ML20205J697

Evaluation of Chlorine Concentration in Vogtle Electric Generating Plant Control Room. W/13 Oversize Drawings

ML20205J697
Person / Time
Site:	Vogtle
Issue date:	02/14/1986
From:	GEORGIA POWER CO.
To:
Shared Package
ML20205J694	List: ML20205J691 ML20205J697
References
NUDOCS 8602260344
Download: ML20205J697 (48)

Text

. - . . _ _ .

p,

.' 4 b

1. .

\

RVALUATION OF THE CHLORINE CONCENTRATION IN THE 4

VEGP CONTROL ROOM i

I

=,

4 DATE: February 14, 1986 i

8602260344 860219 I PDR ADOCK 05000424 E PDR

l l

TABLE OF CONTENTS  !

F.H*

1.0 . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . 1-1 2.0 MVAC SYSTEM DESCRIPTION . . . . . . . . . . . . . . . . . . . 2-1 2.1 Overview . . . . . . . . . . . . . . . . . . . . . . . 2-1 2.2 Control Room HVAC Systems . . . . . . . . . . . . . . . 2-1 2.2.1 Normal Mode of Operation . . . . . . . . . . . 2-2 2.2.2 Isolation Mode of Operation . . . . . . . . . . 2-3 2.3 Control Building HVAC Systems . . . . . . . . . . ... . 2-5 2.4 Dilution of Chlorine Within the control Building . . . 2-6' 2.4.1 Dilution in Spaces Immediately Adjacent to the Control Room, Level 1 (Mode 1) . . . . . . . . 2-8 2.4.2 Dilution in Cable Spreading Rooms . . . . . . . 2-11 2.4.2.1 Dilutior.in Cable Spreading Rooms, Level A (Mode 2) . . . . . . . . . . 2-12 2.4.2.2 Dilution in Cable Spreading Rooms, Level 2 (Node 3) . . . . . . . . . . 2-13 2.4.3 Dilution in Filter Rooms, Level 3 (Modes 4, 5, 6, and 7) . . . . . . . . . . . . 2-14 2.5 Dilution of Chlorine Within the Control Room Zone . . . 2-15 3.0 CONTROL ROOM INFILTRATION MODEL DESCRIPTION . . . . . . . . . 3-1 3.1 Control Building Chlorine Concentration . . . . . . . . 3-2 3.2 Control Room Chlorine Concentration . . . . . . . . . . 3-7 3.3 Conservatisms . . . . . . . . . . . . . . . . . . . . . 3-14 .

4.0 RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1

5.0 CONCLUSION

S . . . . . . . . . . . . . . . . . . . . . . . . . 5-1

6.0 REFERENCES

. . . . . . . . . . . . . . . . . . . . . . . . . 6-1 7003W i

~ . _ _ . - . _ _ _ _ _ , - _ _ - , - . _

0 s

1.0 INTRODUCTION

i Open Item 4 of the Safety Evaluation Report for the Vogtle Electric Generating Plant (VEGP) Units 1 and 2 concerns-the effects of a postulated onsite chlorine gas release on the habitability of the VEGP control room.

General Design Criterion (GDC) 4 of Appendix A to 10CFR50 requires that structures, systems, and components important to safety be designed to accommodate and be compatible with the environmental conditions associated with postulated accidents. GDC 19 of 10CFR50 Appendix A requires that a control room be provided from which the nuclear power plant can be operated safely and maintained in a safe condition under accident conditions.  ;

More definitive guidance on methods to ensure that the control room will remain habitable following a postulated hazardous chemical release is i

provided in various USNRC regulatory guides and standards. Regulatory i

Guide 1.78 (Reference 3) describes a methodology acceptable to the Nuclear  !

l Regulatory Commission (NRC) staff to use in evaluating the habitability of a j i

i i

nuclear power plant control room, following a postulated hazardous chemical '

release. Regulatory Guide 1.95 (Reference 2) describes recomumended measures t

to protect nuclear power plant operators against an accidental chlorine gas j release.

' Standard Review Plan Section 6.4 (Reference 5) presents guidelines for control room ventilation system layout and design to ensure that plant operators are adequately protected against the effects of accidental releases of toxic gases. Standard Review Plan Section 2.2.3 provides l

guidance for determining which types of postulated toxic gas releases should be classified as design basis events.

7003W 1-1

4 A toxic gas hazard evaluation was performed by Georgia power Company (GPC) and included in the VEGP Final Safety Analysis Report (FSAR). The evaluation

! considered postulated releases of 10 hazardous onsite chemicals; 5 hazardous cousso4ities stored and used at offsite chemical' facilities located within a 4

5-mile radius of the VEGP; and 11 hazardous commodities shipped on nearby I

transportation routes. The analysis concluded that postulated releases of all j of the chemicals under adverse meteorological conditions would either not jeopardize control room habitability, or would provide operators sufficient time to don protective breathing equipment before incapacitating

concentrations were reached within the control room. On that basis, it was concluded that the VEGP control room operators were adequately protected J

l against postulated onsite and offsite chemical releases, and that the applicable NRC acceptance criteria were met.

i i

l During their independent technical review of the VEGP FSAR control room i

habitability evaluation the NRC staff identified several concerns about the j analytical models used to predict chlorine gas concentrations inside the VEGP i

control room following the postulated catastrophic rupture of a 1-ton chlorine l gas cylinder located at the VEGP Buclear Service Cooling Water (NSCW) chlorine

storage building. A series of formal BRC questions and GPC responses i

followed, which were helpful in clarifying the basis for the FSAR models and i assumptions, and established the areas of technical differences. As a result of these discussions, a course of action necessary to resolve the open item l was developed. GPC committed to reanalyse two major portions of their original analysis. The first comunitment was to evaluate three different l atmospheric dispersion models suggested by the NRC staff to predict chlorine 7003W 1-2

4

sas concentrations that could occur at the VEGP control room HVAC (Heating-Ventilation-Air Conditioning) system outside air intakes.

