ML20212D678

Supplemental Technical Evaluation Report Based on Review of Additional Responses Concerning Individual Plant Examination of External Events at Point Beach Nuclear Plant,Units 1 & 2. Final Rept

ML20212D678
Person / Time
Site:	Point Beach
Issue date:	03/31/1999
From:	Kazarians M, Sewell R AFFILIATION NOT ASSIGNED, EQE INTERNATIONAL
To:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
Shared Package
ML20212D580	List: ML20212D577 ML20212D596 ML20212D641 ML20212D678
References
CON-NRC-04-94-050 ERI-NRC-98-505, NUDOCS 9909240047
Download: ML20212D678 (22)

Text

.' .' t ERl/NRC 98-505 e

SUPPLEMENTAL TECHNICAL EVALURTION REPORT ,

BASED ON REVIElU OF RDDITIONRL RESPONSES CONCERNING THE INDIVIDUAL PLANT EXAMINATION OF 4 l

.EXTERNRL EVENTS RT POINT BEACH NUCLEAR PLANT, j UNITS 1 AND 2  !

I FINAL REPORT l

March 1999  ;

i 1

Energy ResearCh, inC.

P.O. Box 2034 RoCkville, Maryland 20847-2034 Work Performed Under the Auspices of the United States Nuclear Regulatory Commission Office of Nuclear Regulatory Research Washington. D.C. 20555 Contraet No. 04-94-050 Attachment 2 9 29240047 990915 P1 ADOCK 05000266 P PDR g

c l

l l

ERI/NRC 98-505 ,

l SUPPLEMENTAL TECHNICAL EVALUATION REPORT BASED ON REVIEW OF ADDITIONAL RESPONSES CONCERNING THE INDIVIDUAL PLAhT EXANUNATION OF EXTERNAL EVENTS AT POINT BEACH NUCLEAR PLAST, UNITS 1 AND 2 i

1 FINAL REPORT l

March 1999 l

l M. Khatib-Rahbar j Principal Investigator i Authors:

R. T. Sewell' and M. Kazarians'-

Energy Research, Inc.

P.O.' Box 2034 l Rockville, Maryland 20847 1

J Work Performed Under the Auspices of the United States Nuclear Regulatory Commission l Office of Nuclear Regulatory Research WashinFton, D.C. 20555 l

Contreet No. 04-94 050 l

1 .

l

' EQE International, Inc.,2942 Evergreen Parkway. Suite 302 Evergreen, CO 80439

1. Kazarians and Associates,425 East Colorado street, Suite 545, Glendale, CA 91205

. * , * . - = - . , _ . .. . .,e., .. .

.r . .

TABLE OF CONTENTS EXECUTIVE SUMM ARY (SUPPLEMENTAL) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii i i INTRODu CTiON . .' . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i 1.1 Objectives of Supplemental Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I 1.2 Overview of Supplemental Review Activities . . . . . . . . . . . . . . . . .........I 1.2.1 S eis mic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

1. 2.2 Fir e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 l

3 2 SUPPLEMENTAL REVIEW ISSUES AND CONTRACTOR REVIEW FINDINGS . . . .

2.1 S eismi c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 2.1.1 Summary of Requests for Additional Information . . . . . . . . . . . . . . . . . 3 2.1.2 Summary of Responses to Requests for Additional Information . . . . . . . . 5 2.1.3 Review Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2.2 Fire...................................................7 7

2.2.1 Summary of Requests for Additional Information . . . . . . . . . . . . . . . . .

2.2.2 - Summary of Responses to Requests for Additional Information . . . . . . . . 8 l

2.2.3 Review Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....9 OVERALL EVALUATION AND CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . 11 1

3 1 3.1 Seismi c . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 i 3.2 Fire..................................................11 l SUPPLEMENTAL IPEEE INSIGHTS, IMPROVEMENTS, AND COMMITMENTS . . 13 l 4

4.1 Seismic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 4.2 Fire..................................................13 l

5. REFERENCES . . . . . . . . . '. . . . . . . . . . . . . . . . . . . . . .................14 1

l l

l:

ii ERI/NRC 98-505 Energy Research, Inc.

EXECUTIVE

SUMMARY

(SUPPLEMENTAL)

This supplemental technical evaluation report (STER) documents the approach and findings of additional review consideration of the individual plant examination of external events (IPEEE) conducted for the Point Beach Nuclear Plant (PBNP). This supplemental technical evaluation review was performed by Energy Research, Inc. (ERI) on behalf of the U.S. Nuclear Regulatory Commission (NRC).

The supplemental review was initiated because the Step-l review of the IPEEE submittal and of the licensee's first-round responses to requests for additional information (RAls) concluded that additional information was needed for the NRC to make a determination as to whether or not the intent of Generic Letter (GL) 88-20, Supplement 4, had indeed been met for the plant. The scope of the supplemental review was limited to consideration of seismic and fire events.

