ML18066A222

Forwards Response to NRC 980304 Second Request for Addl Info Re Util 950630 Response to GL 88-20,suppl 4, Ipeees.

ML18066A222
Person / Time
Site:	Palisades
Issue date:	07/02/1998
From:	Haskell N CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.)
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
GL-88-20, TAC-M83653, NUDOCS 9807080020
Download: ML18066A222 (18)

Text

Palisades Nuclear Plant Tel: 616 764 2276 27780 Blue Star Memorial Highway Fax: 616 764 2490 Covert, Ml 49043 Nathan L. Ha*kell Director. Licensing July 2, 1998 U.S. Nuclear Regulatory Commission ATIN: Document Control Desk Washington, DC 20555 DOCKET 50-255 - LICENSE DPR PALISADES PLANT GENERIC LETIER 88-20, SUPPLEMENT 4, INDIVIDUAL PLANT EXAMINATION OF EXTERNAL EVENTS - REQUEST FOR ADDITIONAL INFORMATION (TAC NO. M83653)

On June 30, 1995, Consumers Energy submitted the response to Generic Letter 88-20, Supplement 4, Individual Plant Examination of External Events (IPEEE). On June 14, 1996, a request for additional information (RAI) was received. Consumers Energy responded to that RAI by letter dated September 30, 1996. On March 4, 1998, a second request for additional information was received. This letter provides the response to that second request for information.

The RAI requested a completion date of 90 days from the date of the NRC letter; however, due to the Palisades refueling outage, an extension of the response period to 120 days was granted by NRC letter dated April 1, 1998, to Consumers Energy.

The attachment to this letter lists each of the individual requests for information and provides the Consumers Energy response.

9807080020 980702 PDR ADOCK 05000255

-- P ~-*

PDR

. r . *, of

• ~/

i

. I' l l./..1, 04

2

SUMMARY

OF COMMITMENTS This letter contains no new commitments and no revisions to existing commitments.

athan L. Haskell irector, Licensing CC Administrator, Region Ill, USNRC Project Manager, NRR, USNRC NRC Resident Inspector - Palisades Attachment Enclosures

/

ATTACHMENT CONSUMERS ENERGY COMPANY PALISADES PLANT DOCKET 50-255 Response to Request for Additional Information (NRC Letter Dated March 4, 1998)

Generic Letter 88-20, Supplement 4, Individual Plant Examination of External Events

. (TAC NO. M83653) 11 Pages

ATTACHMENT PALISADES NUCLEAR PLANT RESPONSE TO IPEEE REQUEST FOR ADDITIONAL INFORMATION NRC letter dated March 4, 1998, requested additional information with respect to Consumers Energy's June 30, 1995, response to Generic Letter 88-20, Supplement 4.

Below is each request for additional information and the Consumers Energy response.

NRG Request:

1. Based on the response to RA/ question A. 11 in the September 30, 1996, letter, the 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 5 Hz. The reported HCLPF spectrum is also much lower, over frequencies of interest, than the review level earthquake (RLE) spectrum defined by a NUREGICR-0098 median spectral shape anchored to a peak ground acceleration (PGA) value of 0. 3g. Please provide a discussion on why the plant is considered to have adequate margins against potential seismically induced severe accidents.

Consumers Energy Response:

Because the plant level High Confidence of Low Probability of Failure (HCLPF) (0.217g) is greater than the plant design basis safe shutdown earthquake (0.2g), Palisades believes that there is an adequate margin of safety against potential seismic events.

As part of the Individual Plant Examination of External Events (IPEEE), Palisades performed a seismic PRA (SPRA) as opposed to a seismic margins assessment (SMA).

A SPRA evaluates seismic capacity by using specific component fragilities in the fault .

and event tree models. Some components have detailed fragilities calculated, but mo~t components in the SPRA are screened into fragility bins based on walkdowns and calculation reviews. A component is expected to have a much higher fragility than the screening fragility used in the SPRA if a detailed fragility calculation were to be performed. Therefore, screening fragilities assigned to components are conservative.

