ML20036C243
| ML20036C243 | |
| Person / Time | |
|---|---|
| Site: | Catawba  |
| Issue date: | 06/07/1993 |
| From: | Rehn D DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| TAC-M74394, TAC-M75395, NUDOCS 9306150372 | |
| Download: ML20036C243 (24) | |
Text
II I
s Duke Itwr Corapany (803)S31J000
' Cata :ho Nudear Station 3
4koo Cuncord had
)brk 'SC24?45
=
DUKEPOWER June 7,1993 U.S. Nuclear Regulatory Commission ATTN: Document Centrol Desk Washington, D.C. 20555
Subject:
Catawba Nuclear Station, Units 1 and 2 Docket Nos. 50-413 and 50-414 Review of Catawba Individual Plant Examination Submittal Response to NRC Questions
_j (TAC Nos. M74394 and M75395) i
Reference:
Ixtter from NRC to M.S. Tuckman, dated April 22, 1993 l
Gentlemen:
In reply to the reference letter, please find the attached response to the questions contained in j
. the NRC's request for additional information concerning.the Individual Plant Examination (IPE)
{
submittal for Catawba Nuclear Station.
l If you have any questions, please call L.J. Rudy at (803) 831-3084.
j Very truly yours, I
i j
0 V[ '
/
D.L. Rehn, Vice Pmsident j
Catawba Site U R/s Attachment 1400.!D oil 0306150372 930607 V'
L[\\
ADOCK 05000413 %jj
-PDR PDR g
.p
(
L'
~d,
D6cumerit Control Desk Page 2 June 7,1993 xc (W/ Attach):
S.D. Elmeter, Regional Administrator Region II 1
3 R.J. Freudenberger, Senior Resident Inspector R.E. Martin ONRR D
f l
i J
i I
J l
i
9 I
i I
i I
5 Duke Power Company j
Catawba IPE -
. Response to NRC Questions j
1 s
June 1993 f
i t
i b
f f
I 1
-)
,v I
Catawba IPE i
NRC Questions and Duke Responses i
Ouestion 1 Given that the design and operation for the Catawba facility are essentially the same as that of the McGuire facility, provide any major design and operational differences (along with the differences in l
major assumptions) that are found to impact the frequency of major functional sequences for the above facilities (e.g., TBU sequences).
j Resoonse The major differences in the system and analysis aspects which affect the important sequence frequencies are presented in the following.
Nuclear Service Water /Comoonent Coolinn Systems t
At McGuire, the Nuclear Service Water (RN) System cools the Component Cooling (KC)
System and most of the safety related motor driven pumps (ECCS and auxiliary feedwater pumps). The KC System cools the Reactor Coolant (NC) pump thermal barrier. In Catawba the KC System, which is cooled by the RN System, cools the NC pump thermal barrier and the ECCS and motor driven auxiliary feedwater pumps. Therefore, accident sequences involving loss of NC pump seal cooling and ECCS failure can occur with either the loss of RN System or loss of the KC System at Catawba, while for McGuire only the loss of RN event is ofsignificance.
Although the RN Systems between units are interconnected at both plants, the cross connect is kept isolated at McGuire and open at Catawba.
L For McGuire, the Containment Ventilation (RV) System can be used to support critical RN loads in the event of a loss of RN in one unit.
For both plants, the Safe Shutdown System (SSF) provides an altemate system to proside NC pump seal cooling. In the Catawba analysis, the time available for SSF activation for loss of RN and loss of KC events was determined to be on the order of 50 minutes, while the McGuire analysis assumed a 30 minute time interval. Therefore, the operator action for the i
SSF activation was assessed to be more reliable in the Catawba analysis versus the McGuire analysis. The net effect of these differences is that the core damage frequency from loss of l
RN/KC events is estimated to be 2.9 E-5/ year for Catawba and approximately 1 E-5/ year for McGuire.
Loss-of-Offsite Power Both plants incorporate two station shared auxiliary transformers which can feed the sital AC
[
power loads in one unit from the other unit. In the Catawba analysis, the loss +f-offsite power (LOOP) frequency was determined by considering the industry data invohing loss-of-offsite-power to two-unit sites and loss-of-offsite-power to one unit and considering the availability of power from the unit through the shared transformer. In the McGuire analysis, the LOOP 1
.O.