I' The second task GPC cosmitted to perform was a more detailed analysis of the infiltration of chlorine gas from outside air into the VEGP control building and control room in order to determine whether control room operators would have at least two minutes after a control roce alarm annunciates the presence 1

of chlorine gas in the outside air intake to don protective breathing j apparatuses before a concentration of 15 parts per million by volume (ppsv) of i' chlorine gas was reached within the control room. UsuRC Regulatory Guide 1.95 states that the chlorine concentration within the control room should r.ot exceed 15 ppay within two minutes after the operators are made aware of the 4

presence of chlorine. USNRC Regulatory Guide 1.78 and 1.95 consider two i

I minutes sufficient time for a trained operator to don protective breathing equipment.

i The analysis of chlorine concentrations predicted to occur at the VEGP control a

room HVAC outside air intakes by the three atmospheric models suggested by the, NRC staff was completed and transmitted to the NRC in a letter dated

! November 26, 1985 (Log GN-750) from J. A. Baily of Southern Company Services i

j to B. J. Youngblood of the NRC. Based on the results of this analysis, GPC

concluded that the Regulatory Guide 1.95 atmos),heric dispersion model j

(Reference 2 would be appropriate for use in evaluating VEGP control room chlorine concentrations in the subsequent portions of the analysis.

7003W 1-3 I

- -, N n \ --. w eM ' '

l This submittal provides the results of the~ chlorine gas control room habitability analysis that GPC has committed to provide to the NRC. The infiltration of chlorine gas from outside air into the VEGP control building and centrol room, and the rate of buildup of the chlorine concentration inside the control room is determined using outside air chlorine concentrations 1

predicted by the Regulatory Guide 1.95 dispersion model for worst 5 percentile '

dispersion conditions of F Pasquill stability category and 0.9 meters per second windspeed. Additionally, in order to show the sensitivity of the analysis to a change in windspeed, results are also presented for 0.7 meters per second windspeed.

To assist the reader in understanding the technical basis for the chlorine infiltration model, Section 2.0 describes relevant design and operational characteristics of the control room and control building NVAC systems including: information about areas served by the various systems, configurations of supply and return registers and mixing characteristics within various areas served by these systems, and other pertinent parameters.

Section 3.0 describes the mathematical model and input parameters that are used to simulate the buildup of the chlorine concentration inside the control room considering both direct inleakage from outside air and indirect inleakage through adjoining spaces within the control building. The results of the e

analysis are presented in section 4.0, followed by a discussion in Section 5.0. A package of 13 drawings is included in an appendix at the end of the report to supplement the information presented in the text.

7003W 1-4

• ~ *~

= ~7 . . ?_2_- ._ .- rvr~ . - - - - - -

l i

l

. I l

2.0 NVAC SYSTEM DESCRIPTION l 2.1 OVERVIEW The VEGP control building is served by various HVAC systems. These systems serve the control room zone, adjacent spaces to the control roon zone (

on level 1, the upper cable spreading rooms on level 2 and lower cable spreading rooms on level A. All but one of the control room zone walls are l

located directly adjacent to other interior rooms within the control building. A four foot thick solid concrete wall (with no penetrations) at the north end of the control room zone is the only wall whose opposite face is in direct contact with outside air.

2.2 CONTROL ROOM HVAC SYSTEMS There are two different HVAC systems serving the control room zone. One system operates during the normal mode of plant operation. While the second system operates during a toxic gas event in the isolation mode of plant operation. The air conditioners operating during the normal mode are

located on level 4 of the control building. The filter / conditioners that operate during the isolation mode are located on level 3 of the control building. Each air conditioner and filter / conditioner is sized for the entire cooling load for the control room zone.

W ooms R comprising the control roosisone are listed in Table 3 and illustrated on Attachment 2.

i l 7004W 2-1

_ . _. - _ _ .w - . - - _-- __

f Two outside air intakes for the control room HVAC system are located on the roof of the control building. They are approximately 60 feet above grade and are separated by a distance of 150 feet (attachment 1). These intakes and the outside air ducts for the control room are not shared with any other HVAC system. Isolation dampers are located near each air intake, and in the supply and return ducts between the control room sone and the normal air conditioners.

I 2.2.1 NORMAL MODE OF OPERATION I

! The control room zone norinal HVAC system consists of two redundant air 1,

conditioners, and two redundant return air fans, each with 100 percent

, capacity. Outside air is taken from the outside air intakes previously i

i described. Supply and return main ducts connect the air conditioners with the distribution system in the control room. This distribution system is located above the ceiling of the control room, and is used in both the normal and isolation modes of operation (attachment 3). The control room is j maintained at 1/8 inch w.g. static pressure with respect to the outside.

Isolation dampers installed in the supply and return main ducts are closed when the normal HVAC system is not in operation. One exhaust fan serves the toilet, kitchen, conference room and janitor rooms which are part of the control room sone.

i 1

2.2.2 ISOLATION MODE OF OPERATION i

t l The control roon zone HVAC system that operates during the isolation mode i consists of four redundant filter / conditioners and four return air fans (see 1

7004W 2-2 l i

l

1 attachsent 7). Each filter / conditioner has capacity to cool both units 1 and 2 control rooms. The Unit 1 and Unit 2 control rooms are located in a shared area of the control building (see attachment 2). Two filtog/ conditioners and their return fans are powered from unit 1, and two from unit 2. The air filter / conditioners are located rooms 305, 311, 312, and 321 on level 3 of the control building.

A set of supply and return main ducts from each of the four filter /

conditioners and their return fans connect to the common distribution system in the control room. These ducts have isolation dampers which are closed when their filter / conditioner is doenergised.

Should chlorine gas be sensed during normal operation by either or both of the redundant sets of chlorine detectors located in the outside air intakes (one redundant-1 out of 2 logic-set per intake with set points of 2 ppe by volume), a control room isolation signal is initiated which causes closure of the isolation dampers at the affected air intake (s) (see attachment 10).

Both air intakes will isolate if chlorine contaminated air is detected at each of the two outside air intakes. However, if chlorine contaminated air is sensed at only one intake, the unaffected intake will not isolate and will continue to draw in essentially (containing $2 ppe by volume of chlorine) uncontaminated air into the control room sone. This flow is sufficient to maintain the control room sone at a positive pressure with respect to surrounding control building spaces, thereby preventing the infiltration of contaminated air into the control room sone from surrounding spaces. Thus, inleakage from the adjoining sones into the control room can only occur if both of the control room sone outside air intakes isolate.

