05000341/LER-1986-047, Forwards Requested Addl Info Re 861226 Failure of Div 2 Thermal Hydrogen Recombiner,Reported in LER 86-047-00

Forwards Requested Addl Info Re 861226 Failure of Div 2 Thermal Hydrogen Recombiner,Reported in LER 86-047-00

ML20206A417
Person / Time
Site:	Fermi
Issue date:	04/03/1987
From:	Agosti F DETROIT EDISON CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
VP-NO-87-0046, VP-NO-87-46, NUDOCS 8704080018
Download: ML20206A417 (34)

LER-2086-047, Forwards Requested Addl Info Re 861226 Failure of Div 2 Thermal Hydrogen Recombiner,Reported in LER 86-047-00

Event date:
Report date:
3412086047R00 - NRC Website

text

A '

I/le""pIosI!!nY Nuclear Opmtions Detroit rem fQ s%F

$M hi an 48N (313) 586-4150 Nuclear Operations April 3,1987 VP-NO-87-0046 U. S. Nuclear Regulatory Commission Attn: Document Control Desk Washington, D. C. 20555

References:

1) Fermi 2 NRC Docket No. 50-341 NRC License No. NPF-43

2) LER 86-47-00, dated January 23. 1987

3) LER 86-47-01, dated February 9.1987

4) NRC Inspection Report 87-006, dated February 20. 1987

Subject:

Themal Hydrogen Recombiner Report Reference 2 identified a failure of the Division II Thermal Hydrogen Recombiner. In subsequent telephone conversations, your Mr. John Stefano. NRR Licensing Project Manager-Fermi 2. requested additional information. The requested information in the enclosed report is for your information and use.

If you have any questions regarding this matter, please contact Mr. Frank Sondgeroth at (313) 586-4270.

Sincerely.

~

] hl Enclosures l cc: Mr. A. B. Davis Mr. E. G. Greenman Mr. W. G. Rogers Mr. J. J. Stefano USNRC Region III A

740800g{ g 1

December 1986 Thermal Recombiner Event Introduction The primary basis of combustible gas control at Fermi 2 is containment inerting. The nitrogen inerting system provides and maintains a nitrogen atmosphere inside the primary containment. The Fermi 2 Nitrogen Inerting System is described in FSAR Section 9.3.6 (Attachment 6). An inerted containment is sufficient to assure peak oxygen concentrations below the Regulatory Guide 1.7 combustible gas concentration limit, (General Electric NEDO-22155).

Additional assurance of containment atmosphere integrity has been l provided by the installation of Thermal Hydrogen Recombiners l (Recombiners) to insure that Hydrogen and Oxygen concentrations, in I the primary containment, following a LOCA, will not reach combustible l limits. A purge system that uses the reactor building ventilation system or the Standby Gas Treatment System (SGTS) is also available for back-up hydrogen control.

Thermal Hydrogen Recombiner The Fermi 2 Recombiner system consists of two completely redundant, skid mounted Recombiners and their associated motor control centers.

The Recombiner system was purchased from Rockwell International. No operator action outside the control room or relay room is needed for Recombiner operation.

Each Recombiner consists of a closed-loop piping system, a centrifugal blower with a ecchanical seal, an electrical heater, reaction chamber, a water spray cooler, and a water separator. Details of the Fermi 2 Thermal Recombiner System are described in FSAR Section 6.2.5 (Attachment 7).

A Durametallic Type RA Cartridge Seal is utilized to prevent blower shaft leakage (attachment 1). Cooling water is provided to the blower j

shaft seal (attachment 2 & 3). The inlet cooling water line contains j a strainer and orifice assembly to protect and control flow through the shaft seal. The coolant then returns to the water separator. The orifice is sized to provide a seal coolant flow rate of 1 GPM with an inlet pressure of 10 psig + 5 psi (attachment 2) higher than the gas pressure on the blower seal.

Based on an inlet orifice pressure of 80 psig, Rockwell International determined that an orifice size of 0.082 inches was required to provide a pressure drop of 61.48 psi across the orifice and a 1 GPM I' coolant flow rate to the seal with an inlet pressure of 8.2 psig (attachment 4). The design calculation considers losses between the inlet and outlet piping and across the filter and orifice. The pressure loss across the seal was assumed, by Rockwell, to be raa11 and was not considered in the calculation. Fermi 2 plant data, however, showed the pressure drop across the seal to be approximately 50 percent of the inlet pressure.

Attach:2nt to VP-NO-87-0046 Page 2 The Residual Heat Removal System (RHR) provides coolant to the Recombiner seal at a skid pressure of approximately 190 psig. During pre-operational testing of the Recombiner's the RHR system was unavailable and a test change was implemented to allow use of the Demineralized Water System (Keep Fill) as the Recombiner coolant j source. Rockwell International stated, in 1984, (Attachment 5) that a coolant supply pressure of 100 to 200 psig would provide an adequate safe operating range for the Recombiner spray cooler and seal. Based on the report from Rockwell the Keep Fill system was used during the pre-operational tests, and incorporated into the surveillance procedures.

Event Description During a scheduled functional test of the Division II Recombiner, on December 26, 1986, the recombiner failed to meet the Technical Specification requirements. Technical Specification 3/4.6.6.1 requires, among other things, each Recombiner to attain a temperature of 1150 degrees Fahrenheit within seventy-five (75) minutes and maintained for at least one hour (see comment 1). The test was initiated with the Keep Fill system providing seal coolant (the Keep Fill system pressure is 80 psig at the Recombiner skid). The Recombiner outlet gas temperature rose steadily to approximately 800 degrees Farenheit. While performing the test the RHR pumps were started, for operation in the Torus cooling mode, and the Recombiner outlet gas temperature began to drop.

When the Division II Recombiner heater outlet gas failed to meet the Technical Specification requirements, l oth Division I and Division II Hydrogen Recombiners were declared inoperable. Division I was conservatively declared inoperable pending verification of its operability utilizing the RHR system.

Subsequently, cooling water flow to the Recombiners was reduced to maintain Keep Fill pressure, with the RHR pumps running. The flow reduction was achieved by throttling the Recombiner manual isolation valve in the RHR system. The throttling of the manual isolation valves was an interim action and has since been terminated.

With cooling water pressure reduced to the Keep Fill system pressure both Divisions of Recombiners were demonstrated operable in accordance with the e propriate Technical Specifications. The Division I Recombiner was tested utilizing full RHR pressure and no flow reduction. The Division I test was performed to determine the operability of Division I prior to any modifications. Based on the successful completion of the Division I test, Detroit Edison has concluded that The Division I Thermal Recombiner would have performed its intended safety function.

1 l

l Attechatnt to

! VP-NO-87-0046

! Page 3

. Cause To identify the root causes for the failure of the Division II Recombiner, the seal manufacture, Durametallic, and the Recombiner manufacturer, Rockwell International, were contacted for assistance.

Durametallic was asked to determine the maximum pressure for which the seal would perform its intended function. Rockwell International was requested to review the design of the seal water system to determine how the original orifice was sized and if the original design was capable of operating under the inlet cooling water pressure developed j under full RHR system pressure.

Durametallic determined that the differential pressure between the

blower box pressure and water inlet pressure should be below 39 psig, to prevent seal leakage (Attachment 4). Rockwell International reviewed their orifice calculation and determined the original orifice l size, 0.082 inches, was designed for an inlet water pressure of 80

psig and flow rate of 1 GPM. However, with the inlet pressure

increased to 220 psig, the resultant coolant flow to the seal will be approximately 1.7 GEM with the strainer / orifice absorbing j

approximately 183 psi, resulting in an acceptable presture and flow to r I

the seal.

Field measurements taken at Fermi 2 did not agree with Rockwell's calculation. The plant data showed that with an inlet pressure of 80 ,

psig, the pressure drop across the orifice is approximately 20 psi (Rockwell's calculation ~61 psi), resulting in an inlet pressure to

, the seal of 60 psig, and a seal outlet pressure of approximately 30 l psig. With an inlet pressure of 190 psig the pressure loss across the orifice is 60 psi, as compared viith Rockwell International's l l

calculated value of ~183 psi, utsulting in a seal inlet pressure of 120 psig, and an seal outlet pressure of approximately 60 psig. The 3 measured pressure differential exceeds the Durametallic limit of 39

{ psig indicating probable leakage.

i i Seal water inleakage to the blower can affect the gas outlet tempe rature. When the Keep Fill system is used to provide seal j coolant, the seal inleakage is minimal. When the RHR system is used j inleakage increases and may exceed the heater capacity. The heater is rated at 90 KW. For a design flow 150 CFM at saturated conditions, 50 KW is required to assure adequate recombination of Hydrogen within the Recombiner. The remaining 40 KW is excess heater capacity and I compensates for such items as heater element failure, limited seal j leakage flow and voltage variances.

