ctbuh_tallest korea building

8/10/2019 CTBUH_tallest Korea Building

http://slidepdf.com/reader/full/ctbuhtallest-korea-building 1/9

CTBUH

Technical Paperhttp://technicalpapers.ctbuh.org

Subject: Building Case Study Structural Engineering

Paper Title: Integration of Design and Construction of the Tallest Building in Korea, Tower PalaceIII, Seoul, Korea

Author(s): Abdelrazaq, A1

Baker, W.2 Chung, K. R.

Affi liation(s): 1Samsung Corporation 2Skidmore, Owing & Merrill LLP Dong Yang Structural Engineering Co., Ltd

Publication Date: 2004

Original Publication: CTBUH Conference, Seoul. october 10-13, 2004.

Paper Type: 1. Book chapter/Part chapter 2. Journal paper 3. Conference proceeding 4. Unpublished conference paper 5. Magazine article 6. Unpublished

© Council on Tall Buildings and Urban Habitat/Author(s)



654 CTBUH 2004 October 10~13, Seoul, Korea

Ahmad Abdelrazaq, Vice President / Executive DirectorSamsung Corporation, Engineering & Construction Group12 flr Samsung Plaza bldg. 263Seohyun, Bundang-Gu,

Sungnam-Gu Gyonggi-Do, Korea 463-721Tel: 82.2.2145.5190; Fax: 82.2.2145.5770, E-mail, [email protected]

Integration of Design and Construction of the Tallest Building inKorea, Tower Palace III, Seoul, Korea

A.K. Abdelrazaq 1, W. F. Baker 2, K.R. Chung 3, J. Pawlikowski 4, Insoo Wang 5, K.S. Yom 6

1 Executive Director, Samsung Corporation, Engineering and Construction Group2 Partner, 4 Associate, Skidmore, Owings & Merrill LLP, Chicago, USA3 Principal, Dong Yang Structural Engineering Co., Ltd, Seoul, Korea

4Vice President, 5 Manager, Samsung Corporation, Construction & Engineering Group

Abstract

Tower Palace III was conceived as a 93-story residential tower soaring 320 meters into Seoul’s skyline. However,concerns of the local residents and authorities over the building height resulted in a 73-story tower with the samegross floor area. The early integration of aerodynamic shaping and wind engineering considerations played a majorrole in the architectural massing and design of the tower. The Contractor’s input in selecting the optimumstructural system resulted in a cost effective tower that served the clients needs and the tower was completed ahead

of schedule. This paper presents a brief overview of the structural system development of the tower and its directintegration with the construction aspects, discusses the construction planning of the key structural components ofthe tower, and briefly describes the monitoring program incorporated into the tower for the evaluation of timedependent deformation.

Keywords: indirect outrigger belt-wall system, composite column instrumentation, self compacting/consolidating concrete, matfoundation, wind engineering integration with the architectural massing.

1. Introduction Tower Palace III was part of Samsung Life

Insurance Togok site development plan, which islocated in the Kangnam district of Seoul, South Koreaand consisted of three phases. Phases I and II includedTower Palace I and II development that comprised ofsix high-rise residential towers varying from 42 to 66stories. Phase III of this development, Tower PalaceIII (TPIII), consisted of a 73-story single point towerwith an adjacent eight (8) story sport center, and six(6) levels of parking below grade.

Tower Palace III site development is located in amoderate wind climate and subject to moderate/lowseismic forces. Early planning and concept envisionedthis tower to be 320 meters high and an all-residential

building. The critical design criteria for this very tallluxury residential tower called for controlling thedynamic response of the tower and managing its windengineering aspects. During the design process, the

building evolved into three different schemes, whereeach of the schemes was accepted by the client andfully studied at the Boundary Layer Wind TunnelLaboratory (BLWTL), See Figure 2. Force balance

studies indicated that the tower massing, exterior walltreatment, and the dynamic building characteristicsresulted in a structure that was not sensitive todynamic wind excitations, and the predicted buildingacceleration and torsional velocities were below theinternationally acceptable acceleration and torsionalvelocity criteria.

While the three schemes had the same gross floorarea and approximately the same number of apartmentunits, the 93-story tower was not accepted by the

Figure 1: Tower Palace Development



CTBUH 2004 October 10~13, Seoul, Korea 655

neighbors due to its height and the potential for trafficcongestion in the area. The 73-story tower, scheme 3,was finally selected by the client to satisfy theconcerns of the local authorities and the neighbors.

