Junk Plant Hope Creek: Not Worth It To Fix Safety Components (SRVs), Because We Might Permanently Shutdown

Licensee Event Report 2018-002-01 Safety Relief Valve (SRV) As-Found Setpoint Failure

On April 20, 2018, Hope Creek Generating Station (HCGS) received results that the 'as-found' set-point tests for safety relief valve (SRV) pilot stage assemblies had exceeded the lift setting tolerance prescribed in Technical Specification (TS) 3.4.2.1. The TS requires the SRV lift settings to be within +/- 3% of the nominal set-point value.
During the twenty-first refueling outage (H1 R21 ), all fourteen SRV pilot stage assemblies were removed for testing at an offsite facility. Between April 20 and May 11, 2018, HCGS received the test results for all fourteen of the SRV pilot valve assemblies. A total of eight of the fourteen SRV pilot stage assemblies experienced set-point drift outside of the TS 3.4.2.1 specified values. All of the valves failing to meet the limits were Target Rock Model 7567F two-stage SRVs. 

This is a condition reportable under 10 CFR 50.73(a)(2)(i)(B) as an Operation or Condition Prohibited by Technical Specifications.

This is the wording from above. They are suppose to be shutdown if they find more than one SRVs outside the  valve pressure testing lifting accuracy. They requiring 13 SRVs to be operable implies there is not much safety slack in the SRV. 57% of the valves were unacceptable. The magnitude of these kinds of failures have been going on for decades. The NRC got a industry group studying this failure, and after NRC prodding by me, Hope Creed has admitted this is unacceptable quality assurance. They have been promising to replace these valves for years.    

"Technical Specification (TS) 3.4.2.1 requires that the safety function of at least 13 of 14 SRVs be operable with a specified code safety valve function lift setting, within a tolerance of +/- 3%. Action (a) of TS 3.4.2.1 specifies "With the safety valve function of two or more of the above listed fourteen safety/relief valves inoperable, be in at least HOT SHUTDOWN within 12 hours and in COLD SHUTDOWN within the next 24 hours." Therefore, this is a condition reportable under 10 CFR 50.73(a)(2)(i)(B) as an Operation or Condition Prohibited by TS."

The cause of the set-point drift for the eight SRV pilot stage assemblies is attributed to corrosion bonding between the pilot disc and seating surfaces, which is consistent with industry experience. This conclusion is based on previous cause evaluations and the repetitive nature of this condition at HCGS and within the BWR industry.

CONDITIONS PRIOR TO OCCURRENCE When the reports of the 'as-found' results were received, Hope Creek was in Operational Condition (OPCON) 5, Refuel, at 0 percent rated thermal power. No other structures, systems or components that could have contributed to the event were inoperable at the time of the event.  

DESCRIPTION OF OCCURRENCE During the twenty-first refueling outage (H1 R21) at Hope Creek Generating Station (HCGS), all fourteen Main Steam safety relief valves (SRV) pilot stage assemblies {SB/RV} were removed and tested at NWS Technologies. The SRVs are Target Rock Model 7567F two-stage SRVs. During the period from April 20 through May 11, 2018, HCGS received the results of the 'as-found' set pressure testing required by Technical Specification (TS) Surveillance Requirement (SR) 4.4.2.2. A total of eight of the fourteen SRV pilot stage assemblies had set-point drift outside of the required TS 3.4.2.1 tolerance values of +/-3% of nominal value. 