With the RHR pumps operating, excessive inleakage prevented the l Recombiners from satisfying Technical Specification requirements.

1 The failure of the Recombiner outlet gas to reach the required temperature was caused by excessive cooling water inleakage ,i.e., the l!

vendor supplied orifices did not provide the required pressure drop to reduce the inlet water pressure to be within the seal's design I requirements.

I

Attechasnt to VP-NO-87-0046 Page 4 i

1 Detroit Edison believes the reason the originally designed orifice did not achieve its design pressure drop of 183 psi is because Rockwell International assumed no pressure drop across the seal. Discussions with Durametallic indicated a calculated pressure drop, across the seal, of 0.3 psi. However, the plant data indicated that seal pressure drop is approximately 50 percent of the inlet pressure. The high pressure drop across the seal creates a back pressure on the orifice which in turn reduces the flow and pressure drop across the orifice and results in a higher pressure to the seal. When the pressure increases beyond the maximum seal pressure, leakage may occur. In the Rockwell International calculation the pressure drop i across the seal is assumed to be negligible.

I Corrective Actions To correct the excessive pressure to the seal (120 psig) a new orifice l was fabricated and installed. The new orifice is 0.04 inches. The l new orifice reduced the inlet water pressure from 190 psig to 32 psig

with a flow rate of 0.5 GPM to the seal. After the installation of '

the new orifices, the Recombiners have been satisfactorily tested and

returned to service.

To assure continued operability of the Recombiners, an aggressive j testing program has been implemented. Each Recombiner has been j tested, to approximately 400 degrees Farenheit, every other day for i one week. Each Recombiner will then be tested, to approximately 400 j degrees F, once per week for the next three weeks. Each Recombiner will then be tested to approximately 400 degrees Farenheit monthly until a total of six months has lapsed, at which time the Technical Specification required surveillance testing intervals will resume.

LER 86-047-00 was issued on January 23, 1987, describing the Recombiner failure. LER 860-47-01 was issued in February 9,1987 as an update to LER 86-047-00. Detroit Edison is evaluating this

, situation under 10CFR21. The results of this analysis will be i provided to the Senior Resident Inspector.

l 1

p i i -

wj( hye II j lki lh eh;;- 1

.g b l m ?{ 3 "h g:,  :

j f(e0;;I,i g

j i $

l is ll J!E$C).

L_ ( W'~ J 6 !M in m 4g 'ki- 1 W p'

S .' p '

i 11e p, 1 I  !?

E (m m. i! bdd htid'f E $$h-l e

8 . ,g p,m ~

e ~ , -?

y h 3., je [3 ..

S ,

agg  ; p g?r -

< (in -h - ~s 3

+

( }

a*-

- r e v r-- % i- -

rd

(. !a y j, 3lj ; I j j, I,1, aI 1

/

j

• ;x * ==*\ t F---h-i-- g' 1:

W ]11..I! 'I'

& l ij i . q/._ i I .i .

! 6 "qg ~i}/p]ii d,y tr V ll

] l-----/-i--

il j N; h',

?} g

.{l l -

5 l l

j HQ%:#-"g4 !5 :

<g ..

p$h.l.jjhhd. .. w.

A --

. n.-

j

f. 'n - T-s u gi b

• s't l.I. .---v=-, v 7 .

.rI

/ - t . .. " IYM l lj, I'.ll 4' -

Q3'!l] I, i .

. J' ,

4

.' r f s[ it;ff' L

yt%

/

Q* +

... e

-y- si y..

.. 4 .y f 5!

lj

, 3 g 1

C* ( E' < l . ,  :

i f r -Sm- i' '* S-i 'N d '_M O_h <!D -

E

g- - - - -

~'

(

1% ji y._.;_h] lj gi --f.,

f!f jf g {j -*

Pr 1 L. -

u--i--s 3 I .

T-- r--+J jJ lj

" Jf- AM aa> r- 'hs, - J3 . . ~n=>-

. . 4 t_ . v a  ; k is!

_.'.]........ L. p X m ".

! t [!.mt-St.-- z.

I --

--?- . __

y

,r ,, m m -7c ri A i 4 N --Il k';"E

(( s Ld'l f II --4 r3 j.g gu

-* *e i

N 4!L_w. L- w- wJ r-nN

._J

,, w re ".g'.F

@.-. . 1 j' wf d il' g h;;

s j,<l'l ' .;1.t3) 's gil Ny f-di{t ,

e s .eEN

=

1 g  :', .l q r., ., y y 3

';;:'t.t.ai' "

t  ;

,l A . - s- - .3 *

• , lr. --i{lilp.>3 yh e*

[ 'g j - - . r_1- ._k '

F.5'j" l ..

J!gj,p, ds ;*u 1'

t

=ma iL t

n ib i f 1

. V .

[4 n rt: w! .rN. . gy , 1 ii Q.._\ L ., 7_.1- e A  %-L , .,- -~ zi

.y a .'. .- ... p e  % . . _ _ ..,w- .

.y

" f-$g$ '.5q Et@~r ,r 3- 1 /

li'lllllllll'UllltiWtttttHH , .

\

Attachment 2

. t 1

i .., l.. I l I .,

= 2y

=

25.

2 '4' ps F r's rest waren Fisnes intarr i 3 ,.

~

l C gG p n G 80 958

.s m v a860*

.oTsE1" e oas vassa.as A FAA1". FLO4.s a. howR. ATE .

es asAs. .

3 If -

39 I. 4 ~ '/ 2" si w a s so.svasen

{

• U em SY Z h . c. .

,.2 .DG

• 3 s x - -

N,y \s g

\

A d

-.//,

a_ i

.l--.

W ','

g u

k $ $ .

\ \ j- M '+

s e- a -

j

= - < s. ~ s n ,'

l A W

.. y . 1,, ~ ~

O<  : XN ~+~JW

, fur n-A a

?"

db eeh Aj% h

, n:  : - -

C y a e c  :

e . . - . c.

e, .

a

) -

• a ( ) g i

T .s ne N

8' k l

.h h I l ,

en I

. \ V s A ,.,.==.=s' , . su.tava v.aise .e e se LW I N M x. s 1 x.

y -

3.

-se s..=,e .,s. .se t

_1J

-.r ,.... ..

<. , . e m . a . , . . . .e s s , . . . s . .. . . . .e

.= sei.. .. c s e s a s e. v. sv.e. eave rns ,

i . -o 't .= <. s 1 ' <'

'1

.s. ... v. a vos er is....v. . % Pst? (ts h*to/

os Ps 8

p ast so A ,pensa

g s.a...a...e..

a . as se.. . ... . . .. as.. .evooeras

..... i ..e.s.v. #8. .

._ sumasaa'Yv'E m v. . . . . r c. . s . 3, c, . s_ _ us a gan. u.,ec..s... <._.ri

,,,,,,,.,, p g f wa m y , g.y, an gi, ne.1t o , o ut s.or. M A C.

3 s a m i. . c. r a. c.=,e 3,9 34 s.a . si ..L r... e.e_ erg,

. . . . .i. s c se as

• s e . p. ...r seu ar .., .

s e.s t a =

c . , = . . c. t.

3.

io sva .

sv.

c. e.* a . . <. s - . . s , .a. 4888t _A N W:9' IMIERMMi C^ "E**C ' CAL . 8"*8" Pa'"' "' */*' "o. efu6ADEM 6tho U ~ss Ise.~r ESee, s7 . .- .c , , -

etmaust eseree4Teses v seea s &W nefuesSD~

.. .* - -- ==ca.

) s

-i.e --

.. .u ,6 .

.n ...,,..s .. . . . . o. 3..< .s i. .r.... .

Ese see ' *' 'N

.. .to ~ ~

F_. 5 ** ~.6 .'.*~ 5. ._* ? . .S? a"* _ s.e e-.

a

- ' ' a. _m * -e. S '*- S ' '.:. ._

s e.e s r sas P*e4 **"*.*."."*.. , , , , .

2 .= s_e_a v ._ .__. ., sr e . . ., e_i s...er v.e. _t._1= T _ _

l e= ~as.

o s.., ..<. s - i- .. e. it e . c 3,. . . . . .

to-its a7 aus

". ** *: n ! .. .st e e . . . s a s= r

, l

Attachment 3 .

}

]A],

s.!ll e.$

v:

, .n _

, (fo in O- ~ n , j,] ef' lj -- -

n

, :M .j c .

m L

I l i i i

p$(&

us i 1 V-v -

@i p

di ss .

v OIO , ,

yMO . .

m - . &J

J > [is a

@-' 1f

, .1 j

l ,L .

_A 1 ,' i, .

l  :. ., j ..

1 1 1 jl

!. . ; ii t g <)

> g q -

% 1 1 l t

J@!l .,

t*

N il .

