ML20044B952
| ML20044B952 | |
| Person / Time | |
|---|---|
| Site: | Nine Mile Point  |
| Issue date: | 03/09/1993 |
| From: | Menning J Office of Nuclear Reactor Regulation |
| To: | Sylvia B NIAGARA MOHAWK POWER CORP. |
| References | |
| GL-88-30, TAC-M74437, NUDOCS 9303150223 | |
| Download: ML20044B952 (13) | |
Text
,
.~
We eSMo Q o
p UNITED STATES l
F lj NUCLEAR REGULATORY COMMISSION 1
.W ASHINGTON D.C. 20li65-0001
- +,*'
March 9, 1993 Docket No. 50-410 t
a 1
l Mr. B. Ralph Sylvia
.l Executive Vice President, Nuclear i
Niagara Mohawk Power Corporation
'301 Plainfield Road Syracuse,.New York 13212 l
Dear Mr. Sylvia:
j
SUBJECT:
REQUEST FOR ADDITIONAL INFORMATION ON NINE MILE POINT NUCLEAR i
STATION, UNIT NO. 2 - INDIVIDUAL PLANT EXAMINATION (IPE) i SUBMITTAL - GENERIC LETTER 88-20 (TAC NO. M74437)
.l t
.By letter dated July 30, 1992, Niagara Mohawk Power Corporation-(NMPC) l submitted the Nine Mile Point Nuclear Station, Unit No. 2 IPE results for NRC
'l review. Based on our review of the NMPC submittal, we have determined that we-need additional information to continue _ with our; review. The' enclosed list of questions identifies the information we need. We have found with other
~
licensees that a conference call is very useful prior to the licensee submitting its formal response to such a request for additional information.
.Accordingly, we plan to schedule a-conference call'in' about 20-30 days to discuss the NMPC responses. The purpose of the call is to give your staff and l
ours an opportunity to better' understand the issues and what information the NRC needs to finish this review.
It is likely that the conference call can in some cases result in more direct (and possibly shorter) written answers to the.
j questions. We request your written response within 60 days of the date of -
l this letter.
j Should you have any questions regarding this request, please contact me at 1
(301) 504-1406.
's i
- 6 4
NBC RE C W EB00PV a
190022 a
9303t50223 930309 PDR ADOCK 05000410.
[i p
[
/
- } -
e
.4 1
Mr. B. Ralph Sylvia March 9, 1993 This requirement affects one respondent and, therefore, is not subject to Office of Management and Budget review under P. L.96-511.
Sincerely, b.
John E. Menning, Project Manager Project Directorate I-1 Division of Reactor Projects - I/II Office of Nuclear Reactor Regulation
Enclosure:
As stated cc w/ enclosure:
See next page 8
e
n i s Mr. B. Ralph Sylvia Nine Mile Point Nuclear Station-
- Niagara Mohawk Power Corporation' Unit No. 2 cc:
4 Mark J. Wetterhahn, Esquire Regional Administrator, Region I' Winston & Strawn U. S. Nuclear Regulatory Commission 1400 L Street, NW.
475 Allendale Road' Washington, DC 20005-3502 King of Prussia, PA 19406 Mr. Richard Goldsmith Charles Donaldson,. Esquire Syracuse University Assistant Attorney General College of Law New York Department of Law E. I. White Hall Campus 120 Broadway-Syracuse, New York 12223 New York, New York 10271 Resident Inspector Mr. Richard M. Kessel Nine Mile Point Nuclear Station Chair and Executive Director-1 P. O. Box 126 State Consumer Protection Board
~
Lycoming, New York 13093 99 Washington Avenue Gary D. Wilson, Esquire
. Albany, New York 12210-Niagara Mohawk Power Corporation Mr. Martin J. McCormick Jr.
300 Erie Boulevard West Plant Manager, Unit 2
' Syracuse, New York 13202 Nine Mile: Point Nuclear Station.