The SPRA models (with the component fragilities) are then evaluated over the complete range of ground motions that define the mean seismic hazard curve for Palisades to obtain estimates of the Core Damage Frequency (CDF), the plant median capacity, and the plant HCLPF. All components in the SPRA have a fragility, and no component is screened out of the SPRA.

page 1 of 11

As described in NUREG-1407, a utility has the option to conduct a SMA or a SPRA.

Consum,ers En.ergy chose to complete a SPRA for the Palisades Plant. Given that a utility elects to perform one or the other analysis, Consumers Energy believes there is no technical validity to compare the results of one type of study (e.g., a SPRA) to the criteria for another (e.g., the Review Level Earthquake (RLE) as defined by the NUREG/CR-0098 median response spectrum anchored to 0.3g PGA).

From a technical perspective, Consumers Energy believes it is invalid to compare the HCLPF generated in a SPRA with either the HCLPF derived in a SMA or to the RLE (e.g., the NUREG/CR-0098 median response spectrum anchored to 0.3g PGA). The HCLPF derived from a SMA is not determined in the same manner as a HCLPF computed in a SPRA. As a result, these different estimates of a plant HCLPF cannot be logict;1lly compared (i.e., on a consistent basis). Based on the experience of consultants, had a SMA been performed for the plant, the estimated HCLPF would have been higher. This is because the SMA approach screens everything out at a 0.3g HCLPF and the SPRA evaluates components with any HCLPF (such as 0.1 g HCLPF).

Allowing lower fragility components to be evaluated in the SPRA leads to a lower plant HCLPF. Furthermore, anchoring HCLPF values estimated by different methods to response spectral shapes is also inappropriate.

It is Consumers Energy's view that the comparison between the EPRI or the Lawrence Livermore National Laboratory (LLNL) response spectral shapes and the Palisades SSE response spectrum highlights the significant difference between western United States soil site ground motions, and the central and eastern United States ground motions for rock sites. As can be seen from the central and eastern United States seismic hazard studies (e.g., EPRI and LLNL}, the ground motion is not nearly as pronounced in the low frequency range as motions (for similar size earthquakes) in the western United States. In addition to basic differences between eastern and western United States ground motions, the Palisades SSE response spectrum is also influenced by ground motions recorded on soil site conditions, which further"accentuates the differences at lower frequencies. Lastly, there is also a magnitude effect that may exist as well. The EPRI and LLNL response spectral shapes shown in the response to RAI Question No. A 11 for the Palisades site are dominated by the response spectra for relatively small earthquakes (e.g., 5.0 < mb <5.8), which tend to roll-off faster at low frequencies, particularly for rock sites, than motions recorded for larger, western United States events.

NRG Request:

2. Although the submittal reports numerous anomalous conditions, outliers, and low capacity components (e.g., poor anchorage, unanalyzed [and unqualified] block walls, and interaction concerns), no potential plant improvements have been identified. For each such case where an anomaly, outlier, or evaluated low page 2 of 11

capacity has been identified, please discuss the disposition and basis for such a{sposWon.

Consumers Energy Response:

Anomalous conditions were identified during walkdowns or reviews of calculations and were considered when a component fragility was estimated or calculated. Each anomalous condition, outlier or low capacity component was accounted for explicitly in the SPRA, and none were identified as dominant seismic contributors. No plant improvements were identified as a result.

Certain relays were identified as outliers. All relays identified as outliers in the SPRA were also included in the USI A-46 Seismic Qualification Users Group (SQUG) program. These relays were dispositioned as SQUG outliers. All SPRA outlier relays have been dispositioned (replaced) per SQUG guidance as of June 1998. Therefore, no further plant improvements were identified for SPRA outliers.