1 initiator was not distinguished between the 1-unit LOOP and the 2-unit LOOP. This refinement in the initiator treatment resulted in the LOOP initiated sequence to be 1.2 E-6/ year at Catawba versus 1.1 E-5/ year at McGuire.
The differences in the loss of RN/KC sequence and LOOP sequences largely account for the difference in the TQsU functional sequence (3.3 E-5/yr at Catawba versus 1.8 E-5/ year at l
McGuire).
I Recoverv of Main Feedwater The main feedwater (CF) system would be available following a plant trip not caused by a loss of main feedwater. The recovery probability of main feedwater is estimated based on plant operational data. This recovery probability was estimated to be 0.95 for Catawba and 0.8 for McGmre.
This difTerence is an additional factor influencing the difference in the TBU functional sequence, which involves events such as the loss of CF event, loss of RN/KC events, and loop events.
6900/4160-V Auxiliarv Power Transformer Ancther difference in the two plants is the location of the 6900/4160-volt transformers. At Catawita, these transformers are located in the basement of the turbine building. This makes them susxptible to large turbine building flooding events. At McGuire, these transformers are located outside the turbine building. Although flooding of these transformers does not lead l
directly to a core damage accident, it would result in a failure of offsite power for an extended i
period of time. This difference accounts for the TB flooding sequence at Catawba and not at McGuire.
I FWST Level Transmitters
{
l l
The refueling water storage tank (FWST) at McGuire utilizes a 3-channel level instmment system, while the Catawba configuration is a 4-channel system. The Catawba FWST level instrument system is assessed to be less susceptible to common-mode failure. This difference is responsible for the difference in the sump recirculation failure for the LOCA sequences, and accompanying functional sequences.
Question 2 Provide the following information related to the Catawba plant walk downs performed as part of the IPE:
A.
Types of walk downs performed B.
Objective and scope of c:xh walk down C.
A brief discussion of the process used to integrate findings into the plant model (for further modeling or for deciding plant fixes) 2
i u.
R_ esponse There were three types of walk downs performed during the Catawba IPE study. The first type was conducted during the system familiarization phase of the project. These walk downs were carried out by the system analysts, with assistance from personnel from the plant performance and -
engineering groups. The purpose of this walk down was to gain an over all idea of the configuration of the system and how it was operated.
The second type of walk down was to gain insights on specific questions which arose during the fault tree development and solution phase of the project. These types of reviews were used to determine information such as the accessibility of equipment, the potential for flooding, and the potential for recovery actions. Findings from these walk downs were incorporated into the draft versions of the PRA models.
The third type of walk down was conducted during the independent review phase of the project.
These walk downs were used to resolve questions raised by station personnel concerning modeling assumptions. Any findings of this review were incorporated into the final plant models. Also, potential plant enhancements were sometimes explored during these walk downs.
Ouestion 3 Provide additional infonnation related to the peer review process conducted for the Catawba IPE:
A.
Briefly discuss the personnel involved in the peer resiew B.
Activities conducted (e.g., areas of spot checks and audit calculations)
C.
Tools used (e g., procedures and checklists)
D.
A summary of major findings made during the peer review actisity and a dispositioning of these findings
Response
The PRA was reviewed by independent reviewers from the Catawba site. The review began with kickoff meetings with the individual teams. During these meetings, the objectives of the resiew were discussed along with some PRA basics and a brief discussion of the section of the PRA. Each review team member was also provided a copy of the PRA section to be reviewed. After the individual team members had reviewed the PRA section, the team met again to discuss comments
{
and provide feedback. Comments were generally provided as marked up copies of the PRA and IPE reports, or as notes taken during the meetings. The comments from the independent reviewers were resolved by the individual analyst responsible from the section being resiewed.
The reviewers utilized their knowledge of the present operating plant to provide feedback conecming the information provided in the PRA reports related to their area of expertise. The tools used by the review teams varied, depending on the section being reviewed. For example, the operations and operations training personnel resiewed the human factors sections and provided feedback on the timing of the required actions, the training and knowledge of the crews, and the clarity of the procedures for the action considered. System models were reviewed by the " expert team" assigned to that system. This " expert team" consisted of plant personnel from the performance, operations, 3
.i
V maintenance and engineering groups. These teams relied on drawings, maintenance procedures, surveillance requirements, and their general knowledge of the plant.
