offi ces.

o The supplied air was not known so it was not possible to know how much the effi ciency of the AHU should be decreased. In order to solve the problem the airfl ow was estimated


382

The Table E.1 below shows the ventilation rates of each zone as inserted in IDA ICE 3.0.

o The Supplied air per zone (mech. ventilation) shows the air that is provided by the AHU as mechanical ventilation. In the corridors there is not any supplied air and in the meeting rooms is shown the estimation we made.

o The supplied air per zone (mech. ventilation) is increased in the of-fi ces (natural ventilation in the offi ces, corridor and meeting rooms). In the meeting rooms the airfl ow is the same since infi ltration was added to the offi ces.

o The Total Exhaust air (l/s) per fl oor shows in each zone the exhaust air. There is not any exhaust air in the meeting rooms since they do not contribute in the total exhaust air from the corridors. However in reality it should be considered in order to fi nd the proper amount of total airfl ows which will give the correct decrease of the AHU effi ciency.

Appendix F

383

Table F.1 Ventilation rates for cell type.

Room type Zone Supplied air Supplied air Total Exhaust Supply air/ Input for per zone per zone air (l/s) Exhaust air IDA (mech. vent.) (total) (l/s) per zone (l/s)

Corner offi ce rooms B1 15 17.01 0 1000 0.017008 B4 15 17.01 0 1000 0.017008 B6 15 17.01 0 1000 0.017008 B9 15 17.01 0 1000 0.017008 C1 15 17.01 0 1000 0.017008 C4 15 17.01 0 1000 0.017008 C6 15 17.01 0 1000 0.017008 C9 15 17.01 0 1000 0.017008 A1 15 17.01 0 1000 0.017008 A9 15 17.01 0 1000 0.017008

Double offi ce rooms A2 20 22.04 0 1000 0.022042 A6 20 22.04 0 1000 0.022042 B2 20 22.04 0 1000 0.022042 B8 20 22.04 0 1000 0.022042 C2 20 22.04 0 1000 0.022042 C8 20 22.04 0 1000 0.022042

Single offi ce rooms B3 10 11.34 0 1000 0.011339 B7 10 11.34 0 1000 0.011339 C3 10 11.34 0 1000 0.011339 C7 10 11.34 0 1000 0.011339 A3 10 11.34 0 1000 0.011339 A7 10 11.34 0 1000 0.011339

Meeting rooms (6p) B5 21 21 0 1000 0.022726 B10 21 21 0 1000 0.022726 C5 21 21 0 1000 0.022726 C10 21 21 0 1000 0.022726 A10 21 21 0 1000 0.022726

Meeting rooms (8p) A8 28 28 0 1000 0.03069

Meeting rooms (12p) A4 42 42 0 1000 0.045349

Storage room (1st fl oor) A5 62 70.58 0 1000 0.070576

Corridor (1st fl oor) A11 0 0 577.10 0.001 577.10

Corridor (2-6 fl oors) B11 0 0 699.67 0.001 699.67 C11 0 0 699.67 0.001 699.67

Natural Ventilation (not including the corridors and the meeting rooms) 242.5Natural Ventilation ( including the corridors and the meeting rooms) 462.9Weighted Natural Ventilation) 462.9Total Weighted Natural Ventilation 4075.4Total Mechanical Ventilation 3612Recovered heat (60%) 2398.2Equivalent percentage for recovered (%) 53.773AHU on Value 1AHU off Value 0.1038Weekends off value (=50% mechanical ventilation +100% natural ventilation) 0.551


384

AHU for Open plan typeFor the open plan type the calculations were much simpler since both supply and exhaust air were provided in all the thermal zones. However, the meeting rooms were treated exactly in the same way as the cell type offi ce building. In Table F.2 are shown the airfl ows of the building.

Table F.2 Ventilation rates for open plan type.

Room type Zone Supplied air Natural Ventilation Supply air/ Input for per zone ventilation per zone Exhaust air IDA (mech. vent.) per zone (l/s) (l/s) (l/sec)

Typical corner zones B1 112 19.4 131.4 1 133.0366 B4 112 19.4 131.4 1 133.0366 C1 112 19.4 131.4 1 133.0366 C4 112 19.4 131.4 1 133.0366 A1 112 19.4 131.4 1 133.0366

Reduced corner zone A4 84 15.2 99.2 1 100.4822

Intermediate zones A8 168 32.22 200.22 1 202.938 B8 168 32.22 200.22 1 202.938 C8 168 32.22 200.22 1 202.938

