2

We conclude, based on the attached detailed legal anlaysis, that only one minor aspect of one of the five major interpretations cannot be squared with the rules. Specifically, we believe that the proposed interpretation on fire area boundaries should be modified to make clear that any fire barrier cable penetration identified as an open SER item at the time Appendix R became effective must be sealed to provide fire resistance equal to that required of the barrier the cable penetrates, and that any variation from this requirement requires an exemption. The staff has agreed to modify the proposed interpretation and has provided text for the modification, which is included in the attached analysis.

In one other area, where we support one of the staff's proposed interpretations, we do so with arguments which nonetheless lead us to recommend that the interpretation be clarified because it obscures either the intent behind the rule interpreted or some other rule. Specifically we recommend that the proposed interpretation of the phrase "fire damage" should be modified to make clear that the functional definition of "fire damage" which the staff proposes does not relieve the licensee of the duty to protect hot shutdown equipment against damage by fire suppression activity. The staff has expressed its willingness to modify the interpretation.

One final aspect of the interpretation on fire boundaries, which makes it possible for a licensee to erect a fire barrier with unsealed openings without having to apply for an exemption, causes us concern. Nonetheless, we conclude that our legal concerns could be overcome by a strong technical argument that unsealed openings do not generally pose fire hazards and need not be treated as stringently as fire barrier cable penetration seals and fire doors are treated by the rules.

[Footnote Continued]

responded to the May 1984 Memorandum.

Apparently, despite the fact that the Memorandum was advice meant solely for the Commissioners' technical assistants, the NUFPG obtained access to the Memorandum. On September 6, 1984, the NUFPG sent Chairman Palladino a detailed reply to the issues raised by the Memorandum. On October 9, 1984, the author of the Memorandum, Mr. Trubatch, replied to the NUFPG. We have found nothing in the exchange between the NUFPG and Mr. Trubatch which would cause us to alter our analysis.

3.

In general, we pay somewhat more attention to the history of the rules in question, and their relation to other rules in Appendix R, than do either the Memorandum or the staff's replies to it, and in some instances we are led by that history to conclude that the staff may sometimes have implemented Appendix R with a stringency not supported by the intentions with which the Appendix was first promulgated.

We have had the benefit of comments from the staff and OPE on the attached analysis, and in the light of those comments, we have corrected some things in the analysis and clarified others.

We understand that the Nuclear Utility Fire Protection Group was to have addressed the Commission on this matter this week. We are prepared to evaluate any additional questions of interpretation of Appendix R which they may eventually present, as well as their concern regarding the possible official Commission action on the staff's interpretations, should the Commission so request.

We agree in some respects with the May 1984 Memorandum's
conclusion that the staff's proposed interpretation on fire
area boundaries is contrary to the intent with which the
Commission promulgated the pertinent rules. However,
because we reach this conclusion by a route not taken by the
May 1984 Memorandum, we are able to suggest a narrowing of
the scope of the proposed interpretation such that once
narrowed, the interpretation would both partly satisfy the
staff's aim in proposing the interpretation and more nearly
conform to the intention behind the pertinent rules than
either the interpretation or current practice do. The staff
has informed us that it is willing to make the narrowing we
suggest. Included below is wording the staff proposes to
add to the interpretation in order to narrow its scope.
Section III.G.2 of Appendix R requires that in some circum-
stances fire barriers of various ratings -- 3 hours in one
set of circumstances and 1 hour in another be used to
ensure that where the redundant trains of systems required
for hot shutdown are located in the same fire area outside
of containment, one of the redundant trains would be free of
fire damage in the event of a fire. The staff's proposed
interpretation of this section says that the section can be
met even where the barrier is not completely sealed,
floor-to-ceiling, wall-to-wall, and all penetrations in it
sealed in such a way as to provide the same fire resistance
the barrier provides. In the past, any licensee wanting to
install a barrier which was not completely sealed, or which
did not have the same fire resistance, or "rating," at every
point of its surface -- that is, was not "uniform" -- had to
apply for an exemption from the requirements of sec-
tion III.G.2. The proposed interpretation says, however,
that in such cases, the licensee need not any longer apply
for an exemption (id., Enclosure 8 at 2); instead, the
licensee must perform an evaluation of the adequacy of the
boundary and retain the evaluation for later NRC audit.
Id., Enclosure 4, at 2.