This paper presents an overview of thedevelopment of the tower’s structural system and itsdirect integration with the architectural massing andconstruction planning for the key structuralcomponents of the tower. In addition, this paper

briefly discusses the instrumentation of the tower toevaluate the long term behavior of the compositecolumns and the reinforced concrete core wall. Thestrain measurements at the composite columns and thecore wall correlated well with the predicted strains.

2. Structural System Design Approach

Structural System Design ApproachThe structural design process of the tower was

formulated based on the following goals:

♦ Optimize the tower structural system for strength,stiffness, cost effectiveness, redundancy, and speedof construction.

♦ Manage and locate the gravity load resisting systemso as to maximize its use in resisting the lateralloads while harmonizing with the architectural

planning of a luxury residential tower.♦ Incorporate the latest innovations in analysis,

design, materials, and construction methods.♦ Limit the building drift, acceleration, and torsional

velocity to within the international accepted designcriteria.

♦ Control the relative displacement between thevertical members, especially for composite

buildings.♦ Control the dynamic response of the tower under

wind loading by tuning the structural characteristicsof the building to improve its dynamic behavior andto prevent lock-in vibration due to the vortexshedding. Favorable dynamic behavior of thetower was achieved by:

a) Varying the building shape along the heightwhile continuing, without interruption, the

building gravity and lateral load resistingsystem;

b) reducing the floor plan along the building’supper zone; and

c) creating irregularities along the building’sexterior surfaces, thus reducing the localcladding pressures as well as the overallwind loads on the building structure.

Wind EngineeringWind loads considered in the analysis of the tower

structure were developed using the code defined loadcriteria as well as the results from the wind tunneltesting program, which is based on historical

climatological data. The wind tunnel testing program,conducted at the Boundary Layer Wind TunnelLaboratory (BLWTL) included 1) site proximity windanalysis model, 2) force balance tests, see Figure 3,conducted for all schemes, 3) cladding and pressureintegration test, and 4) pedestrian wind studies.

The climatological study performed for the projectwas determined for a 100 year return period with amean-hourly gradient wind speed of 41 m/s. Thisresulted in a wind pressure of 2.5kN/m 2. Strengthdesign of the tower was based on both the code-

prescribed wind loads with exposure A and the windtunnel developed 100-year return period. Wind tunnelrecommendation included combining the wind loads in

a) 93-Story (320m) b) 77-Story (270m) c) 73-Story (264m) Scheme 1 Scheme 2 Scheme 3

Figure 2: Tower Palace III Massing Studies

93 Story Tower 77 story Tower 73 story Tower

Figure 3: Wind Tunnel Test Models




the orthogonal direction and torsional momentssimultaneously. 1.5% and 2% damping were assumedfor serviceability and strength design respectively.

Seismic ConsiderationsTPIII is located in an area with low seismic

activity and the building is essentially founded on rockfoundation with locally fractured and weak layers ofrocks. The seismic behavior and response of the towerwas evaluated using the modal response spectrumanalysis method. Response spectrum curves andloading conditions required by the Korean BuildingLaw were utilized as a base for the seismic loadingconditions. Since the building is very tall and flexible,the tower was controlled by wind design rather thanseismic design except at the top of the building, wherethe whip lash effects generated forces that are slightlyhigher than the wind forces. The overall building drift

and interstory drift met the Korean and UBC 97 building code requirements.

Foundation System ConsiderationsThe tower superstructure is founded on a 3500

mm high performance reinforced concrete mat overlean concrete slab over prepared rock. The rockquality and mechanical characteristics varied over thesite in general and in particular at the footprint of thetower due to the presence of ancient faults and shearzones. These faults/shear zones started atapproximately 7 meters north of the foundation mat toapproximately 150 meters below the south edge of thefoundation mat. The presence of these fault causedconcerns about the behavior of the foundation system.The geotechnical engineering work, performed byDames and Moore, San Francisco, CA, USA, indicated

that these faults are inactive.

A 3-dimensional finite elements analysis model,using FLAC 3D, was utilized to model the entire rockmass/foundation mat in order to better estimate thefoundation settlement and behavior. See figure 4.

Two FLAC3D foundation analysis models were

performed to evaluate the impact of the faults on thesettlement analysis. Comparison of the analysis results between the two models indicated that the overall building settlement was increased by approximately23% due to the presence of the faults.

Because of the variation of the rock quality, the presence of faults, and the shape of the building/structure, it was prudent to utilize a matfoundation system under the tower footprint in order tominimize the differential settlement, reduce the impactof the differential settlement on the superstructuremember design, and to bridge over the local weak rocklayers and pockets. Based on the 3DFLAC settlementanalysis model, Dames & Moore provided the soilstiffness that was utilized as a basis for soil-structureanalysis model of the tower.

