ML20098F547

Responds to IE Bulletin 84-03 Requesting Evaluation of Potential for & Consequences of Refueling Cavity Water Seal Failure Prior to Beginning Refueling.Precautions Added to Fuel Handling Procedures

ML20098F547
Person / Time
Site:	Byron, Braidwood, 05000000
Issue date:	09/27/1984
From:	George Alexander COMMONWEALTH EDISON CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
9245N, IEB-84-03, IEB-84-3, NUDOCS 8410030110
Download: ML20098F547 (10)

Text

. / Commonwealth Edison

( one First National Plan, Chicago. Illinois

( .7 Addr ss R: ply t;; Post Offic2 Box 767 j Chicago. Illinois 60690 September 27, 1984 Mr. James G. Keppler Regional Administrator U.S. Nuclear Regulatory Commission Region III 799 Roosevelt Road Glen Ellyn, IL 60137 Sub ject: Byron and Braidwood Stations Units 2 and 3 Response to IE Bulletin 84-03 NRC Docket Nos. 50-454/455 and 50-456/457 Reference (a): IE Bul'.etin 84 R. C. DeYoung lecter to All OLs and cps dated August 24. 1984

Dear Mr. Keppler:

Reference (a) requested us to evaluate the potential for and consequences of a refueling cavity water seal failure prior to beginning refueling. Attached is the response for Byron and Braidwood Stations.

To the best of my knowledge and belief the statements contained in the Attachment are true and correct. In some respects these statements are not based on my personal knowledge but upon information furnished by other Commonwealth Edison employees, contractor employees and consultants. Such information has been reviewed in accordance with Company practice and I believe it to be reliable.

Please address any questions that you or your staff may have concerning our response to this office, ga%B8aa 583%g G

(s/

.it

g One (1) signed original with Attachment is being sent

.directly to the USNRC. Document Contrcl Desk in Washington, DC for reproduction and distribution as requested in the Bulletin.

Respectfully, Q t, t.M . v .

G. L. Alex nder Nuclear Licensing Administrator Attachment

.cc: US NRC,: Document Control Desk Washington,.DC 20555 RIII Inspector - BY/8W SUBSCRIBED and SWORN to beforp 'me this 47dl, day of ifrrrn m f_we) , 1984

- MInbjA eks Notary;Public 4

9245N

- . . , , -1 1

, l 1

i I

l i

ATTACHMENT 1 RESPONSE TO.IE. BULLETIN 84-03 ,

REFUELING-CAVITY WATER SEAL BYRON AND BRAIDWOOD STATIONS P

e

, 4 4

i 4

}-

4 i'

t 9

1 t

1 i

,w- c- eme--- ,we,...tes*www w avam-4w . r+v , h are --rw - + e m-+v wrv w o e www -e-*7 w- y ww- mww. . . , , , . . - , = w*yn wwy+-g y*w mwepigy e v c--Wy 1

9 Evaluation of Byron /Braidwood Refueling Cavity Water Seal Summary and Conclusions Important design differences between the Byron /Braidwood refueling cavity seal and the Haddam Neck seal make a loss-of-seal integrity event at Byron /Braidwood significantly less likely to occur.

Discussions with the seal supplier and the Hadoam Neck staff indicate the cause of the seal failure to be misalignment and/or overinflationlof the seal bladder, or improper design tolerances.

Based upon these discussions and the evidence to date, it is our opinion that overinflation of the seal to the pressure reported at Haddam Neck without a corresponding misalignment or out-of-tolerance problem.with the seal would not cause a seal failure at Byron /Braidwood. Since the Byron /Braidwood design does not require alignment of any sort, we conclude that occurence of this event at Byron /Braidwood remains very unlikely.

Although very unlikely to occur,_this event has potentially serious consequences. Therefore we are adding precautions to the station fuel handling procedures to ensure that the seal is inflated to an acceptable pressure range. The seal dimensions and the reactor vessel / refueling cavity floor gap tolerances will be checked against design values, and the design will be reviewed with the supplier.

Unless further information from the_Haddam Neck event warrants it, neither we nor the seal supplier are recommending any design changes to the Byron /Braidwood seal at this time.

Evaluation Seal Design

-The Byron /Braidwood refueling cavity seal consists of a single split-joint type PRS 585 inflatable rubber ring supplied by the Presway Corporation. Figure 1 shows a cross section of the Byron /Braidwood seal. Figure 2 shows a cross section of the Haddam Neck-seal arrangement. The ma jor . dif ference between the two designs is the steel ring in the HN seal which requires that two inflatable rubber seals be in place on either side of the ring. The ring is about one foot wide with a two inch gap on each side of it. The ring is supported by nine hangers (strongbacks) arranged around the reactor vessel flange. It is reported that this_ ring can be displaced horizontally, (there is no alignment fixture for the seal ring) thus changing the width of the gaps on each side of the ring.