7004W 2-3 i

... . . . _ _ _ l

j Following a control room isolation signal from either or both of the chlorine detectors at the outside air intakes, the essential filter /

I i conditioners are energized, and the isolation dampers in the toilet exhaust duct are closed. A no flow condition in the toilet exhaust duct will

}

. doenergize the toilet exhaust fan. The essential return fans are started, i and the isolation dampers in the normal supply and return ducts are closed. +

l When a no flow condition is sensed in the normal supply and return ducts,

the normal air conditioners and return fans are doenergized. ,

l i The travel time for a particle of air from the toxic gas sensors to the j isolation dampers in the normal supply duct is about 20 seconds (see attachments 10, 11, 12 and 13).

a During the isolation mode, the HVAC systems serving the upper and lower cable spreading rooms (see attachments 5 and 6), and the rooms adjacent to j

-s the control room sone (see attachment 4) continue to operate. This provides

! a pathway for contaminated outside air to enter these areas after control I

toom isolation has been achis.ved, where it can subsequently leak into the I c

control room zone (provided that both control room sone nomal HVAC syst1mm 1

outside intakes have isolated--see the preceding discussion).

l I

2.3 CONTROL BUILDING HVAC SYSTEMS 1

J The control building normal HVAC systems serving levels A, 1 and 2 of the control building consist of three air-conditioning systems (attachments 5, 4 -

l l and 6 respectively). These systems utilise duct distribution systems and j outside air intakes which are independent and not connected to the control

! room systems.

j 7004W 2-4

' , - - _ _ _ _ _ - _ - - _ _ . _~_ . _ . . .-- ._.__ _ _ _ _ _ __ _ . - _ . - _

l 4

i When both control room outside air intakes are isolated no outside air is i 1

j introduced into the control room; the control room is at ambient air

pressure. h areas surrounding the control room are at a slight positive pressure, since they continue to bring outside air into their systems.

Chlorine contaminated air could enter the isolated control room from I

adjacent areas in the control building through various leak paths (such as walls, floors, doors, electrical penetrations, ductwork, etc.).

i To model the movement and dilution of chlorine-contaminated outside air as j it travels through the control building prior to inleaking into the control room, the spaces of the control building capable of leaking chlorine contaminated air directly into the control room zone were partitioned into i

l several different areas termed nodes. Criteria used in defining node i

i boundaries included the following: (1) areas included within a node J '

t j boundary should be supplied with air from the same control building HVAC 3

l system outside air intake; (2) rooms included within a given node boundary 1

l should have free air exchange between one another and reasonably homogeneous mixing characteristics; and (3) there should be minimum air cosaunication j across node boundaries. Applying these criteria to areas within the control

! I j building resulted in the delineation of seven nodes. h spaces immediately I

adjacent to the control room on level 1 are defined as node 1. N lower 4 i i cable spreading rooms on level A are included in node 2. h upper cable spreading rooms on level 2 are included in node 3. N four filter rooms on l 1evel 3 which contain the filter / conditioner units are defined as nodes 4 through 7. All rooms which are included within the boundaries of nodes 1 l through 7 are listed on Table 1.

i f,

i a

l 7004W 2-5

_ __ - _ ._ _l_ ~ ~1._ . _._ - 7T~" OZZZi ZL 4- 1" ~~ " ~ ~ ~ '

l 4

2.4 DILUTION OF CHLORINg WITHIN THg CONTROL BUILDING i Frior to infiltrating into the control roor, chlorine-contaminated air which enters, each control building node becomes diluted in the effective not free volume of that node. Starting with total gross volume of each node as scaled from engineering drawings, the estimated portion of that volume occupied by solid objects (euch as furniture, panels, etc.) is subtracted, yielding the estimated not free volume V , of that node. Next, the not free volume is multiplied by a factor, fg (mixing fraction), to derive the effective not free volume. The mixing fraction, f . gcan be viewed as the fraction of air within a node which is available to mix with contaminated air. The mixing fraction is actually a function of many variables, including: the location and orientation of the supply and return registers within the node relative to potential leak paths, the general pattern of air circulation within the node, the occurrence of obstacles to air flow which generate turbulence and mixing (e.g., the high density of cable tray in the j

cable spreading rooms), the functional criteria used in designing the HVAC system for a given area (e.g. , maintaining a comfort zone from head-height downward in occupied office spaces versus design to promote effective head dissipation from areas containing closely-packed electrical cables or other services of heat-generating equipment), and additional factors.

Sections 2.4.1 through 2.4.3 discuss the estimated mixing fractions within each of the seven nodes as well as their basis, and describe dilution in each of the control building nodes. Table 2 lists the gross volume, fraction of unobstructed gross volume, fraction of mixing, not free volume, and V,gg, which correspond to each node.

[

f 7004W 2-6

- . . , _ . - - _ _ _ ___ . _ _ - . _ _ _ _ - . ~ _ _ . . _ _ - . _ _ - ._ , _ _. _ _.__.., _ _ _. , _. . _ . .

. TABLE 1 ROONS INCLUDED IN CONTROL BUILDINC NODES Node Room No. Room Description l l

117 Lobby 118 Air Lock 119 Conf & Lunch Ra 120 Corridor 128 Corridor 138 Corridor 141 Corridor 142 Air Lock 1 143 Radio Chen Lab 144 Sample Room (Spaces adjoining 147 Lobby the control room 149 Corridor zone on Level 1) 155 Inst Repair 159 Air Lock 175 Electrical "A" 176 Electrical-Norm 177 Iloct-Norm 178 Electrical "A" 199 Corridor 199A Air Lock 2 A23 Cable Spreading (Cable spreading A44 Cable Spreading area on Level A) A45 Auxiliary Relay A82 HVAC 3 223 Isolating Auxiliary Relay (Cable spreading 224 Cable Spreading area on Level 2) 225 Cable Spreading 226 Isolating Auxiliary Relay 4 311 Filter Room 5 312 Filter Room 6 321 Filter Room l

7 305 Filter Room 7004W 2-7 L_

. .- --. . . - . _ _ - - - . = _ - .

4 .

