ML17306A745

Forwards Response to NRC 920407 Request for Addl Info Re Concerns Received by NRC Re 911027 Reactor Scram & Primary Sys Cooldown.Surveillance Testing for AFW Sys Adequate to Demonstrate AFW Pump Operability.Concerns Unsubstantiated

ML17306A745
Person / Time
Site:	Palo Verde
Issue date:	05/14/1992
From:	Conway W ARIZONA PUBLIC SERVICE CO. (FORMERLY ARIZONA NUCLEAR
To:	Martin J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION V)
References
102-02148-WFC-T, 102-2148-WFC-T, NUDOCS 9205280167
Download: ML17306A745 (18)

Text

ACCELERATED DISTRIBUTION DEMON+ATION SYSTEM REGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)

ACCESSION NBR:9205280167 DOC.DATE: 92/05/14 NOTARIZED: NO -DOCKET 5 FACIL:STN-50-528 Palo Verde Nuclear Station, Unit 1, Arizona Publi 05000528 STN-50-530 Palo Verde Nuclear Station, Unit 3, Arizona Publi 05000530 AUTH. NAME AUTHOR AFFILIATION CONWAY,W.F. Arizona Public Service Co. (formerly Arizona Nuclear Power RECIP.NAME RECIPIENT AFFILIATION

'MARTIN,J.B. Region 5 (Post 820201)

SUBJECT:

Forwards response to NRC 920407 request for addi info re concerns received by NRC re 911027 reactor scram & primary D sys cooldown.Surveillance testing for AFW sys -adequate to demonstrate AFW pump operability. Concerns unsubstantiated.

DISTRIBUTION CODE IE01D COPIES RECEIVED:LTR ENCL SIZE:

TITLE: General (50 Dkt)-Insp Rept/Notice of Violation Response NOTES:STANDARDIZED PLANT 05000528 Standardized plant. 05000530 D

RECIPIENT COPIES RECIPIENT COPIES D ID CODE/NAME LTTR ENCL ID CODE/NAME LTTR ENCL PD5 PD 1 1 TRAMMELL,C 1 1 THOMPSON,M 1 1 INTERNAL: ACRS 2 2 AEOD AEOD/ DE I IB 1 1 AEOD/DSP/TPAB DEDRO 1 1 NRR MORISSEAU,D NRR/DLPQ/LHFBPT 1 1 NRR/DLPQ/LPEB10 NRR/DOEA/OEAB 1 1 NRR/DREP/PEPB9H NRR/DST/DIR 8E2 1 1 NRR/PMAS/ILRB12 NUDOCS-ABSTRACT 1 1 OE DIR OGC/HDS 1 1 1 MEG FI'" 02 RGN5 FI LE 0 1 1 1 EXTERNAL EG6rG/BRYCEiJ ~ HE 1 1 NRC PDR R NSIC 1 1, D

D D

NOTE TO ALL "RIDS" RECIPIENTS PLEASE HELP US TO REDUCE iYASTE! CONTACT THE DOCUMENT CONTROL DESK ROOM Pl-37 (EXT. 20079) TO ELIMINATEYOUR NAME FROM DISTRIBUTION LISTS FOR DOCUMENTS YOU DON'T NEED!

TOTAL NUMBER OF COPIES REQUIRED: LTTR 24 ENCL 24

Arizona Public Service Company P.O. BOX 53999 ~ PHOENIX, ARIZONA85072-3999

'102-02148-WFC/TRB/RKR WILLIAMF. CONWAY EXECUTIVEVICEPRESIDENT May 14, 1992 NUCLEAR Mr. John B. Martin g

Regional Administrator, Region V I I1 U. S. Nuclear Regulatory Commission 1450 Maria Lane, Suite 210 I'<

Walnut Creek, CA 94596-5368 fD

Reference:

Letter dated April 7, 1992, from R. P. Zimmerman, Director, Division of Reactor Safety and Projects, NRC, to William F.

Conway, Executive Vice President Nuclear, Arizona Public Service Company

SUBJECT:

PALO VERDE NUCLEAR GENERATING STATION (PVNGS)

UNITS 1, 2, AND 3 REPLY TO REQUEST FOR INFORMATION REGARDING CONCERNS RECEIVED BY THE NRC FILE: 92-070-'026 The referenced letter requested Arizona Public Service Company (APS) to respond to certain concerns that the NRC had received regarding (1) the October 27, 1991 events in Units 1 and 3 in which a reactor scram occurred, (2) main steam safety valve testing, and (3) motor-driven auxiliary feedwater pump testing. During a telephone conveisation on May 5, 1992, between H. Wong, NRC, and T. R. Bradish, APS, an extension of the due date for this response from May 7, 1992 to May 15, 1992, was granted. Enclosed is APS'esponse to these concerns. APS has determined that the concerns are not substantiated.

wr r, i v ~(<lf 9205280ib7 9205i4 PDR ADOCK-0500052S S PDR QE- d'/

(

l I

t I

I

Mr. John B. Martin .

