ML20215F368

Safety Evaluation Supporting Util Response to IE Bulletin 86-002 Re Static O-Ring Differential Pressure Switches

ML20215F368
Person / Time
Site:	Oyster Creek
Issue date:	12/15/1986
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20215F347	List: ML20215F344 ML20215F368
References
IEB-86-002, IEB-86-2, NUDOCS 8612230370
Download: ML20215F368 (5)

Text

.

/ 'o,, UNITED STATES v  ! '

o NUCLEAR PSGULATORY COMMISSION a wAsumGTON, D. C. 20555

't l

SAFFTY EVALUATION BY THE OFFICE OF NtJCLEAR REACT 0D REGilLATION DELATING TO STATIC 0-RINC SWITCHES GPli NitCLEAP CORPORATION JERSEY CENTRAI. DOWER AND LIGHT COMPANY OYSTED CREEV,NttCLEAR GENERATING STATION DOCKET NO. 50-919 1.0 INTRODllCTION On January 17, 1986, three out of four reactor low water level sensors at Oyster Creek were found out of specification during the monthly surveillance. The instrumentation is Static 0-Rina (SOR) differential pressure (dpi switches. The "as-found" setpoints for the three switches had drifted downward as much as 6 inches. During the subseouent weeks, the level switches continued to perform erratically. GPU Nuclear (the licenseel continued to perfom additional surveillance, beyond that required in the Technical Specifications (TS), on these switches until about April 12, 1986, when the reactor was shut down for the Cycle 11 Pefueling (Cycle 11R) outage. The S0P dp switches in the reactor low t

water level instrumentation were replaced in March 1986 by the same model SOR dp switches in the reactor low-low water level instrumenta-tion. This is discussed in meeting summaries dated February 24, March 14, and ecgust 14, 1986.

2.0 DISCtJSSION On June 1, 1986, LaSalle 2 experienced a feedwater transient that resulted in low water level in the reactor vessel. One of four low level trip channels which use SOR dp switches actuated, resultino in a half scram. The operator recovered water level and continued power operation. Subsequent reviews by the laSalle licensee raised concerns that the water level may have gone below the scram setpoint and that a malfunction of the reactor scram system may have occurred. Based on this concern, La Salle notified the NRC and subsequently informed SOR Company of possible switch malfunctions. This incident is described in greater detail in IE Information Notice 86-47 dated June 10, 1986 l

Following the meeting of June 12, 1986, with interested parties includino the licensee, on the erratic behavior of these switches, NRC issued IE Bulletin (IER) 86-02 on SOR dp switches, dated July 18, 1986. The Bulletin focussed on SOR Model 102 and 103 dp switches which were 8612230370 861215 PDR ADOCK 05000219 O PDR

installed at Oyster Creek and LaSalle 2. It stated that operation of the SOR dp switches with "0" rings on the cross shaft had been found to be erratic with little correlation between the setpoints established during atmospheric pressure (0 psig) calibrations and switch actuations under reactor system pressure conditions (1000 psig).

In its letter dated July 30, 1986, the licensee identified that Oyster Creek uses SOR Model 103 dp switches for (1) reactor low and low-low vessel water le"el trip input to the reactor scram system and engineered-safety-features actuation, respectively, (2) the safety-related core spray system and (3) Reactor Building to Torus vacuum breaker system. The reactor water level instrumentation is not discussed here because it was replaced in the Cycle 11R cutage by an analog trip system.

For the core spray system, the SOR dp switch starts the alternate core spray booster pump if the primary booster pump has a low discharge head which would indicate its failure to operate. The SOR switch senses the dp across the primary booster pump. Each core spray train has a primary and an alternate booster pump. The switch operating pressure is 140 psig suction to 250 psio discharge which is hioher than the calibration pressure of 0 psig of these switches.

For the vacuum breakers, the SOR dp switch opens the vacuum breakers to equalize the pressure between the Torus and Reactor Buildings. The switch operates if the torus is about 0.5 psi differential pressure (psid) lower than the Reactor Builcing. The switch operating pressure is the trip setting which is slightly less than the 0.5 psid in TS 3.5.A.4.a.

This is very near the calibration pressure at 0 psig of these switches.

3.0 EVALUATION

. By letter dated May 27, 1986, the licensee committed'to replace the SOR switches used in the reactor low and low-low water level applications in the Cycle 11R outage with analog trip systems. .This has been completed in the Cycle 11R outage.

By letter dated September 23, 1986, the licensee provided information on the remaining SOR switches used in the safety-related core spray booster pump failure to operate system (RV-40) and Reactor Building to Torus vacuum breakers (DPS-66) as required by IEB 86-02 Action 5. The licensee stated that based on surveillance data and the special test program data required by Action 3 of IEB 86-02, the surveillance of DPS-66 l switches will be done monthly, instead of quarterly as required by the TS 4.5.I.4.a. for a period of 6 months and the RV-40 switches will continue to have surveillance conducted monthly as required in TS 4.4.A.1 on core spray pump operability. If any test indicates an unacceptable shift in the instrument set point, the licensee stated it will initiate

' interim corrective actions and submit a report to the staff at the end of

'the 6 months of testing these switches.