The activities undertaken for supplemental review cf the Point Beach IPEEE consisted of the following tasks: ,

• Development of second-round RAls to supplement or clarify the licensee's IPEEE submittal or earlier responses, as necessary;

• Examination of additional responses, together with accompanying documentation, provided by the licensee;

• Evaluation of the licensee's responses, and determination of remaining issues and commitments, if any; and

• Documentation.

This STER' updates ERI's earlier final technical evaluation report (TER) (ERl/NRC %505, entitled

" Technical Evaluation Report on the ' Submittal-Only' Review of the Individual Plant Examination of External Events at Point Beach Nuclear Plant, Units 1 and 2"), and serves as a qualitative supplemental assessment of the Point Beach IPEEE submittal, particularly with respect to the objectives described in Generic Letter 88-20, Supplement 4, and the guidance presented in NUREG-1407 pertaining to seismic and fire events.

Seismic The supplemental seismic review focused on considerations of issues or concerns remaining from the

" submittal-only" (Step-1) review, which were presented to the licensee in the form of a second round of l RAls. The licensee responded to all second-round seismic RAls by means of two written communications, the first dated July 24,1998, and the second dated December 17, 1998.

As a rt.sult of the supplemental seismic review of the licensee's responses, the principal remaining review observations include the following:

Energy Research, Inc. iii ERl/NRC 98-505 l

1. As presented by the licensee, the reported plant high confidence of low probability of failure (HCLPF) spectrum does not exceed even the plant's design spectrum for frequencies less than about 4 Hz.

2. There are several reported anomalies and outliers that have not been fully resolved (examples include anchorage concerns, interaction concerns, potential failures of block walls, and potential failures of flat-bottomed tanks).

l

3. The licensee has not fully satisfied the guidance of NUREG 1407 with respect to the IPEEE relay evaluation; however, the information provided by the licensee suggests that it may be unlikely that any additional insights or low-ruggedness relays (beyond those revealed from the unresolved safety l issue [USI] A-46 evaluation) would be encountered if a relay evaluation ofIPEEE-only circuitry I were to be performed.

4. The licensee has not demonstrated a full understanding of the implications of use of the surrogate i element in the Point Beach seismic IPEEE.

Some additional points pertaining to assumptions made by the licensee are noted in this STER.

Fire The supplemental fire review focused on the following four fire-related issues remaining from the Step-l review of the IPEEE submittal: (1) potential for loss of offsite power from fires in various compartments; (2) proper modeling of oil-filled transformers in the cable spreading room and other safety related compartments; (3) use of comparatively small times for automatic detection and suppression; and (4) proper treatment of hot shons in the analysis of cable failures.

Based on a review of licensee *s responses, the review team made the following observations and conclusions regarding the fire IPEEE issues:

• The licensee has stated that the possibility of loss of offsite power from a fire occurring at various points at the power plant had been considered. Power cables that supply offsite power to the Class IE buses have been traced. For control room and cable spreading room fires it was assumed that loss of offsite power would occur. However, it is not clear whether other compartments have been analyzed properly.

• The oil used in the transformers inside safety related areas and the cable spreading room is polychlorinated biphenyl (PCB), which is a non-flammable fluid. Therefore, it can be concluded that the related possibility of a fire is not significant.

• Since the time to cable damage is significantly greater than the time for automatic detection and suppression, the licensee has argued that the final conclusions would not change if the time for automatic fire detection and suppression were altered.

• The licensee has stated that the issue of hot shorts and the possibility of a loss-of-coolant accident (LOCA) and inadvertent steam dump have been addressed.

iv ERl/NRC 98-505 Energy Research, Inc.

From the responses provided by the licensee, it can be inferred that the above fire related issues were properly addressed, and aside from a number of relatively minor concerns that do not have a significant impact on the final conclusions of the analysis, there are no remaining issues regarding the fire IPEEE of Point Beach Nuclear Power Plant.

I

\

)

I I

I i

i

)

Energy Research, Inc.

• ERI/NRC 98-505 1

)

)

1 l

1 INTRODUCTION This supplemental technical evaluation report (STER) documents the approach and findings of additional review consideration of the individual plant examination of external events (IPEEE) for the Point Beach Nuclear Plant (PBNP) [1] The supplemental review, conducted by Energy Research, Inc. (ERI), was initiated because the Step-1 review [2] of the IPEEE submittal and of the licensee's first-round responses to requests for additional information (RAls) concluded that additional infonnation was needed for the U.S.

Nuclear Regulatory Commission (NRC) to make a determination as to whether or not the intent of Generic Letter (GL) 88-20 Supplement 4 [3], had indeed been met for the plant.

This STER updates ERI's earlier final technical evaluation report (TER) (ERl/NRC 96-505, entitled

' '" Technical Evaluation Report on the ' Submittal-Only' Review of the Individual Plant Examination of Extemal Events at Point Beach Nuclear Plant, Units 1 and 2"), and has specifically considered seismic and fire events.