It is believed by Presway that this actually happened and contributed to failure of the seal. The Byron /Braidwood seal consists of a single rubber seal, similar to one of the Haddam Neck seals, but with a fixed, unadjustable gap width. The problem of misalignment therefore does not exist at Byron /Braidwood.

Other important differences between the Byron /Braidwood seal and the Haddam Neck seal are summarized in Table 1.

O

I ?;

il p -

Table 1 Byron /Braidwood vs. Haddam Neck Refueling Cavity Seal Design 1

'Desian Feature . Byron /Braidwood Haddam Neck 4

Supplier _

Presway Corp. Presway Corp.

Seal top. width 4 inches 3 1/2 inches

, :Overall.sep height

Gaswidtht}) 6 1/2' inches 5 1/2 inches 1 13/16 inches 2 inches (each gap)

U LSeal top hardness (21 60 40 Inflation pressure (2) 35 psig. 30-35 psig

'.(1) ?There'are two gaps in the Haddam Neck seal. arrangement a shown in Figure 2.-

> (2) _The durometer readings of the seal top are.60 and 40 for the i' -Byron /Brsidwood and Haddam Neck. seals respectively.

(3) LThe Haddam Neck seal was reportedly inflated to about.40 psig at the

' time of failure.

i I

These dimensions-are all in.the conservative direction against failure of  !

_the Byron /Braidwood_ designs in.the event of overinflation. Overinflation

-of theEseal is thought to cause failure by-causing the seal top to buckle '

downward. This effect combined with the water pressure from the filled n .cavityaforces'the seal through the cavity floor gap.' .The Byron /Braidwood Hv sealttop is made of harder rubber and is wider than- the Haddam Neck 1- design'~and therefore,less susceptable to buckling. Also, the gaps at

-Byron /Braidwood are narrower than.the Haddam Neck gaps, and more

. importantly, - not adjustable.

Emeroency Procedures Emergency procedures exist at Byron for a level loss'in both the '

! refueling cavity and the spent fuel pool, references 3-and 4. The-procedures require that-on rece'ipt of _ indications of level loss, which includes level indication in the control-room, that fuel in transit or in H the RCCA change fixtures be inserted in the reactor vessel or the spent

. fuel' pool, whichever action takes the least amount of time. Isolation of

, 'the spent fuel 1 pool from the refueling cavity is'then performed by

[ closing the sluice. gate and the transfer gate valve. This sequence of actions'will minimize the level decrease rate since the two pools are i

- Econnected until the fuel is in safe locations. , Makeup water can then be-supplied.to the reactor cavity-from the RH System or the FC system if

. necessary. There-are three sources of makeup water available to the

_  : spent; fuel'ppol_ including a borated Safety Category I source, a Category LI unborated fire protection source, and a non-Category I unborated water source. Either of the-three sources can be'used for pool cooling while maintaining a kefffof less than 0.95, (Ref. 6). The spent fuel pool i-could be filled at a rate of about 4 ' inches per minute if 'necessary, with a total. time of just:over one hour to fill it from the top of the active 1

fuel to normal'1evel, using the fire protection system.

, a a ,- w . a -- ~ - . - - . - - . - - , .. . -- - w

. - . -. . - - - - . - - . . - . . . - - ~ ~ . - - - -

w . t

_4_ _

1 ,

Seal' Failures? >

I, LLoss'of Air (The1 air' supply,to the refueling cavity seal at: Byron /Braidwood consists

of:a dedicateo nitrogen. supply so that loss of one of the station air Esupplies will;not. affect?the seal. If the air supply is lost, no-leakage ,

lis expected because the air is essentially a redundant-sealing. mechanism -l

.to the water pressure from the filled reactor cavity.

~

If leakage does occur in'this event is would be well within makeup capacities and.more than adequate. time would be available to return irradiated fuel asemblies

'to a safe location.

Gross Seal' Failure 7 Because the seal is. installed and the refueling cavity is filled long  !

before irradiated fuel is actually moved (on the order of days), any ~

serious problem with'the seal is expected to occur before the potential

, forndamaging fuel exists. This was the case at Haddam Neck.

