McGuire Nuclear StationAn Analysis of Hydrogen Control Measures

at McGuire Nuclear Station

Revision 6February 15, 1983

CHANGES AND CORRECTIONS:

Remove and insert pages in accordance with the following tabulations:

Remove these pages _ Insert these pages

Volume 3 - Chapter 3 -

3.4-2 3.4-23.4-3

Figure 3.4-5 Figure 3.4-5Figure (s) 3.4-7 thru 3.4-9

Volume 3 - Chapter 4

Table (s) 4.4.3 thru 4.4.8

Volum9 3 - Chapter 7

7.0-98 thru 7.0-116

I

i

|

8302170144 330215PDR ADOCK 05000369P PDR

- _ _ . , _ _ - , . .- .... . - , .__ - .__ -. . - . . ... ., . .-

-( N the ice condenser upper plenum. The analysis showed that six igniters spaced

approximately equally will prevent hydrogen concentrations greater than 10%

by volume in the ice condenser at any location if ignition takes place at

8.5% hydrogen by volume and propagation is at one foot /second, for the

maximum hydrogen release rate predicted for the McGuire small break LOCA.

Because 10% is considerably below any plausible detonation limit, this design

criterion is conservative.

Power to the igniter system is supplied by 10 circuit breakers, five 'each on

Emergency Lighting Panelboards ELA1 and ELBl. These circuit breakers, panel-

boards, and associated connections to Class lE power sources are shown on

Figures 3.4-7 and 3.4-8 for ELA1 and ELB1 respectively. The connection

diagram for the igniters themselves, which indicates the number and location

p of !gniters connected to e'ch circuit, is shown in Figure 3.4-9. PeriodicLJ measurements to establish the operability of the igniters are made at Emergency 6

Lighting Panelboards ELA1 and EL81.,|

Because each region of containment is supplied by at least one redundant pair.

!of igniters, a failure which renders a single igniter inoperable, or which causes ,

the failure of all igniters associated with a single circuit or panelboard, will

not affect the ability of the hydrogen ignition system to perform its intended

function.

|

The function of the Hydrogen Mitigation System is to ignite mixtures of

hydrogen and oxygen in the various areas of the containment when the local

concentration of hydrogen has reached 8.5%. The early ignition of hydrogen

| in the containment has several benefits.. C)! < 3.4-2 Revision 6

..

*

l

l'

- - _ _ . - - _ _ _ _ . . _. . - . --

-

:

_/ a. Hydrogen-oxygen recombination reactions occur at hydrogen concentrations('

substantially lower than those shown to produce uncontrolled reactions.

Thus detonation of hydrogen is precluded.

b. Reaction at low hydrogen concentration produces lower temperatures and

smaller energy release rates. Because more time is available for the

containment pressure mitigation systems (ice condenser and containment

spray) to dissipate energy, the total containment pressure rise is

lower. Lower temperatures also have less adverse effects on other

equipment in contai?-ant.

c. The product of deflagration (pure water) does not represent a threat

to containment or system integrity or to personnel entering containment

following termination of the accident.

!

:

|

|

|

O, 3.4.3 Revision 6I New Page

. - - . - . - _ - _ _ - - . . - . _ _. . - - _ _ - - - - . . - _ - - - . - - .

.

|

. . - -

%

, IEHMT,B ST_-

-

,

O_

- x'

' ' - C IEHMTB-49'__O lE,HMTB-64,

IEHMTB-4{ -'HMTB-42-

/ IEHMTB 4 0 O~ lEHM B-41

[// IE -52 W \ ) /'s~ /.' (d N/' \ o IEHMTB-50IE!HMTB-340 0 IEHMTB-32

O O \'

/ 7 IEHMTB-33 EHMTB-31f, jf'M'HMTB-40

'I

- i mwurs-30 0 ft \U lEHMTB-63N \ IEHMTB-33--

i%,

.

\ Ni flEHMTB-35 \ / / , O lEHMTB-66

\ s O CIEHMTB-27,/,

,

I HMTB-36 O O IEHMT@28 .' /

/''

O IEHMTB-45 IENMTB'-38 / ,C 'IE'iMT8-53 O lEHMTB-46 0 OO O I HMTB-4B IEHMTB-47, e4 pIEHMTB-54 O'~QEHMTB-3 ,/

,

' '

lEHMTB-65 ' /,' ''Nsxx - O -.-N. y p'-~~-

; -y-:

1

Note: Igniters in upperplenum of ice condenserare at elevation 840'.

ii

FIGURE 3.4-5

McGUIRE CONTAINMENT-PLAN @ ELEV. 820+0

||

O:Revision 6

- --. .- . - _ . . _ . . _ _ - - - _ _ . . - - . - - - - . - . - - - . - - - - - . - - - _ _ _ _ _ -- . --

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

i -,1

i^. .jr.-

,

.

+

1

it

i @$

j,

5 -^) - m, - - 4. -u,ca-

=._.,

j, '-

m.. . . . ,~

. . ,f- 'Y 4 4j- St.4. &g-= = .

