Resolution of Generic Safety Issues: Item B-26: Structural Integrity of Containment Penetrations (Rev. 1) ( NUREG-0933, Main Report with Supplements 1–34 )
As described in NUREG-0471,3 this issue involves staff evaluations to assess the adequacy of specific containment penetration designs from the point of view of structural integrity, ISI requirements, and new surveillance or analysis methods applicable to containment penetrations which are identified as inaccessible. The issue is applied to all operating plants as well as those plants currently under construction and up for licensing review.
In accordance with a DE memorandum,215 that part of the issue involved with the structural integrity of specific containment penetration design, i.e., forged versus welded design, has been resolved. This resolution is based on a draft report by an NRC consultant. A NUREG is being considered to document this resolution. The second concern which involves the volumetric examination as required by ASME Code, Section XI14 is only partially resolved for: (1) plants under licensing review, where inspection and surveillance problems associated with inaccessible penetrations must be resolved in some manner before startup operations can occur, and (2) operating reactors, where inspection and surveillance problems are reviewed during reviews of licensees' ISI programs.
The staff review should determine whether or not the configuration and accessibility of the welds in the proposed design and the procedures proposed for performing volumetric examination permit inservice examination requirements of Section XI14 of the ASME Code at an augmented frequency in break exclusion regions, as required by SRP11 Section 3.6.2. In the event that penetration designs are found inadequate with respect to conducting current inservice inspections, alternative surveillance or analysis methods would be implemented to ensure that inspections can be completed.215
Upon satisfactory resolution of inspectability concerns, this issue should not affect public risk. However, should it be impractical for a plant to assure the above stated inservice examination requirement in accordance with SRP11 Section 3.6.2, no specific guidance is provided as to what measures provide an acceptable resolution. In these cases, staff approval on a case-by-case review basis may result in inconsistent penetration requirements from plant to plant. Such inconsistencies, should they occur, could result in increased risk to the public. To account for this possibility, the potential public risk reduction is obtained by assuming that the likelihood of radioactive releases from containment may be reduced.
The specific containment penetrations involved in this issue include only the high-energy fluid systems. High-energy fluid systems are defined as those that are in operation or pressurized during normal plant conditions (i.e., during reactor startup, power operation, and reactor cold shutdown excluding test modes) where either or both of the following criteria are satisfied: (1) maximum temperature exceeds 200F, and (2) maximum pressure exceeds 275 psig.
For those plants or penetrations that do not or cannot meet the above inservice examination requirements, the staff should develop guidelines and/or criteria to provide consistent requirements and acceptable conditions to assure the acceptability of the penetration designs and minimum levels of inspectability to meet these criteria.
The PNL analysis64 for this issue assumed that all penetration assembly designs meet code accessibility requirements or approved analysis/surveillance techniques. The result is adequate completion of ISI as well as elimination of unresolved conditions affecting plant startup.
An average of 40 high-energy penetrations per plant are assumed in the following analysis. This number will vary depending on plant type and design and is only an estimate based on information available in several BWR and PWR FSARs (including tables of high-energy lines, identification of systems requiring boundary guard pipes and complete listings of penetration data). It is further assumed that only 20% (8) of all high-energy penetrations per plant need attention as specified by the issue. Since requirements for ISI are known, industry, where possible, attempts to build in inspectability features.
There are analysis and augmented inspection procedures currently available to accommodate many of the inaccessible penetrations. It is estimated that 20% of the 8 penetrations under consideration may require the development of new analysis procedures. Therefore, the number of penetrations per plant requiring new procedures is (0.2)(8) or approximately 2 penetrations per plant. Of the originally assumed 40 penetrations, the 2 penetrations per plant requiring new procedures would be 5 times more likely to fail than the remaining 38. Upon resolution of the issue, all 40 penetrations would have an equal failure probability. This results in a 17% reduction in the containment leakage probability.
For those plants that meet the current requirements (SRP11 Section 3.6.2), this issue results in no change in the core-melt frequency. To determine the potential effect on core-melt frequency associated with inadequate containment penetration designs, the containment failure mode B (containment leakage) is assumed.