Low capacity components were identified during walkdowns or reviews of calculations and were explicitly modeled in the SPRA with their appropriate low fragility parameters. (All SPRA Components With a High Fussell-Vesely Threshold [FV])

contains the list of all components in the SPRA that are in the cutsetS (not truncated out), with a FV>5.00E-3. Enclosure 2 (Seismic SPRA Components With a High

• Fussell-Vesely Threshold [FV]) provides a list of seismic components in the SPRA cutsets with a FV>5.00E-3. The reported importance measures were determined at nine ground motion levels and then combined based on the weighted contribution of each ground motion level to the total seismic GDF. Each seismic event.in Enclosure 2, including the 0.1 g HCLPF events, are discussed further in response to NRG Question 4, below. That discussion demonstrates that important sequences have high seismic capacity components (HCLPF~0.3g) redundant to the low seismic capacity components (HCLPF=O: 1g).

NRG Request:

3. Based on the responses to seismic RA/s questions A.1 and A.3 in the September 30, 1996, letter, the staff believes that the practice of simply assigning a 0. 1g PGA HCLPF to identified anomalies/outliers is inappropriate.

The response to RAJ question A. 1 indicated that some of the components in this category were not walked down. Hence, the statement in the response to RAJ question A.3, that "All equipment assigned a HCLPF of 0.19 [sic] is expected to have a much higher value if detailed fragility analysis were to be performed" does not appear to have a sound basis. Furthermore, it is not clear that the risk achievement worth and Fussell-Vesely thresholds cited in the response to RAJ page 3 of 11

question A. 3 would ensure a sufficiently low contribution to seismic core damage frequency (GDF), such that the components in question could be collectively screened out from the seismic PRA model. Therefore, please provide plant specific analysis of the seismic fragilities for these components, requantify the seismic GDF, and reassess the important risk contributors for those components.

Consumers Energy Response:

As stated in the Palisades IPEEE submittal dated June 30, 1995, components that did not pass a 0.3g HCLPF or a 0.5g HCLPF screen were placed in the 0.1 g HCLPF bin.

Also, components that were not walked down prior to the initial seismic CDF calculations were placed in the 0.1 g HCLPF bin. Eventually, all components in the SPRA were walked down and were screened appropriately. Based on these walk downs and reviews of calculations, a number of components were determined to have a low fragility and placed in the 0.1 g HCLPF bin. These components were conservatively assigned a 0.1 g HCLPF since it is expected that detailed fragility calculations would result in HCLPFs >>0.1g. Since none of the 0.1g HCLPF components are an important contributor to the SPRA results, no further fragility calculations for these components are required. No requantification or reassessment of important risk contributors is required.

After all components were walked down and assigned a HCLPF, initial seismic CDF calculations were performed. The initial seismic CDF results were reviewed and detailed fragility calculations were performed for components that were identified as important contributors (providing high contributions/driving the CDF). For components that were not important contributors, the screening parameters were retained.

Two sensitivity evaluations were performed for components assigned a 0.1g HCLPF:

1) The first sensitivity evaluation consisted of all components assigned a 0.1 g HCLPF (median capacity of 0.22g PGA) with a FV>5.00E-3. These components were changed to have a median capacity of 0.05g PGA; and,

2) The second evaluation consisted of all components assigned a 0.1 g HCLPF and were changed to have a median capacity of 0.1 g PGA.

The combined uncertainty, be, remained the same at 0.46 for both sensitivities. The median capacity of 0.1 g PGA or 0.05g PGA is conservative and an unlikely measure of the true capacity of any of these components. The results of the first sensitivity indicate a seismic CDF increase of approximately 13 percent (to 1.00E-5/yr), essentially no change in the plant level median capacity (0.487g vs .488g), and a plant HCLPF decrease to 0.18g (from 0.217g). The results of the second sensitivity indicate a seismic CDF increase of approximately 12 percent (to 9.95E-6/yr), no change in the page 4 of 11

plant level median capacity (0.488g), and a plant HCLPF decrease to 0.18g (from 0.217g). These sensitivity evaluations demonstrate that reducing the fragility of the 0.1 g HCLPF components does not significantly change the SPRA results. (Seismic SPRA Components With a High FV) provides a list of seismic components in the SPRA cutsets with a FV>5.00E-3. Each seismic event in Enclosure 2, including the 0.1 g HCLPF events, are discussed further in response to NRC Question 4, below. That discussion demonstrates that important sequences have high seismic capacity components (HCLPF;::0.3g) redundant to the low seismic capacity components (HCLPF=0.1 g). Therefore, components in the 0.1 g HCLPF bin are not risk significant contributors and, therefore, do not require detailed fragility analyses to determine their capacity.