The remainder of this section of the STER provides a description of the objectives of the supplemental review, as well as an overview of the supplemental review activities. Sections 2.1 and 2.2 of this report present ERI's findings related to the seismic and fire supplemental reviews, respectively. Section 3 summarizes ERI's conclusions resulting from the supplemental seismic and fire reviews. Section 4 summarizes any additional IPEEE insights, improvements, and licensee commitments, and Section 5 provides a list of supplemental references.

1.1 Oblae*Ives of 4 77 5-m*a*=1 Review The objective of this supplemental review is to evaluate additional information needed to determine the -

extent to which the IPEEE process used by the licensee, Wisconsin Electric Power Company (WE), meets the intent of Generic Letter 88-20,. Supplement 4. Insights gained from this supplemental review are

' intended to provide a reliable perspective that assists in making such a determination. This review has involved a qualitative evaluation of information supplied by the licensee in the form of written responses to second-round RAls, as well as transmittal of supponing documentation. The emphasis of this review has been to resolve, to the extent possible, specific concerns or weaknesses encountered in the Step-l review, particularly with respect to the guidelines established in NUREG-1407 [4].

'I.2 Ovd= of hanlemental Review Activiiles The general activities undertaken for supplemental review of the Point Beach IPEEE consisted of the following tasks:

• ~ Development of second-round RAls [5] to supplement or clarify the licensee *s IPEEE submittal or earlier responses, as necessary:

• Examination of additional responses, together with accompanying documentation, provided by the licensee;

• ' Evaluation of the licensee's responses, and determination of remaining issues and commitments, if any; and Energy Research, Inc. 1 ERl/NRC 98-505

l l

• Documentation.

1.2.I' Seismic The licensee's responses to second-round seismic-related RAls [6,7] consisted of concise written responses.

This written information was examined in detail, and an overall qualitative review of methodological approaches was undertaken. No check of quantitative results was performed in this supplemental review.

No other data or results were available or considered in this review; however, a telephone conversation was held with the licensee on November 16,1998 pertaining to the scope of relay chatter evaluation I (following this conversation, the licensee developed the written response provided in Reference [7]).

1.2.2 Fire )

)

The licensee's responses to the second-round fire-related RAls [6] were examined in detail, and this supplemental TER provides discussion concerning the adequacy of those responses and the overall conclusions regarding the fire IPEEE submittal for Point Beach Nuclear Power Plant.

I l

l Energy Research, Inc. 2 ERl/NRC 98-505

2' SUPPLEMENTAL REVIEW ISSUES AND CONTRACTOR REVIEW FINDINGS The supplemental review focused on resolving issues raised in requests for additional information, by means of a documentation review. The issues, licensee responses, and review conclusions are summarized in the following subsections.

2.1 Estamic

.2.1.1 Summary of Requests for Additional Information The second-round RAls are summarized as follows:

1. Based on the licensee's response to the first-round seismic RAI No. 7 for the Point Beach plant

. (see Reference [8]), the licensee's reported high confidence of low probability of failure (HCLPF) capacity for the plant does not exceed even the safe shutdown earthquake (SSE) spectrum for vibration frequencies less than about 4 Hz. The reported HCLPF spectrum is also lower, over all frequencies, than the review-level earthquake (RLE) spectrum defined by a NUREG/CR-0098 median spectral shape anchored to's peak ground acceleration (PGA) value of 0.3g. Although safety-enhancements were planned that resulted from the USl A 46 and individual plant exammation (IPE) programs, no plant improvements / resolutions have been proposed for several remaming components that have low seismic capacities, and which cause the plant HCLPF to be significantly lower than the RLE (examples include anchorage concerns, interaction concerns, potential failures of block walls, and potential failures of flat-bottomed tanks).

The licensee was asked to provide a discussion as to why, in consideration of the foregoing, the plant is considered to have adequate resistance to potential seismically induced severe accidents.

2. The Point Beach seismic IPEEE submittal reports that release category G (i.e., pontainment isolation failure) was found to be associated with nearly the entire seismic core damage frequency (CDF). Hence, the conditional probability oflarge early release is assessed as being nearly 100%,

given seismically induced core damage. This finding results from the assumption that most sequences result in failure of the automatic containment isolation function. In the containment vulnerability assessment, the study effectively invalidates this finding by assuming that the contamment will be isolated manually by operators at least 90% of the time in the event of a core damage accident where automatic isolation is failed; the basis for this assumption is not provided, yet it is used to reduce the large release risk by an order of magnitude. Despite the arguments in the IPEEE, it seems that failure of automatic containment isolation may be a potential seismic vulnerability.

The licensee was asked to provide the fragility assessment for seismic failure of the automatic containment isolation system. The licensee was asked the following questions: On what basis is the 0.10 buman error probability (HEP) for manual containment isolation justitled? Is the basis consistent with that used in devekving operator error fragilities for the Level-1 analysis? At what locations do the required operations tbr manual isolation take place? How many operators are required to manually isolate the containment? .What are the timings for when the actions are i

I Energy Research, Inc. 3 ERl/NRC 98-505 l

)

i

- e l

required? Will all isolation valves be accessible / operable following an canhquake, or will potential seismic failures possibly lead to some components being inaccessible and/or inoperable?

