BN-97-1-2

Demand-Controlled Ventilation of an Entertainment Club

Gaylen V. Atkinson Member ASHRAE

ABSTRACT

Entertainment clubs, nightclubs, theaters, restaurants, and coliseums, with their highly variable occupancy rate, are excellent candidates for demand-controlled ventilation. The dynamic thermal requirements of both heating and cooling, coupled with the need to control indoor air quality because of the large number of patrons who also may be smoking during the highest occupancy, provide an opportunity to integrate the temperature controls with an indoor air quality control sys­tem. Significant energy savings may be realized by controlling the ventilation of outdoor air to match the heating, cooling, and humidity requirements as well as maintaining acceptable indoor air quality. This paper describes a demand-controlled ventilation system that was Installed in an entertainment club in Boise, Idaho, using a multigas indoor air quality sensor to measure the level of indoor air pollutants, which, when com­bined with a mixed-air temperature sensor to provide econo­mizer cooling, introduces outdoor air at a rate required to adequately ventilate the space.

BACKGROUND

The entertainment club described in this paper is located in Boise, Idaho. It has a seating capacity of244 around tables where drinks and food are served while local and celebrity comedians perform their acts during scheduled showtimes. Smoking is permitted during the performances; therefore, local zoning and code requirements call for a high air change rate to meet indoor air quality requirements. The intermittent operation oflow occu­pancy during the day and very high occupancy during evening performances, coupled with the need to maintain the air quality by diluting the secondary tobacco smoke during the perfor­mances, generated a concern over energy use. It was determined that a mechanical air-handling system was needed with a maxi­mum outdoor air capacity of 5,800 cfin (2,738 Lis) to handle peak crowds yet also take advantage of mixing return air with the

outdoor air for economizer operation, especially during non­showtimes or less-than-capacity performances. This figure came from 290 people total occupancy at 20 cfm (9 .4 Lis) per person. A control method was desired that would provide the outdoor air necessary for thermal loads and indoor air quality during the performances but go back to a standard mixed-air cycle during other occupied times.

DESCRIPTION OF MECHANICAL SYSTEM

Several different mechanical design options were consid­ered. Among them, using up to 100% outdoor air was considered desirable for the performances, with a standard mixed-air econ­omizer cycle during other occupied times. Although it was considered, the size of the mechanical system and project budget did not justify heat recovery for a 100% continuous outdoor air operation. Finally, a single constant-volume package unit was selected with a 5,800-cfin (2,738-L/s) capacity; outdoor air, return air, and relief air damper sections; an evaporative cooling section; and two stages of gas heating to handle l 00% outdoor air at - 10°F (-23°C).

Direct evaporative cooling was selected because it was a good match for the dry Boise climate with very low summer wet­bulb temperatures along with the requirement to cool 100% outdoor air. A direct expansion (DIX) cooling section, sized to handle l 00% outdoor air, would have been considerably more expensive in both first cost and cost of energy for operation. In addition, the DIX section either would have had considerable excess capacity for smaller crowds or nonperformance occupied periods or would have needed an expensive control package to meet temperature control requirements at light loads. It would have been difficult to control the space temperature during the cooling mode without wide temperature swings and short­cycling the mechanical equipment. The potential necessity for I 00% outdoor air during perfonnances was a good match for the direct evaporative cooling selection because evaporative cooling is usually only operated when the unit is in full outdoor air mode.

C11ylcn V. Atkinson is president of Atkinson Elci.::tronics, Im; .. Sall Lakt: City, Utah.

THIS PREPRINT IS FOR DISCUSSION PURPOSES ONLY. FOR INCLUSION IN ASHRAE TRANSACTIONS 1997, V. 103, Pl. 2. Not to be reprinted in whole or in part without written permission of the American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc .. 1791 Tullie Circle. NE, Atlanta, GA 30329. Opinions. findings, conclusions, or recommendations expressed in this paper are those of the author(s) and do not necessarily reflect the views of ASH RAE. Written questions and comments regarding this paper should be received at ASH RAE no later than July 18, 1997.