Presentations were also made to the Catawba station management personnel concerning the results, j
conclusions and proposed plant enhancements. His review and dialogue facilitated the formulation and endorsements of plant enhancements discussed in Section 6 of the IPE Submittal Report.
Appendix A of this response includes the personnel who reviewed each section of the Catawba PRA l
and a brief discussion of the most significant comments.
Ouestion 4 1
- Provide the version and the date of the CAFTA code used for the Catawba IPE/PRA. Is the same i
version of the code used for the Oconee and McGuire facilities? If not, please discuss any major differences (among versions) which could affect the quantification of core damage frequency. Your discussion should include the modeling of train level dependencies (for example, treatment of circular logic loops).
Response
The version of CAFTA used for the Catawba, McGuire and Oconee PRA studies was version 2.0d, April 1989. This version of CAFTA identifies any circular logic within the fault trees so that circular logic can be corrected prior to the solving process. For example, loss of offsite power sequences at Catawba create circular logic because the Emergency Diesel Generator System requires the Nuclear Senice Water System for cooling, and the Nuclear Senice Water System requires the Diesci for power. His circular logic was corrected by deleting the calls from the Diesel Generator tree to the Nuclear Service Water System tree.
Ouestion 5 The staff notes that one of the four actions taken as part of the IPE activity includes improvements to the Catawba training simulator. Please concisely discuss these improvements and the extent to which they refocused training.
Response
The enhancement related to the simulator concerned improvements in training performed on the simulator, not actual improvements to the simulator. This was an enhancement from the original Catawba PRA performed earlier, and was related to the periodic training that operators receive on the simulator. Most of the PRA accidents have always been covered well by the operator trammg program.1lowever, the PRA study did identify several specific sequences which were added as exercise guides to the bank of simulator training scenarios. The exercise guides preside important infonnation for the simulator instmetors, such as the objectives of the scenario, the initial conditions,-
the failures to be modeled, and the expected operator response. PRA team members worked with the simulator instructors to develop these guides. The PRA sequences added included the following sequences; 4
i v
4 Loss of Main Feedwater Excessive Main Feedwater Interfacing Systems LOCA l
Failure of ESFAS to Automatically Actuate Five Small Break LOCA Sequences -
Two SGTR Sequences j
Turbine Building Flood Loss of the Nuclear Service Water System Loss of the Component Cooling System 9
f Ouestion 6 DPC has decided that certain plant improvements considered as part of the turbine flood protection are not cost effective. Describe the analysis performed, the decision process, and the results related to disposition of this issue.
Resoonse As the result of the original Catawba PRA performed earlier, a design study (CNDS-0076) was performed to evaluate the feasibility of several methods to address the turbine building flood concern. The study reviewed the following options; moving all the power equipment to a location not susceptible to flooding,
[
providing a flood wall around the equipment, 3
relocating or changing the operator type for the isolation valves, increasing the sump pump capacity.
The design study concluded that the only marginally cost beneficial fix was st 4.5 foot wall around the 6900/4160-volt transformers. Further review of the cost estimate for thit modification revealed that it did not consider the detrimental effects the wall would have on the nornuti operation of the i
plant. It was also determined that the proposed wall did not protect all the vuherable equipment in
[
the turbine building basement. Therefore, it was concluded that there was no hardware related enhancement that could be justified through cost benefit analysis. Instead, it was decided to put l
additional emphasis on improving the likelihood of the recovery actions, i.e., utilizing the SSF as a
~:
backup for RCP seal cooling and secondary side heat removal.
Ouestion 7 Appendix G to the McGuire IPE indicates that the containment failure pressure distribution i
corresponds to a log normally distributed probability function with a median pressure value of 77 l
psig and a mean pressure value of 76.71 psig. Although the information presented in Appendix G of the Catawba IPE seems to indicate that the' Catawba containment structure is identical to McGuire in shcIl thickness (3/4") and both structures are subject to the same analytical process, the Catawba i
structure is stronger (i.e., the containment failure pressure distribution corresponds to a log normally l
distributed probability function with a median pressure value of 84.5 psig and a mean value of 83.93 s
i 5
}
i V
psig). Please identify and discuss those differences between the two containment structures that result in the increased containment strength for Catawba.
Remonse ne difference in the incan failure pressures is due to the different grades of steel used in the construction of the containments. Statistical averages from actual material properties were u:. d in the structural analyses.
