United States Nuclear Regulatory Commission - Protecting People and the Environment

Resolution of Generic Safety Issues: Task II.E.6: In Situ Testing of Valves (Rev. 2) ( NUREG-0933, Main Report with Supplements 1–34 )

The objective of this task was to evaluate whether existing requirements for valve testing provided adequate assurance of performance under design conditions.

ITEM II.E.6.1: TEST ADEQUACY STUDY

DESCRIPTION

Historical Background

The purpose of this TMI Action Plan48 item was to establish the adequacy of existing requirements for safety-related valve testing. It recommended a study which would result in recommendations for alternate means of verifying performance requirements.

Safety Significance

Valve performance is critical to the successful functioning of a large number of a plant's safety systems.

Possible Solution

It could be assumed that a study would be conducted for both PWRs and BWRs and that it could result in recommendations for additional testing and/or maintenance on all safety-related valves. A program to implement the recommendations would then be required at all plants.

PRIORITY DETERMINATION

Assumptions

In an analysis of this issue by PNL,64 it was assumed that all safety-related valves would be affected by resolution of the issue. Then, since all the dominant accident sequences (of Oconee-3 and Grand Gulf-1, the representative plants) involved failures of such valves, the sequences themselves were assumed to be directly affected. It was assumed that the new program would produce a reduction of 5% in the frequencies of the affected accident sequences (those that involved safety-related valves).

Frequency Estimate

It was determined64 that all accident sequences for Oconee-3, except the following, involved safety-related valves and were thus assumed to be affected: T2MLUO, T2KMO, T1(B3)MLU, T1MLUO, and T3MLUO. For Grand Gulf-1, the only exception was T23C.

For all the affected parameters, the base case frequency was taken as the original value. The adjusted case frequency was then calculated by the 5% reduction. The core-melt frequency reduction was then calculated to be 3 x 10-6 /RY and 10-6/RY for Oconee-3 and Grand Gulf-1, respectively.

Consequence Estimate

Based on the 5% reduction, the public risk reduction was calculated to be 7.1 man-rem/RY and 7.8 man-rem/RY for Oconee-3 and Grand Gulf-1, respectively. The average remaining lives of the 95 affected PWRs and the 49 affected BWRs were calculated to be 28.2 years and 26.2 years, respectively. This resulted in a potential risk reduction of 1.9 x 104 man-rem for PWRs and 104 man-rem for BWRs. Thus, the total risk reduction associated with this issue was approximately 3 x 104 man-rem.

Cost Estimate

Industry Cost: It was estimated that the implementation effort for engineering, etc., would be about 10 man-weeks/plant for PWRs and 8 man-weeks/plant for BWRs. (The difference was due to the fewer number of affected valves in a BWR.) The cost was then calculated as follows:

PWRs: (10 man-weeks/plant)($2,000/man-week) = $20,000/plant

BWRs: (8 man-weeks/plant)($2,000/man-week) = $16,000/plant

For the 95 PWRs and 49 BWRs, this cost amounted to $2.7M.

The annual industry effort for operations and maintenance was estimated to be 16 man-weeks/RY for PWRs and 12 man-weeks/RY for BWRs with resultant costs of $16,000/RY and $12,000/RY for PWRs and BWRs, respectively. For the 95 PWRs with an average remaining life of 28.2 years, the cost was approximately $42.9M. For the 49 BWRs with an average remaining life of 26.2 years, the cost was approximately $15.4M.

Thus, the total industry cost to implement the possible solution to this issue was $(2.7 + 42.9 + 15.4)M or $61M.

NRC Cost: NRC labor for development of the solution for PWRs was estimated to be 1 man-year. Implementation of the solution was estimated to take 1 man-week/plant. Development of the solution for BWRs was estimated to be 0.5 man-year. Implementation time expended was estimated to be the same as for PWRs. Therefore, the estimated NRC costs were $0.43M.

It was also estimated that NRC labor for periodic review of operation and maintenance of the solution would be 1 man-week/RY for PWRs and 0.5 man-week/RY for BWRs. This translated into $2,000/RY and $1,000/RY, respectively, for all plants for a cost of $6.7M. Thus, the total NRC cost was $(0.43 + 6.7)M or $7.1M.

Total Cost: The total industry and NRC cost to resolve this issue was estimated to be $(61 + 7.1)M or $68.1M.

Value/Impact Assessment

Based on a potential risk reduction of 3 x 104 man-rem and an estimated implementation cost of $68.1M, the value/impact score was given by:

equation showing S= 3 X 10 to the fourth power man-rem over $68.1M equals 440 man-rem/$M

Uncertainty

The value/impact score was significantly influenced by the assumption that a 5% frequency reduction could be obtained; this number was highly judgmental.

Other Considerations

(1) Occupational dose would lower (significantly) this value/impact score because the labor required in a radiation zone would be significant. The estimated occupational dose from performing this periodic testing was about 24 man-rem/RY for PWRs and 18 man-rem/RY for BWRs. Over the life of a plant, the overall (total) occupational dose was estimated to be 8.9 x 104 man-rem.

(2) Occupational risk reduction due to accident avoidance was concluded to be small and accident avoidance costs, although large when considered in relation to the other costs, would not significantly change the score.

CONCLUSION

Based on the value/impact score and the additional considerations, this issue was given a medium priority ranking and was later divided into four parts during resolution: (1) pressure isolation valves; (2) check valves; (3) reevaluation of thermal-overload protection provisions of Regulatory Guide 1.1061215 for MOVs; and (4) in-situ testing of MOVs.

The investigation of alternatives to leak rate testing of pressure isolation valves, including check valves, was integrated into the resolution of Issue 105, "Interfacing Systems LOCA." These alternatives included non-intrusive methods to detect check valve disk position and motion, as well as surveillance of internal parts by various means. Any new issue regarding testing of check valves that may be identified in the future will be prioritized as a new generic issue. The results of the staff's study of MOV thermal overload protection were published in NUREG-1296.1216 The staff concluded that, although misinterpreted by the industry at times, the guidelines in Regulatory Guide 1.1061215 were adequate. Several suggestions for improving MOV thermal overload protection were outlined in NUREG-1296.1216 In addition, letters were sent to the pertinent IEEE and ASME subcommittees encouraging the development of standards for MOV thermal overload protection. In-situ testing and surveillance of check valves was being addressed by an industry effort; in-situ testing of MOVs was resolved with the issuance of Generic Letter 89-10.1217 Thus, this issue was RESOLVED and requirements were established.1218

Page Last Reviewed/Updated Thursday, March 29, 2012