Meeting Rooms A2 0 1.5 29.5 1 0 A5 0 1.5 29.5 1 0 B2 0 1.5 29.5 1 0 B5 0 1.5 29.5 1 0 C2 0 1.5 29.5 1 0 C5 0 1.5 29.5 1 0

Storage room (1st fl oor) A9 62 4.83 66.83 1 67.23745

Natural Ventilation (not including the meeting rooms) 426.75Natural Ventilation ( including the meeting rooms) 462.75Weighted Natural Ventilation (the natural ventilation of the meeting rooms is added as nat. vent. In the offi ces) 462.75Total Weighted Natural Ventilation (the natural ventilation of the meeting rooms is added as nat. vent. In the offi ces) 3520.75Total Mechanical Ventilation 3074Recovered heat (60%) 1844.4Equivalent percentage for recovered (%) 52.387AHU on Value 1AHU off Value 0.1314Weekends off value (=50% mechanical ventilation +100% natural ventilation) 0.5679

Appendix G

385

Appendix GFrame Construction for the single skin glazed alternatives

The “improved” frame constructions were suggested by Schüco Interna-tional. For the fi rst to sixth 60% glazed alternatives the frames used are the Royal FW 50+ Hi and FW 60+ Hi as shown in Figure G.1 and G.2.

Gla

ss th

ickn

ess (

mm

)

Fram

e de

pth

(mm

)

Glass thickness (mm)


Frame depth (mm) In

sula

tion

thic

knes

s(m

m)

Figure G.1 System FW 50+ Hi.


386

Figure G.2 System FW 60+ Hi.

The Uframe depends on the frame depth. As shown in Figure F.3

Figure G.3 Relationship between Uframe and frame depth.


Gla

ss t

hick

ness

(m

m)


Appendix H

387

Appendix HPane properties for the glazing used for the double skin façade alternatives

Table H.1 Pane properties for the glazing used for the double skin façade alternatives.

Glazing Thickness Corrected Refl ectance Direct solarUnit (mm) emissivity transmittance indoor outdoor indoor outdoor

Clear pane 4 0.837 0.837 0.078 (s) 0.078 (s) 0.820 (s) 0.082 (v) 0.082 (v) 0.893 (v)Clear pane 8 0.837 0.837 0.074 (s) 0.074 (s) 0.742 (s) 0.081 (v) 0.081 (v) 0.870 (v)Low E coated 4 0.837 0.092 0.207 (s) 0.179 (s) 0.666 (s) 0.059 (v) 0.064 (v) 0.865 (v)Optigreen 6 0.837 0.837 0.053 (s) 0.053 (s) 0.460 (s)(s. c. tinted) 0.067 (v) 0.067 (v) 0.751 (v)(OpGrn6 pgl) Optigreen 8 0.837 0.837 0.053 (s) 0.053 (s) 0.389 (s)(s. c. tinted) 0.067 (v) 0.067 (v) 0.703 (v)(OpGrn8 pgl) Solar control 6 0.837 0.042 0.293 (s) 0.412 (s) 0.201 (s)+ low E 0.146 (v) 0.134 (v) 0.440 (v)(soft coated)(galaxy6gvb) Solar control 8 0.837 0.042 0.268 (s) 0.412 (s) 0.198 (s)+ low E 0.143 (v) 0.133 (v) 0.435 (v)(soft coated) (galaxy8gvb) Solar control 8 0.298 0.837 0.102 (s) 0.088 (s) 0.520 (s)+ low E 0.101 (v) 0.085 (v) 0.669(v)(hard coated)(sunnycl8 gvb) Low E 8 0.170 0.837 0.108 (s) 0.090 (s) 0.677 (s)(hard coated) 0.109 (v) 0.098(v) 0.822 (v)(Kglass6pgl)

(s) = solar(v) = visual


388

Appendix I

389

Appendix IOptical and thermal properties of the double skin façade alternatives

Table I.1 Glazing properties for the “standard” double façade mode (closed cavity).

Case DSF A DSF B DSF C DSF D DSF E DSF F DSF G

U-value 1.93 1.15 1.85 1.85 1.15 1.04 1.14g-value 0.627 0.551 0.516 0.404 0.354 0.301 0.443(total solar energy transmittance)solar direct 0.53 0.447 0.326 0.297 0.264 0.151 0.335transmittancesolar direct 0.153 0.188 0.123 0.0848 0.0867 0.327 0.15refl ectance outdoorsolar direct 0.169 0.248 0.132 0.159 0.241 0.422 0.25refl ectance indoorlight transmittance 0.708 0.683 0.594 0.571 0.551 0.416 0.526refl ectance outdoor 0.194 0.18 0.168 0.141 0.132 0.395 0.16refl ectance indoor 0.201 0.174 0.173 0.191 0.166 0.311 0.177

Table I.2 Properties of the inner and outer skin of the “standard” double façade mode (closed cavity).