NRC Request:

4. In the IPEEE submittal, the most important seismic failures contributing to the seismic CDF were identified as:

• Fire Protection System (FPS) (needed to provide make-up to condensate storage tank [CST] following seismic-induced lost of offsite power [LOOP])

Diesel day tanks (T-24 and T-40) for diesel-driven fire pumps Control panel for diesel-driven fire pumps (EC-137)

Station transformer 13 (EX-13)

• MS/Vs (interaction hazard may prevent them from closing)

• Diesel fuel oil storage tank (needed to supply diesel fuel to both diesel generators following seismic LOOP)

• Bus 1D undervoltage relays (needed to power auxiliary feedwater {AFW] pump P-BC, the only AFW pump available following the Joss of FPS)

In contrast, your response to RAJ question A.3 in the September 30, 1996, letter indicated that the most important seismic equipment failures were:

• Service water pumps P- 7A, B, C

• Auxiliary feedwater pump P-BC

• Diesel generator 1-1 undervoltage relay 127D-1

• High pressure air receiver tanks T-9A, B Please explain why these two lists differ and provide the sequences associated with the equipment failures. Please identify the ranking and contribution of seismic-induced equipment failures that are most important to the seismic CDF.

page 5 of 11

Consumers Energy Response:

The two different lists of important seismic equipment failures comes from reviewing the SPRA results in different ways. The seismic importance measures Fussell-Vesely (FV) and Risk Achievement Worth (RAW) were calculated after the IPEEE submittal dated June 30, 1995, was docketed. The original IPEEE submittal based the important seismic failures on the equipment that appeared a significant number of times in the cutsets (basically FV). In response to RAI question A.3 in the September 30, 1996, letter, important seismic failures were based on calculations that resulted in a FV>5.00E-3 and RAW>2.0. Looking at importance measures provides a better basis for determining risk significance. However, focusing on components with a RAW>2.0 may be too limiting to identify important contributors to seismic CDF. This is because the RAW indicates how important a component may become if degraded to the point of failure, whereas the FV importance measure indicates importance of current contribution to the SPRA results. Therefore, the list of components with a FV>5.00E-3 is appropriate for discussing what is currently important to the SPRA. Enclosure 2 lists all the seismic components with a FV>5.00E-3. Enclosure 2 basically contains all the seismic components considered important in the IPEEE list and the response to question A.3 in the September 30, 1996, letter, plus others that were not included in either list. The only component not in Enclosure 2 but in one of the other lists is diesel fuel oil supply tank T-10, which is not seismically important based on FV measures.

Each of the seismic failures in Enclosure 2 are discussed in detail below along with their important sequences.

Seismic event J1AVAB607 represents common cause (simultaneous) seismic failure of both main steam isolation valves (MSIVs) to close. Based on walkdowns, there is a potential interaction concern that may prevent the MSIVs from closing. Since a potential interaction exists for both MSIVs, a common cause seismic event was developed to represent this potential failure mode. Closure of the MS IVs is important only if an Atmospheric Dump Valve (ADV) opens and fails to close. ADVs are assumed to open on a seismically induced plant trip. If the MSIV on the unaffected steam generator remains open, then both steam generators depressurize as there is a cross-connect between the steam lines downstream of the MS IVs. The blowdown of both steam generators is modeled in the SPRA as an excessive steam demand requiring termination of the Auxiliary Feedwater System (AFW) flow to the steam generators. This is very conservative modeling as, in fact, the operators would continue to supply AFW makeup to the least affected steam generator rather than isolate both of them. Realistic development of two steam generator blowdown sequences would likely page 6 of 11

eliminate seismic interactions with the MSIVs as having any risk significance. The SPRA does . not. credit operators with continuing to feed at least one steam generator.