McGuire is ASME SA-516 grade 60 and Catawba is SA-516 grade 70 steel.
l Ouestion 8 The Catawba analysis indicates that the conditional probability of early containment failure (0.005) is significantly reduced in comparison to the early containment probability (.020) for McGuire.
Please identify and discuss those fac:crs which allow for a significant reduction in the probability of l
carly containment failure for Catawba. The discussion should also address the reasons for a slightly increased probability oflate containment failure for Catawba.
Resoonse Early containment failures are dominated by failures due to the combustion of hydrogen. Sequences that contribute to failures from hydrogen combustion are those in which the igniters are unavailable during the period of core degradation, typically those with no ac power. The frequency of the loss of offsite power initiator is significantly lower in the Catawba PRA than in the McGuire PRA. In the Catawba LOOP analysis, explicit consideration of 2 unit versus I unit LOOP events resulted in a lower frequency than in the McGuire analysis, where such a refinement was not made. Since these sequences usually proceed to core melt only if the onsite emergency power fails, this is an important initiator for plant damage states in which the igniters are not available. This results in a lower proportion of the core melt frequency in plant damage states with no igniters, which in turn leads to a lower conditional probability of early containment failure at Catawba. The difference in the conditional probability of early containment failure is due to the frequencies of the contributing PDSs and not due to any difference in containment performance. Given all the considerations that go into the analyses, the McGuire and Catawba results are not viewed as significantly different.
Late containment failures are dominated by a long term overpressurization from steam generation in the absence of containment spray. Plant damage state (PDS) 19DI is the dominant contributor to the l
late containment failure frequency for Catawba and it's frequency is a larger fraction of the total core melt frequency than was the case for the dominant late containment failure sequences for McGuire.
The slight difference in the conditional probability oflate containment failure is due to the frequencies of the contributing PDSs and not due to any difference in containment performance. The f
table below summarizes some of this PDS information.
% OF CMF -
% OF LCF FREQ.
i CNS-19DI 37 90 I
MNS-22Cl 28 47 MNS-7PI 23 26 6
1 l
.b Question 9 The probability of bypass failure of the containment was determined to be 0.024 at McGuire, and it wa= indicated that bypass failure was dominated by induced steam generator tube ruptures (ISGTR)
I 1
(over 90% of the bypass frequency is due to ISGTR). The Catawba IPE indicates the probability of bypass failure is 0.002 and that it is dominated by ISLOCA. It is our understanding that the steam generators at both McGuire Unit 1 (the unit analyzed for the McGuire IPE) and Catawba Unit 1 (the unit analyzed for the Catawba IPE) are scheduled for replacement in the near future, because of severe tube cracking. Please identify and discuss the differences between McGuire and Catawba which provide such widely different insights into the containment bypass failure mode and the significance of these differences.
Resnonse The ISLOCA and induced tube ruptures contribute to the containment bypass frequency. The induced tube rupture analyses are driving the hight conditional probability in the McGuire results, and these considerations are addressed here.
There are several factors which contribute to the difference in the induced tube rupture contribution to the containment bypass frequency results between McGuire and Catawt t The first is a difTerence in the CET quantification for determining the occurrence ofinduced tube ruptures. The i
CET quantification foi McGuire assumed a high likelihood,0.99, that the operators would start the reactor coolant pumps as instructed by the nadequate core cooling emergency operating procedure.
i This value is contained in PRA Section 6.2, quantification of the basic event NCONBYOPS. This quantification leads to a high probability of having an induced tube rupture for those plant damage states (PDSs) in which power is available to operate the pumps and secondary side heat removal is not available. We concluded as a result of the McGuire IPE, that operation of the reactor. oolant pumps during core degradation with a dry steam generator had a negative influence on plant risk.
He emergency operating procedures are being rewritten at McGuire and Catawba. The revised procedures will not call for reactor coolant pump operation if core temperatures are elevated and the steam generators are dry This change was prompted by the McGuire a t. analysis results. In anticipation of this change, we analyzed Catawba assuming that the reactor coolant pumps would not be operated under these conditions. With this assumption, the quantification for event NCONBYOPS became 0.001. Without forced circulation due to reactor coolant pump operation, the potential for an induced tube rupture is greatly reduced. This quantification change is the dominant reason for the observed difference in the two results.