3. The relay chatter evaluation has not been fully expanded beyond tre scope of USI A-46, in order to address all IPEEE-only systems. The submittal states that relay chatter is acceptable for the unanalyzed systems, on the grounds that operators could reset the relays.

l The licensee was asked to repon what fraction of the seismic CDF is contributed by relay chatter.

l The licensee was also requested to repon the seismic CDF for the case where no operator recovery from relay chatter is assumed. ,

4 The surrogate element has been reponed as a dominant contributor to seismic CDF, which essentially means that an imponant contribution to seismic CDF is not fully understood. This situation raises concerns as to the ability of the seismic submittal to achieve IPEEE objectives.

The licensee was asked to discuss how ins.ights regarding dominant CDF sequences and dominant

- risk comributors would change if the surrogate element was not used, or if the screening level was much higher (e.g.,0.5g HCLPF). The licensee was also requested to discuss the bases, including additional calculations, if any, needed to defend the licensee's conclu.sions.

5. The licensee's response in Reference [8] to the first-round seismic RAI No. 6 for the Point Beach plant was deemed incomplete with respect to seismic inadvenent actuation of the fire protection system and seismically induced loss of fire suppression system capability. The evaluation for inadvenent actuation did not meaningfully describe walkdown findings. The evaluation for seismic loss of fire suppression was limited to potential interactions with safety equipment, and l

did not consider loss of fire suppression capability itself; again, no walkdown findings were l

mentioned. Some examples of items found in past studies include (but are not limited to):

• Unanchored CO2 tanks or bottles l

l

• Sprinkler standoffs penetrating suspended ceilings j

• Fire pumps unanchored or on vibration isolation mounts

• Mercury or " bad actor" relays in fire protection system (FPS) actuation circuitry

. Weak or unanchored 480V or 600V (non-safety related) electrical cabinets in close proximity to essential safety equipment (i.e., as potential fire sources)

• Use of cast iron fire mains to provide fire water to fire pumps NUREG-1407 suggests a walkdown as a means of identifying any such items.

The licensee was asked to provide related results of the licensee's seismic-fire interaction study, i and to provide guidelines given to walkdown personnel for evaluating these issues (if they exist).

Energy Research, Inc. 4 ERl/NRC 98-505

The licensee was also requested to identify equipment in fire suppression systems that may be damaged due to the review level earthquake, and to discuss resolution of these items, if any.

2.1.2'. Summary of Responses to Requests for Additional Information The licensee's responses to the preceding second-round RAls are summarized as follows:

I.. The licensee's principal position is that it has elected to perform a seismic probabilistic safety assessment (PSA), in accordance with NUREG-1407, and it sees no reason or basis to use the PSA results in a seismic margins context. The licensee presents some arguments explaining why it feels it is invalid to compare HCLPF spectra for various cases.

The licensee concludes that is has performed an acceptable analysis, has correctly identified the seismically weak cc,mponents, and has implemented or scheduled many of the needed upgrades, such that the licensee believes that Point Beach will have adequate resistance to potential seismically induced severe accidents.

2. The licensee provided significant clarification regarding the noted issue. The licensee indicated that the action for manual containment isolation requires only that an operator push a single button on a control board located in the control room. The HEP for this action was assessed similarly as for any other action, given a seismic event.

The licensee provided a detailed and meaningful discussion regarding direct containment failures and indirect containment failures contributing to release category G. This discussion constitutes a more realistic auessment as to the seismic capability of the containment isolation function, and 1 gives more detailed information regarding possible anticipated failure modes. Overall, the i discussion is reasonably comprehensive and valid. However, a brief aspect of the discussion j worth pointing out pertains to an assumption regarding the possibility of rupture of component cooling water (CCW) piping that returns from the reactor coolant pumps. The licensee has stated that a large-break loss of coolant accident (LOCA) in the reactor coolant pump cubicles would mpture the CCW piping that runs through the cubicle, and that fission products would enter the CCW system. In this case, the licensee indicates that having a " Closed System Outside of Containment" is needed to prevent radiological releases. The licensee has simply stated (without specific explanation or justification) that large and medium break LOCAs are precluded in the seismic analysis, ao that there is not a credible path for fission products to enter the CCW system, and thus, the system effectively functions as a " Closed System inside of Containment."

The licensee has indicated that several improvements that help avert indirect failures of the containment isolation function have been made or proposed as a result of the USl A-46 program.

4 The licensee has assessed the seismic-related frequency of release category G as being 2.13 x 10

per reactor year, which it concludes does not constitute a vulnerability.

