ML20150A827

Responds to Outstanding Concerns Re Scram Discharge Vol Design & Capability of Plant to Determine Compliance W/ Generic SER Issued on 801201,per Insp Repts 50-321/87-17 & 50-366/87-17

ML20150A827
Person / Time
Site:	Hatch
Issue date:	02/22/1988
From:	Gucwa L GEORGIA POWER CO.
To:	Grace J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION II)
References
IEIN-87-017, IEIN-87-17, SL-4041, NUDOCS 8803160061
Download: ML20150A827 (3)

Text

C Georg-:a Fowsr Company

- 333 Piedmont Avenue Atlanta. Georg's 30308 Telephow 404 526 6526 Matng Address:

Fbst Office Box 4545 l

Adanta, Georg.a 3030 l

Georgia Power i

m L.T.Gucea Manager Nuclear Safey j

and Licenseg

.SL-4041 1914C

'X7GJ17-H120 l

February 22, 1988 U. S. Nuclear Regulatory Commission Region II - Suite 2900 101 Marietta Street, NH

. Atlanta, GA 30323 ATTN: Dr. J. Nelson Grace PLANT HATCH - UNITS 1, 2 NRC DOCKETS 50-321, 50-366 OPERATING LICENSES DPR-57, NPF-5 RESPONSE TO INSPECTION REPORT 87-17 i

Gentlemen:

i The Nuclear Regulatory Commission (NRC) performed a

special i

inspection on July 27-31, 1987, and reported the results in Inspection Reports 50-321/87-17 and 50-366/87-17.

During this inspection, the NRC l

reviewed the scram discharge volume (SDV) design ar.d capability of Plant l

Hatch, Units I and 2, to determine compliance with a Generic Safety l

Evaluation Report (GSER) issued December 1, 1980.

At the conclusion of i

the inspection, Georgia Power Company (GPC) committed to perform evaluations to resolve three outstanding concerns in this area and to i

l provide the NRC a written response.

The evaluations are complete and the I

results are reported below.

1 Item 4.3 of the subject inspection report describes the physical l

inspection of Plant Hatch's Units 1 and 2 SDV.

Information Notice 87-17 alerted BHR owners of long delay times resulting from the piping configuration of float-type level switches (magnetrol switches) used to monitor SOV water levels.

The Unit 1 SOV instrument line piping configuration (the small bore piping which c7nnects the SDV to the j

magnetrol switches) would not affect the system's ability to perform its J

safety function during a fast fill event. (The piping has a very short

)

run).

However, a concern exists on the Unit 2 SDV since the instrument i

line piping run is significantly longer than on Unit I and the response time has not been analyzed.

j eso31600628'so$$$2s PDR ADOCK PDR l

l c

l L

GeorgiaPower A U. S. Nuclear Regulatory Commission February 22, 1988 l

Page Two An analysis has been completed to answer the concern based upon the following conservative assumptions:

1)

All 137 Control Rod Drives (CRD) are leaking at a rate of l

3.4 gallons per minute (gpm)*;

2)

All SDV drain valves are closed; 3)

The worst case magnetrol switch was used (the switch that was analyzed to have the longest total response time); and 4)

The Resistance Temperature Detectors, which are redundant

[

level sensors that will initiate a scram signal in e

approximately two seconds, do not function.

l The analysis also assumed that at time zero, the SDV water level is at an elevation equal to the elevation of the lowest level switch i

attached to the SDV.

Analysis showed that the worst case magnetrol i

switch would initiate a scram signal 17 seconds (inclusive of the one-second instrument response time) after the SDV level increased to the SDV high water level scram setpoint. Using the assumptions listed above, there would still be 495 gallons of empty volume available in the SDV to accommodate a scram.

The GE CRD System Design Specification requires a minimum of 458 gallons of empty volume in the SDV.

Therefore, based on i

the analysis, the Hatch Unit 2 magnetrol switches would initiate a signal such that the reactor is scrammed before the CRD leakage fills the SDV to the point where a complete scram cannot be achieved.

1 Additionally, item 4.3 mentions that the magnetrol switch is susceptible to damage due to the piping arrangement of the SDV.

Plant Hatch SDV magnetrol switches had experienced damage that was attributed l

to hydrodynamic forces created during a scram.

Piping modifications were subsequently performed that connected the switches directly to the SDV rather than the SDV vent thus removing the forces from the switches.

The I

changes were completed in accordance with the recommendations of the BHR l

Owners Group.

Since January 1,1984 (af ter the modification), there have been no magnetrol switch failures attributed to hydrodynamic forces created during a scram.

\\

The justification of 3.4gpm/CRD was sulaitted to the KRC by letter from J. T. Beckham, Jr. to D. Muller dated 01/06/86.

This submittal was reviewed and approved on 05/13/87.

1914C w,

\\

GeorgiaPower A

]

_i U. S. Nuclear Regulatory Commission February 22, 1988 Page Three i

i The third concern was identified in item 4.5, which requested an evaluation of the possibility of water backup from radiological waste i

drains into the SOV.

A visual inspection showed that the SOV vent and drain piping is hard piped to an equipment drain header (located on the i

130 foot elevation of the Reactor Building) which has open drain hubs.

Therefore, if radiological liquid waste backup occurs, the water would drain out of the open drain hubs onto the Reactor Building floor, not into the SOV through the vent and drain piping.

i If you have any questions in this regard, please contact this office ~

3 at any time.

Sincerely,

[

feghe _ :

L. T. Gucwa i

JCJ/lc

?

I i

c: Georaia Power Comoany Mr. J. T. Beckham, fr., Vice President - Plant Hatch GO-No MS

(

l U. S. Nuclear Reaulatory Commission. Washinaton. D. C.

i Mr. L. P. Crocker, Licensing Project Hanager - Hatch Document Control Desk l

[L_S. Nuclear Regulatory Commission. Region II_

i

{

Mr. P. Holmes-Ray, Senior Resident Inspector - Hatch i

i i

i 1

I i

i i

1914C t

f eno

- - - - - - - - - - - - - - - - - - - - - - - - - - - - -l

-