1

l. 97i  :{

yll -} ,

g E@ EP iHiD

l u

,d..e

Pj;-m E 3-0  ;  : .

dg (s .q =

L"'- -- p. 9 --- i 6

.ec.

\ . . e,

.  !!, .c e ih s .-y .n A

_.. 'e(

1

-w in- c3 , , .

- r

@ tw 4

} G. th .ri] 4 .

. m @::-)HO ,g .

HojcR-G l .? 3' t i j

-7 , ,.

' IN j -lhff" .f" cn if

t

,c nu .

i j I -

' k. j I .. .

w 11

.toi, .1 ,

ing g3 ~

!_ :l ; w .c I gf ,

'(- is '

IC ' 9,' 'U 3

-'I -

il 3.if3 a ia '; .

.'- ' /! i ' g i fl

i- '), 'l ys I i6 i'  : I  !

1 1

. ._--- --l

=@[]}s l -w ma -_.q sg i i

, i.

l-

.i.

'g J

f*m

.t /

..  %. If(

is -

l 2 y / __

@,52 ~ -

~ ~ ~ ' '~ ~ -- -

ESGBUILDING102,2NDFLI',II,E, Attachrent 4 o

r 1 of 4 1

Internal Letter i

l Dates February 16, 1987 To: H. Yobs Frems F. Newburn i

635, 055-LB06 635, 055-LB24

Subject:

Fermi !! Recombiner Blower Bearing Seal The calculations made in 1974 to determine the orifice size for cooling the Fermi II recombiner blower bearing meal with

. water are given in Tables 1 and 2 (attached). Table i presents the operating conditions, line lengths and line diameter used to compute the orifice diameters that are given in Tablu 1.

A summary of the parameter values the calculations were based on are as f ollows Parameter Value Water flow 1.0 gpm i

Water supply pressure 30 psig Water temperature 60 to 1BB" Water density 62.4 lb/ft t 60"F 60.4 lb/ft G 199 "F j Inlet line length 3.83 ft Exit line length 4.93 ft Line size 1/4-in DD x .022-in wall Velocity head loss f or fittings 1.5 for both lines l Viscosity 2.736 lb/ft-h G 60 g 0.813 lb/ft-h e 198 F Pipe roughness 0.0018 in/in Table 2 presente the computer run made on the exit line from the bearing seal. The computer run f or the cooling water inlet line has been misplaced or lost. Examination of the delta pressures listed in Table 1, Item B, shows the calculated

! pressure loss in the inlet line is proportional to the equivalent lengths of the two lines. The friction factor in the computer I code is based on the Colebrook equation given in Marks Mechanical l Engineers' Handbook.

1

, _ . _ , _ . - _,.m-_____.-. _ _ _ _ _

P.03 EssguIggy102,2NDFLF.&I.E.

, February 17, 87 Attachment 4 Page 2 2 of 4 A summary of the calculated line Icosas and orifice diameters given in Table 1 are as follows:

Parameter Temperature Value Inlet line pressure loss & F 6.4 psi Exit line pressure loss a F B.2 psi Filter pressure loss 6 F 4.0 psi Orifice delta pressure 6 61.5 psi Orifice diaseter 60,FF 0.0834 in Inlet line pressure loss 18 F 5.8 psi Exit line pressure loss 1 F 7.5 psi Filter pressure loss 1 F 3.6 psi Orifice delta pressure 188 F 63.2 psi Drifice diameter 188 F O.0824 in The calculated water pressure at the gearing seal is as

- follows when the fluid temperature is at 188 F Pressure In Pressure At Recombiner Beal 30 psia 38.2 psia 14.7 psia 22.9 psia

. n Rocky Flats Systems Engineering 1 Attachment (Tabis 1 and Table 2) cc w/ attachment S. A. Itow / I R. M. Mucica K. J. Weston __

f e

2

tsG suILDIN3 102,2NDFL F.8.I.E.

YMELE .E Attachment 4-3of4 ll l i .- .. .

mm. 1 ,,

! Rockwellheemstonal t .

l ~~ .?  :

9 89 l>e/ B A 0 assCet BKAAoN6 , Jt m a' 27 h DIMEN6004S AMD g,MATods sedhtano/s g gepsos __

.m

\

fEGl*Mb**f*1459ffa y1%tJf

( l (p IC$*f

,9(,5 yF4/,8 4/h f I

• gg -4 hatQ 4 f

,sh, b, j< '

) g '

L f t Qnao+ nbr,s* t t i,6)lft ,s*A'Glh N g l efyfy(vts6s,ses/faten glas #

v(

a o $ P!4

.... )k h o,t * ,e :a w l G .nos'J o 1 6* ZhlPUT 1D coMfW13rA CCbK Ahlb C A L G ts L A T R O A P A : c ourss *rr ,

l 1

(se A > f.0 (,+N. /.a i f) (,,do/f 4*Pf, f ( @F) -hi,736 A ,

Mte.f ItV , pr:s k a lof fo r f!N*Nf '

o D A L Ac T r, f*, AP y,,.,

N ' ,0 ** 81V( .511" 48,14 gy l'Y . E 0 to fTI33,

l. s: 6e sE,36 (It gy r,g ,

jpg 60 d $0,y 'I.V5 o

jf ,Re4 , c e

• B)4? dl, , I f' (,e G,3 b t,, >= ,n ..c

• fe fal ~~ '

se Int.$ .

U U

~

C, O M l f t & C 9 / A M E T f 61 CALCI!LAYteW Itf 11.tf gem : ass d,c V'4tr* ger Cr.or<e *ff*** of Fisrods" Vu G!M ' l y , f( # 3 4 * . . ... _

h _# M ", C # ,,

st .se

. . ora t e t* n l $sNP = s,s  : , p .Jeg$8gymto?r<

13,17g 1.W , ,y a s ( p r

• O 0' '9t *f 61 m ,*f EV G 19'$ f Ut. . .

f63.

Y t

Afen w pf* $@ = f jT

• I.36 *J,U. ' AlaY I g M [ '

l .(r..)h

,f ,. pts V , ,y

' ~ '

. s t o :,1,v r- s~s9 &. . .l '5. fit &ltt*f ' ' *'s 3*' -

! I 6kr presserve of 4eero og a 4We't v t a. t 43 1 ye o'a. (t' L fE*g l

. .. o. . . .. .

vwu -

ESS ltuYLWA 8t4PdRLOL FRMts.

_ . _ . . . . . . . _ . - . _ . . . . . . Attachsent 4 __

'['_ 4 of 4 7'4BJ.E E C0hlP 01T R Rua ON UtroT L)NE FRcM vwo TH E BEARING SEAL BN AT - 07 s'05 PN 03 7TY a 04

• • • TODAY (09- 19-74) *** PREV. MAIN 1. FROM 18:00 'l0 Ols 00 HR3 ***

USER I D-= Ritt 3, TY PE SL D OR NEW s RL Ng 'f DPL,1 G . 7,y

/ / DpfTA PRESSURE FeR L19U108' INPUT W. TR Vi sC, N, E e INPUT 00:30 hO 6 FAN W AIe @ 60 I T 499 8,60,2 736,3,.0018 j DI A-IN. A-Fit L-FT M- T I- F RH61 RHe t,-TT 3 DPINW DP P1 P2 MM MF F (IN.3 (PSI) (PSIA) (PSIA) .

A y g #1, INPUTg00910.206,.0009bl45,4.fN3,15,606836,6936

?

996 615 8 18576 psM R.736 13549. 0 000 6 573 0 0405 INPUT 300210 \ APL '8f 1 8 RUNNING TIME: 1 8 SECS 1/9 TIME 1 2 SECS PUNNH DPL10 SK DELTA PRESSURE FOR LIGUIDS INPUT h., TR, V13C,N, E f,, gy gg g / p g 's i

INPU1s00130 l  ? 499.* * = 8 4 56,18 8, . 8139, 2, . 0018 DIA-IN. A-F18 L*F1 K T 1- F RMSI RMS T3 DPINh DP -

Pt Pt KM MF F (IN.) (PSI) ,

(PSIA) (PSIA) ,

INPUT 00210 g y ja m

?.t6,.00$23145,49333,*J5,188,404,604 5 l 206 306 7 44857 psW 0. SIS 44196 0 000 6 163 0 0377 INPUT 00210 A ar#. At

?8 l MUNNING TIME: 1 3 3ECS 1/8 TIME : 1 0 SEC8 l

l SK BYE i, i

. AFF AT,n7 :P -

l e

(~ .

o.n o assen 1

Atomics international Divlelon Energy Systems Group J .

