GATXGENERAL AMERICAN TRANSPORTArION CORIPORATION

C L EA RI N GHO U, EFOR FEDERAL SCINTPIC AND

TECHNICAL INFO-7-MATIoINrfalrdoopy microf ieli-j- -

C-~-cre /DD.A

GENERAL AMERICAN RESEARCH DIV•IGIN.om

7449 NORTH NATCHEZ AVENUE. NILES, I1LINOIS B•084U ,'-s

• .. Copy

Prepared 2orOffice of Civil Defense

Department of the Army, OSA4under

Wor.!- Unit 1214ASRI Subcontract No. B-64220( L9t9A-16) -us

EXI-M-RIMENTAL STUDIES OF FALLOUT

SHELER VETThIMTIOI REQ~UIRE1,03TS

by

H. F. Behis

C. A. Madson

OCD Work Unit 1214'A

CARD Report 1.268-4~o October 10,05

Distrib-uti.Gn of this aci~sixeflt is- a.nliunixeý.

With contributions by:

J. L. Leitherm

B. A. Libovicz

H. A.Meier

Reviewed b-%.: Approved :1y:

Group Leader Genera3. Mlar.agezEnv!-ronr~iental Researcl -I

YhiL re-,_=-, hac fe revi;ýwe-j in --ýhe Otffi.-e of Civil Defense azil apprv..aed Xp15:at1.C ,. xqrrL:.aL e -Lzs ".i.t signifv t1-,h n~t a 3 cessanily r: £lect

ihe -views~ and po: .i-e CýfYice rf MvilŽ 'Defense.

j. iFOREWORD

The forced ventilation tests reported herein were conducted by the General

American Rcsearch Division (GARD), formerly the MvRD Division, of the General

American Transoortation Corporation, Niles, Illinois, during the period from

October 1962 to September 1964 under the Office of Civil Defense Contract

No. OCD-oS-62-134. Mr. Frank C. Allen of GCD's Directorate of Research was the

project monitor. These tests are reported under Stanford Research Institute

Subcontract No. B-64220(4949.A-16)-US. Mr. C. A. Grubb of SRI is the project

monitor. The contracts provided for forced ventilation tests of representative

SI- identified fallout shelters: (a) to evaluate parameters that determine the

nature of the resultant environment in identified shelters in existing

buildings, (b) tc determine minimum equipment requirements for control of the

Senvironm.ent in accordance with limiting critaria, and (c) to obtain and cor-

relate experimental data 1n support of c-.oreii t or modified computational

I methods or for direct use as empi-rical data.

j Pkperimental !nd analytical work regarding natural ventilation tests of

fallout sheite,- bas bei~un ander thi SRI ,zcitract. Such tests have already

1been perfofszia in Batcn Rouge, Louisian, and BRzeman, Montana. A final reporz

o -al- natural venrilation tests will **e available upon czmpletion of all tests.

t

I-

1II

ABSTRACT

The results of two years' field testing of fallout shelters is reported

herein. Simulated occupants (Simocs) and forced flow conditioned air were

used to duplicate emergency environmental conditions. Nine tests have

previously been documented in detailed interim Reports. Based on field measure-

ments of temperature) humidity and heat flux, and supplemented by an analytical

computer program, an "adiabatic" procedure is recommended to predict shelter

environmental conditions. This adiabatic procedure neglects heat transmission

through the shelter boundary surfaces and can predict shelter effective tem-

peratures to within 2°F. The procedure is cor.servative in that it will over-

estimate the shelter temperature for all shelters tested.

_____ ____ __ _ _ __ ____iii _

A

TABLE OF CONTENTS

SECTION Page

FOR0WORD ....................... ............................. i

ABSTRACT .................................................................... i

1 INTRODUCTION ................................................. 1

1.1 Simulated Occupancy ..................................... 3

1 Instrumentation ......................................... 3

1.2.i Air Flow ......................................... 3

1.2.2 Temperature Measurement ..........................

' 1.2.3 Heat Transmission ................................ 6

1.2.4 Shelter Energy Inputs ............................. 6

1.3 Test Vehicle .................................. ..... 6

2 TEST RESULTS ................................................ .. 9

2.1 Houston Basement Tes ................................... 9

2 2.2 Chicago Abovegrotnd Test ............................ 1!

2.3 Milwaukee Aboveground Test ................................. 13

S2.4 M*iwaukee Basemert Test ............................ ... 14

2.5 WJlming-ton Aboveground Test ............................. 13

L 2.6 Wilmrngton Be'owground Test ............................ 20

2,7 Bnzeman Aboveground Test ................................. 25j2.8 Athens Hiurz-n Occupancy Test ............................. 26

2.9 Providence Basement Test ................................. 29

3 SUP14ARY OF PESULTS .................................... .. .. . 33

4 CONCLUSIONS AND RECOMEENDATIONS .............................. 37

L REFERENCES ..................................................... 43

I.iv

LIST OF TABLES

NO. TITLE Page

1 Summary of Forced Ventilation Tests 2

2 Houston Shelter Boundary Surface Properties 10

Chicago Shelter Boundary Surface Properties 12

4 Milwaukee Aboveground Tests - Significant Results 15

5 Milwaukee Basemer,6 Tests - Significant Results 16

6 Milwaukee 13asement Shelter Surface Properties 17

Wilmington Abo-.eground Test - Significant Results 19

8 WiLmington Xboveground Shelter Surface Propert--ie-- 21

9 Wilmington Belowground Test - Significant Results 22

10 Wilmington Belowground Shelter Surface Properties 24

12. Bozeman Aboveground Test - Significant Results 27

12 Bozeman Aboveground Shelter Surface Properties 28

13 Prcvidence Basement Shelter Surface -Properties 31

14 Providence Basement Test - Significant Results 32

15 Summary of GARD Forced Ventilation Shelter Tests .,4

16 Belowground Heat Losses 33

LIST OrF FIGURES

WO_. TITLE Page

f 1 Aggregate Simulated Occupant (Simoc) 4

2 OCD Test Vehicle No. One 7

S3 Aboveground Tests - Effective TemperatureCorrelation 36

41 Aboveground Tests - Dry Bulb TemperatureCorrelation 36

J5 Ventilation Reauirements for 80'F EffectiveTemperature Shelter 38

6 Ventilation Requirements for 820 F Effectivei Temperature Shelter 39

7 Ventilation Requirementz for 85*F EffectiveTemperature Shelter 40

I-

I

II

i.

vi.

SECTION 1

INTRODUCTION

Daring the period from October 1962 to September 1964, forced ventilation

tests were performed by GARD on nine fallout shelters. A brief summary of

these tests is presented in Table 1. All tests have been previously described

in detailed Interim Reports (Ref. 1A to 1H and 2) and these should be referred

,o for specific details. The principle objective of the test series is to

determine the minimum equipment requirements (ventilation rates) for maintaining

habitable conditions of temperature and humidity within building types which

may serve as possible fallout shelters. A secondary objective is to obtain

experimental data to be used to verify enalytical methods of predicting shelter

psychrometric conditions resulting from ventilation with ambient Jir.

Sufficient combinations of shelter geographic locations, tylpes, configur-

ations and sizes were tested to assure a thorough study of the parameters

affecting fallout shelter environment. The shelters were tested with various

occupancy, lighting and equipment loads, ventilation rates, and in some cases

different air distribution systemis were evaluated. Since critical shelter con-

ditions will occur during hot and/or humid weather, all tests except one were

conducted during periods of warm weat:er. Ii possible, the tests were performed

when ambient conditions approached summer design weather, or these conditions

were simulated and sufficient data obtained to correct nondesign test results

to design conditions.