U. S. Nuclear Regulatory Commission Reply to Request for Information Page 2 Should you have any questions regarding this response, please contact Thomas R.

Bradish at (602) 393-5421.

Sincerely, WFC/TRB/RKR Enclosure CC: D. H. Coe Document Control Desk

ENCLOSURE REPLY TO REQUEST FOR INFORMATION

CONCERN NO. 1 Cooldown procedures 41ST-1RC01 and 43ST-3RC01 were not performed when required [...during the October 27, 1991 events in Units 1 and 3 in which a reactor scram and primary system cooldown occurred].

APS RESPONSE At approximately 0722 MST on October 27, 1991, Palo Verde Units 1 and 3 were operating at approximately 100 percent power when a grid perturbation resulted in reactor trips in both units. Immediately following the reactor trips, Safety Injection Actuation System (SIAS) and Containment Isolation Actuation System (CIAS)

Engineered Safety Feature Actuation System actuations occurred on low pressurizer pressure. All required safety system components actuated as designed in each unit.

By approximately 0805 MST on October 27, 1991, the units were stabilized in Mode 3 (HOT STANDBY) at normal operating temperature and pressure. Temperature changes following the reactor trip were within the limits in Technical Specifications (TS) 3.4.8.1 "Pressure/Temperature Limits RCS" and 3.4.8.2 "Pressurizer Heatup/Cooldown Rates." Licensee Event Report 528/91-010-00, dated November 26, 1991, provides a detailed description of this event.

During the period from October 27, 1991, following the reactor trip arid stabilization of the unit in MODE 3, to October 31, 1991, when Unit 1 entered Mode 2 (STARTUP),

reactor coolant system (RCS) steady-state cold leg temperature ranged from approximately 560 degrees Fahrenheit to approximately 566 degrees Fahrenheit.

Normal RCS cold leg temperature during MODE 1 (POWER OPERATION) is approximately 565 degrees Fahrenheit. RCS cooldown or heatup did not occur and was not required as a result of or following the reactor trip.

During the period from October 27, 1991, following the reactor trip and stabilization of the unit in MODE 3, to October 30, 1991, when Unit 3 entered Mode 2 (STARTUP),

RCS steady-state cold leg temperature ranged from approximately 562 degrees Fahrenheit to approximately 566 degrees Fahrenheit. Normal RCS cold leg temperature during MODE 1 (POWER OPERATION) is approximately 565 degrees Fahrenheit. RCS cooldown or heatup did not occur and was not required as a result of or following the reactor trip.

Procedures 41ST-1RC01 and 43ST-3RC01, "RCS and Pressurizer Heatup and Cooldown Rates 4.4.8.1.1, 4.4.8.2.1, 4.4.8.2.2 8 4.4.8.3.1," for Units 1 and 3 respectively, implement TS 3.4.8.1, 3.4.8.2, and 3.4.8.3 "Overpressure Protection Systems." These procedures are required whenever there is a sustained RCS Page 1 of 9

1 1

temperature change of twenty degrees Fahrenheit or more over a period of time. The temperature changes associated with a plant transient (i.e., reactor trip) do not require performance of these procedures. Therefore, performance of procedures 41ST-

. 1RC01 and 43ST-3RC01 was not required during the October 27, 1991 events in Units 1 and 3.

Based on the above information, this concern is not substantiated.

Page 2 of 9

CONCERN NO. 2 MSSV test procedures were not being followed in that the setpoints of some valves were changed following the initial test instead of the third test as required by the procedure.

APS RESPONSE APS has reviewed the records of past performance of the Main Steam Safety Valves (MSSV) test procedures and found that at least two tests were performed prior to setpoint adjustment. The test procedure, 73ST-9ZZ18, "Main Steam PSV Set Pressure Verification," provides for only one additional test after an initial'(as-found) test. There is no requirement for three tests to be performed prior to setpoint adjustment.

Therefore, the records show that the directions in the procedure were followed.

However, it should be recognized that the procedure also permits the test director to deviate from the requirement to perform two tests prior to setpoint adjustment if he feels that the additional testing will not provide relevant data for purposes of performing a setpoint adjustment.