CONTROL CONSIDERATIONS

A standard constant-volume pacl<age uri\t col'!trol system was considered but felt to have several drawbacks." A typical programmable comme~c'!al-grade fnultis~age heatillgtcooling thennostat with the mixed air being t~~ first stage °.f ~ooling probably would have been ad~qyate for,, trye space temperature control if there was not a requirement 6r I 00% outdoor air during capacity perfonnanc.es. Three diffe~ent m6des of opera-tion were determined. ' ·.•

acceptable indoor air quality level. It was decided to pursue this option if an inexpensive yet acceptable air quality sensor could be selected. A carbon dioxide (C02) sensor was considered but not used because even though C02 would be a good indicator of crowd size, in this case it would not necessarily reflect the total indoor air quality problem because of the potential of consider­able amounts of secondary tobacco smoke.

INDOOR AIR QUALITY SENSOR DESCRIPTION

I. Unoccupied Mode: )he unit would n~rmally rem~~ off Therefore, the air quality sensor selected. was the multigas outside ofbusiness hour~ 'because no one would he nre~e~t in thP- type that was sensitive to volatile organic compounds (VOC5) build'hlg. The unit fan, however, ~ould start along ~ith tfl~h-~;~~ It was installed in the return air duct and consisted of a stannic ing section ifthe space temP.erature,?ropped below.~0°F'(i5°C) (tin) oxide element heated to a constant temperature. Molecules and 'the outdoor air damper would remain closed. At the; ~ther of VOC§. .. are catalytically repuced or oxidized on the heated extreme, ifthe space terliperature rose above 85°F (290'c), 'the surface, thus changing the conductivity of the element. This unit fan would start along with the ev~p9rative cooling and the change inxesistance is amplified electronically and converted to

. ... . r 011trlonr ::iir cl11mper wnulrl open . ··' 4 a Oto 10 volt DC1inearnm1loe signal, with 0 volts corresponding

II. Occupied Mode: The unit fan w_ould start, the outdoor air to "poor air qualitx''. and 10 volts DC correSp()ndipg to,_"go0d air damper would open to a l)1it1itnum, ventilali1,>IJ pusiHyp, and the quality." An increased number of VOC molecules indicates a s!"lace ternr>erature wo11lcl he cnntrollecl hy r.yr.lin~ thP hP:.tin~ "poorer air quality." The sensor res~onds well to hutyric 11cicl, stages, when needed, to maintain a 72°F (22°C) heating setpoint. ... · which is a component of body odor along with the components For cooling, a 74°F (23°C) cooling setpoint would be maintained of tobacco s~oke such as carbon monoxide and other secondary by enabling the mixed-air controller (set for 60°F [ 1 S°C]) forthe • --· smoke .. coritarninants. The. sensor is also- 'reactive to VOCs first stage and then starting the evaporntive cooler with opening produced by the offgassing of.new carpet and furnishings and is the outdoor air damper to 100% for the second stage of cooling. calihrated to a 3.5-volt DC otitput corresponding to 50% "good With this cycle, the "standard economizer function" of closing air quality" at 1,000 parts per million methan~ at 70°F (21°C) the outdoor air damper to a minimum position when starting the and 50% RH. . .. - ~ . cooling section would be disconnected as full outdoor air is The sensor does not require field recalibration but must be needed to properly operate the direct evaporative cooling maintained in a continuously powered mode. The manufac-section. turer's literature recommended that the sensor be installed and

IIL Performance Mode: The unit would already be operat- p()wer.ed for four weeks before final occupancy adjustments. ing in the occupied mode but would then have the capability of Intermittent power interruptions require a sensitivity recovery opening the outdoor air damper to a 100% position ifrequired to time ofl 2 hours for a one-hour interruption and two days' recov-ventilate the space with higher air changes per hour to maintain ery time for a power interruption of less than one day. The the desired indoor air quality. Several modes of operation were recommended field calibration procedure is to observe the considered. indoor air quality' and then make gradual setpoint adjustments

(I) A time clock could be installed to open the outdoor air until the desired air quality is achieved. Figure I shows a sensor damper to 100% during performances. This would be simple but output graph. Air quality is measured in "AQs," with 10 AQs would have a cost impact of heating unnecessary amounts of corresponding to lO_volts ~<;--~~--~'perfect air," which would outdoor air when tpere were less~ifian-capaeity -~~owds. only be obtainable with synthetic air. Zero AQs or 0 volts DC