1

Table 1

Summary of Forced Ventilation Tests

Test location Type of Shelter Type of Forced Ventilation Test(s) Test TotalDate Occupancy

a) Programmed cycle test using average hourlyHouston, Texas Partially conditions for a typical August in Houston, 10 Oct

Belowground Texas; (various cfm and occupancy levels) 2 Nov 400b) Constant 750F DBT and 95-100% RH Test 1962c) "No Ventilation" test

a) Ambient air tests 0 3 and 2 cfm/occupant 18-29Chicago, Illinois Aboveground b) "No Ventilation" test March 3.0

1963

a) Ambient air tests at? various cfm usingdifferent air distribution systems

b) Controlled supply air test; constant dew- 30 JuneMilwaukee, Wi;.-' Aboveground point and ambient dry-bulb temperatures 25 Jul 240

c) Dehumidified air tests simulating a 1963refrigerated coil and well water coilsystems.

a) Programmed cycle test using (5%) Milwaukeeasummer design cycle (various cfm)

b) Constant inlet air condition test (ave. of 1-31Milwaukee,5 Milwaukee design) August 200

Belowground c) Simulated air conditioned shelter test. 1963

d) Constant inlet air condition test (hot andhumid climate)

a) Ambient air test: @ 5, 9, and 15 cfm/occ.b) Programed cycle tests at various cfm usirg 9-28

Wilmington, N.C.,. Aboveground summer design cycles of Wilmington,N.C. Oct 320(2-1/2%), Phoenix, Arizona (5%), and Mil- 1963waukee, Wisconsin (2-1/2 - 5%)

a) Programmed cycle test using (5%) Wilmington, 5-13Wilmington, N.C. Belowground N.C. summer design cycle (various cfm) Nov 30b) Constant dehumidified air test 1963

c) "No Ventilation" test

Ambient Air Tests: 11-29 275,Bozeman, Montana Aboveground 1) 460 occ. @ 3, 5, and 7 cfm/occ. June 460

22) 275 occ. @ 3, 7, and 11 cfm/occ. 1964

Human Occupancy Tests (evaluation of various 31 JulyAthens, Georgia Belowground air distribution systems) 2 Aug 300

1964

Providence, R.I. Partially 14-Day Ambient Air Test @ 8.5 cfm/occ. 23-AAgBelowground A Sept $00

196(4

2

For each test extensive data were recorded on shelter, inlet ani ambient

air conditions, as well as heat flux and temperature distributions ir. the

shelter surfaces and in the surrounding soil of belowground shelters.

1.1 Similated Occupancypancy

o0 For all tests in this series (except one), the metabolic load of the

snelter occupants was simulated by aggregate Simocs (Fig. 1) (Ref. 3). Each of

these electromechanical devices can simulate the latent and sensible energy

output of up to sixty sedentary human beings (400 Btu per hr-occupant). The

electrical energy supplied to the Simoc heaters represented the total metabolic

load and gas manually adjusted by means of a variable tiansformer on each Simoc.

The moisture output of the shelter occupants was provided by a humidifier

located in each Simoc. As the shelter dry-bulb temperature varied, the rate

at which water was atomized into the shelter was automatically controlled to

0 match the human output.

1.2 Instrumentation0

1.2.1 Air Flow

For most of the tests in this series, pre-conditioned or u~ntempered ambient

air was supplied to the shelter by means of a 24-inch diameter flexible duct,

a sheet metal air metering station and an appropriate length of polyethylene

duct. The plastic duct was either placed on the shelter floor or hung from

the ceiling and ventilation air delivered to the shelter from one or more

openings in the plastic duct.

3

IFWI

ii III

1�Tii:

IC-)0J HCl)

2-iII C-)/ 0

q

I I

H

02IF::IC-,

HI I

i] I

I -�

/I

IE

A reasonably accurate measurrment of the shelter ve.tiaJix rate was

obtained -from air velocity measurements at the air metering station with either

a termal anemometer (Alrur Mole_ NO. 8500), propeller-type anemometer (Gill

Model B) or a hand-held rotating v'ane anemometer (Taylor Biram's Type No.

3132). The accuracy of all anemometers is estimated at + 5 per cen÷.

1.2.2 Temperature Measurement

Shelter, inlet and ambient wet- and dry-bulb temperatures were measured

with aspirating psychrometers (Sargent Nc. S 42610o equipped with copper-con-

bcantan thermocouple sensors. In some instances, the wet-bulb thermocouples

were replaced by mercury bulb thermometers graduated at I/20F intervals to

obtain more accurate data. For portable indication of psychrometric conditions,

the Bendix Model 566-2 "Psychron" mercury bulb psychrometer was used. The

Minneapolis-Honeywie]l "Dewprobe" (Model SSPl29B) was used to directly measure

the dew point of the shelter air. However, because this sensor required care-

ful handling and frequent maintenance to maintain calibration, its use was

discontinued after the first few shelter teais's.

Additional dry-bulb temperatures were obtained from resistance bulb ther-

mometers and copptr-constantan thermocouples located inside and outside the

shelter. Temoereature measurements employing thermocouples included wall, floor,

ceiling and partition surface ten-peratures; wall interior temperatures and

surrounding soil .emperatures. All thermocouple measurements were recorded

by strip-chart multi-point recorders. Class thermometers and resistance bulbs

are accurate to + I/20F; thermccouples are accurate to + !,*F.

5

1.2.3 Heat Transmission

Heat transmission measurements tllc.ugh -helter surface.3- er• mada at the

inside surface of the geometric center o e•ic', major su:tface oJ T, he sI-elter.

The heat flux. transducers (National Instrument Taot'-esorie-s Mcel 7•-3) co.xsisted

fof a disk with a spiral thermopile moulded in a filler mraterial of po!yvinyl-

chloride. These were neld in contact cih the sheltee 5urfaces by means of an

aluminum plate fastened to the surface by nylon scsews or we:ýre simply tapzed to

"the surface with a sufficient aiount of heat crnducting compound spread between

meter and surface to assure a good thermai nonzact, A third installation method

was to chisel away a portion of thee suorface materxial, position a heat meter in

the recess and plaster aver flush with n surface. The heat transmission

data were recorded ,i-Ith a maltipoipt recorder. Accuracy of the heat meter.,

I which were freauentl-•, recalibrated, is + 5 per ceat.

1.2• LShelter Energ-y MU:s

A kilowatt-hour mete-r me•t-red the total energy input to tte Simocs.

Instantaneous power levelh were also obtained with a precission (1%) ammeter

(with current transfor...ers) axd a voltmeter. An Amprobe Model AVMX recording

voltmeter and- a Model amAA2t recorain; anener provided a record of any voltage

and current fluctuations.

oTM wat•er input to the zbslter from the Simocs was measured on a balanced

beam platform scale, with an esti-.ated acwura.ey of + 1/2 pound.

1.3 Test Vebicle

7he air supply for ;he shelter tests waz obtained from the OCD Test

Vehicle R.. I (Fi-. 22) (Ref. • The test vehicle is capable of supplyixig up

I

'-4-

-�

- I fi �

II?-

I. w

Fig. 2 OtD TEST VEHICLE UO. OI�E

9

Vto 6600 (Pw ;) aIr at a pre6-eterffilned constat!,' dew poin~t tempera-ture a id a diry-

LLLŽ . '1 ~e~-tu ~. ti -,h rnezy '--e aut omatice.lly rveiezi at a. frn-_tion -i th- time

ak -la1y. The dzy-btil tý~e a 3u a corntroU.Je by rt-heating tIve supp'~ i

vdjt'a Phot- vater eoi.- Delum iificat ion was accjjrplI.i-!>ed '!ýy a 20 ton water

ch-i) ler, -and (uiii atnilad reheat) was achieved b,, a i~rt water boiler wit-h

a g'rQSS outpDUt o gu034 Btu per hour. The air flovi was manually controlled byadj~us;ircg the fzn speedA, t'he fan inlet van~es; and the air by-pass ports.

SECTION 2

TEST RESULTS

Individual test descriptions and significant results are summarized in

this section, arranged in the chronological :-rder of the performance of the tests.

2.1 Houston Basement Test

Forced ventilation tests were conducted in Houston, Texas (Ref. 1A) during

October of 1962 in a partially belowground shelter area. High earth tempera-

tures and weather equivalent to that occurring in summer provided ambient

conditions which result in maximum ventilation requirements and very small

heat transmission losses for -his location. Test results indicated that with

average August weather prevailing, a minimum ventilation rate of 13 cfm per

occupant is required to keep the shelter effective temperature from exceeding

85°F. With simulated well water cooled inlet air (constant 75°F dry-bulb tem-

perature and 95-100 per cent relative humidity), a minimum ventilaticn rate

of 9 cfm per occupant would be needed.