Prior to 1992, a Note in procedure 73ST-9ZZ18 stated that "In the event of an initial test failure, DO NOT make any adjustments to valve setpoint. After a minimum 15 minute wait, the test may be reperformed. If results are still unacceptable, adjustment of setpoint is necessary (after 2 or 3 successive out-of-tolerance lifts)... The Test Director shall determine when to make adjustments." Subsequently, procedure 73ST-9ZZ18 was revised on January 15, 1992. The Note in the test instructions section of the procedure was changed to state that "In the event of a failure of the initial (as-found) test, no adjustments should be made to the valve setpoint. After a minimum 10 minute wait, the test may be reperformed. If results are still unacceptable, adjustment of the setpoint may be necessary. Otherwise, the test director shall determine when to make adjustments,"

Based on the above information, this concern is not substantiated.

Page 3 of 9

CONCERN NO. 3 The site and laboratory test director was not qualified for his position in that he did not have a college degree, high school diploma, or BED certificate. In addition, his previous experience was as a pipefitter.

l APS RESPONSE APS engineers who supervise ASME Section XI (pump and valve) testing are required by procedure 73DP-OTR01, "Qualification and Training Requirements for Component and Specialty Engineering," to be qualified to the requirements of ANSI/ANS 3.1-1978, "Standard for Selection and Training of Personnel for Nuclear Power Plants." The documentation for past relief valve testing per procedures 73ST-9ZZ18 (MSSVs),

73ST-9ZZ19 (low temperature overpressure protection), and 73ST-9ZZ20 (remaining ASME Section III valves) has been reviewed and all persons performing the function of the test director or supervisor met these qualification requirements.

Offsite testing is currently performed by Westinghouse personnel. The Westinghouse personnel conducting the testing are qualified to the requirements of ANSI/ASME PTC 25.3 and ANSI/ASME OM-1 "Requirements for Inservice Performance Testing of Nuclear Power Plant Pressure Relief Devices." APS engineering reviewed the Westinghouse qualification procedure and the qualification reports for Westinghouse personnel involved in the testing and confirmed that the Westinghouse personnel met the qualification requirements of ANSI/ASME PTC 25.3 and ANSI/ASME OM-1.

Based on the above information, this concern is not substantiated.

Page 4 of 9

CONCERN NO. 4 The number of valves to be tested in situ had been decreased and it was riot clear to what industry standard the valves were to be tested.

1 APS RESPONSE The MSSVs are required by Technical Specifications Surveillance Requirement 4.7.1.1 and the ASME Code to be tested at least once per five years. The testing program for MSSVs ensures that all MSSVs are tested within the five year interval.

In 1991 APS initiated an enhanced preventive maintenance and testing program for MSSVs that requires that MSSVs requiring testing be removed during a refueling outage and sent to an offsite test facility for testing. The MSSVs are tested using live steam to lift the valve disc off its seat. The "in situ" testing method used a hydraulic assist device to lift the valve disc off its seat. The live steam testing is more representative of the conditions that the MSSVs would operate in than the "in situ" testing. Since the MSSVs are tested at the offsite test facility, they are not tested "in situ". The testing at the offsite test facility does not affect the frequency of valve testing.

APS tests MSSVs in accordance with the requirements of ANSI/ASME PTC 25.3-1976, "Safety and Relief Valve Performance Test Codes." APS has also incorporated some of the requirements of ASME OM-1, "Requirements for Inservice Performance Testing of Nuclear Power Plant Pressure Relief Devices," even though there is no requirement for implerrientation of ASME OM-1 at PVNGS.

Based on the above information, this concern is not substantiated.

Page 5 of 9

CONCERN NO. 5 Main steam safety valves (MSSVs) were being mishandled in a manner which could affect the as-found setpoints.

APS RESPONSE APS is very concerned about the potential for setpoint drift due to mishandling of MSSVs. The MSSV internals are held in place by compression of the main spring. An MSSV would have to be dropped or tilted to affect the internals and result in a setpoint change. The MSSVs are removed and installed in accordance with an approved procedure. APS engineering reviewed and witnessed the handling of MSSVs during the current Unit 1 refueling outage, including removal from the main steam system, loading into shipping containers, shipping offsite, receipt inspection at the offsite test facility, reloading into shipping containers, shipping back to PVNGS, receipt inspection at PVNGS, and reinstallation to the main steam system. There was no evidence of mishandling that could cause the MSSV setpoints to drift. The MSSVs are shipped back to PVNGS in an air ride trailer, and the shipping containers have tilt indicators attached to them. The tilt indicators provide assurance that the MSSVs were not mishandled. If a valve is discovered "tilted," APS engineering is notified and the valve is sent back to the offsite test facility for recertification.