Seismic events J 1EC137 (Fire Protection System [FPS] diesel pump P-41 control panel), J1TKT24 (FPS diesel pump P-98 day tank), J1TKT40 (FPS diesel pump P-41 day tank) and J 1TREX13 (FPS motor pump P-9A power supply transformer) represent seismic failures of individual FPS components. The FPS is credited as one of the methods for makeup to the condensate storage tank (CST) for AFW suction. There are two other methods available: makeup from the Demineralized Water Storage Tank (T-939), using the Demineralized Water Transfer Pump (P-936); and Service Water System (SWS). Pump P-936 requires off-site power, which has a very low fragility and is highly likely to be lost during a seismic event. However, the SWS has a high system fragility and can supply the required flow to support continued AFW operation following*

. a seismic event.

Seismic events J 1HCAB590 (SWS cooling to Component Cooling Water System [CCW]

heat exchangers) and J1TKT3 (CCW surge tank) represent seismic failures of the CCW system. The SPRA only credits the CCW system to provide cooling for the Engineering Safeguards System (ESS) pumps and the shutdown cooling heat exchangers (SDCHxs). SWS backup for ESS pump cooling is credited in the SPRA. The SDCHxs remove decay heat from the containment when ESS pumps take suction from the containment sump (recirculation phase). The ESS pumps provide primary system inventory control/makeup during small break LOCAs or when once-through-cooling (OTC) is initiated. Seismically induced SBLOCAs have a low probability of occurrence and did not contribute to the SPRA results. Initiation of OTC is only required if the AFW system fails to remove decay heat via the steam generators. Also, containment heat removal can be accomplished through the use of containment air coolers and SWS.

AFW, SWS and Containment Air Coolers (CAC) all have a high system fragility.

Seismic events J3HEAB590 (common cause seismic failure of both SDCHxs),

J3RVAB570 (seismic interaction failure of RV-3057, air supply to containment sump suction valve CV-3057) and J3TKAB570 (common cause seismic failure of both high pressure air receiver tanks T-9A&B) represent failure of ESS pumps due to recirculation phase failures. The ESS pumps are only required during small break LOCAs or when once-through-cooling (OTC) is initiated. Seismically induced SBLOCAs have a low probability of occurrence and did not contribute to the SPRA results. Initiation of OTC is only required if the AFW system fails to remove decay heat via the steam generators.

AFW has a high system fragility.

Seismic event J3MORB590 represents common cause (simultaneous) seismic failure of all 4 High Pressure Safety Injection (HPSI) train 1 injection valves, all 4 Low Pressure Safety Injection (LPSI) injection valves, and both HPSI hot leg injection valves. LPSI and hot leg injection are not credited for any success paths in the SPRA. The HPSI page 7 of 11

train 1 injection valves have 4 redundant HPSI train 2 injection valves, which have a high fragility. *Furthermore, HPSI is only required for SBLOCAs and for OTC.

Seismically induced SBLOCAs have a low probability of occurrence and did not contribute to the SPRA results. Initiation of OTC is only required if the AFW system fails to remove decay heat via the steam generators. AFW has a high system fragility.

Seismic event J47HERB590 represents common cause (simultaneous) seismic failure of all containment air coolers (fragility=.47g HCLPF). The CACs remove decay heat from the containment during OTC. Initiation of OTC is only required if the AFW system fails to remove decay heat via the steam generators. The CACs and AFW systems both have high system fragilities.

Seismic event J5PMP8C represents seismic failure of motor driven AFW pump P-8C (fragility=.55g HCLPF). This AFW pump has the lowest fragility of the two motor driven AFW pumps (P-8A=.72g HCLPF). Steam driven AFW pump P-8B had a much higher fragility than either motor driven AFW pump and was represented by the surrogate fragility (meaning its fragility is expected to be >1.0g HCLPF if calculated). AFW is involved in all seismic shutdown scenarios for decay heat removal using the steam generators. AFW pump P-8C has one additional water source for suction than does P-8A&B: the SWS can be used if normal Condensate Storage (CST) makeup (T-939/P-936) or FPS makeup is lost. Both the normal CST makeup and FPS makeup have low system fragilities, which is why P-8C has a higher importance than P-8A&B.