2.4.1 DILUTION IN SPAC33 famannIATELY ADJACENT TO THE CONTROL BOON, LEVEL 1 i

(NODE 1) i 1

! It is. estimated that furniture in the rooms comprising node 1 occupy 0.10 of l

the gross volume. However, corridors have no furniture. Therefore, an estimate of the average fraction of gross volume which is obstructed (i.e.,

occupied by solid objects) is 0.05, and the fraction uhich is unobstructed l is 0.95.

i i

a f HVAC air diffusion in a room is designed to provide o comfortable l environment for the occupants in the stratum within 7 feet of the floor i

! (termed comfort volume). All air in this comfort volume is mixed to j preclude drafts. The ceiling height in these rooms is 10 feet (see i

Attachment 4). The ratio of this comfort volume to gross room volume is j 0.70. The fraction of mixing is therefore estimated as 0.70. The node vo'lume, V,gg , actually available to dilute chlorine contaminated l air is given on Table 2.

2.4.2 DILUTION IN CABLE gPREADING ROOMS i .

i The majority of leakage into the control room sone is expected to occur i

j through the numerous electrical penetrations into the control room sone fron

the cable spreading rooms directly above end below the control room sone.

t l Accordingly, the air mixing which takes place in these spaces was carefully  ;

1 l- and thoroughly analysed. The cable spreading rooms are packed with emble '

i l

j trays throughout 30s of their volume. Air being discharged from the supply '

registers suet traverse a labyrinth before reaching a return register. The 7004H 2-8

-. . - .. . . - - - ,.r:::rr:r-. =: m - ~ ~ ~ ~~~ ~ ~ ' - ~~ "~~ ~ -

i TABLE 2. CONTROL BUILDING PARAMETERS w

I C *** Volume in I Whl'ch chlorine control cross Fraction of** Not Free Contaminated Building Volume, Unobstructed volume

• Fraction Air is Diluted sections Description (ft 3) Gross Volume V (ft 3) Mixing, f Veffective (ft3) i mode 1 Spaces immedletely 49,506 0.95 Vi = 47,030 fg = 0.70 V ,gg.cggy ,_t edjacent to control = 32,921

room, Level 1 i Bode 2 Cable spreading 151,250 0.70 V2 = 105.875 f2 = 0.90 Veffective-2

rooms Level A = 95,287 t mode 3 Cable spreading 117.360 0.70 V3 = 82,150 f3 = 0.M v gg.cgg,,_3 i roome, Level 3 = 65,720 i Y Wode & Filter room 311, 23.850 0.68 V4 = 16,151 f4 = 0.80 Veffective-4

• = 12,921 Level 3 i

mode 5 Filter room 312, 23,850 0.68 V5 = 16.151 f5 = 0.M v gg.cgg,,_$

Level 3 = 12,921 Wode & Filter room 321, 21,600 0.64 V6 = 13,901 f6 = 0.80 Veffective-6 Level 3 = 11,121 l

I i mode 7 Filter room 305, 21,600 0.64 Vy = 13,901 fy = 0.80 V,gg , egg,,_7

, Level 3 = 11.121 Veg = 145,326 V,gg egg,,_cm Centrol control moom Zone 161.474 0.9 Fca = 0.83 l = 120,621 i

j

• set Free Volume = Cross Volume - Volume occupied by solid Objects .

• • Fraction of Unobetructed Gross Volume = (Wet Free Volume / Cross Volume)

• • • Veffective = (Wet Free Volume)(Fraction Mixing) t i

cable trays contribute significantly to the thorough mixing of the air i

supplies to the rooms. h NVAC system was designed to cool the cables uniformly and prevent hot spots in the closely packed cable trays. (see Attachments 8 and 9.)

I 2.4.2.1 DILUTION IN CABLs SPRgADING ROONS, LaVgL A (NODS 2) t I

It is estimated that cable trays occupy 0.30 of the node's gross volume, i

leaving 0.70 of the volume unobstructed.

h penetrations from the lower cable spreading room are located in the ceiling. N supply air duct is located at the east well and runs the entire length of the room from north to south 14 to 15 feet above the  !

floor. There are eight supply registers. Three direct airflow toward the opposite well, and five direct airflow downward. m return air duet is routed along the west well at the same elevation as the supply duct. h re are three return registers mounted on the bottom of the duct which draw air up the wall. N oe supply registers are intended to direct the air across the room and down, with 50% of the supply air being directed vertically down in a wide angle jet configuration and the "horisontal" streams deflected at a downward angle, h tendency of the cold supply air to drop acts to  ;

enhance the pattern. Flow is generally away from the penetrations in the ceiling. h air will follow a semi-circular pattern downward and along the floor and up the west well to the return registers. Air in motion is 7.1 times as much as can be drawn out through the return resisters. h air which has not been exhausted continues to travel upward and to tum parallel with the colling, to be entrair.ed by the air streams from the supply 7004W 2-10 I i

mw_.__ _ _ _

registers. Thus the fraction of mixing contaminated air will undergo is estimated to be 0.90 (See Attachment 5). The node volume, V effective-2' actually available to dilute chlorine contaminated air 3s given on Table 2.

2.4.2.2 DILUTION IN CABLs SPRgADING ROOM, LIVgL 2 (NODS 3)

It is estimated that cable trays occupy 0.30 of the node's gross volume, leaving 0.70 of the volume unobstrw ted.

The supply air duct is located on the longitudinal centerline of the room and runs the entire length of the room from north to south 14'-7" above the flooc. Seven supply registers are spaced along the supply air duct. Four supply r* sisters direct their air toward the east well while three direct airflow downward. Registers are of the " double deflection" type with individually adjustable blades in both the horizontal and vertical planes (with reference to a horizontal discharge) to provide maximum air pattern flexibility and avoid direct paths to the penetrations. The return air duct is located along the west wall of the room at the same elevation as the supply duct. There are three return registers located on the bottom of the duct at the center and two corners of the room, which draw air up along the west wall. Air is thus directed down and army from the return registers.

By engineering judgment we conclude that the air will follow a generally circular path in a vertical plane with air originating at the conter of the room and traveling to the east well, down to and across the floor and up the west well to the return registers. Supply air is always colder than toon air, and the tendency of cold air to drop enhances the pattern. A deep colling been just to the west of, and parallel with, the supply air duct 7004W 2-11

i i

i i

obstructs bypassing of air along the ceiling. Begisters are loested so that t

i complete air mixing occurs before the air stream reaches any penetration or i

i retum register. The duct and beam locations limit the fraction of mixing to 0.80. (See Attachment 6). The node volume, Veffective-2, Se #8 y i

available to dilute chlorine contaminated air is given on Table 2.

j t

l 2.4.3 DILUTION IN FILTER BOONS, LEVsL 3 (NODES 4, 5, 6, AND 7) ,

4 It is estimated that the HVAC conyonents in each of the filter roome 311 and l

312 occupy 7,699 ft (approximately 0.32) of the room gross volume,  ;

leaving 16.151 ft (approximately 0.68) of the gross volums unobotnacted. .