APS also shipped the Unit 1 pressurizer safety valves (PSV) to the offsite test facility for testing during the current Unit 1 refueling outage. Prior to reinstallation in Unit 1, one PSVs'ilt indicator indicated "tilted" and one PSVs'ilt indicator was missing. Both PSVs were sent back to the offsite test facility for recertification.

Based on the above information, this concern is not substantiated.

Page 6 of 9

. CONCERN NO. 6 Laboratory testing of MSSVs would not be as accurate as in situ testing of the valves.

APS RESPONSE APS has determined that the laboratory testing of MSSVs is more accurate than "in situ" testing. This is based on APS and industry experience with "in situ" testing and testing using elevated steam pressure.

APS currently tests the MSSVs at an offsite testing facility (Westinghouse Test Facility) under elevated steam pressure conditions. The testing uses an actual gage reading for the steam pressure required to lift the valve disc off its seat. The published uncertainty'of the dead weight tester is + 0.025 percent.

The "in situ" testing previously performed at PVNGS used a hydraulic assist device with a load cell and a strip chart recorder to determine when the valve disc lifted off its seat. The actual system pressure and the force required to lift the valve disc off its seat (read from the strip chart recorder) were combined to determine the lift setpoint.

Since the test pressure relies on several indications (system pressure, load cell, strip chart recorder, and operator reading of the strip chart recorder), the uncertainty of the "in situ" testing is much greater than the testing performed at the offsite test facility.

Based on the above information, this concern is not substantiated.

Page 7 of 9

CONCERN NO.?

Electric-driven AFW pumps frequently failed their surveillance tests and oscillated when they should have run at a steady speed.

APS RESPONSE

Test results for Unit 1, 2, and 3 for the essential motor driven auxiliary feedwater (AFW) pumps show that these AFW pumps have successfully passed their surveillance tests at a rate of 93.67, 97.14, and 93.75 percent respectively. The tests are performed monthly. The lowest success rate corresponds to one failure for every sixteen tests. APS does not consider this to be an excessive failure rate.

The essential motor driven AFW pump is not powered by a variable speed or frequency motor. Therefore, the essential motor driven AFW pump operates at only one speed. The concern may be addressing pressure pulsations that have been experienced during testing. These pulsations are the result of the hydraulics of the mini-recirculation flow orifice. The flow orifice sets up pulsations which are intensified by pipe resonance frequencies and the pump hydraulics. When these pulsations occur, they are eliminated by changing the pipe resonance frequency by closing a valve and changing the effective pipe length. The pulsations only occur during pump testing and can be manually eliminated. The pulsations do not affect the AFW pumps ability to operate when required. As a permanent corrective action, a design change is being implemented to replace the flow orifice with a drag valve. The design change was implemented in one of the AFW trains in Unit 3 as a temporary modification to verify that the design change will eliminate the pulsations. The Unit 3 AFW train has not experienced these pulsations since the temporary modiTication was installed. The design change is expected to be implemented during the fourth refueling outage for Units 1, 2, and 3 (scheduled for September 1993, March 1993, and February 1994 respectively).

Based on the above information, this concern is not substantiated.

Page 8 of 9

CONCERN NO. 8 The testing methods for the AFW system were poor and unfavorable test results were altered to either produce better results or accept the poor results.

APS RESPONSE The surveillance testing for the AFW system is adequate to demonstrate AFW pump operability. The test method is consistent with industry practice. The test involves running the AFW pump in the minimum recirculation mode and recording the pump discharge pressure (corrected for elevation differences) and pump suction pressure (based on Condensate Storage Tank level). The difference in the pressures is compared to acceptance criteria. Pump initial and final speed are also recorded and must be within acceptance criteria since they affect differential pressure.

In addition, to determine pump operability, the surveillance test also requires that pump vibration readings be recorded and trended. The vibration readings provide an early warning of degraded performance. Unfavorable test results are not altered to either produce better test results or accept the poor results. As discussed in the APS response to Concern No. 7, occasionally there is a surveillance test failure.

This concern may be addressing the establishment of a new baseline for vibration readings. The ASME Code allows for the establishment of new baseline readings whenever any work has been done on the component that could affect vibration readings, whenever it is necessary or desirable, or whenever there are changes in the methods or techniques for data recording. APS establishes a new baseline only when allowed by the Code. Therefore, it may appear that data which was unacceptable or poor one month, is acceptable the next month if a new baseline is established.

Based on the above information, this concern is not substantiated.

Page 9 of 9