Niagara Mohawk-Power Corporation t
Mr. David K. Greene P. O. Box 32 Manager Licensing Lycoming, New York 13093 t
Niagara Mohawk Power Corporation
.(
301 Plainfield Road Mr. Neil S. Carns Syracuse, New York 13212 Vice-President - Nuclear Generation.
Nine Mile Point Nuclear Station Ms. Donna Ross Niagara Mohawk Power Corporation
'I New York State Energy Office P. 0.' Box 32 2 Empire State Plaza Lycoming, New York 13093 16th Floor j
Albany, New York 12223 i
Supervisor 1
Town of Scriba Route 8, Box 382 Oswego, New York 13126
~
l
?
c
(
e b
w er
ENCLOSURE NINE MILE POINT NUCLEAR STATION UNIT NO. 2 (NMP2)
INDIVIDUAL PLANT EXAMINATION (IPE)
REQUEST FOR ADDITIONAL INFORMATION l
GENERAL-1 Please identify the date for which the rodel represents the plant configuration and identify the dates for which plant data used in the IPE were collected.
GENERAL-2 The discussion provided in Section 3.4.2 (Vulnerability Screening) indicates that no vulnerabilities were identified for NMP2, but does not address the criteria used to define vulnerabilities as requested by NUREG-1335, " Individual Plant Examination: Submittal l
Guidance." Provide a discussion of the criteria used to define vulnerabilities.
GENERAL-3 The analysis assesses the core damage frequency (CDF) after the inclusion of the improvements identified in Section 6 of the submittal and has concluded that there are no vulnerabilities.
Provide an estimate of the impact on CDF of the implemented improvements and the basis for the decision to implement.
Please provide, if available, estimates of the benefits (or credit taken in the IPE) for the proposed improvements in terms of both magnitude of reduction of releases and timing of releases.
If not available, qualitatively discuss the impact on IPE findings if improvements are not implemented. Were any vulnerabilities considered to exist prior to implementation?
GENERAL-4 Certain words in the submittal that were used to describe implementation of improvements are unclear (e.g., "it is assumed or the model assumes"; " guidance as appropriate"; and " precautions were identified").
Please provide a discussion clarifying which of the improvements identified in Section 6.2 were or are being implemented and the time frame for implementation.
Include in this discussion the recommended plant improvements listed in Table 6.1-1.
In addition, please provide a discussion of the process used to make the decision on which improvements or insights are chosen to t
be implemented based on the results of the analysis as identified in the tables provided for initiating events, top events, and split fractions.
If a quantitative criterion is not used, please identify what other basis is used.
B.E. - 1.
The two highest frequency, large-release sequences (early/high (E/H) Sequences I and 2 described on page 1-13 of the submittal) were caused by floods that subsequently caused loss of divisional AC power. All injections to the reactor pressure vessel (RPV) and containment were lost. This loss was followed by operator failure to isolate containment (probability of isolation failure was 0.1).
i An early release was assumed because of isolation failure.
However, due to high drywell temperature, containment would fail even if isolation was successful.
It appears that a similar sequence with successful isolation is 9 times more likely than the two cases presented in the submittal.
This is particularly relevant because procedures direct the operators to both restore power and close isolation valves.
Either drywell failure or isolation failure would likely result in a high release. Would drywell failure for these sequences occur within 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />? Similar questions can be raised about other sequences in which isolation failure appears to mitigate a potentially more severe and more likely sequence (e.g., E/H Sequences 4, 5, 6, and 7).
Please discuss the changes in frequency and magnitude and timing of releases for all E/H sequences for which containment isolation was assumed, if containment isolation failure probability is assumed to be (a) 1.0 and (b) 0.0.
B.E. - 2 E/H Sequences 3, 8, 9, and 10 resulted in large releases because of the assumed timing of operator actions to provide service water to the containment. What is the sensitivity of the results to-this timing? For example, would the likelihood and consequences of containment flooding be reduced if flooding was delayed? Would judicious timing for containment venting alter the result?