DSF Case U value g value Tsol Outer skin Inner skin Outer skin Inner skin Outer skin Inner skin

A 5.79 2.9 0.789 0.746 0.742 0.685 D 5.79 2.74 0.532 0.746 0.389 0.685 E 5.79 1.46 0.532 0.655 0.389 0.565 F 5.79 1.31 0.789 0.215 0.742 0.179


390

Table I.3 Glazing properties for the Airfl ow window mode (closed cav-ity).

Case AW A AW B AW C AW D AW E AW F AW G

U-value 1.93 1.15 1.83 1.85 1.15 1.04 0.824g-value 0.627 0.561 0.529 0.404 0.354 0.195 0.191(total solar energy transmittance)solar direct 0.53 0.447 0.326 0.297 0.264 0.157 0.149transmittancesolar direct 0.153 0.201 0.123 0.0848 0.0861 0.279 0.278refl ectance outdoorsolar direct 0.169 0.224 0.132 0.159 0.216 0.311 0.322refl ectance indoorlight transmittance 0.708 0.683 0.594 0.571 0.551 0.357 0.344refl ectance outdoor 0.194 0.173 0.168 0.141 0.128 0.172 0.169refl ectance indoor 0.201 0.182 0.173 0.191 0.174 0.238 0.208

Table I.4 Properties of the inner and outer skin of the airfl ow window mode (closed cavity).

AW Case U value g value Tsol Outer skin Inner skin Outer skin Inner skin Outer skin Inner skin

A 2.87 5.92 0.688 0.846 0.624 0.82 D 2.71 5.92 0.419 0.846 0.338 0.82 E 1.45 5.92 0.369 0.846 0.299 0.82 F 1.30 5.92 0.213 0.846 0.169 0.82 G 1.31 5.66 0.214 0.738 0.176 0.677

Appendix J

391

Appendix JDescription of the opening areas for the box window and multi-storey double skin façade

Table J.1 Description of the opening areas for the box window and multi-storey high cavities.

1: ELA: Equivalent Leakage Area

Zone

type

Cav

ity

Dep

th, m

Cav

ity

Wid

th, m

Dam

per

free

ar

ea, m

²

Dis

char

ge

coef

f. to

p (C

dtop

)

Dis

char

ge

coef

f. bo

ttom

(C

dbot

tom

)

Dam

per

Cop

en

Bot

tom

le

ak, m

²

Upp

er

leak

, m²

Leak

in

betw

een

fl oor

s, m

²

Leak

in

betw

een

rest

fl oo

rs²

ELA

1 M

ulti

stor

ey

Corner offi ces 0.80 4.19 0.87 0.65 0.55 2.94 1.84 2.18 2.35 3.35 1.60(short façade) Corner offi ces 0.80 3.60 0.87 0.65 0.55 2.52 1.58 1.87 2.02 2.88 1.38(long façade) Double offi ce 0.80 4.19 0.87 0.65 0.55 2.94 1.84 2.18 2.35 3.35 1.60rooms (long façade) Double offi ce 0.80 3.60 0.87 0.65 0.55 2.52 1.58 1.87 2.02 2.88 1.38rooms (short façade) Single offi ce 0.80 2.40 0.87 0.65 0.55 1.68 1.06 1.25 1.34 1.92 0.92rooms Corridor 0.80 1.60 0.87 0.65 0.55 1.12 0.70 0.83 0.90 1.28 0.61


392

Appendix K

393

Appendix KShading properties for the double skin façade alternatives

Table K.1 Properties of the venetian blinds.

White venetian Dark venetian blinds blinds

Diffuse refl ectance (shortwave radiation, front and back) 67% 19%Absorbtance (shortwave rad., front and back) 33% 81%Diffuse refl ectance (longwave rad., front and back) 10% 10%Emitance (longwave rad., front and back) 90% 90%Width of slat 28 mm 28 mmSlat spacing 22 mm 22 mmSlat angle 45° 45°


394

Table K.2 Multipliers for the shading devices for the “standard” double façade alternatives.