! i 1

It is estimated that the HVAC components in each of the filter roome 321 and 305 occupy 7,699 ft (approximately 0.36) of the room gross volume, f i .

leaving 13,901 ft' (approximately 0.64) of the gross volume occupied. [

I; The ventilation system in each filter roen is a once-pass-through system; I  !

I outside air enters one end of the room and is exhausted at the other end.

i I

Owing to the large fraction of the gross volumes of these rooms occupied by a

] the fans and filtration units, outside air drawn into the rooms inevitably 1

3 mast pass around the HVAC components and travel through the relatively i

]

restricted f ree air volume to reach the euhaust. The confined specu and i

i relatively high flow velocity induces the mining of room air with i

pass-through air. The mixing frection is estimated to be 0.0. The node

! volumes actually available to dilute ehlorine contaminated air are given on I

l Table 2.

i 1

l i

i 7004W 2-12 l i

4

._---.___,,-..-_-__.-__--~..-_-_..~.m._. . . ~ _ _ _ - _ . .- _ _ , _ - _ . _ . . . _ -

2.5 DILUTION OF CHLORINE WITHIN THE CONTROL ROOM ZONE It is estimated that furniture, panels, etc. in the control room sone (t:ee attac;mont 2) occupy 0.10 of the gross volume. Therefore, the fraction which is unobettveted is estimated to be 0.90.

The control room sone consists of the control rooms for units 1 and 2 and six supporting rooms. All rooms which comprise the control room zone are listed on Table 3. The supporting rooms have ceilings. The control room area occupied by the control board is covered by a suspended colling, while the balance of the control room area occupied by the panels does not have a ceiling.

Supply and return registers in the control room sone are all located at colling level. Air diffusion in the rooms is designed to provide a comfortable environment for the occupants in the volume up to seven feet (termed comfort volume) above the floor. All air in this comfort volume is thoroughly mixed. The ratio of this comfort volume to gross room volume is 0.70 in each of the six supporting rooms. The fraction of mixint in each of the six supporting rooms is therefore estimated as 0.70. The ratio of comfort volume to gross room volume is 0.75 in the area of the control rooms covered by the suspended colling. The fraction of mixing in this portion of the sentrol rooms is therefore estimated as 0.75. A mixing factor of 0.90 is used for the portion of the control rooms not covered by selling in order to account for natural convection to the upper portion of this space. The weighted average of the values of fraction mixing in the rooms whleh j 7064W 2-13 i

comprise the control room sone is 0.83. The fraction mixing in the control room sono is therefore estimated as 0.83. The control room volume actually available to dilute chlorine contaminated air, Veffective-CR'

• I '*" ""

Table 2.

7004W 2-14

-- ~ -

_- m-. . . . - . _ _ . _ . . . _ _

TABLE 3. BOONS INCLUDED IN THE CONTROL 900N EONE Roon No. Room Deecription

.. 156 Jenitor 157 Toilet 158 Kitchen 160 Record storese 161 Emergency storese 162 Conference Room 163 control Room Unit 1 164 Control Room, Unit 2 i

7004W 2-15

3.0 CONTROL ROOM INFILTRATION MODEL DESCRIPTION The chlorine concentration in the control room at any time, t, is calculated using,the concentrations occurring at the outside air intake, xout(t)

(as input). and the infiltration model as described below.

Material balance about the control room (see section 3.2) yields the equation used to evaluate the control room chlorine concentration as a function of time. As can be seen in Figure 2, any control room building area (or " node")'which can leak chlorine-contaminated air directly into the control room, is considered in calculating control room chlorine concentrations. Air which enters each node Q ,

, is conservatively assumed to be undiluted, outside air with centerline chlorine concentrations (x, (t)). This air is subsequently diluted within a node's effective not free volumne, Veffective-1, Prior to leaving into the control room sone.

The control room inflitration model considers mixing and dilution within each of the 7 nodes in the control building, followed by infiltration from each node into the control room. The 7 nodes represent 4 filter rooms, upper cable spreading rooms, lower cable spreading rooms, and spaces immediately adjacent to the control room sone, on Level 1. Chlorine concentrations within each node and inside the control room sono are calculated as a function of time.

700SW 3-1

3.1 Control Building Chlorine concentration The chlorine concentration within each control building node is modeled by treat %ng each node as an isolated box having an effective not free volume of Veffe tive-1 given by equation 1.

Y Y effective-i " i free-i " i i ("I* '

Where l V g= Vg

,,,g = not free volume (ft ) of node = Gross Mode l Volume - Volume occupied by solid objects f

g= fraction of node not free volume available for mixing with contaminated air, i.e, the fraction mixing within a node j (where 1.0 = 100% mixing 0.90 = 90% mixing, etc.).

Assumptions made in determining the chlorine concentration as a function of time within a node are as follows o It is assumed that there is no significant flow between the nodes.

, Node boundaries were delineated to be consistent with this assumption.

Nodes are either located on different levels of the control building or are separated from other nodes by walls (see Appendix attachments 4 through 7). l 1

1 l

i i

7005W 3-2

o At any time the volumetric flowrate of air into a node, Q , is equal to the volumetric flowrote out of a node, Q (see figure 1).

• %ut(t) y V effective i

• where but(t) = the chlorine concentration at the outside air intake Figure 1 NODE CHLORINE CONCENTRATION MODEL o only a fractional amount, fg , of the not free volume of a given node is available for mixing.

Mat.erial balance around node i yields equation 2.

M g (t) = M (t - At) + M g, -M (Eq. 2)

Where M (t) = mass in node i at time t I

t = time after chlorine release M (t - At) = mass in the node at time (t - At) l l

700$W 3-3 N

M = mass entering node i during time At i

M = mass leaving node i during time at and M ( 3 l in-i(at) " *out ' ' Nin-i -

(Eq. 4)  !

