ML20113G662

Submits Info in Response to NRC 920214 Request for Addl Info Re Inservice Testing of Svc Water Pumps

ML20113G662
Person / Time
Site:	Farley
Issue date:	04/10/1992
From:	Woodard J SOUTHERN NUCLEAR OPERATING CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9204150179
Download: ML20113G662 (3)

Text

. .. -. -. .. - . - _ _ - .

4 gt'ggeayO cea f Comprv

=

0,rmutam, uan3ma assot 10lepW@ PO5 fE8 5006 y

J. D. Woodard Southem Nudear Operating Company

&$$l 1pril 10, 1992 the ouumm,+wc wnm 10 CFR 50.55a(g)

Docket Nos. 50-348 50-364 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555 Joseph H. Farley Nuclear Plant Request for Additional Information on

, inservice Testina of Service Water Pumos Gentlemen:

By letter dated February 14, 1992, the NRC submitted its Request for ndditional Information (RAI) on Service Water pump testing performed at Farley Nuclear Plant. The following information is provided in response to the NRC's RAI.

ITEM 1 IndividWLPump Testina NRC REQUEST Provide justification for not individually flow testing the service water pump during cold shutdowns and refueling outages when plant conditions parmit operation e' a single pump.

SNC RESPONSE Cold shutdowns are normally of insufficient duration to perform individual testing of the service water pumps.

Testing of the pumps at refueling is possible; however, the data provided would not be meaningful in trending degradation on the pumps. The individual pump flow reference value required by the ASME Code could only be set during refueling. Subsequently, only like tests performed every 18 months at refueling could be compared to this l reference value to detect degradation. At present, service water pumps are overhauled cvery five years due to service conditions. This overhaul necessitstes the setting of a new baseline in wcordance with IWP-3111 Since this overhaul occurs every five years, the maximurn number of tescs that l could be compared to this baseline would be three tests

! bt fore another overhaul took place. Since three points are required as a minimum to determine a statistical trend, by the time degradation is noted by this test, corrective l

action will already be required due to pump maintenance schedules C

df i

9204150179 920410

.r n* r,

~

hkl p l PDR ADOCK C5000348 P PDR i

, U. S. Nuclear Regulatory Commission Page 2 4

- ITEM 2 Nonintrusive Diaanostic Methodt i

NRC REQUEST Address the feasibility of nonintrusive diagnostic method of flow measuring for each service water pump when plant conditions do not permit operation of a single pump.

SNC RESPONSE Nonintrusive diagnostic flow measuring devices require that the transducers be mounted in areas where the flow profile is well developed. In general this requires a straight run of pipe at it..st 4 or 5 pipe diameters from the nearest flow restricting device. Since the discharge pipe on the service water pumps is 20" in diameter, a straight run of at least 6 feet past the discharge butterfly valve is required to install the measuring device correctly. On the <' harge of a service water pump, the following components are in series: an expansion joint, two incher of pipe, a discharge check valve, and a butterfly valve. After the butterfly valve there is only about 3 feet of straight run pipe where a flow measuring device could be installed. Therefore it is nut feasible to install a nonintrusive diagnostic flow measuring device that could measure flow accurately.

ITEM 3 Masked Pumo Performance When Testina Pumps In Pair NRC REQUEST Quantify tne extent to which the hydraulic performance of a degraded pump may be masked by the other pump when the

. service water pumps are tested in pairs. Take into consideration the effects of the accuracy of the flow instrument used to measure the combined flow and where the pumps are operating relative to the pump curves.

SNC RESPONSE Flow reference values are set taking into acaunt instrument accuracies as specified by Table IWP-4110-1 and the requirement that ell reference values be set when the equipment is operating acceptably.- All reference values shall 'also -insure that the pump is capable of performing its safety related function. The safety related function of a service water pump is defined in the FSAR in terms of two pumps in operation per train. The reference points are set, based on empirical data, at values sufficiently above those required for accident flow such that instrument accuracies are bounded.

In order to prevent masking of a pump that has been in service for a period of time by a-recently overhauled pump a new baseline is set for the overhauled pump. After overhaul on a particular pump a flow test is performed to determine the new baseline. This new baseline is based on actual test data as required by IWP-3110 and therefore increaseJ pump performance is included in the new baseline. Degradation is then trended from that point.

4

[,U.S.NuclearRegstatoryCommission Page 3 To take into consideration where the pumps are operating relative to the pump curves, FNP sets all flow reference values while maintaining the individual pumps at the same  ;

differential pressure. The tolerance on this differential pressure is i 1 psig (or about 1.2%). Since all the service  !

water pumps are held at the same differential pressure, all the pumps are maintained at the same point on their centrifugal pump curves. By maintaining all differential pressures at the same point, an individual pump flow analysis may be performed. Since all pumps are tested at the same differential pressure, this individual analysis will detect any masking that may be occurring.

If there are any questions or if additional information is needed, please advise.

Respectfu'ly submitted, SOUTHERN NUCLEAR OPERATING COMPANY b

N. b . Woodard JDW/C."B: map cc: Mr. S. D. t'bneter Mr. S. T. Hoffman Mr. G. F. Maxwell l-b --

, , , + . . , ,w.. , , , , . . , , - ,,,m_,7,m,.-.,_.,,.~,,,_.,,,,.....,-.. ...~.m.

. . . . ,_.,.,,,..m.,1, ,,-._....,,.-~_,.,,.,__.m.. , . .,. ,