• • • """a7,'nago'*4',,," Roekwell Attachment 5 Canoga Park.Cahforrua 01304 International 1 or 2 Telephone:(213)700 8200 TWX:910 494 1:!37 b. K UM .

w en:ietoi7 gg', is, April 10, 1984 In reply refer to 84ESG-1429 Mr. Gregg R. Overbeck Assistant Superintendent - Startup st Bechtel Power Operations P. O. Box 157 g)dgh Monroe, Michigan 48161 k

Reference:

1. Detroit Edison Letter dated February 2,1984,

" Fermi Recombiner Startup Questions."

2. Telephone Conversation, February 22, 1984, J. Black (D.E.) to W. T. Lee (ESG).

Subject:

Fermi II Hydrogen Recombiners Detroit Edison Company Response to Startup Questions The following information is provided in response to the Reference 1 letter questions.

1. The attached Telephone Conversation Record documents tne discussion between the ESG Material and Processes engineer and the Detroit Edison engineer. No further inquiry has been received to date.

Therefore, this issue is considered resolved at this time.

2. a) Water supply pressure of 100 to 200 psig range with capacity 11 to 25 GPM, respectively, is an adequate safe operating range for the spray cooler and blower seal units.

The resultant flows for the spray cooler and the seal are 10 to 23 GPM and 1 to 1.6 GPM respectively for supply pressure range of 100 to 200 psig and containment pressure of 30 psia.

l

' Attachment 5

'- ' .. 2 of 2 I

B4ESG-1429 Y April 10,1984 Page 2 Rockwell .

International

2. b) The spray nozzle is a patented design. It consists of a spiral full cone spray nozzle insert with a center hole approximately 0.1285 inches diameter. The water from the j spiral anulus is mixed with the center jet to form a full ,

cone. The combined water is reduced in a tapered nozzle  :

discharge section and released through an opening approxi- j mately 1/8-inches in diameter.

3. The high water separator alarm on the control panel annunciator  %

was provided for a specific system as a result of the customers requirement. The recombiner does not require this function provided that the Fenni II installation is designed to prevent the spray cooler water from being turned on unless the recombiner effluent piping isolation valves are open. This may be a procedural step during recombiner operation.

The addition of this function can be accomplished by replacement of the spare window with the reference alarm function by modifying the annunciator logic.

4. a) The annunciator on the Fenni II unit does not have the tempera-ture compensation feature. Therefore, a shift in ambient temperature would result in one (1) to one (1) shift in the annunciation points. It is therefore recomended that an external means of compensating the thermocouple input be provided if such swings are anticipated.

b) The Fermi II Operations and Maintenance Manual, Appendix E, presents the calibration procedure for the annunciator.

c) The DPM is not provided with calibration. The unit is calibrated

• as a part of the annunciator calibration, and as such, its accuracy is dependent on the annunciator output to the DPM.

I Very truly yours,

,' he A .~e v R. J. Cardenas Project Manager Recombiners Enc: Reference 1 and 2.

e 9

. . . - 7 -- m , - . . , - , - - . - . - . - . . - . , , ,_--_----.--w., , - , - - -,--,,,,,w.,,_,--.- r,_mw__,,--,-o. w, <- , ---.~.y,,,-,----,,--...- - - - -

Attachment 6

. EF-2-FSAR 1 f3 9.3.3.4 Tests and Inspections l

The drain lines are all welded, and all required tests for joint l53 R1 soundness were carried out in accordance with applicable codes.

For this reason, the closing field welds are in accessible positions.

. Because spare pumps are installed, no periodic qualifying tests

. need be undertaken. Completed piping will be hydrostatically

tested in the field.

9.3.3.5 Instrumentation 1

- Temperature controls are provided to cool critical sumps by I actuating the flow of sump water through heat exchangers. Each sump is equipped with a high level alarm to signal automatic initiation of the second standby pumps transferring liquid.

This would be indicative of a system leakage.

All reactor building sumps have leak detection instrumentation.

Timers monitor the operation of the sump pumps both for frequency and for length of operation. Leakage is detected by a pump operating before the timers time out or by a pump operating too

long. Leakage is alarmed in the main control room.

9.3.4 Chemical, Volume Control and Liquid Poison Systems l' The only BWR systems that are related to this general class of systems are the standby liquid control system and the RWCU system.

The standby liquid control system is described in

, Subsection 4A.2.3.4 and the RWCU system is described in 50 4

Subsection 5.5.8.

t 9.3.5 Failed Fuel Detection System n'\ In the event of gross rod failure, the increased activity in the coolant would be transferred to the steam and detected by the main steam line radiation monitoring system. Downstream of the steam line monitors are the off-gas radiation monitoring system 54 and the reactor building exhaust vent radiation monitoring system.

l The design bases, system description, safety evaluation, tests and inspections and instrumentation applications for each of these

subsystems are found in Section 11.4. l54 1

9.3-11 Post-OL Rev. 1 - March 1985

1 I

EF-2-FSAR 9.3.6 Nitrogen Inerting System 9.3.6.1 Design Bases , j The Fermi 2 nitrogen inerting system provides and maintains a ni- l trogen atmosphere inside the primary containment and also provides pressurized nitrogen for pneumatic service inside the primary con- l tainment and distribution throughout the plant. The system sche-matic is shown in Figure 9.3-12. l The nitrogen inerting system supply is located outside the reactor building on the south side. The components are shown in Figure 9.3-13. The remainder of the system is located in the reactor building. The nitrogen inerting system receives liquid nitrogen from a commercial source and will supply nitrogen gas at the proper pressure and temperature for inerting the primary containment and for distribution throughout the plant.

The nitrogen inerting system design requirements are the following: ,

a. To provide nitrogen gas at the proper temperature and

. pressure to inert the primary containment to a minimum of 97 percent by volume of nitrogen. The nitrogen gas will be injected into the supply line to the suppres-sion pool and allowed to pass into the drywell through the drywell and suppression pool vacuum breaker valves.

When the primary containment is being inerted, the vac-36 uum breakers will be maintained in the open position.

Mixing of the injected nitrogen will be accomplished by the use of the drywell cooling system (see Subsec-tion 9.4.5). The existing atmosphere will be displaced out through the reactor / auxiliary building ventilation system or through the SGTS.

b. To provide nitrogen makeup for atmospheric leakage out of the primary containment during normal operation and 57l to ensure that a positive pressure is maintained inside the primary containment with respect to the secondary containment. Makeup requirements to some degree will be taken care of by the bleed-off of nitrogen gas from the pneumatic instrumentation inside of the primary containment. Provisions for nitrogen addition to the primary containment atmosphere have been made at the drywell and suppression chamber supply lines through a separate online purge system.

This system controls the pressure of the drywell and torus through vent / makeup of nitrogen.

57l c. To provide nitrogen gas for the pressurized distribu-tion system for the following services:

1. To provide nitrogen to inert the personnel air lock to 97 percent by volume of nitrogen when g 9.3-12 Amandment 57 - May 1984

.Attachrent 6

. . EF-2-FSAR 2 of 3

! personnel access is required into an inerted primary containment.

2. To provide pressurized nitrogen for the pneumatic instrumentation inside the primary containment.

1 During normal operation, nitrogen will be supplied to this instrumentation from the nitrogen inerting system. In the event of a loss of nitrogen supply, i bottled nitrogen will be available for emergency

< use for the pneumatic requirements inside the pri-mary containment. 157

3. To provide pressurized nitrogen to any other re-maining services requiring nitrogen throughout the plant.

Air purging of the primary containment to the breathable limit will be accomplished by the use of the reactor / auxiliary building

. ventilation system or the SGTS.

< 9.3.6.2 System Design 36

! The nitrogen inerting s'ystem primary containment penetrations and the associated isolation valves are classified as ASME,Section III, Class 2. The pneumatic supply system inside primary contain-i ment is classified as ASME,Section III, Class 3. The balance of the nitrogen inerting system pressure vessels are classified as ASME,Section VIII, and the piping is classified as ANSI B31.1.0.

]

The nitrogen inerting system primary containment penetrations and associated isolation valves and the pneumatic supply system inside primary containment are designated as Category I. The remainder of the system is classified as non-Category I.

l I

The nitrogen inerting system has been designed in accordance with

the following criteria: j

! a. Liquid nitrogen requirements are based on the following

p. usage: j

1. To inert the primary containment and personnel air

lock to less than 3 percent by volume of oxygen.

l l

2. To provide additional nitrogen to the primary con-tainment to compensate for leakage.

3. To provide nitrogen for the pressurized distribu- l57 tion system.

l i,

! 9.3-12a Amendment 57 - May 1984 i

- . , - - - . _ , ,=-------..,_n.r_ ,---- - -, - -- ,, . - ---,n , ,c.,- - , - - -,.,..--n._- - - ,, - ..- , _ ,, ,---,n

EF-2-FSAR - -

I b. The inerting and air purging procedures for the pri-mary containment will be completed in approximately 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />. Personnel air-lock inerting and air purging will be completed in approximately 20 minutes.

j 57l

c. The minimum distribution temperature of the nitrogen j gas for all- phases of operation of the nitrogen inert-ing system is controlled. The vaporizing medium

' 57 during the primary containment inerting procedure is

! saturated steam at 15 psig from the plant auxiliary l

i boilers. Heat for the pressurized distribution system l .will be provided electrically.

d. The design capacity of the liquid storage tank is based i on the service requirement of the pressurized distri-bution system for Fermi 2 and the vapor loss from the l

storage tank during the interval of storage. When in-

, . erting of the primary containment is required, the com-mercial delivery vehicle is used as the liquid nitrogen f

57l source.

t 36 e. The receiver usable capacity will be designed to allow a system flowrate of 50 cfm for a period of 5 minutes if the liquid nitrogen source should be out of service.