Heat balance calculations made from heat meter data sometimes showed

unbalance.., ioe., the sum of the heat transmission loss and the heat loss to

the exhaust air was often less than the total heat input to the shelter.

This discrepancy resulted from errcneous measurements by the heat flow trans-

ducers, and consequentlyý no accurate analysis regarding heat dissipation

and transfer rates through the shelter boundaries and surrounding soil could

be made (overall shelter heat transmission could be estimated indirectly).

Table 2 presents a tabulation of shelter boundary surface construction features.

9

Table 2

Houston Shelter Bc.,nid.ry Surface Properties.

Surface Interior Constructiorn Ar f I__ght_ U___or efti)- (lb/ft") -(Btu/Yr-ft"--P

Exterior

-,. NORTH Ext 12" concrete f1454 ulh 0.61

Belowýgradc JIEAST and

I NORTH above- E•xt 12" ccncrete 230 V.4I0 0.53grade walls

SOLTH wall In. ?" x -" studs with 6431 0,071/2" Celotez boardand nominal 3"aluminirn foiled fiber.glasit insulation

Ceiling Int 8" concrete 3867 93 0.52

Floor Ext 6" concrece 3867 7Q 0.71

*U (overall hee.a transfer coefficient) calculated from ASHRAE rcide data,

1IIIII

I ____ l_10 _ -- __- -_ -- _-_ __

t.2 Chlca,'o Aboveground Test

A•rbient air forced ventilation tests were ccxidu-,ed in an abovegrcund

fallout shelter located in Chicago, illincis •ef_ iB) during a deriod o• riild

winter weather in Masch of 1963. Sheller bounoary stu face properties are sum-

marized in Table 3. The shel~tr was tested at ventilation rates of 3, 2, and

zero c-fm per occapant. These tests indicate tnat a minim.,.m ventilattion rate of

3 cfrn per occu.pant was more than adequate to meet Toherma~l environM.ental re-

quirements while the 2 cfm per occupant rentilation rate was 'cund to be

inadeq7_, te in _Lmiting the maximum shelter effective temper;tut're to below 850F.

During the 3 cfm per occupant test (b• days) the shtlter dry-bulb temperature

range was 82"F to 74'F and the effective tempet.atore range was h76F to 68>F

with an average effective temperattre of 73'F. Average etfeetive temperatures

for the 2 cfm per occupant (3-1/2 days) and the zero ventilation (3 days)

tests were 85°F and 806F, respectively. During the latter two tests, ccriden-

sation occurred on all surfaces of the shelte-.

Heat meter measurements showed that approi-imately '0 per ceat Df the total

input energy was lost through the shelter boundary surfaces, hiri2 si.4.•ficantly

reducea the ventilation rate requirements. E3wever, in si•icai summer weather

with higher ambient temperatures, any surface heat Zrans•iassion would oe con-

sLderably smaller, and ventilation requirements nearer tn•e adiabatic) Ta.Le

wou-ld be necessaiy. Alth:ugh adjacent shtlter spaces we••_ heated to the sgme

*Adiabatic refers to a simplified shelter model which neglects any heat trans-

mission through the shelter boundary surf.ces ana assumts all metabolic andinternally generated energy is removed by the ventilstin- air (Ref. 5).

iI,

?Table 3

S, ChiLcago Shelter Boundary Surface -Propertites

Surfa':e2 Surface* f Are- Weigh U *Construction (f (lb/ft (Btu/hr-ft2 - °F)

S .ICeilin f 6" reii'forced 3344 70 0.59j j ~~ znfrete - -

6" reinfrý:zed 3344 70 0.43I cor--ete

NOR¶[ l'," c¢-i•on 7L. 130 0.294al! brickf

SOb2H 13" ý-ommor 7-41 130 0.29wall bric"C

EAST 813" common 42 9 180 0.22

wall brick

ITotal 9028-_ _ .. _ . -r • . . . . _

--J* w1ndow area insignificant

J- U (overall heat transfer coefficient) calculated from dae-a in ASHRAE Guide.

SIi1

dry-bulb temperature as the shelter, about 0O-60 pEr ctf -ctal heat

transfer was through adjacent area surfaces ,floDr and -eilin:). This indicates

that in winter an internal surface canrnot be assumeJ to Le adiabatic even if the

temperature differential across it is reduced to small proportions. This effect

is probably due to slab edge heat losses

Air infiltration tests illustrated the effect of wind velocities and sur-

face quality on infiltration rates. For this shelter (32,000 cubic feet),

with average wind velocities of 22 and 12 mph, the infiltration through the

vialls amounted to 0,10 and 0.08 air changes per hour (5. and 42 cfm),

respectively.

2.3 Milwaukee Aboveground Test

Ventilation tests of a 2hO-man aboveground fallout shelter (Ref. 1C) were

perfor-med with conditioned and widh ambient air supplied to the corridor-type

shelter at rates varying from 3 to 15 cfm per occupant. The "H"-shaped shelter

was supplied with ventilation air at either the geometric center (split-path

system; or at the extreme end tf the corridor single-pass system). Better

air distributioin was obtained with the center supply system than with the

single-pass system. The ambient air test series indicated that some type of

air distribution system is required for a complex aboveground shelter. The

temiperature distrilhtion in the shelter showed that approximately 70 per cent

of the forced ventilation air supply to the shelter exfilti ed prematurely

from the shelter for bcotb air distribution systems and did not leave the

5ýhelter at tht exhaust conditions. Poor quality shelter surfaces and window

fittings were r-sponsitie for this large forcea exfiltration rate.

13

For all ambient tests tU transmission losses from the shelter averaged

from 15 pci cent to 19 per cent cf the total energy inpiit 4 cool sammer weather

prevailingI Lue tc the non-uniformity of temperature within the shelter

resulting from the corridor layout, variations in wall, floor and ceiling con-

struct-lon, and irregularities of wall surfaces caused by large windows, door

and stairwell openings, a meaningful analysis could not be obtained from any

heat meter data,

A summary of significant test parang, ters and results are presented in

Table 4.

2.4 Milwaukee zB.sement Test

Forced ventilation test' performed on a part.al!?f-belowground fallout

sheltei in Milwaukee, Wisconsin .Ref. iD) during Augast of 1963 indicated that

m6 zfm per occupant of ambient ventilating air would be necessary on a Mii-

waukee 5 per cent summer design day- to maintain this shelter at 85'F effective

"temperature When ventilated with hot, moderately humid air, a ventilation

rate of 13 cfm per occupant was needed to limit the shelter to these same design

conditions. Bo significant differences in resultant average shelter conditions

ewere noted when the .shelter was ventilated with cyclic or constant (average

of zyclic ) inlet air co-nditions Tvbles 5 and 6 tabulate significant test

parameters and results, and jivc descriptions of shelter boundary surface.

The energy lost to the below-grade soil backed shelter dur-ing the

*Five per cent summer design day refers to the wet and dry-bulb temperatures,assumed coincident, whach would be exceeded during only five per cent of thetotal summer hours (Ref 6).