Another importent influence is the frequency of PDSs which contribute to induced tube ruptures due to forced circulation. In the McGuire PRA, PDS 14DI, auxiliary feedwater pump room flood, resulted in a high pressure core melt with power available to the reactor coolant pumps and no secondary side heat removal. He relath cly high frequency made this a dominant contributor, 81%,
to the internal induced tube rupture frequency in the IPE. There is no similarly important PDS in the Catawba results. If the McGuire CET is quantified as in the Catawba analysis, PDS 14DI and others like it would play a much less important role in the induced tube rupture analysis and significantly reduce the induced tube rupture frequency.
7 1
1 I
^
5 I
. As in the early containment failure case, loss of offsite power (LOOP) initiators also are important -
contributors to the induced tube mpture results. In these sequences transportation of the hot gases generated in the core to the steam generator tubes occurs due to natural circulation. The CET i
quantification is the same for the McGuire and Catawba analyses for the natural circulation case.
liowever, the higher LOOP initiator frequency does contribute to a higher frequency of containment j
bypass.
These factors which increased the induced tube rupture frequency in the McGuire analysis caused j
the containment bypass frequency to be higher for McGuire even though the ISLOCA frequency is about an order of magnitude higher at Catawba.
3 Ouestion 10 i
In response to our request for additional information on the McGuire IPE, DPC indicated that a strategy for restoring hydrogen igniters in small groups, following loss of critical AC power events, i
is being considered for implementation in the accident management guidance. Please indicate whether this strategy is also being considered for Catawba, and discuss the reasoning for your action with regard [to) this potential strategy.
i Resnonse I
The general strategy of restoring pow et to the igniters in small groups following loss of all ac power events is equally applicable to both plants. The implementation could vary at the plants to the extent that the strategy will be very dependent on exactly which igniters are electrically connected to the same breaker. When energizing a particular circuit, an igniters on that circuit will come on. If too j
broad an area is covered by a group, the burns may not be localized.
i Conceptually, the strategy is to burn the hydrogen through a series of small deflagrations, e.g. one compartment at a time. The pressure response due to the individual bums would be mild compared to any event which consumed all of the contamment hydrogen at once.
No strategy has yet been developed. It.is not clear the.'+cssary analytical tools exist to allow a
A code such as HECTR could be j
development of the strategy with a high degree of core used, but HECTR does have limitations which would
- a ne degree the confidence that the strategy would work as intended. We do plan to perform saue siECTR analyses to evaluate the possibility of develcping an effective strategy.
)
8
9
. ;+.
f i
i Duke Power Company 1
Catawba IPE Response to NRC Questions Appendix A
. i June.1993 8
?
h h
5 e
<r r,
Summary of Catawba IPE Peer Review l
~
Review Subicet Egrsonnel Involved in Maior Comments Resiew General Oveniew M. S. Tuckman This management review group reviewed the of Results and Station Vice President results, conclusions and recommendations of Recommendations the IPE study. Their comments generally (IPE Submittal E. M. Geddie involved questions concerning the methods and Report, and Section Superintendent of assumptions, recommendations for who to 8 of the PRA)
Operations contact for further information, and belpful suggestions for proposed enhancements.
R. C. Futrell Compliance Manager f
J. S. Forbes Engineering Manager T. E. Crawford Systems Engineering Manager W. R. McCollum Station Manager J. D. Wylie Training Manager C. W. Boyd Mechanical / Nuclear Enginecr%g Manager E. W. Fritz Systems Engineering Supenisor D. L. Ward Mechanical Engineering Supervisor S. W. Brown Mechanical Engineering Supenisor S. R. Frye Operations Support Manager r
T. P. Harrall Safety Assurance Manager A-1
l 1
Appndix A j
Summary of Catawba IPE Peer Review i
Review Subject Personnel Involved in Maior Comments Review j
Floods (Section 3.3 J. J. Mackay The draft PRA indicated that after ten minutes, of the PRA )
Civil Engineering the valves necessary to isolate the flood would be submerged. However, comments from K. L. Evans station personnel indicated that the Mechanical / Nuclear transformers would already be submerged Engineering regardless of the isolation attempts. As a result, all references to a ten minute period was K. R. Caraway deleted.
Electrical Engineering Supenisor Station personnel also determined that some important equiptuat was outside the proposed J. E. Herrington flood wall. This made the wall a less desirable Electrical Engineering proposal for addressing the Turbine Building flood concern.