, A full capacity standby receiver is available.

57 f. The pressurized distribution system is designed to allow connection of bottles as a backup source of j nitrogen, i

j g. The design flow of the nitrogen gas to the primary con-i tainment for the inerting procedure is 3000 SCFM.

i 9.3.6.3 Design Evaluation The system fluid will be commercial, 99 percent pure nitrogen.

l The fluid will not be radioactive. System components for the a handling of liquid nitrogen have been constructed of materials suitable for temperatures of -320*F.

The liquid nitrogen storage tank provides the source of supply for pressurized nitrogen distribution. When inerting of the pri-i mary containment is required , the commercial delivery vehicle will i

be used to simultaneously replenish the liquid nitrogen discharged from the storage tank. The tank is equipped with an ambient and electric expansion coil and a pressure control valve. The expan-  ;

sion coil will transfer heat to the liquid nitrogen to generate  !

saturated nitrogen vapor.

l The nitrogen inerting system has a steam vaporizer and electric l heat exchanger. The steam vaporizer will be used only when )

4 4

i 1

l I 9.3-12b Amendment 57 - May 1984 i

i

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

Attachment 6

, , EF-2-FSAR 3 of 3 nitrogen is required for the inerting of the primary containment.

The electric heat exchanger is used to supply heated nitrogen for pressurized distribution.

The nitrogen receivers provide temporary storage to meet sudden demands for pressurized nitrogen throughout the plant. One re-ceiver will be in full standby to allow maintenance without dis-turbing normal plant operation.

The source of system pressure is the liquid nitrogen storage tank.

l The vapor pressure in the tank will be regulated to provide the required system pressure. All pressure-retaining components of the system are equipped with properly sized pressure relief valves.

Piping that is handling liquid nitrogen has pressure relief valves installed in any segment where liquid nitrogen could become en-trapped between closed valves. All liquid nitrogen transfer lines are sloped upward in the direction of flow to prevent vapor _ pocket buildup at the nitrogen source.

The nitrogen inerting system is not required for the safe shutdown of the reactor, and, hence, is not required to protect the health and safety of the public. However, the continuous operation of the plant is contingent upon the nitrogen inerting system maintaining the required nitrogen atmosphere inside the primary containment.

Therefore, to ensure that nitrogen gas is always available to meet the primary containment nitrogen requirements, small amounts of bottled, high-pressure nitrogen will be stored at the site as a p secondary source of nitrogen supply.

36 9.3.6.4 Tests and Inspections The liquid storage and vaporizing facilities for the nitrogen in-erting system are located outside of the reactor building and are accessible for inspection. The nitrogen receiver tanks and bot-tied nitrogen tanks are located in the reactor building and are accessible for inspection during normal plant operation. Initial system checks, valve operability, instrumentation and control l57

~1oop checks, and alarm setpoints for the nitrogen inerting system

('

will be done in accordance with the preoperational test program as discussed in Chapter 14. The temperature and pressure of ni-trogen delivered by the steam and electric vaporizers will also l57 be determined.

Periodic inspections of receiver tanks will be performed. The inspection of instruments will be made to confirm the actuation of relief valves and alarms.

157 9.3.6.5 Instrumentation Requirements When the primary concainment is being inerted, pressure and tem-perature control will be maintained as follows:

a. A pressure control valve located downstream of the 157 liquid storage tank discharge and the steam vaporizer 9.3-12c Amendment 57 - May 1984

_m

EF-2 FSAR - .

57l automatically maintains a discharge pressure of approximately 30 psig.

' I

b. A temperature indicator is located in the condensate 57 discharge line of the steam vaporizer as is a' low temperature switch which shuts down the nitrogen discharge from the vaporizer at preset temperature.

Pressure and temperature control of the pressurized nitrogen dis-l tribution system will be maintained as follows:

a. A pressure control station located between the liquid 57 storage tank and the electric heat exchanger auto-

-! matically maintains a downstream pressure of approxi-mately 110 psig.

b. A variable setpoint temperature controller on the j 571 discharge side of the electric heat exchanger main-tains a nitrogen discharge temperature.

c. A pressure control station located downstream of the 57 receivers maintains a downstream pressure.

d. The drywell makeup station will sense the pressure of the primary containment and the secondary contain-57l ment and will maintain a positive pressure in the

. primary containment.

e. The provision for a bottle backup ctation will include a manually operated pressure regulator to maintain the 57l receiver pressure when required.

f. A pressure indicator is provided to monitor pressure downstream of the receivers. When the pressure of the 4

'~,

57 receiver in operation reaches a setpoint, an alarm is provided to indicate low receiver pressure.

s The primary isolation valves will automatically isolate on high

"\ drywell pressure, low reactor level, or high reactor building radiation.

9.3.7 ( Deleted . )

- l 1

! 4 1

9.3-12d Amendment 57 - May 1984

Attachment 7 1 of 7 EF-2-FSAR 6.2.5 Primary Containment Combustible Gas Control General Design Criteria 41 of 10 CFR Part 50, Appendix A requires that systems be provided to control the concentration of hydrogen or oxygen and other substances that might potentially be released to the containment atmosphere. Title 10 CFR Part 50, Section 50.44 " Standards for Combustible Gas Control Systems in Light Water-Cooled Power Reactors" establishes the standards for these systems. In Fermi 2, no substances of a combustible nature (other I than hydrogen and oxygen) would potentially be released in signif- 1 icant amounts to the containment atmosphere under LOCA conditions. )

To ensure that containment integrity is not potentially impaired due to buildup of combustible gases following a LOCA, Fermi 2 is equipped with redundant and independent combustible gas control systems (CGCS) that consist of two Seismic Category I, Quality Group B thermal hydrogen recombiners. Each recombiner indiv.id-

~

• ually is capable of limiting the amount of oxygen in the con- l36 tainment atmosphere to less than a combustible concentration in it conformance with Regulatory Guide 1.7, including a conservative

• margin for potential instrument error in monitoring the oxygen l36 concentration. The detail description of the recombiner design is given in Reference 11, " Thermal Hydrogen Recombiner System for Mark I and Mark II Boiling Water Reactors," AI-77-55. A purge system that uses the reactor building ventilation system or the SGTS is available as backup for hydrogen control. The 36 purge system is not required to be a qualified system.

6.2.5.1 Design Bases The bases for the design of the CGCS and related equipment are as follows:

a. The CGCS design conforms to the applicable portions of 10 CFR Part 50, Appendix A and General Design Criteria 41, and to the intent of other related criteria, includ-ing General Design Criteria 42, 43, and 50.

b. The system conforms to the design requirements of 10 CFR Part 50, Section 50.44, " Standards for Combus-tible Gas Control Systems in Light-Water-Cooled Power Reactors."

~

c. The CGCS design conforms to Branch Technical Position CSB 6-2, Reference 12, with specific design bases as follows:

1. The capability to measure hydrogen and oxygen con-36 centrations and to control combustible gas concen-trations in the primary containment is provided.

The CGCS is designed to maintain a safe level of oxygen for the inerted containment with suitable l36 margin for instrument accuracy. The hydrogen /

oxygen analyzer nystem continually measures the hydrogen and oxygen concentrations in the primary 6.2-57 Amendment 36 - June 1981

)

i

EF-2-FSAR -

36 containment atmosphere. Oxygen concentration will be controlled to within five volume percent follow-ing a LOCA as required by Reference 12. Additional }

margin has been provided and actual control will 36l be to within 4.0. volume percent.

