Hatch Plant SRV Problems And Clear Indications On Normalization Of Deviance

Updated 4/5/2016

I knew Hatch was in trouble with their SRVs at the end of 2015.

Originally published on 11/6/15...republished

On May 11, 2015 at approximately 0923, Unit 2 was at 100 percent rated thermal power (RTP) when the "as-found" testing results of the 2-stage main steam safety relief valves (SRVs) were received which indicated that two of

Only two violations and it still blew accident assumption models.

eleven of the Unit 2 SRVs had experienced a setpoint drift during the previous operating cycle which resulted in their failure to meet the Technical Specification (TS) opening setpoint of 1150 +1- 34.5 psig percent as required by TS Surveillance Requirement (SR) 3.4.3.1.

The root cause of the SRV setpoint drift is attributed to corrosion-induced bonding between the pilot disc and seating surfaces. This conclusion is based on previous root cause analyses and the repetitive nature of this condition at Hatch and within the BWR industry. All 2-stage SRVs with platinum coated pilot seats were removed from Unit 2 during the 2015 refueling outage and replaced with 3-stage SRVs with a modified pilot. 3-stage SRVs typically do not

They put back the LER 2-2011-002 leaking ones. Based on the failed experimentations of the past, what evidence do they have with the modified pilot 3 stage... 

exhibit set point drift and the modified pilot reduces instances of vibration induced spurious openings and leak-by.

As I asked in Pilgrim, early LERs never spoke about steam line vibrations damaging SRVs. It seems like a new or intensifying phenomena. Please show me the historic trends with main steam line vibrations over time?  

Licensee Event Report 2015-004-00 

Safety Relief Valves As Found Settings Resulted in Not Meeting Tech Spec Surveillance Criteria

This event is reportable in accordance with 10 CFR 50.73(a)(2)(i)(B) because a condition occurred that is prohibited by TS Surveillance Requirement (SR) 3.4.3.1. Specifically, an example of multiple test failures is given in NUREG-1022, Revision 3, "Event Reporting Guidelines 10 CFR 50.72 and 50.73" which describes the sequential testing of safety valves. This example notes that "Sometimes multiple valves are found to lift with set points outside of technical specification limits."

NUREG-1022 further states in the example that "discrepancies found in TS surveillance tests should be assumed to occur at the time of the test unless there is firm evidence, based on a review of relevant information (e.g., the equipment history and the cause of failure), to indicate that the discrepancy occurred earlier. However, the existence of similar discrepancies in multiple valves is an indication that the discrepancies may well have arisen over a period of time and the failure mode should be evaluated to make this determination." Based on this guidance and the fact that the development of the corrosion occurred over a period of time of plant operation, the determination was made that this "as found" condition is reportable under the reporting requirements of 10 CFR 50.73(a)(2)(i)(B).

_________________________________________________________________________________

Wiki: Professionalism/Diane Vaughan and the normalization of deviance 

Diane Vaughan is an American sociologist who devoted most of her time on topics as different as "Tension in private life" and "Deviance in organizations". She states, "I find that in common, routine nonconformity, mistake, misconduct, and disaster are systematically produced by the interconnection between environment, organizations, cognition, and choice. These patterns amplify what is known about social structure and have implications for theory, research, and policy". One of Vaughan's theories regarding misconduct within large organizations is the normalization of deviance. 

Philosophy: What is normalization of deviance? 

Description of Normalization of Deviance 

"Social normalization of deviance means that people within the organization become so much accustomed to a deviant behavior that they don't consider it as deviant, despite the fact that they far exceed their own rules for the elementary safety". People grow more accustomed to the deviant behavior the more it occurs. To people outside of the organization, the activities seem deviant; however, people within the organization do not recognize the deviance because it is seen as a normal occurrence. In hindsight, people within the organization realize that their seemingly normal behavior was deviant.