(2) An "air quality switcli" could be installed ill the stage would correspond to ¥el)! poof air-quality, such as that found in manager's office to open the outdoor air damper to (1 OO%·ifthe a winterril~'fe "thermal iriv~rsiotfWith srrioe ::incl othi:>:r pollutants. air quality during performances· became objectionable. This nie typicaLdesirable indoor air quality lev.el ~ould correspond option had the liability of allowing the indoor air quality to to between5to 7 AQs; lnthis rangethere W<;mld be a noticeable become poor before someone would· have to tike m nual :action freshness to the arr quality: Jilgi:ires lower than this would indi-to correct the probl¢m. ·With-either option,,because Mth_e p,o:si;i- cate that noticeable levels ?f p91lutants would be present. bility of very low supply airte:rriperatures causing c6mfort prob- ·' · ;_ -- - -.. - . !ems, a discharge air low temp~ratur.e.limiUhermostat would be CONTROL SELI;GJipN_· _ --· - ----required to cycle the gas burner to m:atiltain : te~1pering of the _ figure 2 shbws a schemati~ ~lagfiuri 'of the mechanical unit outdoor air until the space thermostat determined that heating ,along with the control hardware that was selected. The VOC air was required. quality sensor is shown in the return duct The packaged rooftop

(3) The third option was installing some kind of sensor to unit's mixed-air dampers and actuators were used, but a direct measure the indoor air quality and override the thermostat, · digital co_ntrol (DDC) terminal unit controller with a built-in modulating the outdoor air damper to bring in the quantity of disch~r~~-:~\r, ~e~sor capabili~ a~~ mµ!tistage ~~ating-cpalip~ outdoor air necessary to ventilate the space and maintain an control with outdoor air/return arr function was installed. A

2 BN-97"1-2

110 AO~ . -

GOdb AIR QUALITY

,. POOR AIR' 11.! QUALITY ' t.. ·

·• r ••

-.

':1(. • I •,Q. ,,

( .

Q\JTSIDE AW; .

.· ·,

~- RELIEF ~Fe· A\fl •.

j .

c; 2:::-1.::.

.. ,

' • I f I l·

I. •

,.

,,

' i-•·· ·,.j

·I"' J... i'J'.

I •

J ,.. .. ~

i, • j• '(' .

• 1

/· 1 11

.... ...

•, 'I·· ·.1... . ...

· . \:

') • ...!•

I r . I

· ; .: t I

r1 I _(_I-'." I I

..,, I ,. · , I

I "I.··:<'• . I ' I

I 'f ·n .. I

1,

I , • lDIJDC ·; '

AIR OU1LJ1Y

:1J ~.

·j

[i] MIXED " AIR

programmable DOC controller was used to input the signal from the air quality sensor am;I cqmpare it against a setpoim in AQs. . , , . . ~

lf~eairquality was less thansetpoint then the mixed-air control signal to the outdoor i\ir damper actuatoi was overridden and the outdoor air damper rrio_dulat~d to a p~sition necessW)' to venti­late th~.1 .~1pce to mairi:rain the desired air quality level. The dischafae air sensoi: .through the erminal unit controller cycled the gas heating to te~iper the outdoor airto avo.id space temper­ature swings as much as possible. The terminal unit controller receiving a signal from its space temperature senso'r: provided all tbe desired occupied and unoccupi,ed ter:nperature control !\me-

" • . I . • tions ' that ·were explained earlier in this paper. A const~nt-

.... ,. • • / 1 ; ., ' I

temperature (60°F [15°C]) mixed-air control loop was used for "occupied '•operation in lieu ~fa·' minimum position control, ptovi<;llng,the benefit of minimum ou~9oor air vent.4ation.

! 1 . . . 1 ••

SYSTEM. 0

0PERA TION

:·r. The system was; irlstaHed and 'bommissioned.; )\• DDC network pa.tie! was· installed tempo'rariiy to store sensor arid

' . \ . - :· ; ~ . .

;1;1 .!

. .d • .. ~ : ')

' ' .. _ ....

·. ' I

. ~_,, I

. ' "' ~ - .~~· P.ACk:AGE ROOFTOP HEATl~~G /COOLING I

·r··. i • "j

·T . -~!(

i ' t:

: . ~ · .

• • r.~ ••

·' .• 1)

• , I'.. . 1'.)

·j · : : .r · 1~. ·~ I._ • .. : ' ,.~ '. .f

l.i · r.t ~ ':. ' ' . I .