I

3. ~ In Reference 16), the licensee clarified that no

• bad actor" relays were found in the USl A-46 systems at Point Beach, and the licensee did not expect to find any had actor relays in the IPEEE-only. systems. The licensee stated that the contribution of relay chatter to seismic CDF is zero, Energy Research, Inc. 5 ERI/NRC 98-505

g- ' ,.

and'that consideration of operator recovery from relay chatter is not required since relays at the plant are not susceptible to chatter.

4. The licensee indicated that it did not believe insights regarding dominant sequences and dominant contributors would change if the surrogate element was associated with a higher screening level.

- The licensee provided some qualitative arguments as support for its position.

5. The licensee provided qualitative discussions regarding an evaluation of seismically induced inadvertent actuation of fire suppression systems, and regarding consideration of seismically

~ induced loss of fire suppression system capability. The licensee indicated that some plant improvements had been made earlier with respect to inadvertent actuation of fire suppression

- systems, including the following: spray shields installed on safety injection (SI) pumps, various motor control centers (MCCs) and equipment control panels; cable penetrations at the tops of control panels sealed to prevent water entry; and one equipment control panel relocated from a high-spray area to a more protected area. With respect to seismic loss of fire suppression system capability, the licensee explicitly confirmed that the IPEEE study made no attempt to examine the seismic capability of the fire suppression system itself.

2.1.3 Review Findings The supplemental seismic review has produced the following overall findings with regard to the licensee's response to each second-round RA1:

1. It should be noted that the concern raised in RAI No.1 is common to many seismic-PRA IPEEEs for central / eastern U.S. plants, where adequate seismic margin has not been effectively demonstrated. For Point Beach, the licensee has not presented sound technical justification to alleviate the concern, and the licensee's response leaves unanswered the principal issue which pertairis to seismic margin over vibration frequencies less than 4 Hz. The licensee did not try to demonstrate that there are no components contributing significantly to the seismic CDF having a fundamental vibration frequency less than 4 Hz.

The licensee cites plant improvements related to USI A-46 as a basis for concluding that Point Beach will have adequate resistance to potential seismically induced severe accidents. Since USI A 46 pertains to design-basis accidents, rather than beyond-design-basis accidents, the licensee's conclusion remains unverified in other words, it cannot simply be assumed that the USI A-46 improvements will also meet IPEEE objectives; rather, specific demonstration is needed to ensure that IPEEE objectives are met.

2. The licensee's response to RAI No. 2 is considered to be generally satisfactory. The licensee's

- claim, however, that large and medium break LOCAs are not possible, and hence, that there is not a credible path for fission-product releases through the CCW system. has not been meaningfully substantiated.

' 3. Based on the licensee's response provided in Reference l6], it could not be determined whether or not the licensee's response to RAI No. 3 resolves the noted concern. Taken by itself, the response appeared to resolve the issue. That is, the licensee stated that no had actor relays were Energy Research, Inc.- 6 ERI/NRC 98-505

l identified in the USI A 46 evaluation, and hence, no bad actor relays were expected in IPEEE-only 1-systems. However, previously, the licensee stated [1] that low ruggedness relays were found in 4.16 kV switchgear, as pan of the USI A-46 evaluation. Thus, there was an apparent inconsistency in what the licensee has reponed, and funher clarification was deemed necessary to

1. resolve this inconsistency before the licensee's response could be accepted. Consequently, a telephone conversation was held 'with the licensee on November 16, 1998, and the licensee followed this discussion by a written response [7]. This subsequent information revealed that the licensee had not strictly met the NUREG-1407 recommendation that the relay evaluation be expanded to IPEEE-only systems, since instances of a low-ruggedness relay (Westinghouse ITH) were in fact encountered in the USI A 46 relay evaluation. However, the licensee undertook analyses demonstrating that the USI A-46 outlier relays had adequate capacity. The licensee implied that a similar finding would apply for the IPEEE-only circuitry, and that it would not be expected that additional " bad actor" relays would be found in the IPEEE unique systems. The licensee also re-emphasized that operators would have time to reset any repositioned relays in the IPEEE unique circuitry, and that none of the associated IPEEE-only systems would be required to function during the strong motion of an earthquake.

Hence, althcugh the licensee has not fully satisfied the guidance of NUREG-1407 with respect to the IPEEE relay evaluation, the information provided by the licensee suggests that it may be unlikely that any additional insights or low-ruggedness relays would be encountered if a relay i evaluation of IPEEE-only circuitry were to be performed.

4. In response to RAI No. 4, the licensee has not presented any analyses or other verifiable support for its qualitative perspectives as to why it believes use of a surrogate element is valid. Hence, the licensee has not demonstrated a full understanding of the implications of use of the surrogate element in the Point Beach seismic IPEEE.

5. With respect to seismically induced inadvertent actuation of fire suppression systems, the I licensee's response to RAI No. 5 is considered to be satisfactory. With respect to seismic loss of fire suppression system capability, the licensee's response simply relied on the assumption that there is no credible way to genote a seismically induced fire at Point Beach.