58

2. Redundant thermal recombiners are provided to ensure that combustible gases do not accumulate within the primary containment in combustible con-centrations following a postulated LOCA. Indepen-dent and redundant sampling and measuring capabil-ity is provided by the hydrogen / oxygen analyzer system. Neither the CGCS nor the hydrogen / oxygen analyzer system introduces safety problems that might adversely affect containment integrity.

i

} 3. The CGCS and the associated hydrogen / oxygen ana-( lyzer system designs conform to the requirements of an engineered-safety-feature (ESP) system which includes the following:

a. Quality Group B classification in accordance with' Regulatory Guide 1.26, September 1974

b. Seismic Category I requirements of Regulatory Guide 1.29, August 1973

c. Sufficient redundancy and separation to meet the single failure criteria and the require- I 36l ,

ments of IEEE 279-1971

d. Emergency power f rom the essential ac power system upon loss of offsite power l l

e. The capability for periodic testing and inspec- l tion of principal system components to the i

extent practical i s f. Applicable quality assurance requirements.

4. The parameters of Table 1 in Reference 12 and other specified analytical assumptions were used in the calculations that form the basis for the CGCS design.

These assumptions are presented and discussed in detail in Subsection 6.2.5.3.

5. Materials within the containment that would yield hydrogen gas due to corrosion from emergency-cooling or containment-spray solutions were identified and evaluated to be of negligible significance as stated in Subsection 6.2.5.3.1.3.

6. Five times the calculated metal water reaction according to Appendix K of 10 CFR Part 50

)

1 6.2-58 Amendment 58 - July 1984

Attachment 7

' ' 2 of 7 EF-2-FSAR (0.7 wtt), is less than the amount of metal water reaction from 0.23 mils clad penetration, 0.77 wtt.

In accordance with Reference 12, the latter quan-tity is used as the basis of the CGCS design.

d. In the event of a LOCA, operation of the CGCS is man-ually initiated and monitored from the main control room. Initiation of the CGCS can occur anytime after a LOCA up to a maximum time as shown in Figure 6.2-22.

If the oxygen concentration-control limit monitor 36 alarms, the system will be initiated at that time.

e. All components of the CGCS are able to withstand the environmental conditions (pressure, temperature, radia-tion, and humidity) to which they may be exposed dur-ing and subsequent to the occurrence of the postulated LOCA. In addition, components of the CGCS are pro-

. tected against postulated missiles and pipe ruptures I to ensure proper operation.

6.2.5.2 System Design 6.2.5.2.1 Principle of Operation The CGCS is designed to ensure that, under conservative assump-tions in the unlikely event of a LOCA, the concentration of oxy-gen in the primary containment (drywell and torus) is maintained l36 at less than the lower combustible limit of five volume percent with sufficient margin. The five volume percent limit is the maximum acceptable oxygen concentration when hydrogen is present 36 in concentrations greater than six volume percent, as shown in Table 1 of Reference 12. This is accomplished by the controlled t

recombination of hydrogen and oxygen by independent recombiner units located outside the primary containment. The recombiner causes the 2H2+O 2 -- 2H2 O reaction to proceed essentially to completion (greater than 99 percent in about 0.5 second) by heat-ing the gases to about 13000F. Hot gases from the recombiner are cooled to a maximum temperature of 2100F by direct-contact spray cooling prior to being returned to the torus.

The CGCS includes two 100 percent-capacity redundant thermal recombiner and systems, independent sets as shown in of controls Figure and power6.2-24, and two completeInaddition,l58 supplies.

the heater capacity in each recombiner is more than is required for operation, and the thermocouples are redundant.

6.2.5.2.2 Design Features I A piping and instrumentation diagram of a thermal recombiner unit is shown in Figure 6.2-24. All active components of the system are located outside the primary containment and are accessible for inspection and testing during normal reactor operation. The 58 r

6.2-59 Amendment 58 - July 1984 I

l

)

EF-2-FSAR recombiner system consists of three packages: the recombiner skid, the control console, and the power panel. The recombiner skid is a welded steel and stainless-steel gas containment system consisting of the inlet piping, flow meters, flow control valves and enclosed blower assembly, heater section, reaction chamber, direct contact water spray gas cooler, water separator, and the return piping. All pressure-retaining components have been manu-factured, fabricated, and tested in accordance with the require-ments of the ASME Boiler and Pressure Vessel Code,Section III, Class 2, 1971 edition, Summer 1973 addenda. The recombiner skid is connected to the power panel and the control console through instrument and power cables.

Each power panel houses the power distribution components for its recombiner system. The power panel contains the 480-V power sup-ply, heater SCRs, control transforme r, blower motor and valve starters, circuit breakers, and control relays and switches.

Each control console houses all the control equipment f or its recombiner system. The control console contains the instrumenta-tion, annunciators, switches, and lights to facilitate remote operation of the recombiner system.

Each recombiner system is designed to recombine 3 SCFM of oxygen with 6 SCFM of hydrogen diluted in a total gaseous flow of 150 SCFM. Oxygen is the ' minority constituent in the inerted contain-ment and is the controlled parameter. The system inlet flow is 39 60 SCFM of the containment atmosphere, which contains five volume percent oxygen. The recombiner system is designed to operate with process gas of 150 SCFM at 30 psia pressure or less.

6.2.5.2.3 Hydrogen / Oxygen Monitoring Because Fermi 2 has an inerted primary containment atmosphere during reactor operation, the oxygen concentration, in the l event of a LOCA, is the limiting parameter. The hydrogen and I oxygen concentrations are continuously monitored following a LOCA, I and are displayed in the main control room. Subsection 7.3'.8.2.3 i contains a description of the hydrogen / oxygen monitoring system.

l To ensure representative sampling, multiple ports allow gas to be drawn into the monitoring system from several locations in the containment. An alarm indicates when the oxygen concentration reaches a preset level thus signaling the operator to initiate operation of the CGCS.

6.2.5.2.4 System Operation Operation of the CGCS may be manually initiated from either the main control room or the local control cabinet which is accessible under postulated LOCA conditions.

l Af ter opening the drywell inlet and suppression-pool return valves, the CGCS is started and the flow is regulated to 150 SCFM. The electric heaters raise the gas temperature to about ll500F when the hydrogen / oxygen reaction begins spontaneously.

6.2-60 Amendment 58 - July 1984

Attache nt 7 3 of 7 EF 2 FSAR The design power level to the heaters is adequate for the system to reach the operating temperature in less than 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />, and dur-ing this time the system is essentially mixing (diluting) the drywell atmosphere by drawing from the drywell and discharging to the suppression chamber.

The flow field in the reaction chamber is highly turbulent with sufficient mixing to bring the inlet gas temperature to a level at which virtually complete recombination occurs. l The single blowers in each recombiner create the pressure dif-ferential necessary to cause the gas to flow from the drywell through the heater section, to the recombiner, then to the cooler and back to the torus. Sufficient head is provided to overcome the 0.5 psi maximum pressure differential between the drywell and the torus chamber.

Based on the hydrogen generation analysis contained in Subsection  !

6.2.5.3, and recombiner heat-up time, the system should be started up within hours af ter a LOCA at a drywell oxygen concentration per 36 Figure 6.2-22. Several hours following the LOCA, the system has started to recombine hydrogen and oxygen and the drywell oxygen i concentration is kept at less than five volume percent. These l36 l starting times and volume percentages were chosen for additional conservatism and to meet the margin requested by the NRC Standard Review Plan 6.2.5. The post-LOCA hydrogen / oxygen concentrations would be controlled by the recombiner. 36 6.2.5.2.5 Containment Purge '

Containment' purge capability is provided for the purpose of re-moving fission product activity from the containment atmosphere and as backup to hydrogen and pressure control. Piping and valves 136 are provided, connecting the containment atmospheres to the SGTS or reactor building HVAC system as shown in Figure 9.3-12. The purge system is comprised of the large purge piping used for purg- 36 ing and inerting and a smaller online purge system used for nitro-gen vent / makeup and pressure control. Isolation valves and pip-ing at the primary containment boundary meet the requirements of l Section III, ASME Boiler and Pressure Vessel Code, Class 2, and are designed in conformance with Category I requirements. The SGTS treats the containment atmosphere prior to its release to the environment. Because purging is initiated under the reac-tor operator's control, and the effluent from the standby gas treatment system is monitored for radioactivity, the incremental dose at the Low Population Zone during the purging will be con-trolled to ensure that the purge dose does not cause the total dose (LOCA plus purge dose) to exceed the limit specified in 10 CFR Part 100. High radiation monitors prior to the reactor building HVAC exhaust fans isolate the containment purgo valves 36 and initiate the SGTS. The purge / inert valves comply with CSB 6.4 of SRP 6.2.4. For further information, see FSAR Subsections H.II.E.4.2.3(b) and (f) and E.5.042.29. 60 l

6.2-61 Amendment 60 - December 1984

EF-2-FSAR i l

6.2.5.3 Safety Evaluation

)

The safety evaluation of the CGCS is considered in terms of the design of the system to perform its intended function. This eval-untion depends on the hydrogen generation analysis in the primary containment So11owing the postulated LOCA and on the design of the CGCS to function.