P 3'
A-2
Appendix A Summary of Catawba IPE Peer Review Review Subiert Personnel Involved in Maior Comments Re3 iew Human Reliability J. E. Teofilak Several comments from the independent (Section 5 of the Operations Procedures reviewers concerned the time required to PRA) accomplish operator actions or the likelihood G. B. Ice oflatent human errors. An example of this Operations Training type comment concerned the time assumed in the PP,A analysis for the operators to manually S. R. Frye scram the reactor during an ATWS event. The Operations Support PRA had assumed a response time of 5 to 10 Manager seconds. The reviewers felt that more time would be needed for the operators to respond.
This comment resulted in a small increase in the ATWS initiated core damage frequency.
Another comment concerned the PRA assumption that a latent human error resulting in a containment isolation failure would be discovered within a short time because of the reduced need for air releases to control normal containment pressure. The reviewers felt that i
this error would be discovered in this manner but it might require several days. This change i
resulted in a significant increase in the containment isolation failure probability for sequences which have AC power available.
i 1
i I
A-3
App:ndix A Summary of Catawba IPE Peer Review Review Subiert Personnel Involved in Mr.ior Comments Reuew ECCS (Appendices J. M. Sawyer A major comment concerned the length of A.1, A.2, A.3 of the Safety Analysis time that the charging pumps could run PRA) without forced cooling to the pumps. The R. Menichelli PRA had assumed that the charging pumps Mechanical / Nuclear would fail fifteen minutes following a failure Engineering of the Component Cooling Water System and result in a loss of reactor coolant pump seal J. A. Kammer cooling and eventually a seal LOCA. Plant Systems Engineering personnel were aware of an incident which occurred prior to plant startup where cooling J. G. Torre was lost to the charging pumps for more than Systems Engineering 30 minutes but resulted in no damage to the charging pumps. This change in the failure time of the charging pumps allows more time for the operators to go to the SSF and start reactor coolant pump seal cooling using the Standby Makeup Pump. This significantly improves the likelihood that the operators will successfully perform this action to prevent a reactor coolant pump seal LOCA following both the failure of component cooling (TIO),
and failure of nuclear senice water (T9).
l l
l l
l A-4
l l
O.
Appendix A Summary of Catawba IPE Peer Review j
i Review Subiert Personnel Involved in Maior Comments j
Review
]
CA System W. R. Tomlinson Comments indicated that failure of the lube oil (Appendix A.5 of Component Engineering cooler would not result in failure of the CA the PRA) turbine driven pump. The model was changed M. L. Edmunds to reflect this comment.
Component Engineering Another comment indicated the internals of I
G. F. Purvis valve 1CA129 had been removed. The failure Systems Engineering mode of this valve transferring closed was removed from the fault tree and deleted from P. W. Barrett the list of cutsets.
Mechanical / Nuclear Engineering B. M. Graves Operations H. J. Nicholson Nuclear Services 1
B B
9 t
A-5
~
^PNUSX ^
Summary of Catawba IPE Peer Review Review S.nbitIl EfJconnel Involved in Maior Comments Reuew RN System B. G. Felker Several comments concemed normal vah e (Appendix A.6 of System Engineering configurations, or operation of valves during the PRA) accidents. These comments resulted in several J. F. McKeown changes to the RN system fault tree. However, System Engineering they had a minor impact of the results of the fault tree solution.
H. D. Mason Component Engineeri:g D. B. Thompson Operations K. L. Bishop Component Engineering L. J. Benjamin Operations R. C. Bucy, Mechanical Engineering S. W. Brown, Mechanical Engineering Supersisor A-6
i 8
Appendix A Summary of Catawba IPE Peer Review Review Subiggi Personnel Involved in Maior Comments Review i
1 KC System M. J. LaForrest One of the most significant was the PRA (Appendix A.7 of Component Engineering assumption that two Component Cooling the PRA)
Water System pumps were required to prevent H. J. Nicholson a Loss of Component Cooling Water initiated Nuclear Senices event (TIO). Station personnel actually tested the pumps in this configuration and C. M. Sahms demonstrated that a single pump could proside System Engineering sufficient flow to cool the plant loads. This change in success criteria had a significant B. S. Dycus affect on the T10 initiator frequency.