1 2.2 Bm 2.2.1 Summary of Requests for Additional Information ,

A set of second-round RAls [5] were submined to the licensee that addressed the following four fire-related .

1 issues:

1. The possibility of loss of offsite power from a fire within the plant had not been addressed properly. Reference [8] stated that "No fire compartmente, were identified that would disable all offsite power." It was not clear that tl.e analysts had considered the control room and cable spreading rooms. It is common that the controls of the breakers associated with offsite and Class IE power are within the same panel next to one another. A small control panel fire may cause control circuit failures that would lead to these breakers failing open (thus, leading to loss of offsite power). Similar failures may occur from a fire inside the cable spreading room, and other Energy Research, Inc. 7 ERI/NRC 98-505 I

\? , ;

i rooms where the cables or equipment associated with the control circuits of these breakers are present. This omission in the analysis could be significant, and therefore, a potential vulnerability may have been overlooked.

2. The cable spreading room (Companment 318) and electrical equipment rooms (Compartments 245 and 246) contain oil filled transformers. If the oil is combustible, there is a potential for a large

. fire from severe transformer failure. References [1,8] did not address the potential for a large fire or explosion in a transformer. The licensee had subdivided the cable spreading room into smaller areas for the detailed fire analysis. This can only apply to small fires. A transformer fire or explosion may have sufficient energy to jeopardize the integrity of the fire boundaries. Potential vulnerabilities may have been overlooked from the omission of fire scenarios involving an energetic rupture of a transformer or an extremely large fire in this room.

3. For several fire scenarios, the licensee had used unreasonably small times for detection and suppression. Specifically, for Companment 156 (MCC room), Scenario 2, time for automatic detection and suppression was 65 seconds. Such a shon detection and suppression time is deemed to be overly optimistic. Similar conditions exist for Companment 166 (MCC room), Scenarios 2 and 3, and Compartment 318 (cable spreading room), Scenario 3. If the timing was modified, the core damage frequency of those scenarios for which short detection and suppression times were used may increase by at least one order of magnitude.

4. In Reference [8), the licensee indicated that hot shorts had been considered in the analysis since Appendix R information, which includes hot shorts, had been used. Theoretically speaking, if one -

conducted a comprehensive analysis of potential failure modes of equipment and cables from a fire 4

and available safe shutdown paths given such failures, all relevant spurious actuations and other failures would be identified properly. However, from the statements made in Reference [8] it could be concluded that the licensee, in its Appendix R effort, did not consider the possibility of .

a LOCA or inadvertent steam dump from a fire event. This oversight indicated that the hot short I (or spurious actuation) phenomenon had not been properly addressed in the Appendix R analysis of this plant. .

l

-2.2.2 Summary of Responses to Requests for Additional Information The licensee's responses to the preceding second-round RAls are summarized as follows:

1. The licensee has analyzed the possioility of loss of offsite power from a fire at various points of the power plant, staning from the breakers in the switchyard to the buses inside the plant. Nine breakers are identified in the switchyard that require air and control power to change position.

The control cables for these breakers are not located within the plant, and therefore a plant fire cannot affect them. For those compartments which contain power cables and buses inside the plant that supply power to redundant huses, it is assumed that a fire leads to loss of offsite power. On-

- site power sources are credited for the recovery from loss of offsite power caused by a fire in these compartments. Lastly, the possibility of opening 4kV and 13kV feed breakers to the Class IE buses was addressed. A specific analysis of the control circuits associated with these breakers was not performed. For control room and cable spreading room fire scenarios, it is assumed that loss Energy Research, Inc. 8 ERI/NRC 98-505 4

i o of offsite power would occur, and the recovery of power is based on using the station's Gas Turbine.

2. Reference [6] states that the possibility of transformer fire had been addressed in the original submittal using " standard transformer oil for the analysis". It is added that the transformers are

• PCB transfonners" and "PCB is a non-flammable substance with no fire point and no flash point listed." The transformers located in the cable spreading roora alarm in the control room on high temperature. These are continuous alarms and would allow the operators to take the affected transformer out of service before it fails. It is also argued that because of the properties of PCb (polychlorinated biphenyl) transformers and the barriers of the cable spreading and switchgear rooms, a fire or explosion which could affect the remote shutdown capability in an adjacent compartment is not credible, in the event of a fire or explosion in these areas, the remote shutdown capability and the Gas Turbine will be used.

3. Regarding the time for automatic fire detection and suppression, the licensee argues that the ,

d critical part of the fire analysis for the compartments referenced in the RAI is the time to cable damage. It is added that as long as detection and suppression succeed prior to cable damage, the CDF will be low. The licensee argues that the time to cable damage is significantly greater than the detection and suppression times. For example, it is stated that for Scenario 2 of Compartment 156, cable damage occurs in,18.5 minutes and automatic detection and suppression in 65 seconds.

Based on this observation, the licensee concluded that "even if the calculated detection and suppression times are low, it is still reasonable to conclude that detection and suppression will occur prior to cable damage."