An alternate method for decay heat removal credited in the SPRA is once-through-cooling (OTC) using the HPSI pumps and Power Operated Relief Valves (PORVs).

AFW, HPSI and PORV's all have high system fragilities.

Seismic event J5PMSWS represents common cause (simultaneous) seismic failure of all three service water pumps (P-7A,B,C). Failure of all three service water pumps*is highly unlikely (fragility=.52g HCLPF), but without service water, the diesel generators fail due to loss of cooling. This coupled with the relatively high probability of losing off-site power in a seismic event, leads to a station blackout scenario.

Seismic event J5RE12702 represents seismic failure of diesel generator 1-2 (OG 1-2) bus undervoltage relay 1270-2 (fragility=.66g HCLPF). This relay is required for proper load shedding and breaker closing .to operate the OG. Failure of a OG is important given the relatively high probability of losing off-site power in a seismic event. The redundant OG (OG 1-1) bus undervoltage relay (1270-1) has a considerably higher fragility (.87g HCLPF). The higher seismic importance of OG 1-2 undervoltage relay (vs OG 1-1 undervoltage relay) is based on the lower fragility of the relay and that 2 of 3 SWS pumps are powered from bus 10 (DG 1-2). Failure of 2 of 3 SWS pumps means less flow from the SWS and that SWS system loads will have to be shed in order to provide enough flow for the essential functions. Both diesel generators (including the undervoltage relays) have high system fragilities.

page 8 of 11

NRG Request:

5. The submittal uses sensitivity analysis to conclude that nonseismic failures are an important part of the seismic core damage frequency. This conclusion hinges on the fact that (for the likely core damage sequences) nonseismic failures must occur in combination with seismic failures, in order for core damage to be realized. The study's conclusion that random failures are important, and yet its implication that seismic failures are not, suggests that the importance of seismic failures may have been missed. The study shows that the largest contribution to seismic core damage frequency comes from motions in the range of 0. 359 to

0. 459 PGA. Since seismic failure rates are more significant at this level than are random failures, it is difficult to see why seismic failures are not deemed to be important. Similarly, the implication that operator errors are more important to seismic GDF than are seismic failures ignores the highly uncertain nature of operator fragility estimates and the potential that such estimates could mask the effects of seismic failures. Please identify and discuss the relative importance of seismic failures as compared to nonseismic failures and operator errors. Also, please discuss what, if any, procedural improvements are planned to reduce the risk associated with human errors.

Consumers Energy Response:

The IPEEE submittal presented the results of several sensitivity analyses to highlight the importance of random failures and human errors in the SPRA. These results were not intended to indicate that random failures and human errors are more important than seismic failures. The IPEEE submittal dated June 30, 1995, section 3.6.5.4.3 provides the results of a sensitivity analysis where all random events had their failure rates set to 0.0. The CDF decreased to 5.83E-6/yr from 8.88E-6/yr. This demonstrates that seismic failures contribute the most to the SPRA CDF, but that random failures also contribute to -35% of CDF.

Also, as importance measures are examined, random and seismic failures and operator errors are all indicated as important contributors to SPRA results. A component in the SPRA that is considered important is defined as FV>5.00E-3. Enclosure 1 (All SPRA Components With a High Fussell-Vesely Threshold [FV]) contains the list of all a

components in the SPRA that are in the cutsets (not truncated out), with FV>5.00E-3. (Seismic SPRA Components With a High Fussell-Vesely Threshold [FV])

provides a list of seismic components in the SPRA cutsets with a FV>5.00E-3. indicates that there is a total of 35 risk significant components: 15 seismic failures; 16 random failures and 4 human errors.

page 9 of 11

The four human errors in Enclosure 1 are adequately addressed in current procedures and training and, therefore, no procedural improvements are planned as a result of human error contributions to SPRA results.