:1E-4

.4-'0 *,O 0 4 SC 0 UC D 0C)0 --2C4-)Ul\'.0 C- 0;c . ut \C j o H H- - CuO 0~

H.~4 Cu rH r- H 4 H 4 H H H H 'H

0

~~~C 0 UC 0t- C-'- Cu Od 0 ý U\ ý Lt ý

Cuj H- H Cuj H H H- H- H H H C\u H C\u V

0 4)

rZ4 4-) &

4- >ý%-. !t~ Lr'0 9 c : - Cu CU Cu- tO.14.`4 E-4 HO4 0 H, co r 0 0 0 t~--'\ t~- CjU" M Co M C\j C:-0

c~~~~~ rlr 14 r- 4 -r4 - 00u___ __ __d 4)k

0 H0

% ~ 0 0 .'q43 4) *>) CO 0 LC\ )\ON '%. t H (Y) Or r\ 0~ CUJ 0t v .r OH)

4 06 Cuý HA H H H HCd P-4-dqP Cu t-: Uý -ý Cr) 4-) H ~ ~ U~(d p C~j -4 ý C~j r-, -4 ý r-4 r4 r4 c

04) 4 .

.rC 4 OS

4-) 0 0

a) -4 E-i c --4 r'1\ H V Lf\C cu I _4 C1;CA)f~ Cu (A) Ar Cu Cu Cu Cr) 04~ow 0 S (a

-P .rC 0 02(a -r .0 0.

Hd 0 ___ -ý -"--- -I

H~., .5 ~ .

tx 0 0 4) 4) k $&a 0 d.. 4r0 4)I 14

4)34ýQ4) 4) 00l 03~z 0 r4H-

) w0 000 V..d O4) 4S~ \.o Cu mr t-~ 00 L\ L(\ t-- cn -: Cui ou (n) n( 0 M.j~ 00 0 .-

OD 4 C) t t- cUl\S

o Idu uL HV rO H, 4. 0 CuI Cu 0' N NO C\ 'v ~0 *ý - * CA ; CA *ý Cr)\0~O. COn c'.- -L C

Cm Cu

r0 N -4 N - I I I I I o

) H mu H Cu4 H4 Cu4 HNu 4) 4) 1) 4)W ' > Hd P-4 H C H r-4 CuI I 1 a I If N Q N' Nn o EA 4

15

.4-

Q) Zo H -HH C\JH RH r'JJ H- r- H- r4 (MJ C C\J N' CMJ CVJ-P 4)-.

a)I P

Is\~ El- 0) L'- a\ 0\ OD CkJ Cr) 7\Yjt 0 LE- 0 H- CrI) tN L- 0*-rf -iE-4

a -4 p IH C;\ 04 _: . rý UN- J o C*\ (7\ t-'-\,O D' \0'. ir*\ c>- t Cr*) 0'\ _14 Cd p rt ri H- H- H- H- H- Hq H- H- -

P <

a) r4H - t- \0 C'1) \0 Cr) \0 \.0 t 0 Lt-' \1 () H- --J CrI) 0 _:f G\ w'H -.r4 E-1 r=40 asxo 0 6 s - L\ 4 C -4 O I I

r1) ..I p A 4EA-4 c.' .~ L* C -1 ' C i C ; 6 r£i\ 12 *r j or -C r) H -CtJ 0. r-4 H H 0

0 V)

r-q 4-Dr H H

0

0) H) H H- .- .. H .\ .\ .

4-1 p 1 o l - 00 - - n -O -t m\. '.0 LD 10L I rC) 0) xj E- . . . . . . .)0C) -r4 \0 -) N r-± i t(M r-4 C r-l -iN -IN r-4 r-4 0 r-fl 1 C) '.4 H 0

0_ - 0 0 - -0 - 0 0 a)

0)=4 0 C

0U 0 0 1\

EO r~ Xi' -Tc ')a)OI ~co

a r

00

4) 4-) 0 10 0-Lr

f4 11 - -4-'or) t-4 0 Cd

CH 0

~0 4) T, -*

-PI PP k

a ~ ~ ~ r -P Q 'j00 0-0) 41- v= C+ 4- P4' 0 pr- d- ý4 -)24 )ý HO OC C x k 9- r4)

'0) (1 C: -D:c+0 V. co~- Ci

Q)4 r.- 10 -4 0c 4-1002I 0 0 )++4l X

4)- H iC -:') + 0) k) 5-i 4-)r *-I 0lCd e. -41\a P C a o Z) .Z 0.i- Q a- +

Q' 4 +f pi -P) 0 : -H) 0. 0,- W C.He-40 .4) ;

rq 0 r-jrr a) tr JC -4c 0 V.4 H 0v - -- Vp+ V 0 4 $r-I Z:4 C) t- j 0" 1 0

U~C kG) H"- HOfd C' )4 ( 4 c A -) 0r p) iý4 W

0d z 4-)0CD r- Z)S C).4 .*P L k 94)3- Ea a t-I.~~ e -.- +U 0S p 03 H0 O -PW

.rqH4a +aCd :c -IC0 H02 -P ' HO\ CUH 'C)HH 4-3 a)

Ea; jC r I :4 HQ Ha CH

4-) + biI r- riI V,01 Cd

H ' 0- -0 .- -PI 0 41Hi 0 L0 -- 40 0H '0 -PH a! U4

CQ) 0 - V- * - -,c ul 3= r c -O

-O a

4-r2)

entire test was 3.9 Btu per hour-ft2 or 12 per cent of the total energy input

to the shelter. No significant reduction of this heat transfer was observed

as the tests progressed, The wide variations in flow rates and inlet conditions

tended to mask any large reduction in heat transfer which may have occurred

as the temperatures of the surfaces and the surrounding zoil increased. The

heat transfer to soil-backed surfaces was shown to be proportional to the

j• shelter minus earth temperature differential, and the heat transfer to "non-

occupied" adjacent areas was somewhat proportional to the shelter minus ambieiit

- temperature difference The overall heat transfer coefficient calculated from

b heat meter and temperatare differential data was 0 23 Btu per Xr-ft 2- 0 F, andI-

is approximately one-half the overall coefficient calculated from the ASHRAS

j• Guide of 0.43 (Pef, 6).

2.5 Wilmington Aboveground Test

Ventilation testb at various fluw rates and inLet conditions were per-

Sformed on an aboveground fallout shelter in Wilmington, North Carolina (Ref. 1E)

dzuring UC7rtober of 1963 f' see Table 7). (Infiltration tests warc also run, but

the results .• r., data have little significance other than to illustrate that

infiltration Y~t -is d-- -n wind speed and `ireczion ) For this shelteae,

a ventilation rate -f approY'ixnteI_- z: 'n per :ccupant- is required to tiS4t

its effective temperaturt to 85"F when 1nta , -: W; -I.at. ?-1/2 pe"

, cent summer design air. Fromi these tes+z, • pr-cedure wade-ve-1ed zo pre-

L dict shelter ventilation reauicement! for e)v auovegrcnid shelter in any

locac±on (Ref. 5' Using this proczdure and comparing the resu)-.Iz -with

adiabatic rates, predicted ventilation rates assurming heat loeses were I to

j_ 2 cfm. per cc..pt lower than those which w,-r computed adiabat-ica'.i1y.

L.L -z j ,. -

F-4 4-I P14I0 00 cu lQ C6C' CUý CU C';

a)~~~ ~ ~ ~ ~ ~ %-U N M CV Uu CU :ýr4rl . Q cu CU

4-)z r~-l M \D C7\ r~-4 -0 O 0 C Lr\ U-\

*Q r

P,*10

two\%~ Hr 0-cc

HHr- -4 C.0 faO 0L- U 0 ~ H C

H 4L1 CU 10C~c U C U~Y ~ (~t\\ ~ C 4.-

Z1--- --

V~~* Q) r- -1t-(

E! 4-) rl

4- 1 \ I '

(D4 4- )-iEiF 4 . . . .. .I- . 04 ;-o WZ .rt j') 'j" m? C~) cc UN%

Erii~ ~2E-4 U.\ A~)JL~ czoc~c

-4E-4 o UN Lr\ -\ -k U\ Lr\ ,Cy y, lD bD

0'~ HC; 0UC 01 CAi~ C JCN (t

Cý 1_ - l- co CC '9 Y

~-.J tv0

- r4-

Heat fluxes through exterior surfaces (4.9 Btu per hr-ft 2) were over

twice as large as those through interior surfaces (2.4 Btu per ar--ft 2 ) due to

the greater temperaturt differential across the exterior surfaces. However,

only 8 per cent of the total input energy to the shelter was transferred

T through exteiior surfaces whereas the larger interior surfaces transferred

approximately 6wice as much total energy. Under design conditions, when adjacent

interior areas would be occupied, the only energy transmri ted -hrough the

shelter boundary surfaces would be transferred to the external surfaces re-

sulting in an in3lgnificant reduction in required ventilation rate. Con-

Ssequently, for this and similar shelters the adiabatic ventilation riate is

recommended .

I The experimental average heat tra nrsfer coefficient for exterior surfaces

of 0.h4 Bta per hu-ft.- 0 F agreed very well with the ASH!RAE Cuide (Ref. 6)

valae of 0.45. The experimental 2cefficient for the entire sheltcr of 0.17,

however; was somewhat less tnan that given by the ASHRAE Guide 0.2-. This

deviation is probab~y caused by the fact that the experimental value is based

upon the shelter minus ambient tempera-ture differential instead of an area-

weighted average temperaturc difference. The latter could io' !e calcul5te

beo?,use the temperatures of all adjacent area spaces were not ine-%surec..

Table 3 presents a tabulation of shelter Zoundary surface constructiun features.

2.6 W.lmiLgtcn Belowground Test

Forced ventilaticn, tects were perforated on a belowground fallout shelter

in Wilmington, North Carolina (Ref. U) using zi-.ilated occupants, Table 9

presents a suimary of tests performed. For P. Wiia::ington 5 per cc.nt summer

:20

Co;

cu4-)

to. U 0 0 0 co -

cd t- LI- cq \0 .r OH H

___I C r- -4 r:Ir-

4,oOP I

;4 - -c- -0 _:-4~

0) CjY4 CO

CH d Ea0 0 CdOH(1 CY2 CU' 41.0

Cd z _ _G\_ _ rIc

C02 0' r00 CN '.\ 0 0 04

Cod co) H0

CC ci) - -__ __4

10 C3

Hj 0, Hu HC - - -Ir )I0 .rq Hr-..- Cd H- riC *r4-P4 .00