B.E. - 3 Please provide the rationale for assignment of the Decontamination Factors (DFs) in the table on page 4.7-28, titled "DF Adjustments for NMP2."
B.E. - 4 The list of references 4-72 through 4-89 cited in the first paragraph of Section 4.6.2.3 (page 4.6-5) could not _ be' located in the submittal. Please supply the list or indicate its location.
B.E. - 5 Please provide a legend for Table 4.6-8 that defines the event designators.
B.E. - 6 Please discuss the purposes of Tables 4.7.6-1 through 4.7.6-5 and their use in the IPE process.
B.E. - 7 In many of the key sequences that resulted in large releases, operators fail to locally close AC powered containment isolation valves before core damage.
(See, for example, E/H Sequences 1 and 2, page 1-13; E/H Sequence 6, page 1-15; and E/H Sequence 7, page 1-16.) What is the conditional probability, given core damage, that the containment was not isolated? Do all these sequences lead to an E/H release or are other outcomes possible?
If the failure to close containment isolation valves'is a-significant contributor to E/H releases, what means have been considered to reduce the probability of isolation valve failure, e.g., DC backup?
. B.E. - 8 It is stated on page 1-18 of the submittal that NMP2 procedures (based on the BWROG Emergency Procedure Guidelines) provide near-optimum guidance for accident management. However, on page 1-13 it is stated for E/H Sequences I and 2 that tne delayed isolation of containment (i.e., the procedurally directed action) would actually result in a potentially higher release than if containment isolation fails.
It is also stated for other sequences regarding Emergency Operating Procedure (E0P) direct containment flooding that if containment vapor suppression is defeated during the flooding proce s (e.g., wetwell is filled with water) before the RPV is breached at high pressure, then containment failure is assured.
It is further stated for. these sequences that such an induced containment failure is both catastrophic and energetic.
(See, for example, E/H Sequence 8, page 1-16; E/H Sequence 9, page 1-17; and E/H Sequence 10, page 1-17.) How do the NMP2 procedures account for these potential failure modes and provide "near optimum" guidance for accident management?
B.E. - 9 It is stated on page 4.2-1 that MAAP (Modular Accident Analysis Program) Version 3.08, Revision 7.01, is used. A reference to MAAP, Revision 8.0, is given on page 4.2-5.
Were both revisions used? If so, please identify where MAAP, Revision 8.0, is used and the reason for using that version.
B.E. - 10 If available, please provide a summary matrix for Figures 4.6-19 through 4.6-32.
A summary matrix would provide for easier comparisons of the contributions of each of the accident classes and release categories.
B.E. - 11 Consistent with Table A.1 of NUREG-1335, please provide the following information:
volume of the reactor coolant system water, mass of the fuel, and the mass of zirconium.
B.E. - 12 If available, please provide a containment matrix similar to Table 4.6-5 that provides containment failure modes instead of release characterization.
For example, instead of E/H, the matrix would contain conditional probabilities that the containment fails according tn the various relevant failure modes.
B.E. - 13 Please provide a description and diagrams of the wetwell and drywell hardened vent systems, including identification of the location and routing of system components.
Identify those valves (by valve number) used to initiate and terminate emergency venting for both the wetwell and drywell hardened vents.
Provide details of your analysis to determine the operability of those valves under beyond design basis loading conditions and specify the conditional
. l i
failure probabilities (both opening and closing) used in the IPE analysis.
The IPE submittal seems to indicate that the valves in question are the large (20-inch) containment isolation valves in the wetwell and drywell purge exhaust lines, 'i.e.,
butterfly valves.
j It is our understanding that the large butterfly containment isolation valves are time limited in their use during normal plant operation and must be " blocked" to prevent them from being opened more than 60 to 70 degrees. The " blocking" was required in order I
to assure that if the valves are open at the time of.a Design Basis Accident (Loss of Coolant Accident) they can be closed against the.
pressure and flow from the containment pressure transient. The Design Basis Accident evaluation determined the valves operable (i.e., capable of closing) against only the maximum containment pressure and resulting flow predicted by the analysis-during the normal closing time for the valves - several seconds. Those loads may be significantly less severe than the pressure and. flow loads the valves would be demanded to open and close against during the severe accident emergency use functions modeled in the IPE.