Component Engineering D. B. Thompson Operations R. F. Flowers Chemistry W. B. Hallman, Mechanical Engineering R. C. Bucy, Mechanical Engineering A-7
0 Appendix A i
Summary of Catawba IPE Peer Review Review Subiert Personnel Involved in Wior Comments Review VI System M. G. Osteen A significant comment concemed the success (Appendix A.13 of Mechanical / Nuclear criteria for the Instrumer.t Air System. The l
the PRA)
Engineering PRA had assumed the configuration and air demand described in the system description.
3 D. C. Ashburn Plant personnel responsible for the system -
Electrical Engineering indicated that leaks in the system had increased the air demand to the point that an additional H. J. Nicholson compressor was required during a significant Nuclear Services part of the year. Although the failure of instrument air is still not a significant T. E. Gaye contributor to risk, this change in success Operations criteria resulted in an order of magnitude increase in the T12 initiator frequency.
C. B. Cauthen Component Engineering W. C. Adams I
Mechanical / Nuclear
[
Engineering V. D King Systems Engineering J. R.Wallace instrumentation &
Electrical i
f b
I A-8 L
6 Appendix A Summary of Catawba IPE Peer Review Review Subject Personnel Involved in Maior Comments Review Safe Shutdown E. W. Fritz, The SSF system team expressed concern over System (Appendix System Engineering the treatment of some manual valves in the A.18 of the PRA)
Supervisor SSF system. Kerotest manual plug type valves have experienced very few failures at D. Davies, Catawba. A scarch of the INPO NPRDS data System Engineering base did not identify any failures of the type described in the fault tree model. Therefore, F. Poley, the probability of failure of these valves was Electrical Engineering decreased by a factor of ten below the typical generic failure rate.
J. Kammer, System Engineering HVAC (Appendix S. B. Putnam The reviewers express concern over the YM A.19 of the PRA)
System Engineering system unavailability values. They believed that the operators would have significant time J. A. Kammer to correct YM related problems. The System Engineering unavailability value for YM was decreased to address this concem.
A-9 1
~
Appendix A
.6 Summary of CatawliaIPE Peer Review Review Subiert Personnel Involved in Maior Comments Resiew Transient, LOCA, G. B. Swindlehurst No Significant Comments SGTR, and ATWS Safety Analysis Analysis (Sections Supenisor 2.2,2.3,2.4 and 2.5 of the PRA )
ISLOCA (Section J. E. Teofilak No Significant Comments 2.6 of the PRA )
Operations Procedures G. B. Ice Operanons Training Seismic (Section D. R. Kulla No Significant Comments 3.2 of the PRA )
Civil Engineering Supenisor W. B. Shoemaker Civil Engineering Tornado (Section D. R. Kulla No Significant Comments 3.4 of the PRA )
Civil Engineering Supenisor NC Pressure Jesse Ray No Significant Comments Control System Mechanical / Nuclear (Appendix A.4 of Engineering the PRA)
NS System W. Brown No Significant Comments (Appendix A.8 of System Engineering the PRA)
Containment G. B. Ice No Significant Comments isolation System Operations Training (Appendix A.9 of the PRA)
Power Conversion P. W. Hallman No Significant Comments Systems (Appendix Electrical Engineering A.10 of the PRA)
T. E. Cook Component Engineering A-10
Appendix A e
Summary of Catawba IPE Pect Review 1
Review Subject Personnel Involved in Maior Comments Review Vital I&C Power R. L. Herring No Significant Comments Systems (Appendix System Engineering A.ll of the PRA)
Containment Air S. S. Davidson No Significant Comments Return and I&E Hydrogen Skimmer System (Appendix J. M. Yandle A.14 of the PRA)
Mechanical Engineering Hydrogen L. R. Wilson No Significant Comments Mitigation System Electrical Engineering (Appendix A.15 of the PRA)
Essential Aux.
R. L. licrring No Significant Comments Power System Systems Engineering (Appendix A.16 of the PRA)
Diesel Generator R. L Herrin; No Significant Comments and Load Sequencer Systems Engineering (Appendix A.17 of the PRA)
D. L. Sweat Electrical Engineering F. C. Poley Electrical Engineering R. A. Kaylcr Systems Engineering W. W. Gallman Nuclear Services W. D. Green Component Engineering E J. Haack Op: rations A-11
Appendix A e e Summary of Catawha IPE Peer Review Review Subiert Personnel Involved in Maior Comments Review Reactor Protection J. C. McCart No Significant Comments and ESFAS Electrical Engineering (Appendices A.20 i
and A.12 of the PRA)
Y t
i i
A-12
)
-