ML20203N587

Submits Revised Final Response to IE Bulletin 84-03.Reactor Cavity Seal Failure Described in Bulletin Not Credible Concern.Requirements of Bulletin Met

ML20203N587
Person / Time
Site:	Harris
Issue date:	09/23/1986
From:	Cutter A CAROLINA POWER & LIGHT CO.
To:	Grace J, Grace N NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
References
IEB-84-03, IEB-84-3, NLS-86-323, NUDOCS 8610090276
Download: ML20203N587 (9)

Text

Y 'c o .

2su$

CD&L Carolina Power & Light Company SERIAL: NLS-86-323 SEP 23 386 Dr. J. Nelson Grace, Regional Admisiiflag A10 : 4 6 United States Nuclear Regulatory Commission Suite 2900 101 Marietta Street, NW Atlanta, GA 30323 SHEARON HARRIS NUCLEAR POWER PLANT UNIT NO. 1 - DOCKET NO. 50-400 FINAL RESPONSE TO IEB 84-03 REACTOR CAVITY SEAL FAILURE

Dear Dr. Grace :

Carolina Power & Light Company (CP&L) hereby submits a revised final response to IEB 84-03 for the Shearon Harris Nuclear Power Plant (SHNPP).

By letter dated July 1, 1985, CP&L provided details of our design review of IEB 84-03 and of the changes that were intended to be made as a result of that review. Subsequent questions from your office have denonstrated a need to provide an update to that response giving clarification where needed or to describe the actual as-installed configuration. Revisions to the original response are indicated on the attachment by a bar in the right-hand column.

These revisions address the questions as subsequently recorded in Inspection Report 50-400/86-61 (Paragraph 8, Report Details).

We still conclude that the seal failure described in IEB 84-03 is not a credible concern at SilNPP, and we have met the requirements of the Bulletin.

If you have any questions concerning this matter, please call Mr. James D.

Kloosterman at (919) 836-8055.

Yours very t ul Ed A .

A. B. Cutter - Vice President Nuclear Engineering & Licensing SDC/pgp (4067SDC)

Attachment cc: Mr. B. C. Euckley (NRC)

Mr. H. R. Denton (NRC)

Mr. G. F. Maxwell (NRC-SHNPP)

A. B. Cutter, having been first duly sworn, did depose and say that the information contained herein is true and correct to the best of his information, knowledge and belief; and the sources of his information are officers, employees, contractors, and agents of Cc.rolina Power & Light Company.

G-t.aqd . ***

y

~ Notary (SeaLY fy .**********e k,,'<,

My commission expires: // '27 NOTAR 8

t -  : n i,6 *, PUBLIC I ci5 411 Fayetteville Street

• P. O. Box 1551

• Raleign. N. C. 27602 ' *** ,.e**

• c n-r z . m _ m. _ ,

<,, 6 ,,,d

,,,,'s CO 8610090276 860923 O k'

- ee ,,,U14

"".t.i PDR ADOCK 05000400 G PDR

Attachment to NLS-86-323 ATTACHMENT TO FINAL RESPONSE TO IEB 84-03 EVALUATION OF A REACTOR CAVITY SEAL FAILURE AT SHNPP A. Potential for a Refueling Cavity Water Seal Failure Carolina Power & Light Company evaluated the SHNPP cavity seal design in regard to IEB 84-03 and the Haddam Neck incident. We have concluded that a gross failure such as that experienced at Haddam Neck is not a credible event at SHNPP based on the SHNPP design and proper installation, as discussed below.

1. Design and Installation The cavity seal ring girder spans the open (annular) area between the reactor l

vessel refueling flange and the refueling cavity liner (Figure 1). The cavity seal consists of two inflatable seals positioned one on each side of the ring l girder which is a circular metal plate structure resting on support beams.

This structure is designed seismic category I.

a. Ring Girder The ring girder is segmented into three equidistant sections for handling purposes; when bolted together, these segments interlock to form a complete ring. A rubber seal used between the segments prevents leakage at each joint. When installed, the ring girder spans the majority of the 3' 11.5" opening, but leaves two-inch gaps (nominal) on both sides for the inflatable seals.

To ensure the ring girder is installed in the proper position on the support beams, guide plates are welded to the bottom of the ring girder, and shims and guide plates are welded to the support beam. When installed, the ring girder guide plates straddle the support beams. These design features assure that the cavity seal gap dimensions are established and remain constant for each refueling outage.

The ring girder also has sides which extend further down into the cavity than the length of the seals. These provide additional surface area for the seals to press against and help to form a water-tight seal when the seals are inflated.

While it is " highly unlikely" that the SHNPP inflatable seal would push through the cavity gap, the consequences of a rapid gross loss of refueling water without adequate makeup capacity caused sufficient concern to revise the ring girder design. To further reduce the possibility of a seal failure incident,~ the design has been modified as follows:

Metal plates welded around the side of the ring girder below the location of the inflatable seals (Figure 1) will act as " bottom stops" for the seals to prevent slippage down through the gap.