P 4,ýLr Cd 4, 0 4-> r-4V 0U+ r4 H

>i -1 -P I4C ) Ca Ir0O - ý P rI 4- U.I 4P ;4, H , -P a H- ai)

94 (' r-4UC 0 kO U t )(a) z rI H1z k., 0- r-4' H Cd Q) a0)

0 -k rIHCD :3 ) &a ' Cd ý> r-01 c.) P0 0P~ 0, O 0cdUQVO0 4-

0 (1) rq ('2 0, H- "1- HO .- 0 0 0'4,> z a) cs Cd H:4 = ' HoUI JP4 4, 0Cd LIS0

Cd 0 PC0O 0 V 0

- 2 O H a) ýQ = )cs U- -P (Mco0 a

0_ _ ____ 9_ 0 -- ) -q 9 k

_j ___ ____ V2 to _t+p \0 + LrCccd a P o

44

Sx

C§

cd 0CE-1 r I, H 9- H oq o-

21

f 4 cd5 o C% CACul

o a)

4-) r=4

__ __ _ - - -

IS" . 4 .

to0

a): Lr\ IUfF \tLAN fl 0'

00

E)-4* -4 - -~~ed 0) 0

V- Cu

0 A0

0'l 0' .f C

CdC

Q1 PL0 4)r

___-_ -- -____4

to be adequate i•o maintain this shelter at 85F effective temperature. When

the shelter was ventilated witn dehumidified aix "63'F dry-bulb and 59°F wet-

bulb temperatures), 5 Pfm per occupant waz su f~cient, W-hen the ventilation

was stopped completely, the snelter effective temperature increased from

85°F to 90OF in one hour, to 92OF in two hours and to 930F in three hours.

The earth temperatures during the first six days of testing increased approxi-

materly 5 to 60 F. This temperature rise resulted in a slight decrease in heat

transmission losses through belowground shelter surfaces, which over the0

entire test period ranged from 1.6 to 6.9 Btu per hr-ft and averaged 3.7 Btu

per hr-ft . The latter "ralue agreed fairly well with the heat loss through

these surfaces predicted by procedures in the AS1IAE Guide of 4.5.

it was noted that this heat flow consistentl:y peaked between 1200 and

1900 hours, and was a minimum between midnight and 0500 hours. Heat losses

through adjacent interior sarfaces was slightly less than the losses through

soil-backed surfaces but may be absent in emergency situations, Heat trans-

ferred from the shelter to the surrounding soil was 14.5 per cent of the total

energy input to the shelter an.d significantly reduced the required ventilation

rate. The overall heat transfer coefficient calculated fprom neat meter data;

an area-weighted average AT, and the total area of the shelter was 0.43 Btu

Der hr-t-tF, which is less than the U value of 0.56 determined from the

ASHiAE Guide. Table 10 lists shelter surface paoperties.

23

r. c~ 'oII L- . -4. 0 o H,~ -r

I ' - C \I Cl)

0

0 *P4 4,

0- (Y)

4-.-) P- ~j c

0 '

0 C)

L I - ) Y0)

4 P4

0 14ria \ 0'-'I +1C) Q .)Cd k)

U V 9 -i '~ C)c~ '.1 H -

0 r-4 ca) 0)4 -PH-cPSHP aj~ -I CH0:: --~ ' 0 '0 - ,4\ 0Q)t

P4 04 P4 - 0 4)--'J F-I 94L I 0)0z CC )

cuJ-P oJ C11 -0 -P-P N Q P4 0

4-)

Ej -4 ,E.. 0i .3-4

C/)4 0

C,

2.7 Bozeman Above,- rourd Test

An eighteen-day, 40-man simulatej occan.y t•s- was 3r';ried on a

second floor fallo7.ut zhslter in Bozeman, i._ontaa CL. 1G, durin. j.e o0

1964, The shelter was ventilated with outside air at flow rates from 3 to 11

cfim per occupant. The energy loss through tLe Lhelter boundary s,&uI*Cces

amounted to 17,4 per cent of the total ener-- input to the shelter. The energy

loss through the interior surfaces kll.9 pex cent) was twice as large as the

exterior suiface loss 5.-5 p.=i cent) For the average June dTy, which approxi-

mated actual test weather, the shelter when loaded with one occupant per ten

squarE 'e:t of floor area, would require L.5 cfm per occupant to limit the

shelter average effective temperature to 85c'F, This is an 108 per cent reduction

from the •diauatic ratE :f 5 5 cmn, per occupant, For the hottest 3one day on

re2ord, the shelter would re4uire 7.0 cfm per occupant, a 7 per cent reduction

from the aaia.ttic rato of 7.5 cfm per occupant. The above rates are based

on the assumption tha; the floors adjacent to tve shelTer were unoccupied.

Under emergency conditions, this would probably not be the ,ase. Heat trans-

mission losses from the shelter would, therefora, be reduced, and the corres-

ponding reduction in air flow rate would be less than 0.5 cfm per occupant.

Consequently, the adiabatic model is recommended fcr selecting ventilation

req,-irements for t-his and similar shelters.

The avterage hea-t transfer coefficient fo i the shelter given by the

ASfHMI, Guide is 04.5 Bta per hr-ft 2-F. This compares favorably with the

calculated oxperimental coefficient of O.O, which is based on an area-

weighted ;sioal! average AT.

25

This tes~t i.rovided shelter and w-,ather data for a com uanion etaudy vnich

mathematically predicted hourly s.:elter psychrometric conditiv±,s. Thcze data

¶ included hoarly weather conditior-s, shelter ecCnstr uvtiori ann ,nfiguation

details, and aijacent irterlor arca temperatures. .e...ted shelte• d'ry-bulb

aand effectivw• temperatures were generally within a'F and ).F of the experimental

sie±lter dry-1,;ib and effvutive tempert res, iespet:rively, when actual adjacent

minerior temperatures were used in the calculatLons. When the temperatures

r o! adjacent non-shelter areas were cstiniated the ýgrezment was generally

within 2*F for both shelter dry-bulb and effectivr, temperatures- It was con-

cluded from these results +-haZ without a more ac.'urare interior temperature

estimation technique, farther refinIement of this anal.ytical .iritbod would not

L- be possible,

A sturmary of significawt test parameter; and reosilts are presented in

Table ii 'Table 12 lists shelter surface constmrw .ion characteristics and

E_ physical properties

2.8 Athens .11nan Occupancy Test

A fifty-hour, 30-btuman occupant forced ventilation test was conducted

in a basement fallout sli~ter in Athens, Georgia (Ref. mH) in conjunction with

the University of Georga. , the test, ventilation ar-r was suoplied to

the shelter by means of three different veritilation- systeems. In the first

portion of the test, thu installed shelter ventilation system was used and

fouad to be ade4uate to maintain habitaŽle shelter c, -,dit .mns if the central

i;efrigeration plant is operating. Following this, two sirr'lified ventilation

L systems constructed of plastic (polyethylene) duct wvas evaluated. One system

I

L26

- _ .o - -.- _ •••, - - •---- - • --- ...- a--•- ----- • --•

1 I r fý0 0 tNU\U- r j%0 0 -

4-)

C.) ( >30 4 0 Lt\o 00 C NL\LOtt\O t1 \P 0 0 I

CdQ 40 O OLC\0ul\ 0no0u\L\ LC\0O0 r t\00

., -() HE-i-..--. ..-'. .

-H r-4 N F4 cy-a _:t mWco .-0 y ~JC'I\ CI 00 L\; rr4AcOO

0) H) P -r-H)4' E-1'~\ 0 L\W 0 WLfQ

- -4 - - - --- 4

OO O O O C Q 4-.C0 0

> Ca2

o 0

S.,4 A. Y ýn

Y)-T r-q ' C\j "7 0 L)-Ot r\~j e i

cHi

M lciys.ý ~0 Lr- r\\O\trn_ MC\00O00c, L\-tCV

a- ~~~co C.) 0 Oz co co C) co(OýOco ~CO:

E-4 E-1

rA m -t trU \D L'-X c C\ 0 H ~ m-: Ur\ \'O Lt- 0 ONO0

27

0I4-1o

co 0)D

0 00 000

r- I- - oc , ,

H~ IV c'

r

z .-4 .-

E-i 0 + + II -

04 0' aD oHo:0 43 +~44 Hi0( rI0

It .) a0 kIk-P La z4,-

10IS ZP a4) rj

A. =o a;P Q)-4 04- Q24 H

0H0 r-4)C\DJ>100Hct v F. 0> 4 k =

Eý rI = i C) a ciz 0 L, - 0 W W r

o -:I r40 c *\

r-4 4,, P4 c :,4- o+c

Hr Irzq 444)4

U) 0

t-o cabý>3.r 0 0- , d rI., a fI(IC