If a~ gross seal failure did occur'during fuel. movement, from an

~

~

unspecified.cause-that was on the order of the Haddam Neck event, we do

- not' expect to sustain any. fuel. damage. The probable sequence of events is'as follows.- A -leak on the1 order of 10,000. gpm occurs which begins to

drain the refueling cavity /SFP system at the rate of about 5 inches per'

minute. Refuelingjprocedures would allow a maximum of -three fuel  ;

assemblies out of storage locations in the reactor vessel and the SFP at

.i-

.any-one time. =There could be two assemblies in the RCCA change fixture,

and one in the tilt machine.- The top of the active fuel in the RCCA change fixture is approximately-one foot above the reactor vessel flange.' .Given this leakage rate it would conservatively take 50' minutes to reach a -level 2. -ft. abcve the top of the fuel.. This is more than adequate time-to return-the three assemblies to safe locations in the

reactor vessel.and the SFP, even if no action were taken to begin  ;

'i

[ supplying makeup water to the pool.

t -If~the fuel for some reason was not removed from the RCCA change fixture, ,

4 and.noiaction was-taken to isolate the SFP and' supply makeup water, the-water. level would reach.its minimum level-about-1 hour after initiating  ;

o -the event. At'this point the SFP would have water over the top of.the (active fuel,. and about 1 foot of the fuel in the change fixtures' would be v uncovered.- This assumes that.the incore instrument shaft cover fails and '

ithe refueling cavity water flows to the recirculation sump. If the cover idid not' fail, the cavity level-would only drop about 3 1/2-feet, so that

'the fuel would remain under:19 feet of water. If the cover fails the RHR 4 -pumps would be put on line (one would already be running for shut down B c icooling-flow):to recirculate water from the sump to the reactor cavity at j a maximum rate of 10,000 gpm. In this case, the two assemblies in the  !'

change-fixture would receive adequate cooling.

=

r i ~9245N ,

p _

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

' **""'rk -

i REF \ Af l K K

s. .

\

I- -

\

\

I N \

s N .:

N - ,

\

\ . - _

t i ,

/ 8 , l

/ g N t \, ,.

j 5 a

I

/ l. ,

I ,

/ i' i r s j \

i

/

f .

t

\

y-,

,, s ,,- .,fi-

' , - -g . ,, i

• I -: >

m %- . /

L- * *

/

,: ' _/ . .

\

• e i i r.

I /

• s' i >

. ,/

l *  ; ,

8 L i

, i

\

l i

s i i a . w b $ ,

N  ! s < s  !

'N, g*

(' 3

\ ' .

\ .

\ l

\5 u

\

'N ,

-4 . >. g I(o fs tuAE 1 l Bva.*>/Baniowooo Stot. 'f i l I,

- I 2 ,/ I(o .2 E :

i

~.

I - Pneumatic I Pneumatic Seal Seal i

1. [AC W /~LANc,[ r oEdtwo &

'lu SEAL em a N D==ti.' pm-.'

} l l k M y e?q;v Ir -Nh':

p7l l .',

l l

3 -

I) y :. .

iI I

i g

Figure 1 - Haddam Neck Refueling Cavity Water Seal i

l 4

+. . .

- References 1.- Telephone conversation between Ted Hollander of Presway Corp. and K.D. Brienzo of CECO' on 9-14-84 concerning refueling cavity seal design details.and the Haddam Neck event.

2. Telephone conversation between Dan Casey of the Haddam Neck staff and K.D. Brienzo of CECO on 9-12-84 concerning the seal failure

-incident of August 21, 1984 and probable causes.

3. Byron procedure number OBOA REFUEL-3, Rev. 1, SPENT FUEL POOL LEVEL LOSS UNIT 0.

'4l.- Byron procedure number 1 BOA REFUEL-2, Rev. 1, REACTOR CAVITY LEVEL LOSS UNIT 1.

5.- Discussion between Andy Mills of Byron Fuel Handling Staf f and

-K.D. Brienzo of PED at Byron on 9-13-84.

' 6. Byron /Braidwood FSAR Section 9.1.3 and 9.1.4, available makeup to the refueling water cavity and the spent fuel pool.

"7 . Fire Protection Report Section 2.1.3.2.3, operation of the fire ,

protection system.

8. Letter to D.M. Crutchfield from Connecticut Yankee Atomic Power Co.

dated August 31, 198A, Docket No. 50-213 811 299.

i t

4

- , r..7_-,-r. , ,.+ ... _

y--.,, , ., , m.. .._., ,, ..,. , . , , _ . - _ . . . , _ -_,-..,,,,y ,