ML20100C286

Submits Listed Evaluations in Response to IE Bulletin 84-03, Refueling Cavity Water Seal. Seal Deflection Measurements Will Be Performed W/Seal Installed in Annulus in Both Deflated & Inflated Conditions

ML20100C286
Person / Time
Site:	Ginna
Issue date:	11/28/1984
From:	Kober R ROCHESTER GAS & ELECTRIC CORP.
To:	Murley T NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
References
IEB-84-03, IEB-84-3, NUDOCS 8412050387
Download: ML20100C286 (6)

Text

-

y.

/7%

"z u rg m

...? ; ~" %

f 2"'l:h k

y

^

Rh$3W GF i""Q ROCHESTER GAS AND ELECTRIC CORPORATION e 89 EAST AVENUE, ROCHESTER, N.Y. 14645-0001 k ROGER W. KOGER g*g _ _

November 28, 1984 y,qhgg Dr. Thomas E. Murley, Regional Administrator U.S.

Nuclear Regulatory Commission Region I 631 Park Avenue King of Prussia, Pennsylvania 19406

Subject:

IE Bulletin No. 84-03:

Refueling Cavity Water Seal R.

E. Ginna Nuclear Power Plant, Unit No. 1 Docket No. 50-244 Gentlemen:

As requested in IE Bulletin No. 84-03:

Refueling Cavity Water Seal, Rochester Gas and Electric hereby submits the following evaluations.

At the Ginna Nuclear Plant, a Presray Model PRS 585 molded fabric reinforced inflatable seal is used to effectively Jeal the annulus between the reactor vessel and the refueling cavity.

The seal has performed satisfactorily since 1977 at which time it was first put into use.

Although the inflatable portion of the seal used at Haddam Neck is of similar design to that used at Ginna, there are some distinct differences to contend with.

.To begin with, the seal centerline diameter at Ginna is 172 1/2 inches as compared to 256 1/8" for Haddam Neck.

This magnitude difference in diameter decreases the potential for pissible seal misalignment at Ginna.

As per discussion with Presray Seal and Connecticut Yankee at Haddam Neck, there is a marked difference in the rigidity of the seals.

The Haddam Neck seal has been noted by Presray to be more flexible than the Presray PRS 585 that is employed at Ginna.

The rigidity of the Ginna seal makes it very unlikely for the seal to displace itself from the annulus area.

The double seal design at Haddam Neck is used to seal approx-imately a two foot gap between the reactor flange and the cavity wall.

This is accomplished by the use of a steel seal ring between the two inflatable seals.

The seal ring is supported by strongbacks at only nine specific locations around the circumference, and due to this arrangement, a scalloping or ripple effect can result.

This condition can introduce a potential for misalignment.

At Ginna, the seal is a single membrane, sealing a nominal 2 1/2 inch gap between the reactor flange and the cavity liner wall.

The depth of the annulus at Ginna allows for greater surface contact of the inflatable seal (full surface cintact on the cavity liner side and 2 1/2 inch surface contact an the reactor vessel flange).

This is compared to a total 1 5/8 inch surface

(

contact with the Haddam Neck application.

8412050387 841128 PDR ADOCK 05000244 gt G

PDR

5:

l ROCHESTER GA5 AMD ELECTRIC CORP.

SMEET NO. 2

DATE November 28, 1984' To Dr.LThomas E.~Murley.

The inflation pressure'at-Ginna'is procedurally regulated.at

'30.psig.,1To prevent the. possibility _ of -overpressurization, a-relief valve set at approximately~40_psig.is._ employed.

The. net effect is ' that.- the sealLwill'not undergo deformation due to overpressurization.

Duelto the complexity in analyzing th'e mechanical properties

- of a fully inflated--PRS 585. seal, short of performing a full-scale mockup, the following' static experiment is offered-(see attached Experimental Job Sheet).:

This study was' performed by the seal manufacturer, Presray Corp. prior to initial seal usage.