0 Z4 0E-4r. CE-1 4 E-E4&- V.- r- H 0ri Zd 4) :rJ~H C:'O -H I 0L C

Q -P -- C

C- -r-

was a snlintarag etinwchail oýf t"e ve~ntilat I )Y nl nzppnlip

at one end of thL shelter and exhausted at the opposite end. The otner arrange-

ment was a distribution-type system, i.e,, air inlet holes were cut into the

plastic duct at uniformly spaced intervals The single-.inlet system provided

adequate air mixing witr a single room, with the stipulation that the air

inlet was designed to eliminate local drafts. The more conventional distribution-

type of system should be used to supply air to partitioned areas, but is an

unnecessary luxury in a single-room shelter. A minimum thickness of 4 mils is

recommended for the plastic air duct to provide adequate resistance to damage.

2.9 Providence Basement Test

A 14-day am~bient air test was run in a partially belowground fallout

shelter in Providence, Rhode Island (Ref. 2) during August of 19 64 at a con-

stant ventilation rate of 8.5 cfm per occupant. Ventilation air was uniformnly

distributed in the shelter by a plastic duct at floor level. During the

test, almost no temperature stratification in the shelter was evident. Sur-

rounding soil temperatures increased as the test progressed, and the temper-

ature rise was greater for soil nearer the wall and caused the total shelter

heat loss rate to decreasz slightly during the test. Transmission losses to

aboveground exterior surfaces were most sensitive to changes in ambient air

conditicns.

Approximately 23 per cent of the total input energy tCo the shelter was

lost through the shelter boundary surfaces. This accounted for a 30 per

cent reduction from the computed adiabatic ventilation rate. Roughly one-

half cf this total heat transmission was conducted thruugh aboveground

29

exterior boundaries and one-third through soil-backed surface_. However. it

is concluded that under design-day conditions, the most impcrtant heat sink

would be the soil-backed exterior surfaces. The heat flux. through these sur-i2

faces was smaller (2.7 Btu per hr-tF) than observed during previous GAPD

shelter tests.

The experimental overall beat transfer coefficient computed from an area-

weighted average b T was 0.25 Btu p~r hr-f't-OF, which is quite different

from the ASHRAE Guide value of 0.38. It is believed that this disagreement

was due to transient heat transfer effects, and the inability to properly

measure the temperature of 'the outside surface of exterior belowground sur-

faces. Physical properties of the shelter boundary surfaces are showm in Table

L 13; a summary of daily test averages is presented in Table 14.

The empirical data from this test was used in conjunction with a computer

program to analytically predict shelter psychromezric conditions entirely

without the knowledge of experimental shelter and adjacent interior area tem-

perature dpta. input information supplied to the computer program consistea

of measured hourly inlet air wet and dry-bulb temperatures, estimated cloud

cover, day of year and latitude of the shelter location. Besides this weather

data, other inputs included ventilation rate, number of occupants, instrunen-

tation load and various shelter construction and configuration details. The

mathematical predictions (both dry-bulb and effective temperatures) were

generally within 15°F of the experimental results. It was therefore con-

J cluded that this analytical method can be used to accurately predict shelter

psychrometric conditions for given inlet conditions for this type of shelter.

I

U 30

I I C\j cu 00 M- enCY0 0 0 0; 0 0 C

-- C ~ : ~ rIC)OC~C,

a) Cd

-14 0- \.0 _:t -D .:t Q1 4 CIO _tS C-) .rq Y 4 CJ WO (1)a)0

H- k- E-1 i~a)

W) 4-)V

1 ceJ- 00 UN ar 0

PP 0)

W) (D) C)

oo 43 0) 0)k) 4-)V 4'0) C-0) 0)4 0)) a) C14 a) F AWa

-P -rj ý4' * 0 ~4-' 0d C l 430 >C) CO>'O ' C 4-')- 00 ;4 0 0) o-~ 00 ; .Ia) c S:: d ~0) V

C) U .0 C) ~ 0) Oc C.)

= r r. -.,- 0 - - 0)

0 D00 \00 0 0 - 0 kC _I ~ 0

CH'0 0 0 0 0 k) 0 0)kH k0 4H 0 k-~ 0 ' Hb)b W)~ X i tO 0 0 .,

a) W e0) 0)i- )-P 0-~ 0 (1)H 0) H a) 03

4-) rJ - - P a

31

r-

LA LA 0 .A .A .A .A 0co UN co L AO H'fl O\\ r4(TC s ;Cý LA:

0 4a

P4 ., HE-iI, - LALA 0 LAO 0 tr\ LA LrO LAO u V

4:: 4 A <C~j r-I Hj H- H4r4 C~J H- H- CUj

bD ) -i -iLA\ 0 0 LA\ in~ 0 0 0 0 0 LA\ 0 0 0

a, c ____ r- \ z- 0-O -cONL\N\0L- -f

a) a) 'H-I - LALALAOO 0 O LALAOO l LALAE-i E-1j~ r-4 \\L-H JCIO HI

Z' H d H H H0 \0 H \H E- H4 C

W a1 a -l 0 C

4) ~-P4 pq o m m \. CUj H- c) CUj A~~1 A CU ý t-

a) .C/ p --

ror

0)-P~ 00 AA LA LA\ LAý L LA ILA

01 F4 14 Cd

4S 'al L f\-o o LA LAý U-\ N U-\L, U :-\O U-\) 0 0

0) tý . \3 L- -\ f L L- CF \- %1 -, -ý

GC3O 0E-4 ci

0)'D 1 1 x 0\- WH n H :IH

32

SfA4R Y FT FEET

A tabulated summarv of important test rl-,ultzt, a. taken from the interim

reports, is presente-d in Table 15. Si4nif acant rezults ar - summarized as

follows:

1) During all aboveground tests, the shtlter always lost energy

by transmission tLrough the boundary surfaces However, in an emergency

occurred on a hot summer day, this heat loss cannot be expected to occur

because: (a) the tests were conductAd during a period when cooler than

design summer weather prevaiJed, thus increasing the heat loss through the

external surfaces, and/or ib; the building areas adjacent to the shelter

were unoccupied .and thus co~ler than the sheltei area); during an emergency

these areas would usually be occupied, eliminating this heat sink.

2) Dulng most teiowground tests, the shelter lost energy to concrete

soil-backed surfaces, as indicated below in Table 16.

Table 16

Belowground Heat Losses

Test 9eat loss "o Belowgrade Initial SoilExterior Shelter Surfaces Temperatures

Btu/hr-ft2 Or

MilwaukeeBasement 4.O6

WilmingtonBelowground 3 7 70

ProvidenceBasement 2.7 73

33

T __

F ____

17� -'--i

Zr -I 0 -. 0

I'--' c�

0 .. ' � CV

* 4' .4 t\ -w

31 - - - - -.-- -

4'-' * -4

= . �0 4' I -

I K __

jr � ___

�&-.-� .'.' _____ CV fifS. I

'.4. CCV I- S. C'- �- -�2 S. C'-

- _____ - -- - -

1' 4-�;:. U IS..)�E�

* - I. I. IIV II

''I 'CS.... C C' O4.�LI CCCCS.CC Cl)

CCCC.C 4 ... > 4 4 C IC

'SC�-�. QIC) C) S.C C2 2 C)

Ca C24. � - -' --

Cl�!� us? C�n z, .r.L

�, I T1I. U CV

CC 4:

C4 - -*- r

____ �

___

(-'C

__�±.� 2'2 '0

.. � ____ :'�;' UOS. 4L 0 C

.44 V- -c S.- 4 C� �C

HCC 54 -- -� .C-� -- 4

________ � A:I-. o�

- - � -�

-- -4 -

The soil s'-Lrrounding the Milwaukee shclter was silt and -7a", while the other

two shelters had a dry 5a.id and gravel f'll mixture. It is suspected (data not

sufficientlY accurate) that the heat loss duxing the Houston test approached

zero, as the soil temperature exceeded 90CF As the above four tests were con-

ducted during the summcr, it is reasonable to assum• That these heat losses

vrould also exist during a summer design day emergency, since scbsurface soil

temperatures show little diurnal variations.

3) The shelter diurnal dry-bu3 b nd ?ffective temperature cycles clos&ly