'J: '::~ . r

. .. ,._.It .".: ! ., '1"

~ .. J , . ~ ~· ( .>~~~;\i;.z;cJ~rl;P~NT . ·; J l i ' '. )"~

:.:.~ '(l, ii . ;:7ij.1'f(Q3,sc}'- -~~C ,• . r:::.~.,, ,; : .:;;1~: ~- '" r;.i; ,,

--!.r1EMP. ,-. • r)i ;.-...~~.i>,·: ;r_,·~.~.1:r.?IN~~~ ~ •. 1~ tT~Ci _, :} ~;- :.1•: ._rjd.G ;~"' \ c .. ' ., ,,,,.... ·-·· .

.. ,.r ·:- ·· '\ ·.-~ ·i·Jr\f1··~. , ! ;-.·0

·t: f.1! r· 1 1!;.v~'°~; i !L Cttlt 1 i' I•J ,::1::i~1 . '; '' . . • - . bO'f' (1'.iscj .. UN c ' . ) ' "'" •

1 · t.~i! ' r~nJ ;;; :iAtJ ·~i ! :·:'.':~~Yj:~~b ':r:· .i.;~~;vaO.f1 1 j·~.r- ~!nf.q •J;:1· ) :J ·nl~Al!G(; ;, • ::.2•;11:?'.f :.($4S;t; ~~~c~t',:. rWfilc«;·, ·1 ·;·i; ;x;;·c, ! •.·.1 -~·;(.'Nf; ·L)l·;J.)j ~~~: .. :J," .. .':iif.'.1'."i ~q ~ ... ·J·!~b'( ~ ~u~ :' 1 .1! :.--.::.,!·\ .. (.i · f(

, ,.: ! 11 11.1 '" l · •· .. ,,..:., ••-·'' · . ..... 1<,ril.n !~' :~ i~ S~.: _i~t~;. It~; , . 1 fi ;

~,· . . f Rq~, ,n:~¥~l~~jtt-}~.c9~~q~tJ,~d :·. ·<.ii A!·-'.J ~;, (,c,,~;1,~'.: . . ' HIL 11,:,- ~~ )0(·;\i1,

l...,: '.!p' · ' . I

·.._ 1.,;'.A ·· .. ·~·! ~j- :·~:· ;!i;;,~ .: ~ ~ .... 1 no ·q._i ·y11.1i; ~).1 ··, L

.~r - f~1 td .· '.i ll''' - !''YIM 11 11: • W':!i:·:E}j R~·ffi.~ .[):J.@FPAL CON TROLt[RS ;;'L'.; : . ; _·.~

F~D'r'd"l:' : isa'hem'ii1i~ ;~j;;;Jfiia,;J~~; :zii;t' ~it~ 2~~lk6fh~,:a~are. ..., ·" .. e · · ~-" ., '\ ' ..... ·; , · !. . ;•!" ·· -·_j ; .~ • ) i1,!: ,I i. : ~;t~ it.

BN-97-1·2 3

control system values and trend the control system operation so that the desired setpoint for air quality could be determined after observing a number of performances. The building thar;was renovated into the comedy club did not have any other DDC controls so the field panel was removed after the desired set'j:>oint adjustments had been made. Figures 3 and 4 show graphs of the data collected by the field panel over several different days \Vhile it was connected to the DDC controllers. The outdoor air temper­ature, indoor air quality; zone temperattfre, -arid outdoor air damper position are· graphed over time corresponding to 24 hours each day on February' l8 and 19, 1995. ..

Figure 3 shows .the .building . .switching into "occupied" mode at 10:00 a.m. The outdoor air temperature only rises to about 60°F (l 5°C) by 2:00 p.m, and then starts to drep. The space temperature cycles in the heating mode around 70°F (21°C), and the outdoor air damper position stays around 60% until the first show starts at 7:00 p.m. The indoor air quality remains around 70% AQs during the same period. When the first show starts, the indoor air quality drops rapidly to Jess than 50% AQs. The outdoor air damper is modulateCl open to I OOo/c{whith'kee'os the air quality from dropping any further. The first show. e~ at about 9:00 p.m. and the air qualitX rapidly impro.y~s, refu.!fling the damper to the mixed-air control position of about 60%. At I 0:00 p.m., the next show starts and the same cycle repeats itself, opening the damper to 100% outdoor air. The second show is ov~r 'at about I I :30 p.m., the crowd leaves, and by midnight the air quality has improved and the outdoor air damper position returns to .t_he mixed-air cqntrQL The system· switches to "unoc­cupied" after.midnight.