Case Orientation White venetian blinds Blue venetian blinds U value g value Tsol U value g value Tsol

north 0.6144 0.2554 0.7524 0.153 northeast 0.6102 0.2332 0.762 0.1294 east 0.6012 0.2058 0.7726 0.1016DSF A,F southeast 0.923 0.5936 0.1962 0.923 0.7732 0.0968 south 0.5938 0.1996 0.773 0.1032 southwest 0.5964 0.1952 0.777 0.0954 west 0.604 0.2032 0.7768 0.0982 northwest 0.61 0.2294 0.7642 0.1248

north 0.7218 0.2512 0.803 0.1524 northeast 0.7198 0.236 0.809 0.1282 east 0.7144 0.2004 0.8176 0.0998DSF D,E southeast 0.923 0.7148 0.192 0.923 0.821 0.0962 south 0.7192 0.1962 0.8236 0.1032 southwest 0.7208 0.1912 0.8264 0.0948 west 0.7212 0.198 0.8242 0.0966 northwest 0.7208 0.2252 0.8122 0.1236

Appendix K

395

Table K.3 Multipliers for the shading devices for the airfl ow window alter-natives.

Case Orientation White venetian blinds Blue venetian blinds U value g value Tsol U value g value Tsol

north 0.6852 0.2692 0.8134 0.1566 northeast 0.6836 0.2468 0.8242 0.1322 east 0.6804 0.2196 0.8388 0.1038AW A southeast 0.955 0.6762 0.2106 0.955 0.8406 0.0998 south 0.6764 0.2146 0.84 0.1068 southwest 0.6792 0.21 0.8444 0.0986 west 0.6844 0.217 0.8434 0.1008 northwest 0.6856 0.2432 0.8284 0.1276

north 0.7454 0.2668 0.8424 0.1562 northeast 0.7442 0.2442 0.8514 0.1314 east 0.7426 0.216 0.8662 0.099AW D southeast 0.952 0.7414 0.2082 0.952 0.8662 0.099 south 0.7442 0.213 0.8638 0.1018 southwest 0.7452 0.2072 0.8698 0.0982 west 0.747 0.2134 0.8668 0.107 northwest 0.7466 0.2406 0.8554 0.1266

north 0.7874 0.2832 0.859 0.159 northeast 0.7876 0.2612 0.8678 0.1344 east 0.788 0.2334 0.8798 0.1048AW E southeast 0.951 0.7882 0.2266 0.951 0.8824 0.102 south 0.7916 0.232 0.883 0.1106 southwest 0.7922 0.2258 0.8858 0.1012 west 0.792 0.231 0.884 0.1018 northwest 0.79 0.2576 0.8714 0.1292

north 0.7926 0.2806 0.8618 0.159 northeast 0.7926 0.2584 0.8694 0.1342 east 0.793 0.2314 0.8808 0.1058AW F,G southeast 0.954 0.793 0.2234 0.954 0.8834 0.102 south 0.7966 0.2276 0.885 0.1092 southwest 0.7978 0.2224 0.8876 0.1008 west 0.7972 0.229 0.885 0.1026 northwest 0.795 0.2548 0.8728 0.1296


396

Appendix L

397

Ener

gy u

se fo

r he

atin

g (k

Wh/

am2 )

Ener

gy u

se fo

r co

olin

g (k

Wh/

am2 )

Use

of e

lect

ricity

for

light

ing

(kW

h/am

2 )

Use

of e

lect

ricity

for

equi

pmen

t (kW

h/am

2 )

Use

of e

lect

ricity

for

pum

ps, f

ans (

kWh/

am2 )

Use

of e

lect

ricity

for

serv

er ro

oms (

kWh/

am2 )

Ener

gy u

se fo

r coo

ling

serv

er ro

oms (

kWh/

am2 )

Tota

l (kW

h/am

2 )

Appendix LEnergy use

Table L.1 Energy use for the reference building alternatives.

SS-30%-Cell-NS-strict 56 21 14.7 22 8 10 5 137SS-30%-Cell-NS-normal 52 11 14.4 22 8 10 5 123SS-30%-Cell-NS-poor 47 7 14.2 22 8 10 5 114SS-30%-Cell-NS45-strict 56 20 14.7 22 8 10 5 136SS-30%-Cell-NS45-normal 52 11 14.4 22 8 10 5 123SS-30%-Cell-NS45-poor 47 7 14.2 22 8 10 5 114SS-30%-Cell-EW-strict 56 19 14.7 22 8 10 5 136SS-30%-Cell-EW-normal 52 10 14.4 22 8 10 5 122SS-30%-Cell-EW-poor 47 7 14.2 22 8 10 5 113SS-30%-Open-NS-strict 50 29 19 21 6 13 6 144SS-30%-Open-NS-normal 45 18 19 21 6 13 6 127SS-30%-Open-NS-poor 38 11 19 21 6 13 6 114SS-30%-Open-NS45-strict 50 29 19 21 6 13 6 144SS-30%-Open-NS45-normal 45 17 19 21 6 13 6 127SS-30%-Open-NS45-poor 38 10 19 21 6 13 6 113SS-30%-Open-EW-strict 51 28 19 21 6 13 6 143SS-30%-Open-EW-normal 45 16 19 21 6 13 6 126SS-30%-Open-EW-poor 38 9 19 21 6 13 6 112