The staff concludes that this is acceptable for short-term operation because it is consistent with Action 5 of IEB 86-02.

i In response to TER 86-0? Action 6, the licensee stated in its latter dated September 23, 1986, that SOR dp switches for DPS-66 and RV-40 applications are not sub.iect to operating reactor pressure and, therefore, the potential for deformation o# the "0" ring and subseouent hinding on the cross shaft in the switches does not exist. Also, the licensee further stated that the observed drift over a period of 5 months for nPS-66 and 16 months for RV-40 switches has been within the specified TS limits, there is no common mode failure, the switches function reliably ar.d, therefore, no replacement or long-term corrective action for these switches should be required.

in addition, by letter dated October 14, 1986, the licensee stated that the DPS-66 switches should be removed from the scope of the IER C6-02 since the prescribed level of confidence has been achieved by the special test program. The licensee stated, for the DPS-66 switches, that (1) the switches are calibrated within only 0.5 psid of the trip setting, (?) the as-found setpoint has never been found to exceed the TS trip setting (3) the two vacuum breakers controlled by these switches are both sized for full design air flow so that only one SOR switch has to function, (41 the switches are installed so that a failure of one will not affect the operation of the other and (5) operator training will be revised to discuss the indications of e.nd compensatory measures applicable to the failure of these switches.

However, the licensee has committed to testine these DPS-66 switches monthly .for a 6-month pariod and to notify the staff if the as-found setpoint of these switches exceeds the TS trip setecint limit.

The staff has reviewed the licensee's responses to Action 6 of the IEP 86-0? and finds that the corrective action taken by the licensee #or Av-40 switches is in accordance with IER 86 07 end is, there# ore, acceptable. Powever, the staff concludes that the licensee has not given sufficient basis for the exclusion of DPS-66 switches fron IER 86-02 for long-term' action without a review of the results of +be additional testing of these switches over the 6-month testino program.

The licensee is reouested to address long-term corrective ections for all 500 dp switches based on the monthly testina when it submits the results to the ste.ff. The licensee agreed to submit the results in the phone conference with the staff on November 4, 1986. This submittal should include a deternination as to whether improvements in calibration testing methods or setpoint methodology could be achieved.

4.0 CONCLUSION

The staff has reviewed the licensee's submittals erd concludes the following:

1. The licensee's commitment to replace all reactor vessel water level SOR differential pressure switches with an analog trip system is acceptable.

P. The licensee's coninitment to conduct surveillance of OPS-66 and RV-40 switches monthly for a period of 6 months is responsive to the short-term actions reouired by IEP 86-0? Action 5 and is, therefore, acceptable.

r e

i

' l t

t

3. The licensee is reouested to address the long term corrective 1 action for both RV-40 and DPS-66 switches used in the core spray  !

system and Reactor Building to Torus vacuum breakers system i respectively, in its submittal on the results of the 6 month I testing of these switches. l 5.0' REFERENCES

1. IF Bulletin No. 86-02, Static "0" Ring Differertial Pressure Switches, dated July 18, 1986.

P. Letter P. B. Fiedler (GPUN) to J. A. Zwolinski (f:PC), Reactor Protection System Switch Peplacement, dated May 27, 1986.

3. Letter P. B. Fiedler (GPUN) to Dr. T. E. Murley (NRCi, IE Pulletin 86-02, Respon e to Action No. 1, dated July 30, 1986.

4 Letter P. B. Fiedler (GPUN) to Dr. T. E. Murley (NRC), IF 86-02, Static "0" Ring Switches, dated September 23, 1986.

5. Letter P. B. Fiedler (r, PUN) to J. N. Donohew (MDC), IF Bulletin 86-02, SOR D/P Switches, dated October 14, 1986.

6. NRC foneting Summary, January 73, 1986, Meeting with GPU Nucleer Corporation to Discuss the Channel Checks for Operability of the Low and Low-Low Raactor Water Level Instrumentation Channels, +

tated February PA, 1986.  !

7. NRC Meeting Summary, January 1986 Progress Peview Meeting on l Licensing Actions, Section 16.0, dated t' arch 14, 1986.

8. NRC Meeting Summary, June 12, 1986, Meeting with GPU Nuclear Corporation and Other Licensees to Discuss the Erratic Bebevior of Static 0-Ring Switches, dated August 14, 1086.

9. Phone conference between J. Onnohew (NRC) and J. Rogers (GPUN),

Static 0-Ring Switches, on November 4, 1986.

Principal Contributers: N. Trehan and J. Donohew Dated: December 15, 1986 i

l

surveillance. You are requested, as agreed to in the conference call on November 4, 1986, with GPU Nuclear (the licensee), to provide these results to the staff. In your submittal, please provide the long term corrective actions for all the SOR switches based on these results. You are requested to provide this within 90 days after the 6 month testing period. Your long terin corrective actions should include a determination as to whether improvements in calibration testing methods or setpoint methodology are practical for these SOR dp switches.

The reporting and/or recordkeeping requirements contained in this letter affect fewer than ten respondents; therefore, OMB clearance is not required under P.L.96-511.

Sincerely, Original signed by R. Auluck for Jack N. Donohew, Jr., Project Manager BWR Project Directorate #1 Division of BWR Licensing

Enclosure:

Safety Evaluation cc w/ enclosure:

See next page "

DISTRIBUTION:

BeshetJHen NRC'PDR'~~'

Local PDR RWD1 Reading RBernero OGC-BETH(Infoonly)

EJordan BGrimes JPartlow JZwolinski JDonohew CJamerson Clainas MSrinivasan NTrehan

• ACRS(10)

OC file GHolahan DBL:BWD#1 1 I)Bt. b 1 :BWD#1 CJamerson6 J0o chew:ac Zwolinski 12/d/86 12 /86 12/ff/86

._ . __