! MT 7-v,-

i; / M, ., H-t

: m ti <:it nm rt? w-.; - ; - - =

23 ',w. 4 .,,~,, i; 7 3 - ~ ~;, wa%C l f ':"9:r.. .t Y-f.

~~ w e ja|

,

)3 * P* in.,5 w ,*cn. g _-- 7< - .

_. ,

;_. ). i2 _-e ;: f- | __; . n. .

,,___3, _ __,

- w> - , . ~ , ,1%.

.e.,4

,.==..o :.,_.. .-.~." i *%

=.:{_~

N _ x.->~

..we4 ue ;j

'

1 eMM=wn'o:em i,, , . , . . . . ~ . :,i .~~

I - - - - -3 .['~

~3..,s(A nar 50 I

! )6

1 .

t>.~

--j

i-,

--1 _ _ ''g ls' :-

,suq - - |1 . . -4

,_ '.i._.ia ) -i ,. 4

--:1 ___. 1:._,_:|,%*____- n 1::2 1-- I *:

:-~-i:r:: .,

~ .q:::i . i{ - ,s._.s, . '! >= i:: = t,ta#Je.4)4 3.m , ,, .

1- ;i =>x. ..

ih.. % -t 3. t. c .Ua,. rv.' t.::." entrurm ., m v:: . .9 - - - -- - ;.

< ,

l. i'

LIGHTINGONE LINE DIAGRAM |REACTOR BUILDING

0i McGUIRE NUCLEAR STATION1f. Figure 3.4-7 i| ,j Revision 6f ,

d'

i

i i ti- - - _ -- _ __ _ _ _ _. _ _ . _ _ . _ _ ___ _ _ _ , 't

as a u.ama a a __ - -,a

4

1

0

,

d

,

.

O|

|

9

_ -_- __ - __ _ _. -

- _ - _ _ _ _ _ _ _ _ _

s w ~. e , ... c.s n mi.x, , u. ,,), , . o..--.~.-.~,.-

;_Qm - mm ,r ;_.,

P

L_ _ -: g;j'h .y

^. %NUW N NdN MM^,g- -^W .* * M.v1.ae.ss** *% estSN em

m - aonem w o. e some.ses.es,|l"es'e'='' 4+-w - % sr - y g* *' tg .s - 'W *** w = gu m o uw _ _ .,C == * WSv= -7 ..;> (

Lifr a_, ri >' I "l-tot].+---- -

.- ..R

."Ar

i , i

h4C b#f NE/,

Lrs .Rxao.TlB.I.**

i

i'

r__re_ ew_

|

i , ;

U.! ? W-

4 .

6

[ [W 'N

d d &- J'

,, c, d i.

..

4 Se. ;

D. N __ . ..__4,.. 4 4|3.d 44 i 1_ _4c4 tm .< .

t^ ' ~

i i :,&j ,,| 43! *Aw =w! i j

i

eB +5

,,, ,s >>

g*am%

' I f

e - e- 1

e.,.. c, -,,.-)

s . A.%.was%i,L L J

| P.s.'

|- 'u===..a.y .. ' = = , , ., Meast W .i

i,

, i0' -.e.,c.,=- B |- S =<

- a ork I

m as s. e.

.. .- - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _

. 6 F g

8 6 @ @ @ 9>

;g ^ %K-,.< ns C & b T 'y,v-4w. ;~

i|- u-< u ,<

[ wa--n c- n_,__m_....-1 a

t O O O $ @ O

)

i C 'Na

e o 9 e _ e e-

1 ifLA.DPCs

I

i

, ,

3'

I e,..-.o -~ u ew .og .,3 cgg;g, .v. o;g

I . ,.. . .r.~

s' is ,:g 2 Qy - !!t.--WP 6y}e a qI| c

!- d%2$e.j,,7 r-- #ww w.ms ,%; ! 4 | 1 4- tj g- e ,.m so .a = vg o. g .o ~.,u-

. .\ ~, . n . . v,

qf,gg,c % % ~- ua,.3 g ,,6 4 oo 'v- | 1

Ii____J _ j

-

.

t?n 9 . " 2. '5:*

=n - ., ,,,a. =-. .- ,

JM L % . 4XML ~;! k

W- g";" (. ].,

|-

L74 L *,, W A w " ,

yy g .~ A * g g , *"' ~. % L

8.M IMA A.; WK

Q.fA. 0-

-, - .-.

&b

=- .ss Y N.'ammsb

GJITOR DOMFYE - itITEACONIECT -

SCHD.LE '*** *'* " * * * *

1y - .= g g,rg,= = = _ .'.

%3 K3,.

=_.g= =" . = > ._4

F% FJ.

w e.us

k 'm,

*"g g.",1- .3y; . . . . .4

Yd" 7thb

-

._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ _ _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ _ ,

)

_

HvDAOGEN ICApT@ OOxEO.5%O.'. SUICD4NT MTA4.5 FCft TB'$

men--ame

[o 'lo/ rd _ . _ _ i . '- /~ ~ ~.

b

e

! *i ,f+

3e- - '

'-

Q- -y,-,T " . ' '

k :~~ f.m . ... :=.. =v..w;.:r"-... .Qgy .,~~~

>a

- ~ ,,

lh3% .'