A circular metal ring attached to the underside of the reactor vessel seal ledge provides additional surface (parallel to the edge l (4067sDC/ cts)

~

Attachment to NLS-86-323 of the reactor vessel refueling ledge) for the inflatable seal to press against. This essentially prevents the seal from extruding, or " ballooning", to the side.

The bottom stops and metal support ring enhance the design's ability to prevent the inflatable seal from pushing through the cavity gap, and .

thus, prchibit a gross leakage of refueling water incident such as occurred at Haddam Neck.

4

b. Inflatable Seals An inflatable seal is installed in the two-inch gaps on each side of the ring girder. While each inflatable seal is manufactured as a single component, it is not continuous like an O-ring. When a seal is positioned in the gap opening, the ends of the seal interlock together to form a water-tight barrier when inflated.

The angular portion of the seals (Figure 2) rest on either side of the gap, preventing the seal from falling through the opening. The seals are wide enough in the cross-sectional plane to handle variations in nominal gap width. When inflated, the lower portions of the seals expand to press tightly against the reactor vessel refueling ledge and ring girder on one side, and refueling cavity liner and ring girder on the other side

to form a water-tight seal (Figure 1).

l The SHNPP seal design is significantly different from the Haddam Neck seal in two critical areas: the width of the upper seal (flange) and the hardness of the rubber material. The width of the SHNPP upper seal (flange) is 4 inches; the Haddam Neck Seal was 3.5 inches. The durometer (material hardness) value of the rubber in the SHNPP

! inflatable seals is 65; the Haddam Neck seal was only 30. This means that the SHNPP seals are less flexible in.the upper regions, increasing i

the resistance to " push through" the opening, such as what occurred at.

Haddam Neck.

2. Seal Testing i

The potential for seal failure due to varying seal gap width conditions and equivalent water head _ pressures was both tested using a segment of spare H. B. Robinson Unit 2 inflatable seal which is similar in cross-section dimension and material properties to that used at SHNPP. With simulated gap openings of 2 and 2.25 inches, the seal withstood a pressure equivalent to four times the normal SHNPP refueling water head. Even with a 2.5-inch gap, the seal withstood a pressure equivalent to 3.8 times the normal SHNPP refueling water head before the top of the seal began to bend excessively.

3. Installation Procedure Installation of the seal structure is controlled by a corrective maintenance installation procedure. This procedure includes inspections to assure the seal is properly in place.

Additionally, steps are being added to the general operating procedure for

! Fill, Refueling, and Drain Down of the Refueling Cavity (GP-9), to require (4067sDC/crs)

I

. l Attachment to NLS-86-323 monitoring of the seal for leakage, and actions to be taken if significant leakage is detected.

In conclusion, the potential for a refueling cavity seal failure is not a credible concern due to the design features of the seal structure and mechanisms to

' assure proper installation.

B. Scenario for Gross Seal Failure Although a gross seal failure is not a credible event at SHNPP,'a." worst case" scenario was postulated in which all of the considerations enumerated in IEB 84-03 are included (i.e., leak rate, make-up capacity, time to cladding failure, effect on stored fuel and fuel in transfer, and emergency operating procedure).

The following assumptions were used in this scenario:

The seal structure design modifications (described above) are not present.

The refueling water is initially at normal refueling water level (El. 284.5').

The refueling cavity, fuel transfer canal, and spent fuel pool are interconnected.

The transfer tube remains open throughout the scenario.

A quarter section of each seat ring falls through cavity opening.

No operator action is taken.

1. Maximum Leak Rate The maximum leak rate for the above scenario would be approximately 34,000 gpm under the full water head. Approximately 512,000 gallons of refueling water would flood past the seal into the bottom of containment.

2. Makeup Capacity Makeup water from the following sources would be available to ' replace the water lost in the gross seal failure scenarios:

a. Refueling Water Storage Tank (RWST)

The RWST could supply water to the Fuel Transfer System using the residual heat removal pumps. Two pumps are available, having pump capacities of 3,750 gpm. The RWST is designed to hold 470,000 gallons of water; however, only a minimal water supply may be available after initially supplying the water needed for refueling.

b. Containment Sump - Emergency Core Cooling System (ECCS)

The containment sump, once sufficient water has leaked, will be the long-term source of water to the fuel transfer system. Two. residual heat removal pumps, capable of being aligned to the containment sump are available to utilize this water ' source.

(4067sDC/crs)

Attachment to NLS-86-323

c. Emergency Service Water System (ESWS) l The ESWS could supply water via the Fuel Pool Cooling System by using emergency connections. This system has an almost unlimited water supply from either the Auxiliary or Main Reservoirs.

d. Reactor Makeup Water Storage Tank (RMWST) l This system could supply water to the Fuel Transfer System using one of the Reactor Makeup Water Pumps and utilizing the Chemical and Volume Control System. The RMWST is designed to hold 85,000 gallons.

e. Demineralized Water System l

This syst2m could supply water to either the RWST or the RMWST using two pumps that can supply up to 600 gpm each. The demineralized water storage tank is designed to hold 500,000 gallons of water.