Figurei4 shows the system operation onthe next day, Febru­ary 19~ On' this ·day the outdoor air-temperature rises t'o almost

I

, I·

r; 1 ;":,.I!:' •1:

1· t ..•

-, t.·rr1JJ ·.J: :-' ,:' •· ···~r:r:

. I • -~ ·.t '

~> :.

''· .. . -<. .. Figi1re J'·

I .•. • I I ' I 1

IAQdata- February 18. 1995. i ...

4

.. 70°F (21 °C). The mixed-air controller opens the damper to the l 00% position during the afternoon, but by 5 :00 p.m. the outdoor

··airtemperatureha:S-dtopped to below 60°F (15°C), which allows the damper position to return to the mixed-air controller. At 7 :00 p.m.,tti.eshow starts arid the indoorarrqualitydropsrapidly. The damper position modulates again to I 00% outdoor air, which prevent~ the air q1iaii€ from gehing any worse. The single show is over a.bout !O:OQ p.m: and.·Jhe indoor air quality improves rapidly, returning the outdoor air damper to mixed-air control. Tbe unit once again switches-to the "unoccupied" mode around midnight.

From observmg the rapid change m' ''indoor air quality on multiple days it became apparent tl)at the.::;ytpoint adjustment for the indoor air quality in AQs was about right. The temperature cpnlrols functioned normally, keeping .. the space temperature within desired limits. The air quality sensor rapidly detected the show beginning with the crowd showing up and starting to smoke. The instantaneous response of the demand-controlled ventilation system pre:Vented the air quality from dropping bei~·w acceptable level~ d.1:1.~ the performances. Once the cr~w(fieft, the ~door . atr.·· quality improved rapidly and the system reverted tO normai mixed-air control.

' ' . \ .. ~ '

One could e,stimate ~p~rgy saviµgs from.nm, havii}.g to heat l 00% outdoor air for non-showtime occupied periods on Fe.bru­ary 18, 1995, as shown in Figure 3. Assuming that the unit is on for 14 houts from 10:00 a.m. to midnight, 6ne could take the average outdoor air temperature for the period along with the av9rage out~oor

1air damper p9si~ion and calculate t.7e s~yings_ in

heating a re4~ced amount o~ q)Jtdoor air. up 19 ~e ~pac;:e .tr~pc;r­aMe. A more accura~e figur~ could 'be 9~~<,1 in~d by piowing.the

""' a ..

"\.. : . .1

I 'o f

t \ \.if. t•: I

-ZONE TEMP

I • ' '·

+H DAMPS! POSITION

;1 )t! i f, . , C,

, . .::• 'l'

. '

. ,:..,,, .. I ,o

(o · , . . ..

•'

• t .•l

.. .,,._ , ~ .. • 1 .. :::>: •<

I• ', . ,, •.,•

.'•I •.·

I • •_..•

BN-97~1-2

I •

; • . I •

.·.;

. , ) ( f1 ·.: •; ... ' .,.,.

, r: ; :1

:1·

. '· , ... ..,. 1 .• ·rY · :-:· ·\"

Figure 4 IAQ dafci_:._February 19, )995.

supply air temperature, ·M't unf6rti.iifately ·' tlihe data w~~e not collbt:ted. '· · ·· ·. · :<· ,.·; ·: 1.:: · · : ..

IQ(no_n~how~ours).x 5800 cf!ll .x.40%(0A),

, , : ,., ~ .(7,0 - 55°F) = 34SMBH/Dayfe11ergy savings ,·• ·_., •• -. J • • •1 .. ••' t ; ' • I ·· 1 .

T.~.i~ c~~espo~~ . to I ~2: ki l ~w·~~s . , Qne sho~J~ p <?t use'f pis da1ly figure to alculate the savings for a year, as tlie outdoor afr;teltlperaru·re and h;e'ating loaci' v:;buld vary each~ day. ,; ! '

CONCLUSIONS

After installing the control system descr[bed ~b~ve ~~d observing its perforinance,.,t~e...followi!)g f.QQclu~ofu call.be made .