M ,

=x g (t) Q _g&t i M (t - 6, -) = x (t - At) V f (Eq. 5)

Mg (t) = xg(t) V f g (Eq. 6)

Substitution of equations 3, 4, 5. and 6 into equation 2 followed by division by V f yields equation 7.

i xg (O a xg(t - ot) + V f I'N in-i A -A(i Nout-L A (Eq. 7)

[out ,

l 1

I Where l

x (t) = The chlorine concentration in node i at time t d

1 I

x (t - At) = the chlorine concentration in node i at time (t - At)

I i

l l

700$W 3-4

_ . _ - I

1 l

I l

xg j (t) = the chlorine concentration at the outside air l I

intake at time t g (t - At) = the chlorine concentration at the outside air intake at time t - At x (t) = [xout ~

+I out ' ,

S*

xg (t) = [x g(t - At) + x g(f.)]/2 (Eq. 9) ,

At = time increment = 1 see All other variables as previously defined.

Substitution of equation 9 into equation 7 and simplifying leads 'co equation 10 which is used to calculate the chlorine concentration in a node versus time.

3 XgR - A @ out-i

< X g R-AO+yg xout N in-i ~

2

~ ~'

xg(t) =

Qout-iA ',

ii (Eq. 10)

The talues of the parameters needed to calculate the concentration of chlorine within each node (using equation 10) are given on Table 2 (columns 5 and 6) and Table 4.

l l

7005W 3-5 t

s . _. _ _. .

..-.--~~.-n.--~_., ,

TABLE 4*

VOLUMETRIC FLOWRATES OF AIR INTO CONTROL BUILDING NODES I

Node so. Description Qi n-i(Cf")

1 Spaces launediately adjacent to 3,535 control room, Level 1 2 Cable spreading rooms, Level A 604 3 Cable spreadi..d rooms, Level 2 340 4 Filter room 311, Level 3 4,100 5 Filter room 312. Level 3 4,100 6 Filter room 321, Level 3 4,100 7 Filter room 305, Level 3 4,100

• See Appendix attachments 4 to 7 for node delineation. Rooms included in nodes are given in Table 1.

t 7005W 3-6 l.

3.2 Control Room Chlorine Concentration l

Leakage from each node into the control room after isolation is considered in calculating the concentration of chlorine seen in the control room as a i

function of time (See figure 2).

l The control room chlorine concentration versus time is modeled similar to the node chlorine concentration with the following exceptions:

o No chlorine is assumed to enter the control room directly via the control room outside air intakes. The reason for this assumption is that the time from detection (2 ppm) to isolation is 17 seconds, of which 11 seconds account for the chlorine monitor delay time, and 6 seconds are required for the isolation dampers to close. The time for chlorine to travel between the monitor and downstream isolation damper is about 20 seconds. Hence, the dampers are closed before chlorine contaminated air arrives. (See Appendix attachments 10 through 13.)

o Until control room isolation is accomplished, no contaminated air from the adjacent nodes will enter the control room since the control room exists at a positive pressure relative to the surrounding nodes. This situation will persist until both control room outside air intakes isolate. For modeling purposes it is conservatively assumed that both intakes are exposed to centerline concentrations of the chlorine cloud such that both intakes isolate.

700$W 3-7

~7:: ~ -- - . ,- , . - ,: , .. ; a .. ._ ,;7 7 v ;; T L ..

J/4 NORMAL OUTSIDE AIR INTAKE Ut*1 ; CHLORINE DET8CTOR

? NODE 1

-4 6-ISOLATION DAMPER I F leek-1 Q ---

A  ; F,. .R O in-2 y NODE 2 out-2 l ;F

'~-g gg.

'+ CONTROL ROOM += = -- E biraet in X,ut (t)

O O in-3 m out 3 , pp $ $ 4' r NODE 3 -F

' id 3 J *1 lF hd4 hd4lI lI

-- F 1

I ind ^

Oout-4 ,

L'- 38*4- - ,J ld l 6 d '

l (F leek 7 d '

Oout 7 v

NODE 4 Oout4 Oout4 NODE 5 NODE 6 NODE 7 NOTE: F hdi IS PART OFOout i Oin4 O in-6 O in-7 Figure 2 CONTROL ROCH CHLORINE CONCENTRATION MODEL 4546t 3- 8

_ _ ~ _ .- - -- _ - . . . , - - _ . - . - - . _ ,

o No outside air is assumed to enter the control room zone directly due to ingress or egress since the control room doors are either part of an airlock (i.e., two-door vestibule) or are accessed by passing through an airlock (refer to Appendix attachment 2.)

o To conservatively account for a "short-circuit" leak path from a supply register pointed directly at a control room. door on Level 1, 5 cfm of undiluted, direct outside air (F ) is assumed to enter the l control room sone (see Section 3.3) without undergoing any dilution within the intervening node.

o The total leakage out (cfa) of the control room zone, F ,

, following isolation is the same as the total inleakage rate, i.e.,

I 7

out-CR " in-direct + g, leak-i I

where, F = the inleakage rate into the control roon zone from 6 node i o The total inleakage rate into the control room zone after isolation is assumed conservatively to be 750 cfm to account for the possibility of wear of door seals, aging of penetration seals, etc. (The calculated inleakage rate is less than 300 cfm for the VEGP control room zone.)

i 7005W 3-9

I l

l Material balance around the control room zone yields equation 11.

l l

~

~ 7 nodes M " + * -

AR CR( - in-i(At) in-direct (At) out-CR(At)

_i=1 ,

(Eq. 11)

Where M (t) = The mass in the control room zone at time t M - " " * -

CR 7

I Min-i(At) = The mass entering the control room zone from the i=1 surrounding nodes during time at M = The mass of undiluted outside air entering the control room zone during time at M = The mass leaving the control roon zone during time at and M

in-i(At) " i leak i i

in-direct " Iout in-direct 7005W 3-10

t (Eq. 14) out-CR " *CR out-CR (Eq. 15)

Mg ,(t - At) = x (t - At) V f M (t) = x (t) V f ( I' }

CR xg (t) = [x (t) + x (t - At)]/2 (Eq. 17) x (t) = [x (t) + x (t - At)]/2 (Eq. 18) x g(t) = [x (t) + x (t - At)]/2 (Eq. 19)

Substitution of equation 12 through 16 and equation 19 into equation 11 followed by division oy V f leads to equation 20 which can be used to calculate the chlorine concentration in the control room zone as a function of time for any time, t, greater than the isolation time.