Please discuss the impact on CDF and containment releases (both i
magnitude and. timing) if the valves in question are assumed to~
(a) fail to open on demand or (b) fail to close on demand.
F.E. - 1 The IPE analysis only appears to address flooding that deals with submergence, and does not describe (as requested in NUREG-1335) scenarios wherein water intrusion from spray or direct impingement such as' splashing can take place. Provide a discussion of how~ your assessment considered water intrusion and/or the rationale used.for its exclusion.
Include in the discussion, as an example, consideration given for systems with " limited" volume such as the Reactor Building Closed Loop Cooling and Turbine Building Closed Loop Cooling systems which were determined to offer no significant challenge to the operating crew on a submergence basis.
F.E. - 2 The assumption was made that "a procedure that outlines effective mitigating actions is available to the operating crew" even though
" currently NMP2 has no event-oriented procedures for mitigating a flooding scenario." Since the value used for the magnitude of the operator failure probability is significantly dependent on the existence of, and training on, procedures for specific actions, it is expected, as indicated in NUREG-1335, that written procedures will be provided for those operator actions. What is the' impact on CDF if this beneficial assumption is not correct?
F.E. - 3 In the review of initiating events it was noted that partial closure of the Main Steam Isolation Valves (MSIVs) was considered as a transient with the power conversion system available. Other Probabilistic Risk Assessments (PRAs) have considered partial
m I
-5 closure of the MSIVs as a transient with loss of the power conversion system (PCS) (i.e., all MSIVs close).
Please. discuss i
the consideration given to the choice between these two categories j
and provide the rationale for the categorization of this event and i
its impact on the results of your analysis if it could be categorized as an event with loss of PCS.
F.E. - 4 The frequencies of some of the initiating events in Table' 3.1.1-1 were reduced because of credit given for the reduced number of
-l transients per year of commercial' plant operation. This decrease had been based on the more recent operating experience (1984 -
1990) representing the " current operating trend of U.S. BWR reactor i
plants."
Provide a description of how this information was t
combined with the NMP2 operating experience to arrive at the identified lower event frequencies, and discuss how the experience of NMP1 correlates with this more recent industry experience. What effect, if any, does this experience have on.your assessment of frequency for NMP2 events? Please include in your discussion Divisional AC failures and partial loss of offsite power events.
F.E. - 5 The description of top event LA in Section 3.1.2.1 indicates that failure of top event LA results in _a guaranteed failure of top event HA. riowever, according to the' example rule provided in Section 3.1.4, the split fraction for HA should be HAF when LA-F, d
as the split fraction for HB is HBF.when LB F..It is not shown-this way in Fig. 3.1.2.1-3.
This is also the case with PA and other top events where the' guaranteed failure is to exist dependent on previous failures, such as CIF is'on HSF, but-is not'shown in the logic as such.
Provide as discussion to clarify the rules used j
for assignment of split fractions in general. and in this-case specifically.
In addition, please discuss the'use and meaning of the negative sign before the system designations in the logic.
)
F.E. - 6 The station blackout tree indicates that c'redit is taken for i
injection of fire water from the diesel driven pump..However, the general transient tree description indicates that the success j
criteria does not credit fire protection due to uncertainties regarding its capability to provide low pressure injection. This appears to be inconsistent.
Please provide clarification as to why-5 it should or should not be credited in both events.