 ________________________________________________________________________________

The results from this MSIVF event analysis was performed by the Nuclear Fuels Department in order to bound the "as-found" results of the U2 Cycle 21 2-stage SRVs pressure setpoint drift. The results from this analysis showed a small increase in peak pressures relative to the Hatch-2 Cycle 21 reload licensing analysis (RLA) results. The higher peak pressures were due to the fact that eight of

Man, look at the runaway "normalization of deviance"? Higher pressure than assumed in the RLA. There should be big alarm bells going off in people heads when actual plant condition goes past nuclear safety analysis assumptions. I exceeded the spending limit but I am safe because a police officer didn't give me ticket. This is related with shifting from plus or minus 1% to plus or minus 3% setpoint accuracy tech spec limits. 

the eleven SRVs opened at pressures higher than that which was assumed in the RLA. It should be noted that in this analysis, the larger actual valve bore size was used in the calculations for nine of the valves rather than the smaller bore size which was conservatively assumed in the RLA. Therefore, higher steam flow capacities than those assumed in the RLA were used in this analysis for those nine valves.

Based on the analysis, the calculated minimum margin to the 1375 psig ASME Boiler and Pressure Vessel Code overpressure limit for peak vessel pressure would have been 27.7 psig and the minimum margin to the 1325 psig Tech Spec Safety Limit for the reactor steam dome pressure would have been 2.9 psig during an MSIVF event during Cycle 21 operation. Therefore, these test results show that in this case, where two of the eleven SRVs would have opened at pressures

Man, "normalization of deviance" on steroids with accident dome pressure going past accident assume valves. What about just fixing the valves so the plant stay way clear of any assumed accident analysis or limits?

higher than that which was assumed in the RLA, the peak pressure at the bottom of the vessel would have remained below the ASME Boiler and Pressure Vessel code limit and the peak RPV dome pressure remained within the TS Safety limits.

PREVIOUS SIMILAR EVENTS:

LER 1-2014-003, identified multiple SRV setpoint drifts for 5 of the 11 two-stage SRVs installed on Unit 1. The two-stage SRVs with platinum-coated pilot discs were removed from Unit 1 during the 2014 refueling outage and replaced with 3-stage SRVs that have a modified pilot. The modified pilots will help reduce spurious openings and leak-by due to system vibration.

LER 1-2012-004, identified multiple SRV setpoint drift for 8 of the 11 SRVs. Corrective actions included replacement of the 2-stage SRVs with 2-stage SRVs whose pilot discs had undergone a platinum surface treatment which was considered at that time to be the long term fix for this corrosion bonding issue.

This is what irks me about Pilgrim's 3 stage problems. Why don't these guys learn from each other? These guys below had terrible troubles with the new install 3 stage (leaks), two forced shutdowns, then immediately yanked them out. Pilgrim danced with their over two operational period and got themselves deeply in trouble with the NRC.      

LER 2-2011-002, identified multiple SRV setpoint drift for 8 of the 11 SRVs. Corrective actions included replacement of the 2-stage SRVs with 3-stage SRVs during the Unit 2 Spring 2011 refueling outage which was considered at that time to be the long term fix for this corrosion bonding issue. Subsequent to that outage the 3-stage SRVs exhibited signs of unacceptable leakage which resulted in two separate outages that involved changing out four SRVs during the first outage and the remaining seven SRVs during the subsequent outage in May 2012. The 3-stage SRVs were replaced with 2-stage SRVs containing pilot discs that had undergone the platinum surface treatment.

LER 1-2010-001, identified multiple SRV setpoint drift for 5 of the 11 SRVs. Corrective actions included refurbishment of the pilot valves and included the replacement of the pilot discs with discs made from Satellite 21 material. Additionally, the insulation surrounding each SRV was upgraded to improve resistance to corrosion-induced bonding. These were the same actions that were taken following similar failures reported in LEA 2-2009-001, since improved results had been seen to some degree in the industry for at least one operating cycle when these actions were implemented.