3 f7 5 x CR(t - At) + g y x (t) F in-direct

• A(

i leak-i CR CR ,

(i=1 {

l -xg, R - AWout-CR 2

( -

xCR(t) = F q out-CR 2f (Eq. 20)

CR CR 7005W 3-11

,.- . - . . ____ _ , --m-, -. -

For t > isolation time, where >

1 x (t) = the chlorine concentration in the control room zone at

. time t x (t - At) = the chlorine concentration in the control room zone at time t - At V = net free volume of control room zone j F = fraction of control room zone not free volume available for mixing i

1 All other variables as previously defined.

4 i

i

] The concentration of chlorine in the control room zone at any time less than the isolation time is given by equation 21.

x (t) = 0 , for t i isolation time (Eq. 21) i i Values of parameters needed to determine the control room zone chlorine l

! l I

concentration using equation 20 are listed on Table 2 (columns 5 and 6) and Table 5. Volumetric leakage rates from each node. Which correspond to a total control room zone inleakage rate of 750 efs, are listed on Table 5.

These leakage rates were calculated using methodology presented in l References 7 through 10.

l l

! 7005W 3-12 t . . . ,_ . _; r .. . _ . - .

, ._ _ _ y1,- . _ .

, ;, ~ ,- r .- . _ . _ ,

l TABLE 5 j VOLUMETRIC LEAKAGE RATE OF CHLORIDE CONTAMINATED AIR l INTO THE CONTROL ROOM ZONE

• Volumetric Leakage Rocet or Area Leakage Paths Rate, F(cfa)

~

1 1

Mode 1 Concrete walls, floors, doors, Fleak-1 = 163 dampers, duct penetrations Node 2 Electrical penetrations Fleak-2 = 259 Node 3 Electrical penetrations Fleak-3 = 259 Wode 4 Supply fan enclosure, filtration Fleak-4 = 16 unit, return fans, duct work Mode 5 Supply fan enclosure, filtration Fleak-5 = 16 unit, return fans, duct work Node 6 Supply fan enclosure, filtration Fleak-6 = 16 unit, return fans, duct work Mode 7 Supply fan enclosure, filtration Floak-7 = 16 unit, return fans, duct work Outside Air Supply register pointing directly Fin-direct " S at control room door control Room Zone Susumation of all inleakage rates Fe,g = 750 (given above)

• Appendix Attachment 3 shows the control room zone pressure boundary, anu air distribution.

l 700$W 3-13

- n. . .n.--.

3.3 Conservatisms Conservatisms used in determining chlorine concentrations in the VEGF control room are given below:

s o No credit is taken for the physical separation of the control room zone

- outside air intaker, nor for the fact that they isolate independently (see Section 2.2 and attachment 1 in the Appendix).

o Centerline C12 e n entrations are assumed to occur at all control room zone and control building HVAC outside air intakes sismiltaneously >

despite the fact that the intakes are physically separated.

f 4

l o No credit is taken for a remote (non-1E) C1 detector 2 at the NSCW ,

, Chlorine Building which annunciates inside the control room zone. This detector could provide advanced warning of an accidental chlorine release at the NSCW chlorine storage area.

o The wind direction is assumed to be directly toward the control room air intakes from the release point.

o A positive 1/8" wg pressure differential is assumed to exist across all leak paths driving contaminated air fruet the control building into the i

control room zone following isolation.

i i

i 700$W 3-14 4

---.m.y ., ,c-,.y--.,-r-y ,ye,. , , p. . . ,i..,., y- ,, y,,.e-,...,,,. ., - + , .,w.--..~we,*..e,,ip,,.,n.c-,_w.-.e,.>-,e+%- - . - .w w-- , . e - - - . . .m. . . . . - - =w=-. -.-se=

l o No credit is taken for the removal of chlorine by air which is recirculated through charcoal filters after the essential HVAC units have been actuated. (Charcoal has an extremely high affinity for chlorine).

o Air which enters each node, Q , is conservatively assumed to be undiluted, outside air with centerline chlorine concentrations, x (t). (Some dilution would undoubtedly occur in the ducting.)

o To conservatively account for a "short-circuit" leak path from a

\

supply register pointed directly at a control room door, 5 cfm of undiluted direct outside air is assumed to enter the control room.

o Less than full credit has been taken for the free volume of air within the nodes and control room zone that is available to dilute incoming contaminated air.

o A total inleakage flowrate of 750 cfm has been assumed to enter the control room zone after isolation. The calculated inleakage rate is substantially smaller.

l 1

l l

7005W 3-15

. ._ -r- m=~- _

o It is assumed that even moment ary exposure to a C1 concentration exceeding 15 ppa during the first two minutes of chlorine entry into the control room zone would incapacitate the control room operators. Literature cited in NRC-sponsored studies (e.g.,

NUREC/CR-1741, Reference 6) suggests that higher C1 concentra-tions can be tolerated for longer time periods without producing incapacitating effects.

l 7005W 3-16

. _ m_ _

___m..,

_ . _.z .

l 4.0 RESULTS Chlorine concentrations predicted to occur inside the VEGP control room zone were analyzed using the infiltration model described in Section 3.0.

Chlorine concentrations occurring in the VEGP control room outside air intakes, and input to the infiltration mode 1, were calculated using Regulatory Guide 1.95 atmospheric dispersion model (Reference 1). Two ,

different seta of meteorological conditions were considered: (1) Pasquill F I stability and a 0.9 meter per second windspeed and (2) Pasquill F stability .

and a 0.7 meter per second windspeed. The results of both analysws are summarized in Tablo 6. Plots of the concentration profiles are shown on j Figures 3 and 4.

i t

I i

i i

4 l

1 7006hl 4-1

- r --  : ---

-;-.-.. . = =2  : - -. ::::.:= r'7:_ . ,.._:.__

TABLE 6 PREDICTED CHLORINE CONCENTRATIONS INSIDE THE VEGP CONTROL ROOM ZONE FOLLOWING THE POSTULATED RUPTURE OF A 1-TON ONSITE CHLORINE GAS CYLINDER Time Available for Control Room Control Room