follow the a.bjent cycles, but at diminished amplitudes. For aboveground

shelters, there is a fair coirelation between "he ratio of tnr aimplit:des and

the air flow rates (see Figs. 3 and 4). For belowground sheliters., no similar

correlation could oe made, since the shelter cycle ic apparently a stronger

function of the heat storage mass exposea to the shelter air.

4) Possibly the most significant results of this test series is the

verification of the mathematical shelter model (Ref. 2). Dry-bulb and effective

temperatures can almost without exception be predicted to within 20F, usually

much closer.

5) Large open shelter areas will n-ot need elaborate conventional-type

distribution systems. (Since aggregate Simocs inherently disturb the air cir-

culation pattern within a shelter, only one test (Athens) provided an oppor-

tunity to evaluate air distribution.)

35

42

OR ' Pa LIS ~. RA t F4TIS VL tJ'.U V !!

0 -

1-

0.0

1.s, wi:.'1o w " sd z.a.... t.....

Fig. 4 ABOVEGIROUND TESTS - EFFECIVE TEMM'PATURE CORREIATION

______ ________________ __ __ __ ___36_

SECTION L

CONCUJSIONS AND RECOMEIDATIONS

The temperature and humidity that will develop in a shelter are determined

by the state of heat and moisture balance at any time. The energy quanta enter

or leave by the ventilating air, may be generated by metabolism or equipment

in the shelter, or may be transferred by convection, conduction, radiation and

eondensation at the shelter boandary surfaces. A comprehensive analysis of

these transient heat and m'isture flows can be performed for any shelter if

the extensive data are available. Computer programs ha- been developed which

numerically treat many aspects of this comprehensive-model. The major problem

in the application of these models to shelters in general, is the amount of

detailed input information required to analyze any given shelter.

Only a small percentage of the total metabolism energy generated within

most laige shelters will be lost by heat transfer to the shelter walls during

hot summer weather. This is especially true for the second week in a below..

ground shelter and for every day in an aboveground shelter. Therefore, it is

possible to obtain reasonable estimates of the shelter conditions developed at

various ventilation rates during hot weather by neglecting the wall heat loss

or gain. Reference 5 presents the details of a graphical method to make such

predictions.

The results of such solutions are displayed in Figures 5, 6, and 7 for average

shelter effective temperatures of 80, 82, and 85°F. These solutions are based

upon the metabolic heat load of sedentary people and do not include any other

37

NI

I.,i -" I -i P

it i'

zK e

:34f

t4~ L- ~ ~ 'I~Ji}Pf

0 0

ILI

r:0

I. -

tJit[T 1,

z *g7E t

ý38

-~ -

tlot

'-44

S0

4 If ! I ' "

e I E-iE- 4) 0-L_ lC

cccIvl 00

o-

CO Z 2-f

6-/4& A

co

0~

IfY'

.1-I t

rz

3

U0

0 E-1

E39

X \D

IfI,,Il 0, 4,f I I

44 0r .""10 a 00

Ofi4j 4.,~-~~Žt~ 4

I rIN

0I co

11142 1H

0

I--Coll

IXCAIq~3ZO/n

Ir "0aitwa 101W sIl

4o1

heanft 1cAds- These •hs.art,- may be "seEd to Iei.ermin& the minimum re-,iireI shelter

ventilation rate pr shelter occupant as follows:

1. Select the limiting shelter effective temperature; 80, 82, or 850F.

2. Locate the point on the psychrometric chart which reprecents a

24-hour average of the inlet air diy-bulb and dew point tempera-

ture (or wet-bilb temperature).

3. Read the required ventilation rate in cubic feet per minute per

shelter occupant.

Neglecting wall heat losses greatly simplifies the determination of

the shelter conditions. Using 24-hour average inlet conditions results in an

overestimate of the 24-houx average shelter conditions, provided that all

energy sources within the shelter are included in the analysis (Ref. 7).

For partially or comp]letely belowgrade shelters, a reduction in required

ventilation rate may be possible because of heat transfer to cool soil-backed

surfaces. Since soil comDosition around identified shelters is almost impos-

sible to predict, this heat sink must not be considered unless reliable soil

information is available.

The shelter diurnal dxy-bulb and effective temperature cycles follow

ambient cycles, at amplitudes from 15 to 50 per cent of the ambient cycle,

depending on the ventilation rate and heat storage mass exposed to the shelter

air. Considerable research is needed to determine the physiological response

to cyclic effective temperature.

Adequate natural ventilation of most aboveground shelters will probably

be possible. This will be the subject of a future report under this contract.

41

md JaRecent GARD analytical sL. r studies have indicated that the adiabetic

-model -may not be the limit-i - Ee. Foi a shelter in a light-constructed

Fbuilding, solar radiation on the shelter walls may during extreme weather

conditions raise the shelter dry-bulb and effective temperatures above those

U predicted by'the adiabatic model. Present work is directed toward further

r definition of this effect.

4

I

I,

I

L•

REFERENCES

i. The following GARD Interim Reports, prepared under Contract OCD-OS-62-134,

Subtask 1214A:

A. Report MRD 1195-52, "Ventilation Test of an Identified Basement

Shelter in Houston, Texas", January 1963.

B. Report MRD 1195-53, "Ventilation Test of a 330-Man Aboveground

Shelter in Chicago, Illinois", June 1963.

C. Report MRD 1195-54-1, "Summer Ventilaticn Test of a Corridor-

Type Fallout Shelter in Milwaukee, Wisconsin", September 1964.

D. Report MRD 1195-54-2, "Summer Ventilation Test of 200-Occupant

Basement Shelter in Milwaukee, Wisconsin". April 19614.

E. Report MRD 1195-55-1, "Ventilation Test of a, 210-Man Aboveground

Fallout Shelter in Wilmington, North Carolina". February 1964.

F. Report KED 1195-55-2. "Ventilation Test of a 200-Man Belowgrade

Fallout Shelter in Wilmington, North Carolina", April 1964.

G. Report MRD 1195-56-1, "Summer Ventilation Test cf an Aboveground

Sheter in Bozeman, Montana", November 1964.

H. Report MRD 1195-57, "Environmental Conditions During a 300-Occupant

Shelter Test in Athens, Georgia". November 1964.

2. GATC Interim Repo-.rt MRD 1268-30, "Ventilation Test of a 500-Man Basement

Fallout Shelter in Providence, Rhode Island", prepared under SRI sub-

contract B-64220(4941 9A-16)-IUS, March 1965.

3. GATC Report MRD 1191-2, "Operation and Maintenance Manual for OCD Shelter

Test Equipment", Volumes 1 and 2, prepared under Cc.ntract OCD-OS-62-99,

December 1963.

43

- ~ 7-

4. Ibid., Volume 3.

5. G. Engholm, "A Simplified Method for Predicting Shelter Ventilation Require-

ments". Environmental Engineering for Fallout Shelters, Office of Civil

Defense, TR-23, June 1G64.

6. "ASHRAE Guide arnd Data Book - Fundamentals and Equipment (3963)", published

by the American Society of Heating, Refrigerating, ana Air-Conditioning

Engineers, New York, N.Y.

7. G. Engholm, "Physiological and Meteorological Aspects of Shelter Ventilation",