4. Regarding the hot short issue, the licensee states that the oversight was corrected and addressed in Reference [8] in response to RAI No. 4. It is stated that LOCA and inadvertent steam dump were included in the IPEEE fire analysis, it is added that, the Appendix R safe shutdown analysis was used, in part, for determining recovery methods, which considered the capability for recovering from potential hot shorts. The licensee states that 'since none of the fire scenarios analyzed for the IPEEE are more severe than those contemplated in the Appendix R program, the 1 hot short analysis contained within the Appendix R program should be bounding." It is added that a "rebaselining project" is underway for the Appendix R safe shutdown program, which will be re-evaluating the effects of hot shorts.

2.2.3 Review Findings Based on the discussiom provided in Reference (6), the following conclusions regarding the supplemental i

fire review issues can be reached:

1. Per Reference (6], the possibility ofloss of offsite power from a fire at various locations at the l power plant had been considered. From the discussions in Reference (6), it can be inferred that j power cables have been traced, and for those compartments where the power cables and associated 4kV Class IE buses are located, it was assumed that loss of offsite power would occur. The ,

control and instrumentation cables, however, h::ve been treated differently. It can be assumed that l the control and instrumentation cables that could cause loss of offsite power were not traced.  !

From the discussions provided in Reference [6), it is clear that for control room and cable j i

Energy Research, Inc. 9 ERl/NRC 98-505 l

spreading room fire scenarios, it was assumed that loss of offsite power would occur. However, it is not clear whether other compartments have been analyzed properly.

2. Per Reference [6], the possibility of transformer fire had been addressed in Reference [1]. It is' also stated that the oil used in those transformers is composed of PCB-based fluids, which are considered non-flammable. If indeed the transformers are filled with a fluid that is composed almost entirely of PCBs, it is correct to conclude that the possibility of a fire is not significant.

If, however, the transformer fluid is a mixture of PCBs and other substances, depending on the concentration of the mixture, the potential for a fire may exist. It must be noted that PCBs are barmed substances and have not been manufactured for over a decade. The PCB transformers 'are typically refilled with oser fluids. The licensee does not indicate whether or not the transformer fluid is a mixture of PCBs and other substances, nor whether the plant has been capable of maintaining the non-flammable characteristic of the fluid.

The licensee, in Reference [6], presents a summary of an analysis of transformer fires. There are several features that provide protection against such fires and mitigate the effects of energetic failure of a transformer. The control room receives a high-temperature alarm from the transformers in the cable spreading room, which allows the operators to take the transformer out of service before funher deterioration of its condition. It is also argued that the possibility of affectmg an adjacent compartment in the case of an energetic failure of a transformer is unlikely, given that the cable spreading room is a reinforced concrete room with 3-hour rated fire barriers.

3. For the majority of fire scenarios for fire Compartments 156,166, and 318 (the cable spreading room), the licensee has used 9 to 142 seconds Oess than 3 minutes) for automatic detection and suppression of the fire. This timing is too short and is not supported by industry er.perience. In Reference [6], the licensee argues that the time to cable damage is significantly greater than the detection and suppression times for those fire scenarios. Therefore, the final conclusion n ould not change if the detection and suppression times are altered. Although the statement made by the licensee is not completely valM (since the likelihood of suppression failure depends on the fire detection and suppression timin.a the final conclusions of the IPEEE would not change if more realistic values were employed.

4. The licensee, in Reference [6), states that the issue of hot shorts was addressed in Reference [8].

The licensee stated that LOCA and inadvertent steam dump have been addressed. It is added that the Appendix R safe shutdown analysis was used, in part, for determining recovery methods, which considered the capability for recovering from potential hot shorts. The licensee believes that the possibility of LOCA and inadvertent steam dump had been addressed in the Appendix R effort and its resulting procedures. No specific details were provided to support these statements.

Energy Research, Inc. 10 ERl/NRC 98-505

, v ,

3 OVERALL EVALUATION AND CONCLUSIONS 3.1 Seismic In addition to the discussion previously provided in Section 3.1 of Reference [2), the present supplemental review has confirmed the following observations:

1. As presented by the licensee, the reported plant high confidence of low probability of failure (HCLPF) spectrum does not exceed even the plant's design spectrum for frequencies less than about 4 Hz.

2. There are several reported anomalies and outliers that have not been fully resolved (examples include anchorage concerns, interaction concerns, potential failures of block walls, and potential failures of flat-bottomed tanks).

3. The licensee has not fully satisfied the guidance of NUREG-1407 with respect to the IPEEE relay evaluation; however, the information provided by the licensee suggests that it may be unlikely that any additional insights or low-ruggedness relays (beyond those revealed from the USI A-46 evaluation) would be encountered if a relay evaluation of IPEEE-only circuitry were to be performed.

4. The licensee has not demonstrated a full understanding of the implications of use of the surrogate element in the Point Beach seismic IPEEE.

Some additional points pertaining to assumptions made by the licensee have been noted in this STER.