NRG Request:

6. The response to RA/ question A. 10 did not provide adequate information on seismically induced loss of fire suppression system capability. The evaluation apparently focused on potential interactions with safety equipment, rather than on loss of fire suppression capability itself Also, no walkdown findings are mentioned, and the discussion was limited to FPS piping. Some examples of relevant items found in past studies include (but are not limited to):

• Unanchored C02 tanks or bottles

• Sprinkler standoffs penetrating suspended ceilings

• Fire pumps (unanchored or on vibration isolation mounts)

• Unrestrained batteries/rack for diesel-driven fire pumps

• Block wall interactions with fire pumps or batteries

• 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.

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

Provide guidelines given to walkdown personnel for evaluating these issues (if they exit).

Consumers Energy Response:

The Palisades SPRA identified the Fire Protection System (FPS) as low seismic capacity as several of the FPS components credited in the SPRA are assigned 0.1g HCLPF fragilities. As such, the FPS is expected to fail at relatively low ground motions in a postulated seismic event. However, no piping or suppression system failures are expected that would affect other SPRA components.

All of the FPS, including the suppression system, was walked down during the SPRA walkdowns. FPS components that support a SPRA credited function were assigned a fragility. Other parts of the FPS were walked down to identify potential failure mechanisms that would affect SPRA components (interaction and flooding concerns).

No failure mechanisms were identified that would affect other SPRA components. Also, the suppression system was not credited in the SPRA to perform any safe shutdown functions. The only function credited in the SPRA for the FPS is makeup to the CST for AFW suction, as discussed in detail in the response to NRC question 4, above.

page 10 of 11

The Palisades seismically induced fire analysis evaluated the likelihood of initiating a fire in einy area; the equipment affected by a fire; and the effect on the safe shutdown of the plarif following a seismic event with a fire. The results of the evaluation conclude that the Turbine Building is the only location for which a credible fire may initiate following a seismic event (Section 3.5.2.3.3). All equipment (including cables) in the area in which the fire was assumed to initiate was assumed unavailable for plant shutdown following a seismic event with a fire in the Turbine Building. There was no credit for fire suppression to terminate a fire in the Turbine Building.

page 11 of 11

ENCLOSURE 1 ALL SPRA COMPONENTS WITH A HIGH FUSSELL-VESELY THRESHOLD [FV]

ENCLOSURE 1 ALL SPRA COMPONENTS WITH A HIGH FUSSELL-VESELY THRESHOLD [FV]

Basic Event 11;escr.1 .. ,* ... .... :**** .. .

IST-1

• AFW MOTOR DRIVEN P-8C FAILS TO START OR RUN IST-13 SWS TO AFW MV-FW750 FTOIFTRO IST-162 DG 1-2 BKR 152-213 FTC IST-205 ADV CV-0781 FAILS TO CLOSE IST-210 ADV CV-0782 FAILS TO CLOSE IST-215 ADV CV-0779 FAILS TO CLOSE IST-219 ADV CV-0780 FAILS TO CLOSE IST-367 NO FLOW FROM FPS DIESEL DRIVEN P-98 IST-368 NO FLOW FROM FPS DIESEL DRIVEN P-41 IST-4 AFW P-8C TO SG B CV-0736A FAILS TO OPEN DIG 1-1 FAILURE IST-492 DIG 1-2 FAILURE IST-5 AFW P-8C TO SG A CV-0737A FAILS TO OPEN IST-6 AFW MOTOR DRIVEN P-8A FAILS TO START OR RUN IST-75 NO FLOW FROM SWS P-78 IST-78 NON-CRITICAL SW FAILS TO ISOLATE J1AVAB607 SE:iSMIC FAILURE 0:17 MSIVS TO CLOSE/