398

Ener

gy u

se fo

r he

atin

g (k

Wh/

am2 )

Ener

gy u

se fo

r co

olin

g (k

Wh/

am2 )

Use

of e

lect

ricity

for

light

ing

(kW

h/am

2 )

Use

of e

lect

ricity

for

equi

pmen

t (kW

h/am

2 )

Use

of e

lect

ricity

for

pum

ps, f

ans (

kWh/

am2 )

Use

of e

lect

ricity

for

serv

er ro

oms (

kWh/

am2 )

Ener

gy u

se fo

r coo

ling

serv

er ro

oms (

kWh/

am2 )

Tota

l (kW

h/am

2 )

Table L.2 Energy usefor the 60% glazed alternatives (cell plan type).

SS-60%-Cell-NS45-strict (1) 76 31 14.7 22 8 10 5 167SS-60%-Cell-NS45-strict (2) 54 36 14.7 22 8 10 5 150SS-60%-Cell-NS45-strict (3) 58 28 14.7 22 8 10 5 146SS-60%-Cell-NS45-strict (4) 59 22 14.7 22 8 10 5 142SS-60%-Cell-NS45-strict (5) 53 48 14.7 22 8 10 5 161SS-60%-Cell-NS45-strict (6) 58 27 14.7 22 8 10 5 145SS-60%-Cell-NS45-strict (7) 62 14 14.7 22 8 10 5 137SS-60%-Cell-NS45-normal (1) 72 20 13.4 22 8 10 5 151SS-60%-Cell-NS45-normal (2) 50 24 13.8 22 8 10 5 133SS-60%-Cell-NS45-normal (3) 54 18 14.2 22 8 10 5 131SS-60%-Cell-NS45-normal (4) 55 13 14.4 22 8 10 5 129SS-60%-Cell-NS45-normal (5) 48 36 13.8 22 8 10 5 143SS-60%-Cell-NS45-normal (6) 54 16 14.2 22 8 10 5 130SS-60%-Cell-NS45-normal (7) 59 7 14.4 22 8 10 5 126SS-60%-Cell-NS45-poor (1) 66 14 13.2 22 8 10 5 138SS-60%-Cell-NS45-poor (2) 44 17 13.7 22 8 10 5 120SS-60%-Cell-NS45-poor (3) 48 12 14.1 22 8 10 5 120SS-60%-Cell-NS45-poor (4) 50 9 14.3 22 8 10 5 118SS-60%-Cell-NS45-poor (5) 43 28 13.6 22 8 10 5 130SS-60%-Cell-NS45-poor (6) 48 12 14.2 22 8 10 5 119SS-60%-Cell-NS45-poor (7) 53 5 14.4 22 8 10 5 118

Appendix L

399

Ener

gy u

se fo

r he

atin

g (k

Wh/

am2 )

Ener

gy u

se fo

r co

olin

g (k

Wh/

am2 )

Use

of e

lect

ricity

for

light

ing

(kW

h/am

2 )

Use

of e

lect

ricity

for

equi

pmen

t (kW

h/am

2 )

Use

of e

lect

ricity

for

pum

ps, f

ans (

kWh/

am2 )

Use

of e

lect

ricity

for

serv

er ro

oms (

kWh/

am2 )

Ener

gy u

se fo

r coo

ling

serv

er ro

oms (

kWh/

am2 )

Tota

l (kW

h/am

2 )

Table L.3 Energy use for the 60% glazed alternatives (open plan type).