. / .

a,-t

....

. _ . . . -. .- _ .- ..

t

Table 4.4.3,

MCGUIRE CLASIX RESULTS SUMMARY

CASE 1 - REFERENCE CASE

Flame Speed = 2 ft/sec

Number of burns LC 5-

UP 16,

Total H burned (1bm) 957; 2

H remaining (lbm) 5812'

Peak Temperature (F) LC 1138; LP 228

UP 1517UC 162DE 238

Peak Pressure (psig) LC 11.8LP 11.7UP 11.7

|UC 11.6

! DE 11.85

| Ice remaining (1bm) 4.27 x 10

!

Revision 6

- _ _ _ _. __ - - - _

.. ._ _ _ - - . - _ _ .

4

4

O Table 4.4.4,MCGUIRE CLASIX RESULTS SUMMARY

CASE 2 - MINIMUM SAFEGUARDS

Flame Speed = 2 ft/sec

Number of burns LC 7'UP 10

Total H burned (1ba) 9562i

H remaining (1bm) 5822.4

Peak Temperature (F)' LC 1101LP 218

: UP 1520UC 175DE 245

| Peak Pressure (psig) LC 12.7LP 12.7 :

. UP 12.7|

- UC 12.7DE 12.7

5Ice remaining (1bm) 5.44 x 10

l

!

l

i,

O "-""":

I

l'

|'

._ ____.. ~ _.__ - _ _ _ _ _ . _ _ _ _ . _ _ . . _ _ . . _ _ . . _ . _ . _ _ _ _ . _ _ _ _ _ _ _ . - . _ _ _ _ _ . - - .-

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

j

; Table 4.4.5

MCGUIRE CLASIX RESULTS SUMMARY

CASE 3 - HIGH H, RELEASE RATE

Flame Speed = 2 ft/sec

Number of burns LC 11UP 22

Total H burned (1bm) 18292

H remaining (lbm) 5322

: Peak Temperature (F) LC 1193; LP 549

~

UP 1522UC 164DE 254

1

Peak Pressure (psig) LC 10.4LP - 10.4UP 10.4 'UC 10.6,

| DE 10.4i s| Ice remaining (lbm) 2.58 x 10.

!i

Revision 6

- - _ _ - - - - -_ -- - _-

- - _ . _ _

Table 4.4.6

i MCGUIRE CLASIX RESULTS SUMMARY

CASE 3 - ICE MELT 00T BEFORE BURN

Flame Speed = 2 ft/seci

Number of burns LC 7UC 1

Total H burned (lbm) 10042

H remaining (1ba) 5332,

Peak Temperature (F) LC 1167LP 261UP 282

i UC 407; DE 225 ,

Peak Pressure (psig) LC' 19.2 -LP 19.2,

i UP 19.2

O UC 19.2DE 19.2Ice remaining (1bm) 0.0

i

!

|

Revision 6

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

.. . - - _. .- . . - . - -

,

,

;

'

Table 4.4.7

MCGUIRE CLASIX RESULTS SUMMARY

CASE 4 - INERTED LC

Flame Speed = 2 ft/sec

|

Number of burns UP 15; DE 2

Total H burned (1bm) 9582

H remaining (lbm) 5792

Peak Temperature (F) LC 225or 690

;UP 1427UC 160DE 796

)

Peak Pressure (psig) LC 10.3LP 10.9+UP 10.9

O "C 10.6DE 10.3'5

: Ice remaining (lbm) 3.99 x 10

;

i

4

ii

Revision 6;

-

i

f

s

.

- - - , - - . ,-,-y,.,,,,-, - - ---, - , w.. ,,n.n------.- - - - - , - - , - - - , - . . - - - - . . - , - - .- - - . - _ _ . - . - - _ - - - . -_---,----.--,-.------a --an, ,-

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

;

,

Table 4.4.8 -^

MCGUIRE CLASIX RESULTS SUMMARY'

|

CASE 5 - INERTED UP

Flame Speed = 2 ft/sec

!

Number of burns LC 8UC 1

Total H burned (lbm) 9502! H remaining (1bm) 5422

Peak Temperature (F) LC 1250LP 270

'

UP 167UC 388DE 208

; Peak Pressure (psig) LC 14.4i LP 14.4i UP 14.4

UC 14.4'

,DE 14.4

'S

Ice remaining (1bm) 4.92 x 101

I!

|

+

'

Revision 6

,

,

w-- - - - - -*r 't f*-' ' * " - ' " - "'- " * - '- w+~&w"- 7" ''"W w--'*-**'* e-*-- * - +w-- " ' - ~----*-~=--r--= - - - - - -- - - - - - - - --

Response to Request For Information On Hydrogen Controls) Transmitted By Ms. Elinor G. Adensam's Letter Of January 24, 1983,

1. Recent discussions with Duke suggest that the upper plenum igniters in

McGuire are located alternately between the crane wall and side and the

containment shell side of the upper plenum in a staggered fashion, rather

than all on the containment shell side of the upper plenum as depicted in

Figure 3.4-5 of the McGuire submittal. Verify the "as-installed"

locations for these and all other igniters in the permanent system.