F.E. - 7 In the station blackout model credit is given for battery operation in the 2 to 8-hour' time frame if the operators shed all non-essential loads. Does the probability of failure of the batteries increase from time frame to time frame due to the decrease.in battery capacity? Please provide a discussion of your -
i assessment of battery' failure rates during these successive time frames.
t We v
~
. r F.E. - 8 Section 3.2.1.25.1 discusses the significance of a reactor recirculation pump (RRP) seal LOCA. This discussion indicates that the impact of a potential RRP seal LOCA is considered in the station blackout analysis which considers how long Reactor Core Isolation Cooling (RCIC) can operate from the condensate storage tanks with a RRP seal LOCA. However, the station blackout event tree (Section 3.1.2.2) does not appear to address this condition.
Provide a discussion of your assessment of this condition in this event.
F.E. - 9 In the assessment of the Heating, Ventilating, and Air Conditioning (HVAC) systems credit is taken for the performance by operators of mitigating actions such as opening doors, using other equipment, bypassing high temperature trips, and opening water tight doors.
Please discuss the implemented procedures and training at NMP2 used to support the argument that HVAC events are not considered likely and therefore are not modelled.
P Some licensees have found that opening doors to rooms on loss of HVAC may not be sufficient duc to room arrangements.
Identify the basis (e.g., calculations, testing, or prior experience) for your assumptions that the actions proposed and the amount of time for operator response would be sufficient.
F.E. - 10 The IPE submittal indicates that on loss of chilled water to the coolers it is possible for the operators to provide service water as a backup to mitigate this failure. However, there is no discussion of the impact of common cause loss of the fans for which there is no in-place backup system. Discuss your assessment of this type of failure including the frequency of such events.
F.E. - 11 The system description for the fire water crossties indicates that the cooling water trip is bypassed when the diesel fire pump is running. What system is the diesel fire pump dependent upon for cooling, and is it available during station blackout? Discuss the impact of loss of this cooling water system on the diesel fire pump.
F.E. - 12 The station blackout event tree description allows operator action (01, 02, and 03) to depressurize the RPV upon RCIC failure.
However, since the safety / relief valves (SRVs) are dependent upon DC power which is modelled in the RCIC top event (UI, U2, and U3),
how is this dependency treated if DC failure is the cause of the RCIC failure? Discuss how this dependency is tracked through the analysis.
P k.
. F.E. - 13 NUREG-1335 requests the use of plant-specific data for important equipment such as ECCS and feedwater pumps. Table 3.3.2-1 in the submittal identifies plant-specific failures by component but uses
" generic" terminology to describe the component, such as " standby centrifugal pump fails to start." This does not differentiate between the pumps for which data is available.
Please describe your plant-specific data collecting process, and its application to systems for which NUREG-1335 requests use of plant-specific data.
H.F. - 1 Consistent with NUREG-1335, please identify the sequences that, but for low human error rates in recovery actions, would have been above the applicable CDF screening criteria.
Please discuss any sequence that drops below the CDF criteria because the frequency has been reduced by an order of magnitude by credit taken for human recovery actions.
Include in this discussion information on the timing and complexity of human actions.
H.F. - 2 Please discuss the degree on involvement of NMP2 personnel in the quantification of the Human Error Probabilities (HEPs) for the IPE.
W
" technology transfer" in this area a goal ano will HMP2 personnel perform the human reliability analysis for any IPE updates?
H.F. - 3 Discuss the formal approach, including the process and structure, used to obtain, document, and incorporate plant-specific data derived from interviews with operating, training, and maintenance personnel and from walkdowns and simulator data.
H.F. - 4 Describe the use of and identify the type of plant-specific performance shaping factors (PSFs) applied to the human interactions (HIs) in each of the methods used (ASEP, THERP, EPRI) to quantify the HEPs.
Please discuss the rationale for the selection of PSFs and any weighing factors applied to the PSFs.
H.F. - 5 Section 3.3.3.4 (Pre-Initiating Event Human Errors) states that
" human induced failure modes affecting single train systems, caused by maintenance related error, are subsumed by equipment unavailability due to test and maintenance activities and equipment failure rates." It is also stated that "this assumption appears to be supported by industry and plant-specific data."
