Section 50.59 of the Nuclear Regulatory Commission’s regulations allows a nuclear power plant to make changes related to its updated final safety analysis report (UFSAR) without a license amendment, provided the changes do not trip any of the eight criteria in 10 CFR 50.59(c). The Nuclear Energy Institute (NEI) has issued guidance (NEI 96-07) for implementing 10 CFR 50.59, and the Nuclear Regulatory Commission (NRC) endorsed that guidance in Regulatory Guide 1.187.

As a result of a lessons-learned review in 2015 related to 10 CFR 50.59, the NRC has identified clarifications that it desires to make with respect to the language in NEI 96-07 regarding some of the eight criteria in 10 CFR 50.59(c). The clarifications are described in a draft revision to Regulatory Guide 1.187, which has been published as Draft Regulatory Guide DG-1334:

  1. The fifth criterion requires a license amendment for changes that create a possibility for an accident of a different type than any previously evaluated in the UFSAR. NEI 96-07 states: “Accidents of a different type are credible accidents that the proposed activity could create that are not bounded by UFSAR-evaluated accidents.” DG-1334 states that this language should be clarified to state: “Accidents of a different type are credible accidents that the proposed activity could create that have an effect on the plant that current UFSAR accident analysis do not evaluate (i.e., a different accident analysis would be needed for this different type of accident).”
  2. The eighth criterion requires a license amendment for a departure from a method of evaluation described in the UFSAR used in establishing the design bases or in the safety analyses. Section 50.59 defines such a departure as (i) changing any of the elements of the method described in the UFSAR unless the results of the analysis are conservative or essentially the same; or (ii) changing from a method described in the UFSAR to another method unless that method has been approved by NRC for the intended application. NEI 96-07 states that such a departure does not include “[u]se of a methodology revision that is documented as providing results that are essentially the same as, or more conservative than, either the previous revision of the same methodology or another methodology previously accepted by NRC through issuance of an SER [safety evaluation report].” DG-1334 states that this language should be clarified to clearly distinguish between the two parts of the definition of departure from a method of evaluation. Somewhat confusingly, DG-1334 goes on to state that a licensee “may” document a “methodology revision” as a change in method of evaluation.

In general, the clarifications in DG-1334 are consistent with the manner in which the industry has applied NEI 96-07. Additionally, DG-1334 is unobjectionable to the extent that it states that a license “may” document a “methodology revision” as a change in method of evaluation. However, if that statement is interpreted as requiring rather than allowing a licensee to evaluate use of a new revision of a method described in the UFSAR under definition 2 rather than definition 1, then DG-1334 represents a significant change in the manner in which 10 CFR 50.59 has been applied in the past.

Although DG-1334 itself is confusing, NRC’s 2015 lessons-learned review provided a discussion that is clear. Specifically, the review report states with respect to NEI 96-7:

  • It could be read to infer that ‘methodology revision’ results could be compared to ‘the previous revision of the same methodology’ instead of being compared to the revision currently specified in the FSAR which may be an earlier revision than the ‘previous revision’; and
  • It could be read to infer that ‘methodology revision’ results could be compared to ‘another methodology previously accepted by NRC through issuance of an SER’ instead of being compared to the methodology currently specified in the FSAR.

For some reason, DG-1334 did not directly use these clarifications from the lessons-learned report, and as a result DG-1334 itself could use some clarification.