Complete descriptions of these systems can be found in the SHNPP FSAR.

3. Time to Cladding Damage Without Operator Action The only condition under which cladding damage could occur is if a fuel assembly in transit was in the manipulator crane. Calculations were developed to determine the time to fuel failure after being uncovered with water using the following assumptions:

1 Fuel had 48-hour cooling period prior to refueling.

The fuel assembly modeled was the peak assembly for three consecutive cycles.

A peaking factor of 1.34 was used; the peaking factor times the core average assembly power equals the modeled assembly power.

Reactor power is 100 percent during the assembly life in core. The assembly was irradiated at the full-power level for 930 effective full-power days, with no credit taken for shutdowns.

The fuel rod heatup is totally adiabatic; all heat generated in a rod stays in the rod.

Cladding failure is at 1300*F due to burst rupture (NUREG-0630, Cladding Swelling and Rupture Models for LOCA Analysis (Draft),

November 1978).

The coolant temperature and the cladding temperature are equal just prior to fuel uncovering.

The operator takes no action.

The time to fuel failure af ter uncovering would be approximately 17 minutes for 48-hour cooled fuel.

(4067sDC/crs1

~

Attachment to NLS-86-323

4. Effect on Stored Fuel and Fuel in Transit A gross-seal failure resulting in a loss of refueling water down to the elevation of the reactor vessel refueling ledge and cavity seal ledge (El. 260.2) has been evaluated for the effect on sto' red fuel and fuel in transit. The top of active fuel in the fuel upender, RCC change mechanism

. .and spent fuel storage racks would remain covered with water, as shown in Figure 3. As long as the fuel remains covered with water, it is not expected to fail.

Only a single fuel element in transit in the manipulator crane would have the potential of losing its water cover. l

5. Emergency Operating Procedures While CP&L considers separate emergency operating procedures for a gross loss of refueling water event unnecessary because the potential for a refueling cavity seal failure is not a credible concern, fuel in transit is addressed in a plant abnormal operating procedure.

l l

(4067sDC/crs) l l

-- - r - - - -

O

. , n

'N 'N 7 g s \r:

5 sts

• \ r's v . c<.! '. <$$,

o

! )/ * ,)N\ \ U6 x

m I!!) $

m x '

x 3 0 g yj! d k o 0 w

5

- N

\ e a u

=J .

\\  %

3 L

t y

t M

O K e ke 9

s\ o

?

e N

h w 3  ;

e 2

I \\ 5 # A D Q

k i $

s o 2 o il: E a

\

s\

\\ f s

jo gs <e sa.

5 z

- \N

\ = .-

p ,'l gn -

i 8"3 h 4mmv ye.;sss:& '

t ,-cs N

2 i) J .gll-

.I , _ WA .f

'\' \ 5,11 , '\\;..

hj '\ N '.

e.,c l

F I G O 9 E. 1 It4 WLRTAE.LE SERL CitAEN5 toms

=

a t." -

r W

s

+

g" t

- =

t,'/z" l

= f Ig---+

I

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

e

. e 9 e

= g G s 'C * *

, O  %

, , .n g

o

. 1 . ko

s g nb

(.s$.a cI a

1; a

• E - ,. N-i , - 3

.r.-._._._. . . w-s #

J gqm. ,

S#

o ,

, 0 U .g

• I

& g' L..-.-. . .

J t i

• W ,* =

• ,J l

) 4 Y

&l, 1 -

w =,

e- , e r.:

• v5 e- .. -

, n ,~oy

.:r

.e

=

,2 ,.........;,.....

b . ao

@ g fut se g W**

, 3 i , e2 ~a W of

.:l .' ' ' '

g l 8 '.~.T

• ..e..

. ,. ., I S. ..

• p.- e*

e

• ..** .e e,s '

, f 4- g 8 .% 8

. 1 . e # .O 4 p . .

• r--

l 7:ec.. . i (s' *. * .* y ** . g -,,, e * -* , l , .

y... e e -e i

, s., .e-

• **e I ,.. e e

t I f

J er e O

3 w0 32 a

T l hC

• ' 3

$ g YW.

%e.u i *

• - 9 ,*,

a z p , 60 i

4 -e, a

<I

' u- -

wa O m, 8 c ** . N g r

oc ,a

,i

..c - -

3 sn ate o' ~ 1 b

• 1 .so- 9' p

Eb hi u;

e d u - -

ad s

s y

e

=

.s2 :

N 8

% O e, *f

= e D = k

} r *o 0 9

e e

t N

q e t G A ,

• O J '

g W,

O W .'.

6 d '8 4 *

• p =s

, e en '

d i

1 sy a

' e *

• s 0 # *

< >

• O s6

• O w 30, e e s

s 0

• M -

z-e we f a;

N 3 y Vo J b d

' w W M 0 o ww g7 l- . .

s . *

• O s , ,

y .

. - % *, e *es q

'  % 2 ,

, e e,3 .

N e . = 6 e e o

• e e s -

9V t

. , . . _ .