J1EC137 *$i;.1$MIC FAIQJRE OF*FP$J:>IE.$E.L:PQMP*.P41** ¢0NTRO.L***P:ANE.W*********>>****************

J1HCAB590 *SEISMIC*FAILOREOFALLSWCLG diJTLEtVLV:S FOR CCWHE:At:**EXCH<*****./

J1TKT24 SEISMIC FAiLDRE OF FPS P~98 blESELbAY.tANK /

J1TKT3 SEiSMiC F.ArtUR:E 6FcCWs0R:GETANK

> / **.*. *.*.*,*,**,****

J1TKT40 .$!;1$Mi¢ F.AiLURi;>OF F.P$ P.4.tf of~$!;L bAY**T:ANi(/**************.*.*.*****

J1TREX13 $E;f$Mi¢ F.AiLDRi; OF F.P$MotoR<PR!VE;N P~9AP.oW.E;RtRAN$f:ORM~R /

J3HEAB590 J3MORB590 J3RVAB570 *sEiSMIC F.Aii::DRE()F:*cdNTSUMP.sUctibN cV~3o57*.AiR SP.LY R\i. /** * * * * * * * * * * *,*.*.*

J3TKAB570 J47HERB590 *sEISMic**FAILURE OFALL*cdi\JT'AiNMENTAIR*.bdbLERS**** / . . . . . . .*.*.*,'.'.'.'.' '

J5PMP8C J5PMSWS J5RE127D2 SEISMic***FAILURE dFoG*1~2 BUS UNDERVdi..T:AGE RELAY}*********************************,*.****

QPMOEFPS OPERATOR FAILS TO START FIRE PUMP(S)

SFNOECNT OPERATOR FAILS TO RESTART CONTAINMENT AIR COOLER FANS UOOOBNCHDR OPERATOR FAILS TO ISOLATE NON-CRITICAL SWS

. UPMOEPUMP OPERATOR FAILS TO START A SW PUMP

ENCLOSURE 2 SEISMIC SPRA COMPONENTS WITH A HIGH FUSSELL-VESELY THRESHOLD

_,-_c - , - - [FV] - --* *

• ENCLOSURE 2 SEISMIC SPRA COMPONENTS WITH A HIGH FUSSELL-VESELY THRESHOLD

[FV]

Basic Event *Uesc .......

J1TKT24 *SEISMlC***FAILURE***oF***F*PS P.~9s**DIESELDAY**TANK>/*****************. . . .*.*.*.****

J1TKT40 SEISMIC..FAILURE OF FPS P.4tDJl::Sl::CDAYTANI(

• s****E***,**s***M**:**1*c****:F***A***:1*L**:u*::*R****E***:.o****:F***:F***p****s***::o***.,**E***s****E***L**:::p***u*:**M*:**p****:p****>>X*1**::c****a****N**:T****R****a****L*:::p***A*:*:N****E***L**:::::::::::::::::::***

J1EC137 ....:.::: .............. :*:*.:::...:..........:*.:*: ...... :.::*:::.":*....:-::::... :-:.: . :.. :.*:::.:* ::.: :: .":*:::.:..:.. :'. ... ." ....:..:::: ......~~:(.. :::: ...:*... :*: ... :...::.::.... :.. :*: ... :*::::..:..:'.*:.".:... :: ..":*::::::::-:-:*:::::***

J1AVAB607 SEiSMiC FAiLURE OF .. Ms1VS to cLosf:>/ * . .*** .

J1TREX13 $j;i$Mi¢ F'AilQR~ OF FPS MOTOR DRIVEN P~~APb.WJ;R:T8AN.$FORMSR /

J1HCAB590 *SEISMte**FAILURE bF.ALL*SWCLG bUtCET:Vtvs*FoR ccWHEAT EXCH*****//

J1TKT3 $EIS.Ml¢ F'AiLUR:g OF ccW SURGE TANK*******/<********************* * * * * * *

• J3HEAB590

. J3M_QBB.59Q. ____

J3RVAB570 J3TKAB570 J47HERB590 J5PMP8C *s6isMidf:A1i:iiRE OFMOTORbRiVEN AFVV*PUMP. piac* /*******************>>**>*.*.*.****

JSPMSWS J5RE127D2