SS-60%-Open-NS45-strict (1) 70 38 19 21 6 13 6 172SS-60%-Open-NS45-strict (2) 49 44 19 21 6 13 6 157SS-60%-Open-NS45-strict (3) 52 34 19 21 6 13 6 151SS-60%-Open-NS45-strict (4) 53 29 19 21 6 13 6 147SS-60%-Open-NS45-strict (5) 47 54 19 21 6 13 6 166SS-60%-Open-NS45-strict (6) 52 33 19 21 6 13 6 150SS-60%-Open-NS45-strict (7) 56 21 19 21 6 13 6 142SS-60%-Open-NS45-normal (1) 65 26 19 21 6 13 6 156SS-60%-Open-NS45-normal (2) 43 31 19 21 6 13 6 139SS-60%-Open-NS45-normal (3) 46 22 19 21 6 13 6 133SS-60%-Open-NS45-normal (4) 48 17 19 21 6 13 6 130SS-60%-Open-NS45-normal (5) 42 41 19 21 6 13 6 148SS-60%-Open-NS45-normal (6) 47 21 19 21 6 13 6 132SS-60%-Open-NS45-normal (7) 50 10 19 21 6 13 6 125SS-60%-Open-NS45-poor (1) 57 18 19 21 6 13 6 139SS-60%-Open-NS45-poor (2) 36 22 19 21 6 13 6 123SS-60%-Open-NS45-poor (3) 39 17 19 21 6 13 6 120SS-60%-Open-NS45-poor (4) 41 10 19 21 6 13 6 116SS-60%-Open-NS45-poor (5) 35 31 19 21 6 13 6 131SS-60%-Open-NS45-poor (6) 39 15 19 21 6 13 6 119SS-60%-Open-NS45-poor (7) 43 6 19 21 6 13 6 113


400

Ener

gy u

se fo

r he

atin

g (k

Wh/

am2 )

Ener

gy u

se fo

r co

olin

g (k

Wh/

am2 )

Use

of e

lect

ricity

for

light

ing

(kW

h/am

2 )

Use

of e

lect

ricity

for

equi

pmen

t (kW

h/am

2 )

Use

of e

lect

ricity

for

pum

ps, f

ans (

kWh/

am2 )

Use

of e

lect

ricity

for

serv

er ro

oms (

kWh/

am2 )

Ener

gy u

se fo

r coo

ling

serv

er ro

oms (

kWh/

am2 )

Tota

l (kW

h/am

2 )

Table L.4 Energy use for the 100% glazed alternatives (cell plan type).

SS-100%-Cell-NS45-strict (1) 95 43 14.7 22 8 10 5 198SS-100%-Cell-NS45-strict (2) 63 50 14.7 22 8 10 5 173SS-100%-Cell-NS45-strict (3) 69 37 14.7 22 8 10 5 166SS-100%-Cell-NS-strict (3) 69 38 14.7 22 8 10 5 168SS-100%-Cell-EW-strict (3) 69 34 14.7 22 8 10 5 164SS-100%-Cell-NS45-strict (4) 71 29 14.7 22 8 10 5 160SS-100%-Cell-NS45-strict (5) 75 16 14.7 22 8 10 5 151SS-100%-Cell-NS45-strict (6) 92 30 12.9 22 8 10 5 180SS-100%-Cell-NS45-strict (7) 59 37 13.5 22 8 10 5 155SS-100%-Cell-NS45-normal (1) 65 27 14.0 22 8 10 5 151SS-100%-Cell-NS45-normal (2) 68 19 14.2 22 8 10 5 147SS-100%-Cell-NS45-normal (3) 58 54 13.4 22 8 10 5 170SS-100%-Cell-NS-normal (3) 66 24 13.9 22 8 10 5 149SS-100%-Cell-EW-normal (3) 72 9 14.3 22 8 10 5 141SS-100%-Cell-NS45-normal (4) 84 22 12.7 22 30 10 5 186SS-100%-Cell-NS45-normal (5) 53 28 13.3 22 8 10 5 140SS-100%-Cell-NS45-normal (6) 59 20 13.8 22 8 10 5 138SS-100%-Cell-NS45-normal (7) 61 13 14.1 22 8 10 5 134SS-100%-Cell-NS45-poor (1) 52 44 13.2 22 8 10 5 154SS-100%-Cell-NS45-poor (2) 59 17 13.7 22 8 10 5 136SS-100%-Cell-NS45-poor (3) 66 6 14.2 22 8 10 5 131SS-100%-Cell-NS45-poor (4) 95 43 14.7 22 8 10 5 198SS-100%-Cell-NS45-poor (5) 63 50 14.7 22 8 10 5 173SS-100%-Cell-NS45-poor (6) 69 37 14.7 22 8 10 5 166SS-100%-Cell-NS45-poor (7) 71 29 14.7 22 8 10 5 160