' 'I

• t • •

•. ' ·

' t. :r

!~ ' thl~ ~p~per has been. installed for more than two yeai;~ .. . . ,operati~ as designed, '¥,i,~hout requiring any rec~J,ibi;a-

tion. r .•i· .. ·. '1~ . II! .' ·! ,

5., The mu'ltigl'is voe indoor air ·quality sensor u'setl in the project appeared to react favorably to ·tpeople load. Whetl~.tfr other sensors, sucb,a~ the, ~Ortype sensor, track o,ccupancy load more accurately is npt for di~cussion in this paper. Combining the requirement to sense both peo­-i:rntcount and' overall indoor· air quality, the multigas sen­sor appeared to be an acceptable selection for a sensor to

-cfrlve the ·demand-contr~lled ve~tilation system.

ACKNOWLEDGMENTS 1. Volatile organic C::°J?-pound,. multi_.g;a~ _a.iy qua)i_t)f sen.sor.s . . . ~ - -~ ~~ :-""'' - -- ·· - . . . .

are an acceptable, relati.Y.elx_:inexpensive ·way to ¢easure · The a~uthor wishes to thank the followmg md1v1duals who indoor air quality in envir-0Rm'eius where there·are .vari'-" ~ .. t-~bav~ ·Glrntrjbute~~~stanfiirt~-y-t<Fthe· s\.locess of this project: able people loads al~n~ . 'Yith air quality~c~in~tf6n.. R6bert Tikker of eifgineering Inc., Boise, Idaho, mechanical sources such as sntcflfolg: -- +· ::.-. - ~'."- . -·- · ---;!- ·- -·- ·system designer, Tony S.~tton- of-Atktl)son Electronics, Inc.,

2. Demand-controlled ventilation .'with outdoor air is an Boise, Idaho, control engios;~\ngi and commissioning; and acceptable c.ontroT ieeliiruq~~- that' admits-only the quantnY·-· ---sfycef\tKmson ori\-~kin~91lE1ecfronlcs;·1nc ., Salt Lake City,

of outdoor air n~eded_ !2._.!!!~_i!,1.!_ajn a_ g~sired ~[ _q@JjJY _ .. ~~1··-~~~i:_ationa~ dat~~~al~~is '!!l~_Pr_eJ_~ inary write-up. when there is a means of measuring the air quality with a ,. control sensor. ···-- ·--· .. _ - · _ -·--· ... ___ . ___ ._.BJB.LIO.GRAPHY ___ ···-·· ~':. __ ---·-~ t

3. Energy savings can be realized by using demand-con- <,

trolled ventilation ill-lie1H>flOO% continuous -eutdoor·air .. _. ASH.RA.~. J9.~~ . Al!S.Jl,{Sf1ME..Sw.wiard 62-1989, Ventila­systems. The energy saved is that not needed to heat or rion for acceptable indoor air .if~ality. Atlanta: Ameri-

cool excessive am~)l1pt~~iR}l~?W~a'!!.:.::b!)iQ"~'~ '£1U'lfll =--rryrr~:/IT1.:f·~;;-~~j~t.~ ~-: -~efJ"~~~ .. . ~~igerating and Air­tity necessary to maintain .!!.I.!~~_p-~gle.Jnc!Q.QI .air Jl!.l_aliD'._ .. . _ --~~-~~~t~~~!".~ . __ n_g!~~'.~!fic . level. ~ ,·.: .. T;e:~ '' >? '"'' n •-'c ~1.r!.' ~ :::: · ·.

4. The long-term stability of th-evoc:ry'i)e"rurquaHtys'e-n: .....

sor appears to be adequate because the system described

BN-97+2 '

S:~!.P?rl.~<:r; ·-s~~ ·--1. ~.915":~.~M'fgy and IAQ impacts of C02 demand-controll~d ventilation.,s_ystems. {ISHRAf TrPIJ.S.f actions I 02(2). . ' · · · . · · ., · · , · r .

5 '

Elovitz, O.M. 1995. Minimum outside air control methods for VAV systems. ASHRAE Transactions 101(2): 613-618.

Haghighat, F., and G. Donnini. 1992. IAQ and energy-man­agement by demand controlled ventilation. Environmen­tal Technology 13: 351-359.

6

Levenhagen, J.I. 1992. Control systems to comply with

ASHRAE standard 62-1989. ASHRAE Journal 34(9):

40-44.

Staefa. 1995. Application bulletin S 1-02.50. San Diego,

Calif. : Staefa Control Systems, Inc.

BN·97-1·2