6.2.5.3.1 Rydrogen Generation Analysis In establishing the design and assessing the capability of the CGCS, an analysis was performed to determine the primary contain-361 ment (drywell and torus) hydrogen and oxygen concentrations as a function of time following the postulated LOCA. The analysis was performed in accordance with the assumptions and criteria provided in Reference 12. The results of the analysis indicate 36 that hydrogen and oxygen can be safely and effectively controlled "Y by the CGC8 to the limit of Table 1 in Reference 12.

• Plant operating procedures will prohibit recombiner initiation un-

' til containment pressure is less than 30 psia. In addition, plant 36l Procedures will indicate when the recombiner needs to be initiated.

6.2.5.3.1.1 General Following a postulated LOCA, hydrogen gas may be generated from the following sources: )

a. ' Metal-water reaction involving the zirconium fuel clad-ding and the reactor coolant

b. Radiolytic decomposition of the post-accident emer-gency cooling solutions

c. Corrosion of containment materials by solutions used for emergency cooling or containment spray.

6.2.5.3.1.2 Assumptions The following assumptions, consistent with the requirements of

. Reference 12, were made in assessing the hydrogen generation and subsequent primary containment concentrations following the postu-lated LOCA:

a. Conditions and Data

1. Drywell Free volume = 163,730 ft3 Pressure = Figures 6.2-11, 13, of the FSAR I

Temperature = Figures 6.2-12, 14, of the FSAR 6.2-62 Amendment 36 - June 1981

Attachment 7

^

'. 4 of 7 EF-2-FSAR i

2. Torus )

1 Free volume (min) = 130,900 ft3 i

Preesure = Figures 6.2-11, 13, of the FSAR l Temperature (air) = Figures 6.2-12, 15, of the l FSAR l 3. Fuel parameters' (8x8 with 2 water rods pre-pressurized) i Number of fuel pins = 47,368 1

) Active fuel pin length = 150 inches l .

j Fuel pin outside diameter = 0.483 inches Cladding thickness = 0.032 inches l2'7

4. Reactor thermal power = 3,430 MWth

) b. A metal-water reaction of 0.77 wtt (i.e., 0.23 mils I cladding penetration depth) was assumed. Hydrogen j

evolved from the metal-water reaction was considered

) to be released to the drywell within two minutes fol-l lowing the postulated LOCA.

}' c." A total of 16 moles of gases present in the drywell and suppression pool is based on the temperature and pressure responses in Figures 6.2-11 to 6.2-15, Case "a," and the ideal gas law.

! d. The valves given in Table 1 of Reference 12 were used j in the analysis.

i.

j '- e. The postulated released fission products that are inti-

mately mixed with the coolant are assumed to be swept out of the core as the core cooling water exits the

.! break and flow through the containment downcomers to l the torus.

) f. Hydrogen released to the drywell and torus atmospheres j

is considered to mix homogeneously with the gases present.

l l

) g. An operating control limit of 4.0 volume percent has '

! been established for the oxygen concentration in the  !

! drywell or torus. The margin of 1.0 volume percent 36 l l allows for conservatism to ensure that the five volume l percent oxygen limit will not be exceeded.

l

' )

\

h. Recombiner flow rate is 150 SCFM. I l

6.2-63 Amendment 36 - June 1961 i

i

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

EF-2-FSAR ,

l

i. Recombiner heatup time is 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />. During this heatup time, the recombiner blower is assumed to be operating, that is drawing suction from the drywell I

{

, and discharging to the torus atmosphere.

l j. Recombiner efficiency for removing hydrogen in the presence of sufficient amounts of oxygen is assumed to be 100 percent.

6.2.5.3.1.3 Analysis and Results Sources The significant sources of hydrogen following the design basis LOCA have been considered in detail and the hydrogen generated from each of these sources has been determined.

- a. Metal Water Reaction ,

As a result of a LOCA, fuel cladding temperatures begin to rise beginning after blowdown and continuing until the core is reflooded. Zirconium will react with steam according to the following reaction:

E r + 2H20 --> Z rO2 + 2H2 (6.2.19)

Thus, for each mole of zirconium which reacts with steam, 2 moles of free hydrogen are produced. The }

extent of the metal-water reaction and associated hydro-

. gen generated depends strongly on the course of events assumed for the accident and on the effectiveness of emergency systems. Reference 12 conservatively stip-ulates that the amount of hydrogen assumed to be gen- i erated by metal-water reaction in determining the performance requirements for the CGCS should be five times the maximum amount calculated.in accordance with 10 CFR Part 50, Section 50.46, but no less than the amount that would result from the reaction of all the metal in the outside surfaces of the cladding cylinders surrounding the fuel (excluding the cladding surround-ing the plenum volume), to a depth of .23 mils.

The metal-water reaction calculated in accordance with 10 CFR Part 50, Appendix K for the Fermi-2 8x8R fuel design is 0.14 wtt. Five times this calculated value is 0.7 wtt. Based on the GE 8x8R fuel design, a teac-tion that results in 0.23 m13s cladding penetration depth is equivalent to 0.77 wtt. Therefore, because the metal-water reaction based on five times the calcu-lated value in accordance with Appendix K is less than l the 0.23 mils penetration, the latter reaction was as-sumed as a basis for determining the amount of hydrogen generated by the me,tal-water reaction.

) l l

6.2-64 Amendment 20 - March 1979 i

Attachrent 7 5 of 7 EF-2-FSAR The analysis assumes that hydrogen generation from the metal-water reaction begins at the end of blowdown and continues until core reflood. The hydrogen, thus evolved, is assumed to mix homogeneously within the drywell.

b. Radiolysis The post-accident fission product release specified in Table 1 of Reference 12 was assumed in calculating the hydrogen generation due to radiolysis. Of the fuel equilibrium fission product inventories, all of the noble gases were considered to be released to the dry-well atmosphere, 50 percent of the halogens and one percent of the solids were considered to be released to the coolant, and the remaining fission products were considered to remain in the core. Except for the noble gases, the released fission products are assumed to be swept out of the core as the core cooling, water exits the break and is carried over to the torus through the downcomers.

Hydrogen produced by radiolysis from coolant adjacent to the core was assumed to evolve to the drywell atmo-sphere. Hydrogen produced by radiolysis from fission products initially mixed with the suppression-pool water was assumed to evolve to the torus atmosphere.

The noble gases released to the drywell atmosphere pro-duced insignificant quantities of hydrogen and were not considered as a significant source.

The decay energy release rates were calculated using the curve fit equations and methods given in Refer-ence 12.

Gamma energy from in-core fission products attenuated by a factor of 10 was used to calculate the radiolysis i of the coolant adjacent to the core. Unattenuated gamma and beta energy from the released halogens and solids was used to calculate radiolysis of the sup-pression-pool water. In all cases a constant yield value of G(H 2) = 0.5 molecules /100 eV absorbed (from Reference 17 Table 1), was used in the calculation of radiolysis.

. c. Corrosion of Containment Materials The corrosion of containment materials was considered as a potential source of hydrogen. The corrosion of aluminum, zinc, and zine-base paints located either in the drywell or torus was evaluated for a potential source of hydrogen. It was determined that these po-tential sources were insignificant for the following reasons:

6.2'-65 Amendment 20 - March 1979

EF-2-FSAR - -

s

1. The containment spray solution, if used, does not contain any chemical additives. The pH of the spray solution is 6.5 to 7.0.

I

2. Aluminum corrosion is highly pH dependent. The

, Oak Ridge National Laboratory (ORNL) experiments described in Reference 13 have determined that at high pH (approximately 9.3), the corrosion of aluminum was about 100 times greater than at a pH 6.5 to 7.5, which was shown to be negligible.

t

3. Although the corrosion of zinc does not exhibit I the same pH dependence as aluminum, the corrosion I of both sinc and aluminum is highly temperature-dependent. The post-LOCA time / temperature profile in the drywell and torus is much less severe than that experienced in typical PWR containments. The

. magnitude, as well as duration of elevated tempera-

, ture, is short-lived as shown in Figures 6.2-12, 6.2-14, and 6.2-15.

Because of these reasons, the corrosion of aluini-num and zine is relatively insignificant and does not represent a significant source of hydrogen.

Time-Dependent Hydrogen Concentration The hydrogen concent'ation r in the drywell rises quickly due to the generation of hydrogen from the postulated metal-water reac-tion which is assumed to occur within two minutes af ter the blow-down ends. 'As shown in Figure 6.2-25, the hydrogen concentration in both the drywell and torus gradually increases. The oxygen concentration is the controlled constituent with the inerted containment. In order to allow for a margin, a conservative 36 control limit of four volume percent has been established for the oxygen concentration. As shown in Figure 6.2-25, the hydro-gen concentration reaches 4.5 volume percent in the drywell at

, about four hours following a LOCA. The recombiner requires a 1

I 6.2-66 Amendment 59 - September 1984

Attachment 7 '

• 6 of 7 EF-2-FSAR l

! 1.5 hour5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> heatup period and the CGCS will be activated at four 36 i

hours when the oxygen concentration is about 3.8 and 4.1 percent in the drywell and torus, respectively. No credit has been taken I i

for recombination until the heatup period has elapsed, but credit

! is taken for the blower operation on the hydrogen concentration.