The May 1984 Memorandum argues that, since Appendix R was
promulgated to prevent another fire like the one at Brown's
Ferry, a fire which clearly illustrated the consequences of
having non-uniform barriers, the staff's proposed interpre-
tation, by treating non-uniform or incompletely sealed
barriers as possibly being in compliance with Appendix R,
"directly contradict[s] the Commission's intent in pro-
mulgating Appendix R." May 1984 Memorandum at 2.
Memorandum also claims that, because the proposed

The

2

interpretation does not provide for NRC review of

non-uniform and incompletely sealed barriers before they are installed, the interpretation permits delay and

inconsistency in the implementation of Appendix R. Id.

at 2-3.

In reply, the staff (OELD) and the proposed interpretation itself put great store by the fact that Appendix R nowhere defines the word "area" in the phrase "fire area." See SECY-85-306, Enclosure 8, at 1-2. The staff's argument seems to be that the fire barriers spoken of in Appendix R by definition separate fire "areas," and that were it the intent of Appendix R to require that fire barriers be floor-to-ceiling, wall-to-wall, and everywhere fire-resistant to the same degree, the word "area" would be defined in such a way as to make those requirements clear.

Nothing in Appendix R suggests that every fire barrier is a boundary of some fire area, and every area delimited by fire barriers. On the other hand, nothing in Appendix R standing alone says that fire barriers must be uniform and completely sealed.

However, the history of Appendix R provides a better measure of the soundness of the proposed interpretation than does the lack of definition of some of the terms in the Appendix. The history of Appendix R clearly shows that, with one important exception, fire barriers need not be uniform.

1Indeed, there seems to be a paradoxical logic in linking "barrier" and "area" too closely: if every barrier is a boundary of some fire area, then if two redundant trains of hot shutdown systems located in the same fire area and insufficiently separated by horizontal distance are separated by a fire barrier, the redundant trains are no longer in the same fire area since the barrier separating them puts them in different fire areas. But then they are not required to be separated by a barrier, for the requirement that they be separated by a barrier applies only if they are in the same area. Under this reading, section III.G.2.a and c would be without any practical force.

3

Section M of the proposed rule contained this paragraph: Penetrations in these fire barriers, including conduits, cable trays, and piping, shall be sealed or closed to provide fire resistance rating equivalent to that required of the barrier. Door openings shall be protected with doors, frames, and hardware that have been tested and approved by a nationally recognized testing laboratory to have a fire resistance rating equivalent to that required of the barrier.

45 Fed. Reg. 36089, May 29, 1980 (emphases added).

Section M of the proposed rule required barriers with 3-hour ratings. In addition, Section N of the proposed rule imposed on fire barrier cable penetration seals a rather elaborate set of rules which amounted in large part to a requirement that the seals have the same fire rating as the rating required of the barriers through which the cables penetrated. Id., at 36089-90. The standards enunciated in these sections are the ones the May 1984 Memorandum argues section III.G.2 now embodies.

However, the final rule contains neither the paragraph quoted above nor any other part of section M of the proposed rule. The comment accompanying the final rule said:

Other comments to the effect that the requirement
[section M] was excessively restrictive with regard to
fire barrier penetrations, including fire doors and
their associated frames and hardware, and ventilation
systems have been acted upon by the staff and the
requirement, as it had affected these items, was
deleted.

45 Fed. Reg. 76608, November 19, 1980.

Nonetheless, the final rule retains in simplified form the requirement in section N of the proposed rule that fire barrier cable penetration seals have the same fire ratings as the ratings required of the the fire barriers in which they are installed:

The acceptance criteria for the test [used to rate fire barriers] shall include: 1. The cable fire barrier penetration seal has withstood the fire endurance test without passage of flame or ignition of cables on the unexposed side for a period of time equivalent to the fire resistance rating required of the barrier. . .

Appendix R, section III.M. The comments accompanying the retained section unequivocally restate the requirement for