- Operators to don Chlorine Atmo' spheric Pasquill Protective Concentration Dispersion Stability Windspeed Breathing 2 Minutes Model Category (meters /second) Equipment (1) After Alarm Regulatory F 0.9 10 min 26 sec 2.74 ppmv Guide 1.95 Regulatory F 0.7 8 min 10 sec 3.18 ppmv Guide 1.95 (1) Determined tjy subtracting the time when an alarm from the chlorine detector (located in the outside air intake) sounds inside the control room zone from the time thst 15 ppmv is first exceeded.

l l

l

)

i i

l 7006W 4-2 l --. - --- . . . . . .. _ .. . . , _ . . . . ._

l 16 .

t = 827 SECONDS, LEGEND

^ 15 PPM IS FIRST EXCEEDED %

REG. GUIDE 1.95, F STABILITY.O.9 M/SEC. 750 CFM INLEAKAGE j 14 I' 12 ,

m -

.a o

E

• 10 r 8

u r

z-8 8

G 4 -

N

$ 6 W

8 _

w E

8 4

/

a 5 -

2 t = 201 SECONDS, ALARM SOUNOS i i e i

' ' e e e 0 500 600 700 800 900 I 100 200 300 400 l

TIME AFTER RELEASE.SEC CHLORINE CONCENTRATION IN THE VEGP CONTROL ROOM FIGU.RE 3 1

4-3 l l I i i

- . - - - - _ _ _ . . , . , _ _ , , . , _ _ _ _ _ , i w-e ,y _

9 16 ., , , t = I45 SECONDS, LEGEND

- REG GUIDE 1.95, 15 PPM IS FIRST EXCEEDED 'N

- F STABILITY,0.7 M/SEC, 750 CFM INLEAKAGE 14 7 E - 12 I= -

J K

10 Y

g 8 -

E w

8 1 ,

E -

% f

=

6 -

8 -

z .

i b

a

,4 ~

5 2

t = 255 SECONDS,

_ ALARM SOUNDS

' ' I I I ' I 0

O 100 200 300 400 500 600 700 TIME AFTER R5 LEASE,SEC.

CHLORINE CONCENTRATION IN THE VEGP CONTROL ROOM FIGURE 4 4-4

5.0 CONCLUSION

S According to Regulatory Guide 1.95 (Reference 2, page 1.95-2) "The chlorine concentration within the control room should not exceed 15 ppm by volume within two minutes after the operators are made aware of the presance of chlorine. Two minutes is considered sufficient time for a trained operator to put a self-contained breathing apparatus into operation." As can be seen in Table 6, the time available for workers to don protective breathing equipment as predicted by Regulatory Guide 1.95 model with F stability and either a 0.9 m/s or 0.7 m/s windepeed (10 minutes, 26 sec. and 8 minutes, 10 sec. respectively) is greater than the two minute guideline. Therefore, this analysis shows that the VEGp control room is in compliance with NRC regulatory guides regarding the protection of control room operators against an accidental chlorine release.

1 i

l l

i 7007W 5-1 l

--- = - ,- - - - _

- _ _ _ _ _ . : :_ :7 - _ -

= .

J E

l

6.0 REFERENCES

1. Georgia Power Company, " Evaluation of the Chlorine Concentrations in the i

VEGr CC,ntrol Room" November 25, 1985 Submittal (Log GN-750).

i

2. Regulatory Guide 1.95, " Protection of Nuclear Control Room Operators Against an Accidental Chlorine Release", Revision 1. U.S. Nuclear ,

Regulatory Comunission Of fice of Standards ' Development, Washington, D.C. ,

, 1977.

1

3. Regulatory Guide 1.78, " Assumptions for Evaluating the Habitability of a Nuclear Power Plant Control Room During a Postulated Hazardous Chemical ,

Release", U.S. Atomic Energy Conumission, Directorate of Regulatory Standards, June, 1974.

4. EUREG/CR-3786, "A Review of Regulatory Requirements Governing Control Room Habitability Systems", Sandia National Labs, prepared for the U.S. Nuclear l

Regulatory Commission, Washington, D.C., August, 1984. i l

l l

l

5. NURRG-0800, Revision 2, Section 6.4, " Control Room Habitability System", i l

! USNRC Standard Review Plan, July, 1981.

l l

l

6. NUREG/CR-1741, SAND 80-2226, R-4, "Models for the Retimation of Incapacitation Times Following Exposures to Toxic Gases or Vapors", Sandia l National Laboratories, prepared for the U.:*. Nuclear Regulatory k

' Commission, Washington, D.C. , December,1980. ,

l

\

l 700SW 6-1

-. _ m. - -

.__- _ _ mm - r-~_ 7.--- _ry 7 - ;- r~y --- - ,

l

)

)

7. Design Basis No.1531. "CB Control Room HVAC Systein" Revision 3, dated 5-6-85.

8. " Conventional Buildings for Reactor Containment", published by Atomics International; Catalog No. "NAA-SR-10100", dated 6/15/65.

9. ASHRAE Handbook, 1977 Fundamentals

10. ANSI N509-1976 Nuclear Power Plant Air Cleaning Units & Components; Page 8 Paragraph 4.12.1.2, Housings.

7008W 6-2

.. -.n. . . . . __

. l l

l APPRWDIX DRAWING ATTACMENT >

Control- Room Outside Air Intakes 1 2

Control Room Air Locks 3

Level 1 - Control Room Air Distribution Air Distribution in Areas Adjacent to the 4

control Room (Wode 1) )

Level A - Lower Cable Spreading Room Air Distribution 5 (Node 2) ,

Level 2 - Upper Cable Spreading Room Air Distribution 6

(Mode 3)

Level 3 - HVAC Air Distritution in the Filter Rooms (Modes 4, 5, 6 and 7) 7 Typical Locations of Supply Outlets and Returns in Cable Spresding Rooms 3 Typical Locrations of HVAC Ducts in Cable Spreading Rooms 9 Toxic Gas Tiavel Path Sheet 1 of 4 10 sheet 2 of J. 11 Sheet 3 of 4 12 Sheet 4 of 4 13 1

A-1 4531t

OVERSIZE DOCUMENT  ;

PAGE PULLED I i

SEE APERTURE CARDS  :

! NUMBER OF PAGES:

ACCESSION NUMBER (S):

? [o 0 0 & fn 0 W &f- Q /

l 4-+> n u -/4 i

l l

l j

l APERTURE CARD /HARD COPY AVAILABLE FROM RECORD SERVICES BRANCH, i

FTS 492 = 8989 l

l l

,