1 Figures 6.2-22 and 6.2-26 show the ability of the CGCS to control l the oxygen and hydrogen concentration in both the drywell and 36 l j torus.

i

! Model i

A two-region model was used to determine the time-dependent hydro-j gen concentration in the drywell and torus atmosphere. The in-core and suppression pool radiolytic decomposition was calculated

using the equations and methods given in Reference 12. All hydro-l gen from the metal-water reaction and core radiolysis evolves to

, the drywell and all hydrogen from sump radiolysis evolves to the

torus.

4

! For the calculation of hydrogen concentrations, a total of 16

' moles of gases in the drywell and torus are determined from the temperatures and pressures shown in Figures 6.2-11 through 6.2-15.

Case "a" was used because it is the worst case as far as hydrogen

concentration is concerned. It results in the least amount of j dilutent. Credit is taken for recombiner flow during the neces- ,

sary 1.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> heatup period, but no recombination is assumed to occur.

I After the heatup period has expired, recombination is assumed to i occur. Because sufficient amounts of oxygen are available for i the hydrogen to recombine, the recombiner is considered 100 per-

! cent efficient. Therefore, it is assumed that there is essen-1 tially no hydrogen being returned to the torus by the recombiner j after heatup.

1 i 6.2.5.4 Testing and Inspection s

i' i

The thermal recombiner system is based directly on a demonstration test system which was operated at Atomics International Test Facil-1 ity. This test was performed on a similar, but lower capacity l thermal recombiner than the Fermi 2 unit. Results of these tests

] have been completed and reported in Reference 11.

Before the thermal recombiner units were shipped, the skid-mounted

} units were installed at Atomics International. A system accept-4 ance test was performed on each unit to test the complete syht.em and demonstrate proper mechanical, thermal, and control system performance with hydrogen input for hydrogen recombination.

Edison joined wi

' capacity (150 f tgh three

/ min) other utilities recombiner. to run Results teststests of these on awere higher reported in Reference 11.

i l36 i

l 6.2-67 Amendment 36 - June 1981 I

e

EF 2 FSAR -

The thermal recombiner system active components are capable of be-ing tested under simulated operating conditions. (The recombiner system manufacturer - Atomics International Division of North Amer-ican Rockwell - recommends full system testing at a frequency no greater than once every two months to avoid excessive and unneces-sa ry thermal cycling. )

All active components of the redundant thermal recombiner units of the CGCS are located external to the primary containment, and are accessible for inspection during normal operation. The sys t em testability and test frequency is adequate to ensure that the CGCS will operate and function properly in the unlikely event of a LOCA that might require its use. Tests are itemized in the Technical Specifications.

Preoperational tests of the CGCS are conducted during the' final I stages of plant construction prior to initial startup. These tests ensure correct functioning of all controls, instrumenta-tion, recombiners, piping, and valves. System reference char-acteristics, such as pressure dif ferentials and flowrates, are documented during the preoperational tests and are used as base points for measurements in subsequent operational tests.

l 6.2.5.5 Instrumentation Requirements '

The automatic sequence of operation of the CGCS is manually initi-ated from the main control room. In addition, initiation, moni-toring, and control functions may be exercised at the local control panel which is also accessible under postulated LOCA conditions.

In the event of a LOCA, the oxygen and hydrogen concentrations are continuously displayed in the main control room, and should the 58l oxygen concentration reach a preset limit, an alarm alerts the operator to the necessity of initiating operation of the CGCS.

Thereafter, the operator alternately selects the atmosphere (d ry-well or wetwell) from which suction to the recombiner is taken, on the basis of the concentrations actually existing in the con-tainment atmospheres. Meters and lighted indicators provide the necessary information for the operator to determine the operating 58l condition and performance (temperatures, flows, and valve posi-tions) of the recombiner units. For a description of the inctru-mentation and functional requirements, see Subsection 7.3.E. The redundant hydrogen / oxygen monitoring system is described in Sub-section.7.3.8.2.3. The recombiner does not require any instru-j mentation inside the primary containment for proper operation after a LOCA.

6.2.5.6 Materials There are no materials in the CGCS subject to radiolytic or pyro-lytic decomposition under the conditions that would exist following a postulated LOCA. The principal materials used are

a. The heated components forming the containment boundary of the system are type 304 (or equivalent) stainless steel in i i

6.2-68 Amendment 58 - July 1984

Attacha nt 7 7 of 7 EF-2-FSAR accordance with the appropriate ASME material specifica-tions, and Section III, Class 2 requirements

, b. Unheated components forming the containment boundary i

conform with Section III, Class 2 of the ASME Code.

Carbon steel, per SA 106, Grade B or SA 333, Grade 6, is used for piping, SA 216 for castings, and code

allowable carbon steels for plate, forgings, weld rod, i

and other components, as appropriate

c. Heated structural components with design temperatures <

' in the 15000F range are type 316 stainless steel, meet-

)

ing commercial requirements I d. Heater sheath materials are alloys of Inconel or Incoloy i

e. Shell and structural frame materials are carbon steel, ,y f* commercial grade .

I i f. Insulation within the steel shell of the recombiner package external to the recombination chamber is B&W Kaowool with hardener, H. T. Banroc, or Johns-Manville ceraform

g. A Schutte & Koerting Co. direct-contact water spray gas-cooler is associated with each recombiner unit

h. Inco'nel-clad Type K thermocouples (Mgo insulated) are used.

Generally, the materials used in other containment systems are the same as those used in the CGCS. Inconel is the only material unique to the CGCS that is exposed to the containment atmosphere gases. None of these materials or components decompose radiolyti-cally or pyrolytically under the post-LOCA conditions.

l  !

i

.i I

l I i 1

l l

j 6.2-69 Amendment 36 - June 1981 l

)

1  !

EF 2 FSAR i 1,

6.2 CONTAINMENT SYSTEMS I

REFERENCES

! 1. Joint G. E. and PG&E Report, Pressure Suppression Test Pro-gram, Appendix I, Preliminary Hazard Summary Report, Bodega Bay Atomic Park, Unit 1, December 1962.

2. "The General Electric Pressure Suppression Containment Ana-l lytical Model," NEDO-10320, General Electric Company, San l Jose, California, April 1971. ,

3. F. J. Moody, Maximum Flow Rate of a Single Component Two-Phase

) Mixture, Journal of Heat Transfer, 8_7, p. 134.

I 4. J. D. Duncan and J. E. Leo.ard, Emergency Cooling in BWRs Under Simulated Loss-of-Coolant (BWR FLECHT Final Report),

j

• GEAP-13197, General Electric Company, June 1971.

! 5. B. C. Slifer, Loss-of-Coolant Accident and Emergency Core l Cooling Models in General Electric Boiling Water Reactors,  ;

I NEDO-10329, General Electric Company, Nuclear Energy i

{ Division, April 1971.

6. D. P. Siegwarth, M. Siegler, " Detroit Edison Standby Gad

Treatment System Gasketless Filter Test Series," NEDC-12431, i Class I, General Electric Company, January 30, 1974. r

}

7. J. O. Henrie, Thermal Recombiner Demonstration Test, Report AI-72-61, Atomics International Division of North American

• Rockwell, October 25, 1972.

)

I 8. "Enrico Fermi Atomic Power Plant, Unit 2, Plant Unique Anal-54l 45 ysis Report," DET-04-028-1, 2, 3, 4, 5, Nuclear Technology

! Incorporated, San Jose, California, April 1982.

t l 9. D. W. Pyatt, Enrico Fermi 2 Reactor Vessel - Sacrificial 1- Shield Annulus Pressurization Analysis, Report NUS-3129, i j March 1978.

i j 10. Structural Design Assessment for Safe-End Break Enrico Fermi Atomic Power Plant - Unit 2, Report SL-3647, Revision 2, 29 March 14, 1980.

! 11. J. O. Henrie and S. A. Itow, Thermal Hydrogen Recombiner Sys-l tem for Mark I and II Boiling Water Reactors, Report AI a 55, Atomics International Division of North American Rockwell, j September 12, 1977. ,

t

12. USNRC, " Control of Combustible Gas Concentrations in Contain-ment Following a Loss-of-Coolant Accident," Branch Technical i

l 36 Position CSB 6-2, Regulatory Standard Review Plan, Section i 6.2.5, March 1975.

1. )

i i

1 6.2-70 Amendment 54 - March 1984 1

l I _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . _ _ . _ _