ML18106A954

Forwards Response to NRC 980910 RAI Re Request Relief from Certain Requirements for IST of SI Accumulator Discharge Check Valves

ML18106A954
Person / Time
Site:	Salem
Issue date:	11/02/1998
From:	Storz L Public Service Enterprise Group
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LR-N980515, TAC-M98259, TAC-M98260, NUDOCS 9811120022
Download: ML18106A954 (6)

Text

..

• Public Service Electric and Gas Company Louis F. Storz . Public Service Electric and Gas c~Yny 2p Y.~~x 236, Hancocks Bridge, NJ 08038 609-339-5700 Senior Vice President - Nuclear Operations LR-N980515 United States Nuclear Regulatory Commission Document Control Desk Washington, DC 20555 Gentlemen:

RESPONSE TO REQUEST FOR ADDITIONAL INFORMATION REGARDING TESTING OF ACCUMULATOR CHECK VALVES SALEM GENERATING STATION FACILITY OPERATING LICENSE NOS. DPR-70, DPR-75 DOCKET NOS. 50-272, AND 50-311 TAC NOS. M98259 AND M98260)

On December 26, 1996, the Public Service Electric and Gas Company (PSE&G) submitted a request for relief from certain requirements for the inservice testing of the

• safety injection accumulator discharge check valves. On February 13, 1997, the U.S.

Nuclear Regulatory Commission (NRC) granted interim relief to PSE&G for the current operating cycle, while the NRC staff conducted a more thorough review of the relief request.

By letter dated September 10, 1998, the NRC requested PSE&G to provide additional information relative to its request for relief. Attachment 1 to this letter contains the NRC's request for information, and in Attachment 2 the NRC's questions are re-stated in boldface type as written in the request for additional information, followed by the PSE&G response in regular (non-boldface) type.

Should you have any questions or comments on this transmittal, please do not hesitate to contact me.

\

\\\

Sincerely,

,., n (l )U

..-. r\ r.. \ 'i, - :) ...) ~ '

~\.i'** ~/ G~

Attachment (2)

~ - ---- -

9811120022 981102 PDR ADOCK 05000272 P PDR

@. Printed on

~ Recycled Paper

• • • NOV 2 1998 Document Control Desk LR-N980515 C Mr. H.J. Miller, Administrator - Region I U.S. Nuclear Regulatory Commission 475 Allendale Road King of Prussia, PA 19406 Mr. P. Milano, Licensing Project Manager - Salem U.S. Nuclear Regulatory Commission One White Flint North 11555 Rockville Pike Mail Stop 14E21 Rockville, MD 20852 Mr. S. Morris USNR'C Senior Resident Inspector (X24)

Mr. K. Tosch, Manager IV Bureau of Nuclear Engineering 33 Arctic Parkway CN 415 Trenton, NJ 08625 95-4933

• ATTACHMENT 1 Response To Request For Additional Information Regarding Testing Of Accumulator Check Valves Salem Nuclear Generating Station, Units 1 And 2 LR-N980515 NRG Request For additional Information On sheet 12 of Attachment 1 to the December 26, 1996, letter, PSE&G references Equation 3-1 from EPRI Report NP-5479, Application Guidelines for Check Valves in Nuclear Power Plants, dated January of 1988. Equation 3-1 is the minimum flow velocity (Vmin) to cause full disk lift in a swing check valve. On page D-5 of the EPRI report, there is a discussion of the disk shielding effect in the Clearway check valve body design. The guide states that if the disk does not protrude in the flow path, then it is expected that a significantly higher flow velocity will be required to achieve full disk opening than predicted by equation 3-1. Since the accumulator discharge check valves do not employ disk stops and experience the disk shielding effect, PSE&G uses a flow test to verify that the valve disk is free to move to the safety position.

1. Provide the percent or degree of valve disk swing to attain the safety position for these valves.

2. Describe PSE&G's rationale for concluding that the V min equation is valid for the analysis of the accumulator check valves.

For the flow test, PSE&G calculated a minimum pressure decay time of 24 seconds, then added 15 percent margin to this number to arrive at the acceptance criteria of 27 .5 seconds. It appears that this factor was added to account for some variation in the test conditions and data collection methods.

3. Describe the basis for the selection of this margin along with a confirmation that the valve disk will attain the safety position at the limits of the acceptance margin. Also, discuss the effect of instrument inaccuracy on the test acceptance criteria.

Page 1 of 3

• ATTACHMENT 1 Response To Request For Additional Information Regarding Testing Of Accumulator Check Valves Salem Nuclear Generating Station, Units 1 And 2 LR-N980515 NRC Questions # 1

1. Provide the percent or degree of valve disk swing to attain the safety position for these valves.

PSE&G Response to Question #1 The swinging disks of the accumulator check valves attain a position depending on the water flow rate through it. However, there is a flow rate above which the valve disk does not open any more and has essentially reached the full open position. In Reference 1, Westinghouse determined that a flow rate of 3537 gpm corresponding to a velocity of 20 ft/sec in a 10 inch schedule 160 pipe is the limit above which the valve disk does not open any more. Equation 3.1 of Reference 2 was used to back calculate the disk angle corresponding to a velocity of 20 ft/sec. Using this equation the valve maximum swing (measured from the vertical position) is 72.4 degrees. In determining the accumulator pressure decay time history during the check valve testing, PSE&G assumed that a free valve disk continues to open as the velocity increases to 20 ft/sec and does not open any more when the velocity exceeds 20 ft/sec.

NRC Questions # 2

2. Describe PSE&G's rationale for concluding that the V min equation is valid for the analysis of the accumulator check valves.

PSE&G Response to Question #2 The V min equation (Equation 3-1 of Reference 2) was not used to determine the minimum flow velocity required to open the check valve disk to its maximum position.

Reference 1 provided a value for the minimum velocity required to keep the valve in the full open position. The V min equation was used to back calculate the disk swing angle at all disk positions from 0 to 72.4 degrees. This was necessary because during the test, the check valve opens from the closed position to its maximum open position and returns to a partial open position. The valve swing angle at various times during the test was used only to determine the friction loss coefficient for the check valves at partial and full open positions. Based on PSE&G's engineering judgement, for the disk open angle in the range of 0 to 72.4 degrees, the disk is significantly in the flow stream and the use of equation 3-1 of Reference 2 is justified in determining disk open angle for a given flow velocity. Additionally for open angles greater than 72.4 degrees, the Page 1 of 3

• ATTACHMENT 1 Response To Request For Additional Information Regarding Testing Of Accumulator Check Valves Salem Nuclear Generating Station, Units 1 And 2 LR-N980515 valve friction loss coefficient remains essentially unchanged. Therefore, the full open friction loss factor is relatively independent of the calculated value of the full open angular position of the valve disk and the use of Equation 3-1 of Reference I in determining the disk angle for a given flow velocity is justified.

NRC Questions# 3

3. Describe the basis for the selection of this margin along with a confirmation that the valve disk will attain the safety position at the limits of the acceptance margin. Also, discuss the effect of instrument inaccuracy on the test acceptance criteria.

PSE&G Response to Question #3 The accumulator check valve testing is done by measuring the accumulator gas space pressure decay time for predetermined accumulator pressures and accumulator and reactor water levels. For a free check valve, which does not stick at an intermediate position, the calculated pressure decay time is 24 seconds. The measured pressure*

decay time for this situation was approximately 23.9 second. Therefore, PSE&G concludes that the mathematical model that represents the accumulator pressure decay scenario of the test is accurate.

During the accumulator pressure decay scenario under the test conditions, the flow rate through the check valve increases from zero to a maximum value and then decreases.

The calculated value of the maximum flow rate through the check valve during this transient process for a situation when the check valve is free to open is approximately 4900 gpm. This value is greater than the minimum flow of 3537 gpm (provided in Reference 1) required to fully open the check valve. Thus, PSE&G concludes that the valve attains its full open safe position.

As stated in the first paragraph, the calculated value of the pressure decay time for a successful test is 24 seconds. A 15% margin (3.5 seconds) was added to this value to provide the operator some flexibility in data collection and therefore, an acceptance criterion for pressure decay time of 27.5 seconds or less was chosen. According to calculation, the maximum value of the flow rate through the check valve, even if the pressure decay time is 27.5 second, is approximately 4400 gpm. This lower flow is much greater than the minimum flow of 3537 gpm required to lift the check valve to its maximum position. Thus, PSE&G concludes that the 3.5 second margin added to the Page 2 of 3

• • ,L "'

• ATTACHMENT 1 Response To Request For Additional Information Regarding Testing Of Accumulator Check Valves Salem Nuclear Generating Station, Units 1 And 2 LR-N980515 calculated value of pressure decay time is reasonable. Meeting this acceptance criterion provides reasonable assurance that the check valve will function as designed.

INSTRUMENT INACCURACY: The four primary instruments that are used during the test are the accumulator pressure and level transmitters, the reactor cavity level indicator, and a hand held stopwatch.

The total loop inaccuracy of the accumulator pressure indication is 0.28% of 1OOOpsig (2.8psig) .. The total loop inaccuracy of the level indicator is 0.28% of 159 inch (0.45 inch). The inaccuracy of the reactor cavity level measurement is 1 inch, and the inaccuracy of the hand held stopwatch is 0.001 sec. However, the reading is rounded to nearest 0.01 sec. The combined inaccuracy of the instruments, as well as the minor difference in "operator response time" is bounded by the 15% margin added to the pressure decay time.

References:

1. Westinghouse letter to PSE&G, PSE-90-530, dated May 21, 1990.

2. EPRI report entitled "Application Guidelines for Check Valves in Nuclear Power Plants", Report No. NP-5479, January 1988.

Page 3 of 3