It is noted _ that. the ~ experimental' test was performed to. determine the-

-effectiveness of the inflatable seal with a 25 foot head of water

-if the seal becomes deflated.

This appears to be.a more credible

" event at Ginna than the displacement of the seal mainly due to 1)

~

the rigidityLof the seal, 2) the specific application for which

.the seal.has been designed by the manufacturer, 3) measures taken

.which preclude overpressurization.

The results of the experiment conclude'that the seal will not push through the opening.

If it can be postulated that the same event could occur at

~ Ginna,.., inherent plant design characteristics will minimize the effect of seal failure.

Assuming _a quarter.circumferential seal failure similar to that experienced at Haddam Neck, the resulting flow rate would be considerably less.

This is due to the smaller

. annulus diameter-at Ginna.

Assuming worst case, with flow rates

~ approaching 1 those experienced at Haddam Neck and with a fuel assembly in the _ manipulator mast, operator action will be necessary to prevent fuel cladding failure.

System design at Ginna assures thate even with a seal failure and subsequent draining of the refueling cavity, sufficient wat.er inventory will remain in -the lower cavity.and fuel transfer slot area to cover fuel that may be. stored there.

Since fuel assemblies can only be stored in

~

1.-

either the RCC change fixture area, the fuel. transfer.upender or the manipulator u.ast, the only potential concern would be that of having an assembly in the mast.

Recent analysis at'H.B'. Robinson

' indicate that. fuel cladding damage will result in approximately twelve' minutes for an exposed, irradiated assembly.

This assumes total uncovery approximately fifty hours after irradiation.

This fifty hour window is' extremely conservative since normal system cooldown and cleanup precludes fuel movement within that short of a time frame.

We have administrative 1y addressed this concern of having an asaembly in the mast by instructing refueling personnel to position the' manipulator crane over a previously selected and marked' location in the fuel transfer slot and to lower the assembly -_

to the floor.

This instruction was first introduced as part of the concern addres sing' -the installation of the S/G nozzle. dams two years ago.

To date,_ these instructions have not been formally fdocumented._

A permanent change to the cycle specific refueling procedure will be incorporated-into the 1985 procedure to address

-this-area.and formally present the emergency operation procedure to be'followed.

For the other two areas mentioned above, approx-

~

imately one foot of water will remain covering the top of the vertical standing assemblies.

i

M1-

=*

ROCHESTER CAS AND ELECTRIC CORP.

SHEET NO. 3

o4TE November 28, 1964 TO - ~ Dr. Thom'as E. Murley

.It?is-noted that the above. postulated. event will-have no effect on the integrity of the fuel ' located in the core.

Residual heat: removal capabilities will not be effected and the core will'

_ 1 experience no adverse effects.

The effect ~ of seal failure-on the stored' fuel in the spent' fuel pit without any operator action would be that no fuel uncovery would occur.- Approximately 8 inches of water would remain ~ covering the top of the fuel assemblies.

With operator. action, the manually.

~

. operated fuel transfer gate valve would be-closed to prevent unnecessary draining of.the' spent fuel pool.-

The most' limiting case. for. fuel. in transfer ~.between the spent fuel pool and the Erefueling' cavity would be that_of placing an irradiated assembly in the upender and positioning the upender in-the vertical position.

' Other than'.radiolog: cal consequences, no fuel or cladding damage will result.

Approximately one foot of water will be covering.

the_ top of the assembly.

Again assuming worst; case,' without any.

operator action,-the minimum time frame for the.beginning of fuel uncovery in the SFP would be approximately eight hours.

This is based on-present storage configuration volumetric dime'nsions and.

a maximum allowable decay heat rate of 12.0 MBTU/hr.. Additionally,-

this eight hour criterin is based on the heating of the remaining U

' water from 80 F to 212 F, with no credit taken for ambient losses.

With operator ~ action, the two fire pumps can be arranged to re-flood _the pit at the rate of six inches per minute to a level

- above the suction piping for the spent fuel. cooling cycle.