Provide revised drawings as necessary.

Response:

See revised Figure 3.4-5.

2. Section 3.4 of the McGuire submittal cites results of an analysis

performed by Duke to ensure that 6 of the 12 upper plenum igniters will

provide adequate coverage of this region. Based on the information

ssupplied, this analysis appears to differ somewhat from one performed by

TVA to justify the use of 16 upper plenum igniters in the McGuire and

Sequoyah plants, provide details and justification of the McGuire

analysis including (1) assumptions regarding vertical and horizontal

velocities of the rising mixture and propagating flame, and (2) the

method for computing the maximum hydrogen concentration for a given

ignition concentration. Address in your response the possibility of the

gas mixture bypassing the igniters. If the analysis relies on

horizontally propagating flames being carried into the upper compartment

by rising gases, discuss the effects of turbulent mixing and dilution of

the mixture with the upper compartment atmosphere.

i

1

7.0-98

__ _- -- -_ -._

. _ _

Response

O The design objective for the igniter spacing in the upper plenum is toensure that the hydrogen concentration will remain below 10% at all

points in the upper plenum when hydrogen is being added at the maximum

rate. The following are the assumptions used in this analysis:

1) Hydrogen burning begins at 8.5% hydrogen and propagates in all

directions.

2) The burning velocity in the upper plenum is one foot /second in all

directions.

3) The flow velocity of the hydrogen, steam, air mixture entering the

upper plenum from the ice bed is one foot /second.

4) The maximum hydrogen addition rate to the upper plenum is 70 lbm/

minute. This is identical to the hydrogen release rate predicted by

MARCH for the S D accident sequence and thus takes no credit for2

O hydrogen removal by burning in the lower compartment.t The igniter spacing used at McGuire (assuming only one train of operable|

igniters) is 50 feet. With this spacing, the burn time in the upper

plenum is 24 seconds, corresponding to a hydrogen consumption rate of 190

| lbm/ min for hydrogen ignition at 8.5% by volume. Thus, the upper plenum

igniter location and spacing presently used at McGuire'results in

hydrogen consumption rates approximately three times greater than the

maximum addition rate. At the time that hydrogen ignition occurs in the

upper plenum, the upper plenum contains 76 lbs of hydrogen (an 8.5%

concentration). During the 24 second burn period, the maximum hydrogen

addition, basec' on the 70 lbs/ minute release rate, is 28 lbm. Since

hydrogen burning causes a pressure increase in the upper plenum, some

unburned hydrogen will be pushed into the upper compartment. Based on

!O,

7.0-99

- -_- -._ . - -_ - _- . _ _ _ . _ _ . ._.

.

A the 12 foot width of the upper plenum, we have estimated that 14 lbm of

hydrogen will be pushed into the upper compartment during the burn

period. This is the basis of our conclusion that the hydrogen

concentration in the upper plenum remains below 10% by volume.

Concerning the effect of igniter spacing on hydrogen bypassing the upper

plenum, it should be noted that whenever the hydrogen concentration in

the upper plenum is less than about 6%, r.o ignition will take place in

the upper plenum, and therefore all hydrogen entering the upper plenum

will pass into the upper compartment. For hydrogen concentrations

greater than 6%, ignition will occur at the upper plenum igniters, and

there will be some flame propagation upward into the upper compartment.

Although until the upper compartment itself reaches approximately 6%

hydrogen, these flame: will extinguish quickly. Once the hydrogen con-

centration in the upper plenum reaches 8.5%, propagation will be down-

ward in the upper plenum, and the quantity of hydrogen able to reach the

upper compartment wil' be reduced. Note that in the CLASIX analysis for

McGuire, it is assumed that no hydrogen is consumed in the upper plenum

until the upper plenum is at 8.5%, thus all hydrogen which escapes the

lower compartment reaches the upper compartment. This amount of hydrogen

is substantially greater than the amount of hydrogen which could be

pushed into the upper compartment during a burn in the upper plenum. It

is concluded that, cased on this analysis, the amount of hydrogen which

bypasses the igniters in the upper plenum during the time that the con-

centration is below 8.5% is much greater than the amount of hydrogen

which could be pushed unburned into the upper compartment during a burn

in the upper plenum. It appears, therefore, that once a high rate of

O7.0-100

- .- . - . .

. - - . _ ___ __ - -

hydrogen consumption has been established by the spacing of hydrogen

igniters approximately 50 feet apart, hydrogen bypassing is no longer a

consideration in the spacing of igniters.