- 3.2 Bre

.The licensee has expended considerable effort in the preparation of the fire IPEEE. Based on the review of References [1, 6, and 7], it can be concluded that the licensee has employed proper methodology and data for conducting the fire analysis and the fire IPEEE addresses all major elements specified in NUREG-1407. The strengths of the licensee's fire IPEEE effort mentioned in Reference [2] is further validated with l' the information provided by the licensee in Reference [6). However, the concems listed in Section 3.2 of Reference [2] are only partly addressed in Reference [6]. Concems regarding lack of sufficient information in the submittal to allow significant verification of the analysis. cross zone analysis. sensitivity of the final conclusions to the basic assumptions regarding fuel loading and heat transfer mechanisms, modeling of transient combustibles, correlations among numerical results presented in the submittal, modeling of suppression system unavailability and possibility of dual-unit impact weaken the presentation of the analyses carned out in support of the submittal. While these concems are noted as observations. they are not expected l to contribute to masking of a major vulnerability in the plant. In addition to the above listed concems, the following two. which were raised in the course of reviewing Reference [6] should also be noted:

1. Although the licensee has stated that the possibility ofloss of offsite power has been addressed for '

comf=Tmsts where the associated power cables are present (i.e.. switchgear rooms, the control room and the cab!c spreading room). it is not clear if other compartments have been analyzed properly.

Energy Research, Inc. I1 ERl/NRC 98-505 e

6.

4 -

However, given the extent of analysis that the licensee has conducted. it is assumed that the other compartments were treated properly.

2. For several fire scenarios, the licensee has used unreasonably small time periods for automatic fire detection and suppression. However. given that time to cable damage for those scenarios is signi6cantly longer than automatic detection and suppression. the impact of the optimistic assumptions regarding automatic detection and suppression may be mininul.

Although the level ofconservatism in the final conclusions, and ranking of fire scenarios. may be impacted by these concems, overall it can be concluded that the licensee has conducted a reasonable analysis of the potential for fire events and has gamed a reasonable perspective of potential fire vulnerabilities at Point Beach Nuclear Plant.

s Energy Research, Inc. 12 ERl/NRC 98-505

. v 4 SUPPLEMENTAL IPEEE INSIGHTS, IMPROVEMENTS, AND COMMITMENTS 4.1 kmic In addition to the summary provided in Section 4.1 of Reference [2), it is noted here that the licensee has provided some clarifications of significant importance to this supplemental seismic review. Furthermore, in terms of additional IPEEE insights, the licensee reports that the seismically induced frequency of release 4

category G (i.e., containment isolation failure) is 2.13 x 10 per reactor year. The licensee has indicated that USI A 46 improvements have been implemented or planned, but reports no new seismic-related IPEEE plant improvements or licensee commitments.

4.2 Bre The licensee has provided some clarifications of significant importance to this supplemental fire review, but reports no new fire-related IPEEE plant improvements or licensee commitments. However, in Reference [6], as part of the discussion on hot shorts, the licensee does state that "A rebaselining project is underway for the Appendix R Safe Shutdown Program. This project will be re-evaluating the effects of hot shorts to ensure that recovery is addressed."

I l

l 1

Energy Research, Inc. 13 ERl/NRC 98-505

., .w l

l 5 REFERENCES l 1. " Point Beach Nuclear Plant Individual Plant Examination of External Events (IPEEE)," Wisconsin Electric Power Company (WE), June 30,1995. i

2. " Technical Evaluation Report on the ' Submittal-Only' Review of the Individual Plant Examination of External Events at Point Beach Nuclear Plant, Units 1 and 2," ERl/NRC %-505, Final Report, March 1998.

3. " Individual Plant Examination of External Events (IPEEE) for Severe Accident Vulnerabilities -

10CFR50.54(f)," U. S. Nuclear Regulatory Commission Generic Letter 88-20, Supplement 4, June 28,1991.

4. J. T. Chen, et al., " Procedure and Submittal Guidance for the Individual Plant Fxamination of External Events (IPEEE) for Severe Accident Vulnerabilities," U. S. Nuciear Regulatory Commission, NUREG-1407, May 1991.

5. " Request for Additional Information on Point Beach IPEEE Submittal," memo from M.

Cunningham to C. Carpenter, March 3,1998.

6. " Response to Request for Additional Information, Individual Plant Examination of External Events

- Point Beach Nuclear Plant, Units 1 and 2," Wisconsin Electric Power Company, July 24,1998.

7. " Response to November 16,1998, Individual Plant Examination of External Events - Point Beach Nuclear Plant, Units 1 and 2," Wisconsin Electric Power Company, December 17, 1998.

8. " Response to Request for Additional Information, Individual Plant Examination of External Events y

- Point Beach Nuclear Plant, Units 1 and 2," Wisconsin Electric Power Company, August 15, 1996.

l l

I I

l r

Energy Research, Inc. 14 ERl/NRC 98-505