Appendix L

401

Ener

gy u

se fo

r he

atin

g (k

Wh/

am2 )

Ener

gy u

se fo

r co

olin

g (k

Wh/

am2 )

Use

of e

lect

ricity

for

light

ing

(kW

h/am

2 )

Use

of e

lect

ricity

for

equi

pmen

t (kW

h/am

2 )

Use

of e

lect

ricity

for

pum

ps, f

ans (

kWh/

am2 )

Use

of e

lect

ricity

for

serv

er ro

oms (

kWh/

am2 )

Ener

gy u

se fo

r coo

ling

serv

er ro

oms (

kWh/

am2 )

Tota

l (kW

h/am

2 )

Table L.5 Energy use for the 100% glazed alternatives (open plan type).

SS-100%-Open-NS45-strict (1) 86 48 19 21 6 13 6 199SS-100%-Open-NS45-strict (2) 58 56 19 21 6 13 6 179SS-100%-Open-NS45-strict (3) 63 41 19 21 6 13 6 169SS-100%-Open-NS45-strict (4) 64 33 19 21 6 13 6 162SS-100%-Open-NS45-strict (5) 56 71 19 21 6 13 6 193SS-100%-Open-NS45-strict (6) 63 39 19 21 6 13 6 167SS-100%-Open-NS45-strict (7) 68 21 19 21 6 13 6 154SS-100%-Open-NS45-normal (1) 82 36 19 21 6 13 6 183SS-100%-Open-NS45-normal (2) 53 44 19 21 6 13 6 161SS-100%-Open-NS45-normal (3) 58 30 19 21 6 13 6 152SS-100%-Open-NS45-normal (4) 59 22 19 21 6 13 6 146SS-100%-Open-NS45-normal (5) 51 58 19 21 6 13 6 174SS-100%-Open-NS45-normal (6) 58 27 19 21 6 13 6 150SS-100%-Open-NS45-normal (7) 63 11 19 21 6 13 6 140SS-100%-Open-NS45-poor (1) 75 27 19 21 6 13 6 167SS-100%-Open-NS45-poor (2) 45 34 19 21 6 13 6 144SS-100%-Open-NS45-poor (3) 51 21 19 21 6 13 6 136SS-100%-Open-NS45-poor (4) 52 15 19 21 6 13 6 132SS-100%-Open-NS45-poor (5) 44 47 19 21 6 13 6 157SS-100%-Open-NS45-poor (6) 51 19 19 21 6 13 6 135SS-100%-Open-NS45-poor (7) 56 6 19 21 6 13 6 127


402

Ener

gy u

se fo

r he

atin

g (k

Wh/

am2 )

Ener

gy u

se fo

r co

olin

g (k

Wh/

am2 )

Use

of e

lect

ricity

for

light

ing

(kW

h/am

2 )

Use

of e

lect

ricity

for

equi

pmen

t (kW

h/am

2 )

Use

of e

lect

ricity

for

pum

ps, f

ans (

kWh/

am2 )

Use

of e

lect

ricity

for

serv

er ro

oms (

kWh/

am2 )

Ener

gy u

se fo

r coo

ling

serv

er ro

oms (

kWh/

am2 )

Tota

l (kW

h/am

2 )

Table L.6 Energy use for the 1st and 5th (cell type) 100% glazed alternatives (increased heating and cooling capacity).

SS-100%-Cell-NS45-strict 99 43 14.7 22 8 10 5 202(1-cor)SS-100%-Cell-NS45-normal 95 31 12.9 22 8 10 5 184(1-cor)SS-100%-Cell-NS45-poor 87 22 12.7 22 30 10 5 189(1-cor)SS-100%-Open-NS45-strict 90 48 19.2 21 6 13 6 203(1-cor)SS-100%-Open-NS45-normal 85 36 19.2 21 6 13 6 186(1-cor)SS-100%-Open-NS45-poor 77 27 19.2 21 6 13 6 168(1-cor)SS-100%-Cell-NS45-strict 62 68 14.7 22 8 10 5 190(5-cor)SS-100%-Cell-NS45-normal 58 55 13.4 22 8 10 5 171(5-cor)SS-100%-Cell-NS45-poor 52 45 13.2 22 8 10 5 155(5-cor)

Appendix M

403

Appendix MNumber of working hours with PPD values lower than 10% and 15% for the 90% of the working hours

Table M.1 Percentage of working of hours with PPD below 10% and 15% for the cell type alternatives.