3. Figure 3.4-2 of the McGuire submittal shows a typical igniter circuit but

does not include details of the overall power distribution. In this,

regard provide a more comprehensive description of the power distribution;

to the igniters for each train. Specifically, describe the circuit

branches, starting from the Class IE power supply and proceeding through

I the control switches, fuses and emergency lighting panel board, to the

igniters. Identify in your response: the number of control switches and

the circuits which they control, the location of the fuses in the system,

the number of igniters supplied through each fuse, the means by whichi

' system status can be monitored during an accident, the location in the

; system at which voltage and current readings are taken for surveillance

purposes, and the number of igniters on each tested circuit. Assess the

potential of a short circuit in a single igniter to inhibit operation of

the deliberate ignition system over a critical region of containment.

,

; Response: ,

See revised Section 3.4.

|

|! 4. Verify that the GM glow plug will reliably initiate combustion in a spray

environment typical of the ice condenser upper compartment. Acceptable

| tests for demonstrating igniter operability should be characterized by al

<

O;7.0-101,

. .__ _ _ - . - _ _ _ - ____ -_ __ __ _. _ _ - - _ _ . _ ,- _ _ _ _ _ --

spray droplet density equivalent to that in the upper compartment,

droplets at terminal velocity with induced turbulence to simulate upper

compartment mixing, and a sufficiently uniform mixture such that hollow

cone nozzle effects are eliminated.

Res' mise:4 ,

The McGuire Hydrogen Mitigation System design, described in Chapter 3,

does not directly expose any ignitors to a water spray. However, as part

of the hydrogen research program jointly funded by AEP, Duke, and TVA,

ignitor assemblies either similar or identical to those installed at

McGuire were exposed to a water spray environment. In all cases, the

presence of a water spray did not prevent combustion from occurring.

!

In order to compare the test conditions with the upper compartment

environment under post LOCA conditions, spray densities were calculated.

Assuming both spray trains are operating in the upper compartment and

that the spray is evenly distributed throughout the volume, a spray

density of 0.015 lbm/ft3 was calculated. This value was obtained by

I using a mean droplet diameter of 700p and a flow of 15.2 gpm per nozzle.

the assumed atmosphere temperature was 125*F. A description of the

McGuire containment spray system is found in Section 6.5 of the FSAR.

!

!

The test environment selected for comparison was from the combustion

tests performed by Acurex corporation. See Section 2.6 for a detailed

description of the test program and facility. There were two reasons for

selecting this project for comparison: 1) the ignition source used was

identical to the McGuire ignitor assemblies and 2) the ignitor was

O,|

7.0-102

I_ __

exposed to three different spray environments. Three tests were

conducted with a Spraco 1713 nozzle. Operated at a flowrate of 15 gpm,

the mean droplet diameter is 700p. Thus, by assuming an even spray

distribution throughout the test vessel, a spray density of 0.004 lbm/ft3

was calculated. All three tests were conducted with a continuous

injection of a hydrogen / team mixture.

Ten tests were conducted with a spray header consisting of nine micro-fog

nozzles in the test vessel. Four of these tests consisted of igniting al premixed atmosphere (quiescent tests) and will not be discussed further.

The remaining six tests were conducted with a continuous injection of either

hydrogen or hydrogen / steam. Two tests were conducted with a nozzle AP of

20 psi. This yielded a flow per nozzle of 0.12 gpm and a mean droplet

diameter of 35p. The calculated spray density was 0.039 lbm/ft2

Four tests were conducted with a nozzle AP of 30 psi. This resulted in a

flow per nozzle of 0.16 gpm and a mean droplet diameter of 21p. One of

these tests was conducted with a mixing fan operating. The calculated

spray density was 0.14 lbm/fts,

The following table summarizes the calculations discussed ir, the

preceding paragraphs.

Droplet Diameter Spray Density(microns) (lbm/f t3 )

Upper Compartment 700 .0151

( SPRACO 1713 700 .0041

Micro fcg (20 psi AP) 35 .039

Micro fog (30 psi AP) 21 .140

7.0-103

|

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

The data presented above shows that an ignitor assembly identical to that

installed at McGuire has been exposed to and successfully performed in

combustion tests with calculated spray densities greater than the spray

density calculated fer McGuire. The fact that no ignitors at McGuire are

directly exposed to spray further ensures that operation of the

containment spray system will not adversely affect the operation of the

hydrogen ignitor assemblies.

5. Provide the results of the AECL-Whiteshell combustion test with top

ignition, 8.5% H , and 30% steam, as committed to by response to question2

1C of the September 17, 1982, NRC Request for Information.

Response:

Five additional low hydrogen concentration tests have been conducted by

AECL-Whiteshell. All tests were quiescent with the ignition source

located at the top of the test vessel. A description of the test

facility is presented in the test reports provided in Appendix 2J. The

test results are summarized in the following table:

Hydrogen Concentration Steam Concentration AP(v/o) (v/o) (psi)8.5 15 09.0 15 0.59.0 30 09.5 30 19.58.5* 15 23

The last test was conducted with fans operating. This test is more

representative of the dynamic post-accident conditions than the four'

quiescent tests. The Acurex dynamic test with top ignition yielded AP's

of 13-20 psi. Thus, it has been experimentally demonstrated that ignitor

lociations near the top of a compartment perform successfully in

conditions typical of postuated degraded core accidents.