This will take approximately 12 minutes to re-flood the SFP to allow normal' spent fuel cooling to be reinstated.

.As a. concern. to bettier understand nature of deflection of the inflated seals,. we will commit to' performing seal deflection -

measurementsifor the upcoming refueling outage.

These will be taken'with the seal installed in the annulus in both the~ deflated andcinflated condition.

This will provide us with valuable

. information regarding the direction of seal de flection, if any, with the seal installed.

k 4

(

1 9~

ROCHESTER GAS AND ELECTRIC CORP.

SHEET NO. 4 DATE November 28, 1984 ro Dr. Thomas E. Murley In conclusion, at the Ginna Plant, although the potential exists for seal failure, existing system configuration and operating procedures negates the consequences of fuel uncovery.

Again it is restated that operator action is necessary to prevent fuel' uncovery in the event that a fuel assembly is in the mast of the manipulator crane at the time of the event.

For all other refueling activities, with seal failure, the lack of water shielding will result in high radiation fields, however, any effect to the health and safety to the general public will be negligible.

V[ytrulyyours, l

4) A%

Ro r W.

Kober Subscribed and sworn to me on this 28tg ay o November 1984

. h[

LYNN 1. HAUCK

(/ /

NOTARY PUBUC, State of N.Y., Monroe County Enc.

My commission Expires March 30. IM

-xc:

U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.C.

20555

ROCHESTER GAS AND ELECTRIC CORP.

SHEET NO. 5 DATE ' November 28, 1984 To Dr. Thomas E. Murley EXPERIMENTAL JOB SHEET PURPOSE:

To determine the deflection of a PRS 585 seal under a 25 foot head of water when seal inflation pressure fails.

PLAN:

Prepare a test fixture that will accept a PRS 585 seal with a 2 1/2 gap and a means to simulate a 25 foot head of water.

APPROACH:

The net hydrostatic load on the seal is distributed over the 2 1/2 inches of the seal flange directly over the 2 1/2" gap hydrostatic pressure is.4335 psi per foot of water.

Using a 6 inch long test fixture, the total test pressure is determined by:

2 1/2 x 6 x.4335 x 25 = 162.562" or 27.094 lbs per linear inch Using a conservative approach, a 180 lb. force was applied to a 1 inch section along the seal %.

TOOLING:

1 Test fixture 6" lg. (See Sketch, Sht. 2) 1 PRS 585 sample 6" 1g.

1 3/4 x 1 x 6 bar PROCEDURE:

(See Sketch, Sht. 2) 1)

Set fixture at 2 1/2" gap 2)

Place seal in fixture 3)

Place bar on seal centerline 4)

Apply force 5)

Measure seal deflection at base of bulb RESULTS:

Seal deflection 1/4 inch CONCLUSION:

The PRS 585 will seal effectively against a 25' head of water if the seal becomes deflated.

The wedge shape of the seal flange will act'as an ef fective " stopper". As the water prer,sure inereases the wedge will become more firmly set.

Under the conditions outlined, the seal will not push through the opening.

..,,.. ~

... r a,

V L..

. :.. Q j. y....rgCE l'" :.....i' *** N :I]I ' '

..i.,,,..

,. s.

4 t-

..~..

..., _,;..c

u. <; ;,. y,..

a 2

.c.

..... e.

a, a.

, 7.

.i,

,?

..i ;

t...i

~.

1

~.

z.

.. ~

.,y zu

,s I

' /

/.

~ TEST. F'IX URf v

p!

N,

..L 1

g

~

t

.2, hA[

q

.# e N

N i.

N

. r.

..,..: y 6

.4 - -.....

7

....4>

ky y

v, yu s.~

/

p-z.,.

1

. 1 r.i h...

q

.it.

r.

4

.w

(

. '.'... ;G.IYk'.'w..

,s i.

.yy ; :. n :

.,. v

. a..

=.u.

....~ d '.....

... :..i. ' 3h...,..,..xM Y:t: k.:& '.

t...n..

y.

'*^*~'..*i-t