It is not evident that this is true in all cases or that all data bases capture all this information.
Please explain how the data that is used for NMP2 specifically captures conditions that relate to maintenance, test, and calibration errors for systems modelled for NMP2. Has this type of data for Nine Mile Point Nuclear Station Unit No. I been captured over its operating history and applied to the NMP2 analysis? Is it being captured for NMP2 for future IPE updates?
i i
i i
H.F. - 6 Section 3.3.3.4 indicates that human induced common cause errors-for the Standby Liquid Control _ system and ECCS due to miscalibration are considered unlikely. Discuss the rationale and your assessment of miscalibration failures:in comparison with common cause failures identified for these systems in Section.
3.3.4.
r H.F. - 7 Section 3.3.3.2.2 indicates that all the HHSA human action events were quantified the same way, i.e., as a response to a pump trip.
i annunciator. However, as pointed out in the discussion, event' HHSA 4 (control service water flow to prevent loss of. service water), which occurs as a result of loss of one of the offsite sources, is estimated the same-(HEP = 2E-4) as the other service water events even though the annunciators will be indicating different problems. The justification for this is.that the operators are highly conscious of the need for service water and
.i establishing appropriate flow is considered a top' priority action.
This appears to be based on~ engineering judgement'~and it is not i
apparent that the assumed value could be arrived at through the structure of the referenced methodologies considering possible conflicting actions dictated by various alarms, alarm response procedures, and other performance shaping factors, including timing that many have an impact.
Please discuss the rationale for stepping outside the structure of the chosen methodology for the quantification of HEPs and, in this case, without consideration of the other PSFs which could apply.
Are there.any other HEPs that were quantified in a similar manner?
H.F. - 8 The human reliability section of the submittal does not discuss human actions for recovery of.the emergency diesel generators or.
offsite power. Has credit been taken for human recovery actions via procedures and training? If.so, provide a discussion describing how the HEP values for these actions for each time phase f
were derived.
l H.F. - 9 Section 4.6.2.5 (Human Intervention) indicates that an alternate-approach to the EPRI methodology was also used.. Provide a i
description-of this alternate approach and identify the human-actions quantified using this approach. Discuss the basis for the application of the EPRI data from the EPRI Operator Reliability Experiments Program. -Provide a copy of the EPRI documents containing this data.
H.F. - 10 Section 4.6.2.5 also indicates tha't one operator action was 4
quantified using results.from an unidentified PRA for suppression pool cooling initiation. Please identify the PRA used for this Human Intervention and describe the basis for its use.
_g_
r H.F. - 11 Please provide a breakdown of the human actions associated with the Level 11 analysis, and provide a qualitative discussion of the sensitivity of these actions to severe accident mitigation as appropriate.
h B
6 i
L Mr. B. Ralph Sylvia March 9, 1993 This requirement affects one respondent and, therefore, is not subject to Office of Management and Budget review under P. L.96-511.
Sincerely.
Original signed by:
John E. Menning, Project Manager Project Directorate I-l Division of Reactor Projects - 1/II Office of Nuclear Reactor Regulation
Enclosure:
As stated cc w/ enclosure:
See next page Distribution:
. Docket File WMilstead, NLS324 NRC & Local PDRs CAder, NLS324 PDI-1 Reading JFlack, NLS324 SVarga LWheeler, 12/G/18 JCalvo RHernan, 14/C/7 RACapra WBeckner, 10/E/4 CVogan JChung, 12/G/18 JMenning OGC ACRS (10)
CCowgill, RGN-I PDI-1:LA PDI-1:PM.
PDI-1:D k
CVogan 14 JMenning:snun RACapra Eb/i/93 3/ 1/93 3/ 1 93
/ /
/ /
//
/
OFFICIAL RECORD COPY FILENAME: NM274437.LTR