7.0-104

.

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

-- .

6. Tables summarizing CLASIX results are presented in the McGuire submittal

for only the base case and the flame speed sensitivity case. Provide.

similar tables for all other sensitivity cases analyzed.

Response:

Refer to new tables 4.4-3 thr.ough 4.4-8.

7. With regard to the structural capability of the McGuire containment,

provide the following information:

(a) the ASME Eoiler and Pressure Vessel Code, Service Level C Pressure

Limit for the McGuire steel containment shell.

(b) a brief description of the calculation method and mate. rial

properties used to determine the ASME Boiler and Pressure Vessel

Code Service Level C Pressure Limit for the McGuire steel

containment shell.

(c) the pressure retention capabilities of the penetrations through the

McGuire steel containment shell.i

(d) the pressure capacities of the operating concrete floor fori

! resisting pressures above and below the floow.

Response:

(a) The ASME Service Level C pressure limit for the McGuire Containment

Vessel is 45 psig based on the analysis described below.

'

O7.0-105

) .L

(b) The analysis to determine the McGuire Steel containment structural''

capacity has been described in the Duke Power Company reporti

entitled "An Analysis of Hydrogen Control Measures at McGuire

Nuclear Station Volume 4", dated February 13, 1981, which has been-

previously submitted. Material properties used are those

established by a statistical analysis of mill test reports for

containment plate. The referenced report provides the detailed

analysis methods and material properties. The results of this

analysis are used to establish the ASME Service Level C Pressure

Limit for the McGuire Steel Containment.

'

Using the 1980 Edition of the ASME Code, Section III, Subsection NE,

Service Level C limits are met when the Primary Membrane Stress

Intensity (Tresca stress intensity) reaches the material yield

stress Sy. Using a 98% confidence level, this statistical yield

point has been established to be 42.1 ksi for McGuire Nuclear

Station. Since the stress intensity due to dead load is negligible

compared to that due to internal pressure, and tends to negate the

effects of internal pressure, the contribution of dead load is

conservatively neglected. Inspection of the results of the

containment capacity analysis described above shows that for a

containment internal pressure of 45 psig, Tresca stress intensities

are less than the statistical yield point for containment plate.

For a containment internal pressure of 50 psig, the calculated,

Tresca stress intensity exceeds the statistical yield point for the,

containment plate at isolated points, though yielding has not yet,

begun baset upon the maximua distortion strain energy (Von Mises)'

yield criterion.

7.0-106

__. _ _

.

] (c) All penetrations through the containment shell have been reviewed to)

verify their structural capability. In all cases it was determined

that these penetrations were capable of withstanding a pressure of

at least 67.5 psig.

(d) The calculated differential pressure capacity of the cperating floor

is 41 psid.

8. In response to Item 2 of the NRC letter dated February 10, 1982, Duke has

stated that the basis for the assumptions that the equipment did indeed

reach an equilibriuc temperatura at least equivalent to the MSLB peak

te.mperature (qualification temperature) was engineering judgment. Please

confirm that the judgment and/or analysis was performed for all the

equipment required for the hydrogen burn event. Also, provide the

reference to individual summary component evaluation worksheets (SCEWS)

together with the qualification profile for all the equipment whose

survivability is demonstrated based on the qualification test performed

in accordance with NUREG-0588. For any other equipment which is required

for the hydrogen burn event but is not in the EQ program, provide the

justification for survivability during the hydrogen burn event.

Response:

To establish the temperature response of the cables in question to the

LOCA qualification test chamger environment, a model of the cable

identical to that contained in Section 5.4 was used. This model uses

cable parameters obtained from the vendor. The following assumptions

were used in this analysis:O~i

7.0-107

-- . . .__ _- _

.~ _

!

1. The test chamber is heated using a flow of steam. Accordingly, a

forced convection flow coefficient based on a velocity of 1'

foot /sec. was used. No credit was taken for test chamber preheating

- it was assumed that the cable began the test at ambient

temperature.

2. Heat transfer to the cable was considered only as convection.

Radiation from the test chamber walls and from the steam in the

chamber to the cable were neglected for conservatism. Note that the

radiation contribution to the heat transfer is a significant

component during the early part of the thermal transient and its

neglect is a particularly conservative assumption when temperature

rise of the cable exterior is considered.

O The results of the calculation may be summarized as follows:

1. The Duke supplied RTD Cable (0.810 inches 00, Okonite type FMR) -

Jacket temperature rise was very rapid, reaching thermal equilibrium

within a few seconds of the start of the steam flow. The insulation

and conductor layers under the cable armor come into approximate

thermal equilibrium less than two hours later. The cable remains in

this environment an additional four hours. For the core exit

thermocouple cable, we have shown by previous analysis that its

response to the temperature profile created by hydrogen burning is

very similar to that of the duke supplied RTD cable. Accordingly,

its response during the LOCA qualification test will also be similar

(its total exposure was for eight hours rather tnan six) and

; therefore was not analyzed explicity.

| 7.0-108

. ._- .- - . . . .. . - . - _

2. The Westinghouse supplied RTD cable - because this cable has lowers.)

heat capacity and smaller diameter than the Duke supplied RTD cable,

its response was expected to be much more rapid. The cable reached

approximate thermal equilibrium through its cross section in

approximately one minute. /.dditional exposure to the same

environment continued for nineteen additional minutes.