The mat foundation analysis indicated that themaximum anticipated settlement under the towerwould be approximately 15mm. The building surveyindicated that the actual foundation settlement wassmaller than anticipated.

Differential Axial Shortening ConsiderationsWhile optimizing the lateral load resisting system

of the tower, minimizing the differential shortening, between the composite columns and the core wall, wasone of the critical issues considered during thedevelopment of the structural system of the tower.

Figure 4: FLAC3D - Foundation Analysis model(Courtesy of Dames & Moore) Figure 5: Soil Structure Interaction Analysis Model




3. Structural System Description

Floor Framing SystemSeveral Floor framing system were considered for

the tower, including a flat plate system and compositesteel framing. However, because of the clientmarketing requirements, a composite floor framingsystem was selected. The composite floor framingsystem consisted of a 150mm composite metal deckslab spanning 3 to 4 meters between 400mm deep hotrolled composite steel beams, see Figure 5. The400mm beam depth was selected to allow for MEPopenings so that the ceiling sandwich is minimized.The composite structural steel beams were arranged to

reduce the number of embedded plates into the corewall and were fireproofed with cementitious materialto achieve the applicable fire rating.

Lateral Load Resisting SystemThe lateral load resisting system of the tower

provided resistance to wind and seismic forces andconsisted of a high performance, reinforced concretecore wall system, from the foundation to the roof thatwas linked to the exterior composite columns by anindirect outrigger belt wall system at the mechanicallevels (16 to 17, and 55 to 56). The interaction of thecore wall system and the exterior columns was

provided through deformation compatibility, resultingin significant forces in the belt wall system

components, which included the belt wall and the floorslabs.

The core wall system While the exterior flanges of the core wall vary in

thickness from 550 mm at the bottom to 400 mm at thetop, the interior core wall web thicknesses weremaintained at 300mm throughout the building height.In a typical level, the core wall vertical wallcomponents are rigidly connected by a series of750mm deep composite or reinforced concrete link

beams. The link beam widths typically match theadjacent core wall thicknesses.

a) Structural System Diagram b) Floor Framing Plan

Figure 5: Structural System Diagram & Typical Floor Framing Plan

55 Belt Wall

ExteriorColumns

Core Wall

MatFoundation

ExteriorColumns

InteriorColumns

R/C CoreWall S stem

CompositeFloor Framin

16 Belt Wall

Figure 6: Composite Link Beam Details




Due to the link beam depth limitations, acomposite structural steel beam was introduced at thelocations where large shear and bending momentforces existed. See Figure 6. The composite link

beam consisted of a built-up wide flanged structural

steel beam or a single structural steel plate that wereembedded in the concrete section. The structural steelweb section, where required, was designed to resist themajority of the shear forces and the composite sectionwas utilized in resisting the bending moments. Sincethe composite link beam section provides significantshear ductility, the bending moment forces in the link

beams can be limited. Thus the maximum forces thatthe core wall system can attract in a seismic event can

be managed, and the overall building behavior can becontrolled without significant damage even in severeseismic event.

The exterior columnsThe exterior columns of the tower are typically

steel reinforced concrete columns (SRCC) that varyfrom 1350mm Diameter at subgrade levels, to amaximum of 1000x1000 from levels 4 to 56, to900x900 from level 57 to 65, to 800x800 from the 66level to the roof. In order to minimize the column size,concrete filled (CFT) with high strength concrete wasconsidered for the exterior and interior columns at theconceptual design stage. However, SRC columns wereadopted instead, in order to minimize the differentialrelative movement, due to immediate and long termdeformations, between the exterior columns and the

core walls throughout the building life, and during theconstruction period.

The indirect outrigger belt wall systemThe indirect outrigger belt wall system is

essentially similar to an outrigger wall system;however, the mechanism of force resolution betweenthe core wall and the belt wall system is indirect,through the floor slabs rather than the direct wallconnection. The indirect outrigger belt wall systemconsists of an exterior reinforced concrete perimeterwall that rigidly connects the exterior compositecolumns and the very stiff floor slabs.

The indirect outrigger belt walls for TPIII werelocated at the mechanical levels (16 and 55) andconsisted of an 800mm wide by 8 meter high wall

perimeter wall and a 300mm thick floor slabs, at the

top and the bottom of the perimeter walls. While thehigh perimeter belt wall bending and shear stiffnessconnected the exterior composite columns rigidly, thefloor slab high in-plane bending and shear stiffnessforced deformation compatibility and shear forceredistribution between the core wall and the exterior

belt wall frame system.