Cell type plan Open type plan 30% glazed 60% glazed 100% glazed 30% glazed 60% glazed 100% glazed PPD PPD PPD PPD PPD PPD PPD PPD PPD PPD PPD PPD 10% 15% 10% 15% 10% 15% 10% 15% 10% 15% 10% 15%

strict(1) 66 93 61 83 46 65 63 97 69 90 50 70 normal(1) 73 93 57 82 31 60 84 97 65 90 36 64 poor (1) 39 63 27 45 15 30 44 64 27 42 19 34 strict(2) 72 94 68 87 81 97 75 92 normal(2) 70 94 55 84 73 96 60 87 poor (2) 31 52 27 42 32 49 34 49 strict(3) 68 93 62 84 78 96 70 91 normal(3) 70 93 57 82 79 97 67 90 poor(3) 34 57 27 45 32 51 34 52 strict(4) 66 92 60 83 77 96 69 90 normal(4) 71 93 59 83 82 97 71 91 poor (4) 38 61 30 49 34 56 35 56 strict(5) 71 93 64 84 82 97 72 92 normal(5) 65 90 50 78 68 93 56 80 poor (5) 30 49 25 40 31 46 32 47 strict(6) 68 93 62 84 78 96 70 91 normal(6) 70 93 58 83 80 97 69 91 poor(6) 33 57 27 45 32 51 34 53 strict(7) 68 93 57 82 78 96 66 90 normal(7) 71 92 60 83 80 97 77 91 poor (7) 33 57 40 58 78 96 46 71


404

Figure M.2 Percentage of working of hours with PPD below 10% and 15% for the cell type alternatives (strict set points, zone level).

1st alt. PPD 10% 37 57 59 56 39 53 53 64 PPD 15% 74 80 86 88 90 93 96 1002nd alt. PPD 10% 49 71 72 71 48 66 66 81 PPD 15% 90 95 97 100 100 85 62 193rd alt. PPD 10% 42 65 68 64 44 59 59 74 PPD 15% 87 93 97 100 100 87 68 224th alt. PPD 10% 46 64 65 61 41 57 57 73 PPD 15% 87 93 97 100 100 87 69 215th alt. PPD 10% 34 62 67 70 46 66 65 66 PPD 15% 82 87 91 88 87 75 60 296th alt. PPD 10% 46 66 67 64 44 59 59 75 PPD 15% 88 93 97 100 100 87 67 217th alt. PPD 10% 44 62 62 57 33 54 54 70 PPD 15% 86 93 97 100 100 88 70 23

Figure M.3 Percentage of working of hours with PPD below 10% and 15% for the cell type alternatives (normal set points, zone level).

1st alt. PPD 10% 24 47 48 40 24 44 43 39 PPD 15% 70 77 83 88 90 85 78 562nd alt. PPD 10% 32 59 60 51 30 55 54 51 PPD 15% 85 92 95 98 99 89 71 293rd alt. PPD 10% 30 59 59 57 34 59 59 49 PPD 15% 82 89 94 97 97 87 73 324th alt. PPD 10% 33 62 61 58 36 59 61 53 PPD 15% 85 92 96 100 100 89 72 255th alt. PPD 10% 25 53 53 50 27 52 51 43 PPD 15% 75 80 85 82 80 71 57 316th alt. PPD 10% 31 61 60 57 35 59 60 51 PPD 15% 85 92 96 100 100 89 72 267th alt. PPD 10% 42 68 65 59 34 56 59 67 PPD 15% 87 93 97 100 100 87 69 22

Cor

ner o

ffi ce

(s

outh

wes

t)

Dou

ble

offi c

e (s

outh

wes

t)

Sing

le o

ffi ce

(s

outh

wes

t)

Mee

ting

room

(n

orth

wes

t)

Cor

ner o

ffi ce

(n

orth

east

)

Sing

le o

ffi ce

(n

orth

east

)

Dou

ble

offi c

e (n

orth

east

)

Mee

ting

room

(s

outh

east

)

Cor

ner o

ffi ce

(s

outh

wes

t)

Dou

ble

offi c

e (s

outh

wes

t)

Sing

le o

ffi ce

(s

outh

wes

t)

Mee

ting

room

(n

orth

wes

t)

Cor

ner o

ffi ce

(n

orth

east

)

Sing

le o

ffi ce

(n

orth

east

)

Dou

ble

offi c

e (n

orth

east

)

Mee

ting

room

(s

outh

east

)

Appendix M

405


406

ISSN 1671-8136ISBN 978-91-85147-23-6

energy and building design | energy and building design ......1.2 energy efﬁ ciency in the...

Documents