In order to provide an additional check on this calculation, another

calculation was performed using the unrealistically conservative assump-,

tion that the LOCA qualification test took place in a quiescent chamber

where the only heat transfer to the cable was by natural convection from

still air. The results of this calculation showed that thermal equilib-

rium occurred for the Duke supplied RTD cable after five hours, thus

allowing an additional hour of exposure with the whole cable at the LOCA

qualification temperature. For the core exit thermocouple cable, as

previously justified, similar exposure would permit at least three hours

at the LOCA qualification temperature. For the Westinghouse supplied RTD

cable, the short duration of the test does not permit any conclusion to be

drawn - the cable temperature after 20 minutes reached 70% of its total

rise indicating a time constant in still air of approximately 16 min ~1

However, the justification for survivability of the Westinghouse supplied

RTD cable is based on testing of this cable by TVA in a hydrogen burn

environment as described in section 5 4.2.4.

It is concluded that for reasonable assumptions concerning the test

environment of the cable during the LOCA qualification test, thermal

equilibrium between the cable jacket and the environment is reached very,

,

l

7.0-109

quickly. Therefore, the use of the LOCA qualification temperature as

measured by a thermocouple close to the cable during the test is a reason-

able standard by which to assess cable survivability. It is concluded

that all of the necessary cables will survive hydrogen burning.

With regard to the instrumentation transmitters, the method used to

assess survivability is described in Section 5.4.3. As noted, hydrogen

burning has very little effect on the interior temperature of the

transmitter. The total temperature rise, even if the environment of the

transaitter were to include hydrogen burning (which is does not), would,

be well below any relevant test temperature attained during qualification

testing. If more massive equipment is considered, e.g., an air return

fan, hydrogen burn events ca not transfer sufficient energy to massive

equipment to have a significant effect on the temperature of thed equipment. The effect represents only a small fraction of the total

! temperature rise created by the environmental effects of the LOCA,

including the large steam release. It is concluded that adequate

consideration has been given to all aspects of equipment survivability

and that appropriate action has been taken to ensure operability of all,l

essential equipment.

Summary component evaluation worksheets (SCEWS) were not submitted as

part of the McGuire NUREG-0588 submittal. Environmental conditions to

which equipment was qualified are specified in the appropriate attachments

in the NUREG-0588.

7.0-110

, _ .__ -- _ .__ _ _ __ _ _ . _ _ _ . . _ _

- __ .

I

9. The lists of equipment provided in Section 5.2 of the McGuire submittal,

do not include all essential equipment, e.g., isolation and FORV valves.

In this regard, provide a summary table, such as Table 2.2-1 in

Att.achment 4 to TVA's October 1, 1981 submittal for Sequoyah, and

justification for the survivability treatment of each item. As a

minimum, add the following equipment to the list of the equipment

required for the hydrogen burn event and provide the analysis to

demonstrate the survivability of this equipment during the hydrogen burn

event:

(a) Hydrogen Recombiner

(b) Reactor Vessel Vent Valves

(c) PORV and Block Valves

Response:

In establishing the list of essential equipment, it was determined that

equipment should be included if the answer to either one of the following

questions is yes:'

,

1. Is the equipment essential to achieve and maintain the reactor

coolant system in a safe shutdown condition?

2. Is the equipment essential co maintain containment integrity?

The key word in each question is " essential", implying that the function

could not be performed if that particular equipment were not available.

O7.0-111

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

In the case of the hydrogen recombiner, it was concluded that it is not

essential to operate this equipment in order to maintain containment

integrity. The design of the hydrogen recombiner is based on the

requirements for hydrogen control within the design basis as defined in

10CFR50.44. The total hydrogen release involved is less than 10% of that

considered in the beyond design basis sequence of events which led to the

installation of the hydrogen mitigation system. The hydrogen mitigation

system has been shown to be adequate to protect the containment from the

effects of hydrogen for the beyond design basis event, by consuming

hydrogen at rates far in excess of the capability of the hydrogen

recombiners. The hydrogen recombiners do not make a contribution to the

control of the beyond design basis hydrogen release event and therefore

are not essential for the maintenance of containment integrity.

|

c The second consideration is whether the equipment is essential to achievel'

and maintain safe shutdown and adequate core cooling. The essential

function of any active component on the reactor coolant pressure boundary

in this case is to assume and remain in a closed position so that the

integrity of the reactor coolant pressure boundary is maintained. Fori

that reason, the reactor vessel head vent valves fail closed on loss of

i

power. There are no essential functions of these valves which require

the valves to be opened in oraer to promote natural circulation cooling.