Since the perimeter belt walls reduced the relativedisplacements between floors, the lateral systemrotations at the belt wall levels were significantlyreduced, and thus reducing the building lateraldisplacement significantly. The restraining effects ofthe exterior columns against the belt wall rotationresulted in axial loads in the exterior columns. These

Figure 7: Belt wall system load flow diagram between the core wall and the exterior belt wall system




axial forces were counter balanced by force couplesinto the belt wall system, which were then counter

balanced by force couples into the slabs that are finallyresisted by the interior core wall system.

The advantages of the indirect outrigger belt wall

system over a more conventional direct outrigger wallsystem can be summarized as follows:

♦ Placement of the belt wall system at the perimeterdid not restrict the mechanical floor space andallowed for freedom in placing the mechanicalequipment in the plant space, thus reducing theamount of coordination work required betweentrades.

♦ The Construction of the belt wall system was notin the construction schedule critical path;

♦ The belt system did not have a direct link betweenthe core wall and the exterior columns, and

therefore eliminated one of the most difficulttechnical problems encountered in the design anddetailing of the direct outrigger wall systems,which is the potential of generating large forcesdue to the differential movement between the corewall and the exterior columns; and

♦ The extensive detailing and construction sequencework required for the direct outrigger wall systemis significantly reduced.

4. Construction Planning

Samsung Construction was involved with thedesign team from the early design stage in evaluatingall building systems (structural, architectural, buildingservices, etc.). Alternate systems and details werediscussed and incorporated in the constructiondocuments. In addition, the General Contractor issueda detailed construction and design schedule that wasutilized by the design team as a base for providing thenecessary documents on time so that the GeneralContractor could proceed early in the construction

planning and work. The core and shell of the 73 storytower and the six (6) subgrade levels were completedin approximately 18 months. The entire project wascompleted in 28 months, including the residential fit-out space and finishing work. A 3-day cycle wasutilized for the tower superstructure and all subsequentconstruction activities.

Mat Foundation ConstructionThe tower is founded on 3500mm thick, 40Mpa

reinforced concrete mat. The mat foundation systemrequired significant planning effort while preparing forthe 8000 cubic meter pour of high performanceconcrete. See Figure 8. The General Contractor usedself-compacting self-consolidating concrete (SCC) forthe mat foundation. The advantages of using SCCincluded:

♦ saving in casting time by placing the mat in asingle pour and within 12-hours;

♦ reduction of labor force;♦ consistent concrete quality and uniformity,

especially in areas with high rebar congestion.♦ Increasing the concrete bond to rebar due to the

elimination of bleeding under rebar;♦ reduction of concrete bleeding;♦ eliminating the vibration noise.

A cooling pipe system was suggested by thecontractor to control the heat of hydration for themassive mat concrete pour and to reduce the curingtime. The curing time was reduced from 45 days to 15days and the maximum temperature at the center of the

mat was reduced from 93o

C to 82o

C. The maximumdifferential temperature between any two points withinthe mat was less than 20 oC.

Figure 8: Casting the foundation mat

AB

C

0 1 2 3 4 5 6 7 8 9 1 0 11 1 2 1 3 14 15

Time (days)

10

20

30

40

50

60

70

80

90

T e m p e r a t u r e

( C )

B

ambient

o

ambient+20 C

C

A

o

0

10

20

30

40

50

60

70

80

90

10 0

0 24 48 72 96 120 1 44 168 192 2 16 240T im e (h)

T e m p e r t u r e ( ℃ )

20 H 5 2 H 9 6H 1 39 H 1 6 3H 1 7 7H 2 40 H

Center

0. 5m from topT op

am bient

Hydration Analysis

Figure 9: Hydration analysis and measurement

Actual Temperature




The cooling pipe system consisted of 25mmstainless steel coils that were spaced vertically at750mm on center, and horizontally at approximately1500mm on center, at 128 locations. , The water flowrate in the cooling pipe system was limited 18liters/min. The cooling pipes were connected to acentral cooling tower, located at the job site, whichwas used to temper the water. The water entered thecooling pipe system at 34 oC, existed at 47 oC.Comparison between the predicted and the actualtemperature of the mat, due to heat generated by thehydration process, confirmed that the predicted heat ofhydration analysis results were consistent with theactual temperature of the mat. Thus testifying to thealready established analysis procedures utilized for

predicting the heat of hydration of mass concrete.

Core Wall ConstructionFigures 10 and 12 depict the construction

sequence of the tower and shows that the reinforcedconcrete core wall is constructed at four (4) floorsahead of the structural steel framing. The core walland the structural steel construction is followed by thecomposite deck placement, the floor slab concrete

pouring, the MEP system installation, and finally theexterior cladding system installation. The central corewall construction utilized a climbing formwork systemdesigned and provided by Doka (DOKA SKE 100).