a paper presented at the Scientific Working Party of the MATO Civil Defense

Comrittee meeting, June 29-July 2, 19655 Paris, France.

4

F-

L

LI

L 1~

UNCLASSIFIEDSecurity Classification

DOCUMENT CONTROL DATA- R&D(Security claeslfcattein, of titft. body of obetect and IndexlbW annetation, seit be entered w*en aie overatl report to laaalled)

I. ORlGINA IING ACTIVITY (Cooper*94 auder) i20. REPORT SlECURItTY C LASSIlrICA&TION

General American Research Div., GATC• Unclassified

7449 North Natchez Avenue 2, GROUP

Niles, Illinois 606h8,3 REPORT TITLE

E(f.-•RIMENTAL STUDIES OF FALLOUT SHELTER VENTILATION REQUIREMENTS (U)

4. DESCRIPTIVE NOTES (TIp' of repoti &Wd thnooutelv date*)Final Report

5. AUTHOR(S) (Last 0i4e. Brat nine. inttlat)

Behls, H. F.Madson, C. A.

S. REPORNT DATE 7.TTLN.0PGI b o rRpOctober 1965 4ý 7

to. CONTRACT OR GRANT NO. UCD-PS-64'201 98. ORIINATO'w , RPORT NUM9E(S)

(SRI) D-6+220 ýh9+9A -16) -U 1268-40A. PROJaCT NO. 1200

C. Task 1210 Sb.. &rR,• ,,PORT NO(S) (Any .,. n.en, that May be as.e..d.

. Work Unit 1214A

10. A V A IL ASILITY/LIMITATION NOTICES

Distribution of this document is unlimited.

II. SUPPL•MENTARY NOTES 12. SPONSORING MILITARY ACTIVITY

Office of Civil Defense (OCD)Department of the Army (OSA)Watmhington. Tf.f.. •O'flO

IS. ABSTRACT

The results of two years' field testing of fallout shelters arereported herein. Simulated occupants (Simocs) and forced flow conditionedair were used to duplicate emergency environmental conditions. Nine testshave previously been documented in detailed Interim Reports. Based onfield measurements of temperature, humidity and heat flux, and supplementedby an analytical computer program, an "adiabatic" procedure is recommendedto predict shelter environmental conditions. This adiabatic procedureneglects heat transmission through the shelter boundary surfaces and canpredict shelter effective temperatures to within 2*F. The procedure isconservative in that it will overestimate the shelter temperature for allshelters tested.

D D , am *. 1473 UNCLASSIFIEDSecurity Classification

LNC LAS i? iDSecurity Classification

14. LINK A LINK R LINK CKEY WORDS- KEY W ROLE WY ROLIE WT ROLC j WT

COIING AITD %.NT.IIATING ,')."-t '¥'FVITRCIATION

FA:C.)UIT 3HrrTFRS

s]-ELrE9 OCCUPANCYSUJRVIVATJ

INSTI•CTIONS - - -

1. ORIGINATING ACTIVITY: Enter the name and address 10. AVAILABILITY/LIMITATION NOTICES: Enter any lir-of the contractor, subcontractor, grantee, Department of De- itations on further dissemination of the report, other than thosetense activity or othe, organization (corporate author) issuing imposed by security classification, using standard statementsthe report. such as:2a. REPORT SECUIUTY CLASSIFICATION: Enter the over- (1) "Qualified requesters may obtain copies of thisall security classification of the report. Indicate whether report from DDC.""Restricted Data" is included. Marking is to be in accord- rance with appropriate security regulations. (2) "Foreign announcement and dissemination of this2b. GROUP: Automatic downgrading Is specified in DoD Di- report by DDC In not authorized.rective 5200. 10 and Armed Forces Industrial Manual. Enter (3) "U. S. Government agencies may obtain copies ofthe group number. Also. when applicable, show that optional this report directly from DDC. Other qualified DDCmarkings have been used for Group 3 and Group 4 'as author- users shall request throughized.

3. REPORT TITLE: Enter the complete report title In all (4) "1. S. military agencies may obtain copies of thiscapital letters. Titles In all cases should be unclassified. report directly from DDC. Other qualified usersUI a meaningful title cannot be selected without classifica-"tion, show title classification In all capitals in parenthesis shall request throughImmediately following the title. ._,._

4. DESCRIPTIVE NOTES: If appropriate, enter the type of (S) "All distribution of this report is controlled. Qual-report, e.g., interim, progress, summary, annual, or final. Ifled DDC users shall request throughGive the inclusive dates when a specific reporting period Is .,;covered.I

If tfie report has been furnished to the Office of TechnicalS. AUTHOR(S): Enter the name(s) of author(s) as shown on Services, Department of Commerce, for sale to the public, mndl-or in the report. Enter lost name, first name, middle Initial. cate this fact and enter the price, If known.If military, show rank and branch of service. The name ofthe principal author Is an absolute minimum requirement. IL SUPPLEMENTARY NOTES: Use for additional explana-

6. REPORT DATE. .Enter the date of the report as day, tory notes.month. year; or month, year. If more than one date appears 12. SPONSORING MILITARY ACTIVITY: Enter the name ofon the report, use date of publication, Ithe departmental project office or laboratory sponsoring (pay-on te rport us dat ofpublcaton.ind for) the research and development. Include address.7.. TOTAL NUMBER OF PAGES: The total page countshould follow normal pagination procedures. Le., enter the 13. ABSTRACT: Enter an abstract giving a brief and factualnumber of pages containing information. summary of the document indicative of the report, even though

it may also appear elsewhere in the body of the technical re-76. NUMBER OF REFERENCES: Enter the total number of port. If additional space Is required, a continuation sheetreferences cited in the report. shall be attached.

Ba. CONTRACT OR GRANT NUMBER: If appropriate, enter It is highly desirable that the abstract of classified re-the applicable number of the contract or grant under which ports be unclassified. Each paragraph of the abstract shallthe report was written, end with an indication of the military security blsssification8b, Sc, & 8d. PROJECT NUMBER. Enter the appropriate of the information in the paragraph, represented as (71S), (S).military department Identification, such as project number, (C), or (U).subproject number, system numbers, task number, etc. There is no limitation on the length of the cbstract. How-

9a. ORIGINATOR'S REPORT NUMBER(S): T;nter the offi. ever, the suggested length in from 150 to 225 words.cial report number by which the document wil'. be identified 14. KEY WORDS: Key words are technically meaoingful termsand controlled by the originating activity. This number must or short phrases that characterize a report and may be used asbe unique to this report. index entries for cataloging the report. Key words must be9b. OTHER REPORT NUMBER(S): If the report has been selected so that no security classification is required. Iden-assigned any other report numbers (either by the orijinator fiers, such as equipment model designation, trade name, mili.or by the aponisor), also enter this number(s). tary project code name, geographic location, may be used as

key words but will be followed by an indication of technicalcontext. The assignment of links, rules, and weights isoptional.

UNCMASSFIKEDSecurity Claissifi cation