In fact, there are several disadvantages associated with the operation of

these valves, including reduction of system pressure, increasing the

; release rate of steam and hydrogen into the containment, and interferingI

with reactor vessel level monitoring. In summary, it is concluded that

|

7.0-112

.

O the reactor vessel vent valves cannot be considered essential, that theiruse during a period of recovery of core cooling should be avoided, and it

is therefore inappropriate to include them in equipment survivability

consideration:i.

In previous analyses of equipment survivability, it was established that

large, massive objects experienced only small temperature increases due

to hydrogen burning. This is ode to the relatively small amount of

energy available in a given volume when hydrogen combustion occurs at low

concentrations. For example, instrumentation transmitters, with

dimensions in inches and weights of a few pounds, experienced temperature

increases of 20-30 F per hydrogen flame passing over them. Extending

that analysis to the pressurizer PORV or electric hydrogen recombiner,

both of which have dimensions measured in feet and weights on the order

of tens or hundreds of pounds, it is concluded that the temperature rise

due to hydrogen burning would be an insignificant fraction of the total

temperature rise due to the loss of coolant accident. The probability of

hydrogen combustion at the location of the pressurizer PORV in the top of

the pressurizer cavity was examined. The combined action of the lower

compartment igniters burning hydrogen closest to the source and the

hydrogen skimmer fan drawing hydrogen out of the top of the cavity make

it unlikely that hydrogen flames would-initiate in the pressurizer

cavity. Hydrogen flames propagating into the cavity from the lower

compartment would be burning at concentrations around 6% by volume and

therefore would have much less effect than the 8.5% flames previously

considered in equipment survivability analysis. It is concluded that

O7.0-113

.

neither the pressurizer PORV nor the electric hydrogen recombiner, evenV

if considered to be essential equipment, would be adversely affected by

the deliberate ignition of hydrogen in cantainment.

10. In response to Question 10 of the February 10, 1982, NRC Request for

Information, Duke justified exclusion of the T B scenario fromB2consideration on the basis of offsite and onsite power reliability. A

description of the utility grid system was provided to support the Duke

position. To further justify the reliability of AC power, provide the

probability values for the station blackout (total loss of AC power) at

McGuire Nuclear Station lasting longer than 90 minutes, but shorter than

140 minutes.

Response:

Duke Power has not developed a quantitative assessment of the probability

of loss of all AC power at McGuire. The basis of our previous response

was that the compounding of low probability events required to produce

the accident sequence specified by NRC resulted in a probability much

less than that of the events which lead to the degraded core and

production of hydrogen in the first place. Accordingly, additional

failures beyond those which lead to the degraded core are not considered.

in assessing the performance of the hydrogen mitigation system. The

quantitative measure of probability in this case is not necessary to

justify exclusion from consideration of a transient involving loss of all

AC power.

O7.0-114

._ . . . -

11. Provide a description of the procedural instructions for turning on and

turning off the hydrogen igniters to clarify and supplement information

provided in Section 3.7.1 of the McGuire submittal. Your description

should include the following items:

(a) the procedures in which operator actions are required to turn on the

igniters.

(b) the conditions which would require the actions (e.g., any diagnosed

1 loss of coolant, any safety injection actuation signal, only if

inadequate core cooling is diagnosed).

(c) the procedures in which operator actions are required to turn off or

verify the igniters are turned off.

(d) the conditions which would require those actions (e.g., upon

reaching cold shutdown, return to power operation, diagnosed

spurious safety injection).

Response:

Operator action to turn on the igniters is contained in Energency

Procedure 1/A/5000/02, " Loss of Reactor Coolant". This emergency

procedure is initiated by the operator on actuation of Safety injection

if the following conditions indicate that a loss of reactor coolant has

occurred: rapid decrease in pressurizer level and pressure, increase in

containment sump level, change in containment atmosphere thermodynamic

state, containment high activity alarms, etc. Based upon analysis, the

time of actuation will be approximately one hour prior to the first

release of hydrogen to containment under S D conditions.2

O7.0-115

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

.

The operator is directed to take the following action with respect to

deenergizing the igniters. If any indication of inadequate core cooling

exists or has existed, it is required that the hydrogen igniters remain

energized for 24 hours after the establishment of adequate core cooling.

If adequate core cooling has always existed for the duration of the

transient, the hydrogen igniters are deenergized when containment

pressure drops below 0.25 psig. If at any subsequent time indications of

inadequate cere cooling exist, the hydrogen igniters are reenergized.

|

|O,

i

!

|

|

|

O'7.0-116

|

|- . _ . _ - . . - - - _ . _ . - . . - ,

17180-17917180-18017180-18117180-18217180-18317180-18417180-18517180-18617180-18717180-18817180-18917180-19017180-19117180-19217180-19317180-19417180-19517180-19617180-19717180-19817180-19917180-20017180-20117180-20217180-20317180-20417180-20517180-20617180-20717180-20817180-20917180-21017180-21117180-21217180-213