Structural Steel Erection and Floor FramingThe structural steel erection of the tower was

divided into 3 wings. In order to reduce the structuralsteel erection time and to improve the constructioncycle, all the steel columns were erected as four storytiers, and the five (5) different balcony framing aroundthe perimeter of the building were prefabricated anderected as single units. This erection sequence reducedthe number of pieces handled at the site. Each tierconsisted of four (4) floors and all the structural steelin these floors were erected in 12 days, thus resultingin 3-day cycle. The concrete slab, at a typical floor,was divided into 3 equal pours, one for each wing.

Belt Wall ConstructionBecause of the complex behavior of the belt wall

system, the size of the concrete pour, and the geometryof the tower, the project called for special detailing,construction sequence, and construction planningeffort. Discussion between the General Contractor andthe design team resulted in eliminating the belt wallconstruction sequence from the construction schedulecritical path. Construction sequence analysis of thetower was performed and its impact on the buildingdesign was incorporated into the design process.While the belt wall system was cast in two separate

pours, the floor slabs were cast in three different poursat each floor in order to eliminate the potential foraxial load cracking in the slabs, resulting from

shrinkage effects, because of the restraining effects ofthe core wall and the exterior belt wall system.

Major EquipmentThree (3) self climbing cranes, supported directly by the core wall, were utilized for the towerconstruction. Two concrete pumps (one foremergency) were available at the site and were usedfor concrete placement. All concrete was pumpedfrom the ground level to approximately 264m abovethe ground.

5. Axial Shortening of the Tower

One of the key issues considered during the

development of the structural system of the tower wasthe differential shortening between the core wall andthe exterior/interior composite columns duringconstruction and after significant completion time ofthe project. The complex behavior of the compositecolumns called for special considerations incalculating the elastic and long term deformation forthe core wall and the exterior columns. The axialshortening of the building was done using the state ofthe art ACI 209 committee draft report for predictingthe creep, shrinkage and temperature effects ofreinforced concrete structures. This draft report is

based the Gardner-Lockman GL2000 Model, which is based on RILEM base of test data.

Steel ErectionPlumbingBolting,

DeckingStopperStud Bolt

MEP workReinforcingCon’c casting

Curtain wallFire proofing

Tier N

N-1

N-2

N-3

Figure 10: Tower Construction Sequence




The predicted composite columns and core wallaxial shortening by the new analysis program werecompared to the actual in-situ strain measurements atTPIII, and was found to correlate well, thus testifyingto the accuracy of the new analysis models andapproach in predicting the time dependent deformationfor axially loaded members. See Figures 11 forcomparison of the predicted strain to the actualmeasured strain.

A compensation program was developed to makeup for the overall building shortening and thedifferential shortening between the columns and thecore wall. The building shortening was calculated to

be between 270mm to 300mm for the core wall andthe exterior columns respectively. However, themaximum differential shortening, at level 60, betweenthe columns and the core wall was estimated to be40mm, at the completion of the building, and 20mmafter 20 years. Therefore, the column elevations wereadjusted during construction for 20mm so the relativemovement between the core wall and the columnswould be limited to a maximum of 20mm. The actualmeasured differential shortening between the core walland the exterior columns have been found to be lessthan the predicted shortening.

At TPIII, several high frequency strain gages wereinstalled at several SRC columns and at two locationsin the core wall at several floors. These strain gageswere connected to a single data logger from which theinformation has been down loaded automatically for

processing by the researchers and the design team.

In addition to the in-situ elastic and inelastic timedependent deformation monitoring programs, amonitoring program has been installed to monitor the

building dynamic response to dynamic excitations,especially as it relates to wind effects. It is anticipatedthat the instrumentation programs will be expandedand integrated into a single system that couldessentially provides a health monitoring program to the

building structure, which could finally be integratedwith the permanent intelligent building system.

6. Conclusion

This paper provided a brief overview of some ofthe issues considered in developing the structuralsystems and construction planning of Tower Palace III,the tallest building in Korea. The monitoring

programs incorporated into the design of the tower arevery unique, could provide invaluable information andreference to the engineering community, and should be considered as part of the intelligent buildingsystem of important building in the world. Tower

Palace III is a landmark tower for the city of Seoul.

Figure 11: Measured vs. predicted strains at atypical Interior composite column

Figure 12: Measured vs. predicted strain atreinforced concrete core wall

Figure 12: In progress construction of the towershowing the construction methods of the tower.

ctbuh_tallest korea building

Documents