The 'as-found' test results for the eight SRVs not meeting the TS requirements are as follows:
Valve ID As Found TS Lift Setting Acceptable Band % Difference (psig) (psi g) (psig) Actual F013B 1210 1130 1096.1-1163.9 7.10% F013D 1191 1130 1 096.1 - 1163.9 5.40% F013F 1146 1108 1074.8- 1141.2 3.40% F013G 1197 1120 1 086.4 - 1153.6 6.90% F013H 1200 1108 1074.8-1141.2 8.30% F013L 1155 1120 1086.4- 1153.6 3.10% F013M 1161 1108 1074.8-1141.2 4.80% F013P 1199 1120 1086.4- 1153.6 7.10%
Technical Specification (TS) 3.4.2.1 requires that the safety function of at least 13 of 14 SRVs be operable with a specified code safety valve function lift setting, within a tolerance of +/- 3%. Action (a) of TS 3.4.2.1 specifies "With the safety valve function of two or more of the above listed fourteen safety/relief valves inoperable, be in at least HOT SHUTDOWN within 12 hours and in COLD SHUTDOWN within the next 24 hours." Therefore, this is a condition reportable under 10 CFR 50.73(a)(2)(i)(B) as an Operation or Condition Prohibited by TS. 

DESCRIPTION OF OCCURRENCE (Continued)
YEAR
2018
SEQUENTIAL NUMBER
• 002
The extent of condition for this event is to expand the scope of the SRV Group 1 valve testing, per ASME OM Code Section 1-1320 for Class 1 Pressure Relief Valves. However, since all fourteen SRV pilot stage assemblies were removed and replaced with tested spares during the refueling outage (H1 R21), the extent of condition scope was satisfied. 

CAUSE OF EVENT
REV NO.
• 01
The cause of the set-point drift for the eight SRV pilot stage assemblies is attributed to corrosion bonding between the pilot disc and seating surfaces, which is consistent with industry experience. This conclusion is based on previous cause evaluations and the repetitive nature of this condition at HCGS and within the BWR industry.
One of the eight SRVs that experienced set point drift, F013H, was determined to have a second failure mechanism present. The H SRV was the only valve that failed its second (informational) lift test. Disassembly of the SRV pilot revealed steam-cutting of the pilot disc and valve seat, as well as a build-up of corrosion products on the seating surface of the pilot valve. Leak-by on the pilot disc resulted in damage to the pilot seat which affected the lift setpoint. This is the cause of the second test lift to remain outside of the acceptable tolerance. This steam leak also caused the corrosion products to be seen on the seat base material.  

SAFETY CONSEQUENCES AND IMPLICATIONS There were no instances during cycle 21 that resulted in any of the fourteen SRVs being declared inoperable and there were no events during that cycle that required operation of the SRVs. All SRVs lifted well below the Safety Limit, providing reasonable assurance that accident analysis conclusions would remain valid. The industry has recognized that corrosion bonding occurs during the operating cycle. Once an SRV lifts, the corrosion bond breaks and subsequent openings occur very close to the set point as demonstrated during testing, with the exception of the H SRV I as described above. 

Since the eight as-found setpoint SRVs are within their Maximum Allowable Percent Increase (MAPI) above SRV nominal setpoint criteria established in GE document NEDC-32511 P, "Safety Review for Hope Creek Generating Station Safety/Relief Valve Tolerance Analysis", the SRVs are bounded by their MAPI value and no formal Technical Evaluation is required.  

SAFETY SYSTEM FUNCTIONAL FAILURE A review of this condition and a previous technical evaluation documents this is not a functional failure, therefore it was determined that a Safety System Functional Failure (SSFF) as defined in Nuclear Energy Institute (NEI) 99-02, "Regulatory Assessment Performance Indicator Guideline," did not occur. 

PREVIOUS EVENTS Similar events occurred during the 2015 (H1 R19) and 2016 (H1 R20) Hope Creek refueling outages when multiple SRVs were found out of the TS required limits of +/- 3%. These events were reported as LER 354/2015-004-00 (ten inoperable SRVs) and LER 354/2016-003-00 (ten inoperable SRVs).  

CORRECTIVE ACTIONS 1. All 14 SRV pilot stage assemblies were removed and replaced with pre-tested, certified spare pilot valves (H1 R21 ). 2. Evaluate options for the replacement of the currently installed Target Rock two-stage SRVs with a design that eliminates setpoint drift events exceeding +/-3% and improve SRV reliability. The replacement schedule will be developed after a suitable valve is identified. 

Why can't they find suitable valves?