For PWRs, PNL64 selected the Oconee 3 reactor as their representative model. The base case core-melt frequencies for the PWR-4 and PWR-5 releases were 9.7 x 10-8 /RY and 4.6 x 10-7/RY, respectively. The reduced core-melt frequencies for PWR-4 and PWR-5 type releases were 7.9 x 10-8/RY and 3.8 x 10-7/RY, respectively.
For BWRs, PNL used the Grand Gulf reactor with a BWR-4 release as the representative model. The base case core-melt frequency was determined to be 2.4 x 10-7/RY. The potential reduced BWR core-melt frequency was 2 x 10-7/RY.
Assuming PWR-4 and PWR-5 releases result in 2.7 x 106 man-rem and 1 x 106 man-rem, respectively, the potential risk reduction is 1.3 x 10-1 man-rem/RY for PWRs. The average remaining life for the 47 operating and 43 planned PWRs is 28.8 years. As a result, the potential public risk reduction considering 2,592 RY is 337 man-rem for all PWRs.
For BWRs with a potential BWR-4 release that results in 6.1 x 105 man-rem, the risk reduction is 3 x 10-2 man-rem/RY. The average remaining life for 24 operating and 20 planned BWRs is 27.4 years. The potential risk reduction considering 1,205 RY is 36 man-rem for all BWRs. Therefore, the total public risk reduction is 373 man-rem.
NRC Cost: Development of guidelines/criteria are assumed to take one man-year ($100,000). If this cost is divided among all operating and planned reactors (134 plants), the per plant cost is $750. Approximately 5 man-wks ($9,620) are currently required to complete a plant-specific review. The developed guidelines/criteria can be expected to reduce the plant-specific reviews to 3 man-wks/plant ($5,770/plant). Therefore, a net NRC cost benefit of approximately $3,100/plant is obtainable by development of guidelines/criteria for this issue.
Industry Cost: It is estimated that 8 man-wks/plant is currently needed to develop supporting analyses and procedures on a plant-specific basis. Appropriate guidelines/criteria can be expected to reduce this effort by 3 man-wks/plant ($5,700/plant). Assuming that the guidelines/criteria may require new inspections or analysis every ten-year inspection period, an additional 4 man-wks/plant/10 years over an average remaining plant life of 30 years would result in an additional cost of $7,700/plant.
Considering the potential cost savings of $5,700 afforded by the guidelines and the potential cost (impact) of additional requirements ($7,700), the net cost (impact) is $2,000/plant. However, if the initial plant-specific reviews without the guidelines were to result in similar inspection requirements (which is likely), the above impact cost of $7,700 would be moot and the result would be a plant cost benefit of $5,700.
Based on the above cost estimates, the combined NRC and Industry Costs result in a net cost benefit (value) ranging from $1,000/plant to $9,000/plant.
The values associated with the issue are: (1) a small potential public risk reduction of 373 man-rem, (2) a net NRC and industry cost benefit of $1,000 to $9,000 per affected plant, and (3) a potential reduction in ORE of 1,200 man-rem for the 63 plants not yet operational (see Other Considerations below). No impacts should result from development of guidelines/criteria for alternate surveillance or analysis methods for inaccessible penetrations.
PNL estimated that at most one failure per year occurs in all (71) operating plants. The time to repair the failure involves about 20 man-wks/failure in a 0.25 R/hr field required (2.8 man-rem/RY).
No reduction in exposure will be credited to plants that are already designed and operating, since repairs would be predicted in existing designs, requirements, and failure rates. Development of the guidelines/criteria, as described in the above assumptions, were estimated to result in a 17% reduction in containment leakages. The potential reduction in ORE for the 63 planned reactors over a 40-year design life is [(0.17)(63)(40)(2.8)] man-rem or 1,200 man-rem.
Based on the above value/impact assessment, this issue was identified as medium priority. However, after a reevaluation of the issue by DE, it was concluded that further expenditure of resources was unwarranted. DE believed that the increase in ORE from additional inspections would negate the small potential risk reduction associated with the issue.647 Thus, the issue was RESOLVED and no new requirements were established.