ML050480580

E-mail from Richard Emch Regarding Need for Conference Call on SAMA RAI Responses

ML050480580
Person / Time
Site:	Millstone
Issue date:	09/02/2004
From:	Emch R NRC/NRR/DRIP/RLEP
To:	Annino E, Gallagher R, Watson R Dominion Nuclear Connecticut
References
FOIA/PA-2005-0115
Download: ML050480580 (3)

Text

Richard Emch - Need for Conference Call on SAMA RA! Responses Page 1 From: Richard Emch To: Edward_Annino~dom.com; Rich Gallagher; William_RWatson~dom.com Date: 9/2104 12:12PM

Subject:

Need for Conference Call on SAMA RAI Responses Rich, Ed & Bill, Attached is a list of questions to be used as an agenda for a conference call to discuss your responses to the SAMA RAls. We are proposing to conduct the conference call on Wednesday, September 8.

I am out of the office this week; if you want to talk to me about these questions or to set up the call, call me on my cell phone at 301-526-8716.

Rich Emch CC: John Tappert; JXD10; Leslie Fields; Robert Palla

Richard Emch - Followup~uestionsonMNPSRAlresponsesU2andU3.wpd Page-1 1 Page 1II Richard Emch FollowupQuestionsonMN PSRAiresponsesU2andU3.wpd QUESTIONS ON RESPONSES TO MNPS RAls

1. (U2 & U3) Different revisions of the PRA were used for the identification of SAMAs and the quantification of benefits. The response to RAI 6a lists the highest importance basic events from the PRA used for SAMA identification (Rev. 2 for U2; 10/99 for U3) and the importance of the same basic events from the PRA used for quantification (Rev. 3 for U2; 10/02 for U3). Confirm that the highest importance events from the later PRA are included in the list. If not, identify those basic events and the SAMAs that address those events.

2. (U2) The response to RAI 1b (p. 5) mentions results of a PRA model more recent than the version used for SAMA quantification. Confirm the existence of this update, identify the major changes (models/assumptions and results/risk profile), and discuss any potential impact on the SAMA analysis.

3. (U2) Relative to peer review F&O AS-5 (Item A.2 in Table 2), confirm that manual control of AFW after loss of air or loss of DC is credited in the PRA. (For example, is success in manual AFW control included in the top success branch in the event tree provided in response to RAI 2c?) If so, what is the failure probability and its importance? Is unavailability of indications due to dependency on power considered in determining this HEP? How was the evaluation of SAMA 113 performed in response to RAI 6g (i.e., what events were revised)?

4. (U3) Relative to Level A peer review F&Os SY-4 and HR-1 (Items A.2 and A.3 in Table 2), please provide a more detailed discussion and support for the conclusion that the incorporation of model changes in response to this F&O will have a negligible impact on the SAMA analysis.

5. (U3) Relative to peer review F&O TH-8 (Item B.19 in Table 2), the impact in Table 2 says that the DWST will provide only 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br /> of water for the AFW pumps. Considering the high importance of the AFW system (the AFW is involved in 3 of the top 4 CDF sequences and the turbine driven AFW pump has a FV importance of 0.235), and the potential for a dependancy between operator action to initiate bleed and feed, justify further why the failure to provide alternate sources of water for the AFW after the DWST is emptied has a negligible impact on the SAMA analysis.

6. (U3) The date provided for Rev. 4 is 10/99. The ER states that the WOG peer review took place in 9/99. What version of the PRA was used in the peer review? The ER implies that it was the version used for the SAMA analysis (10/99). Table G.2-1 indicates a 8/99 version, but this is not included in the response to RAI Id. Please clarify.

7. (U2 & U3) The truncation value used has a significant impact on the CDF. Please provide the truncation values used for obtaining the CDFs given for U2 PRA Revisions 0, 1, and 2, and U3 PRA Revisions 0 (12/95), 2, and 3.

8. (U2 & U3) Of all of the PRA changes listed in response to RAI 1d, indicate which ones (1 or 2) were the major contributors to the changes in CDF from one revision to the next.

1

Richard Emch - FollowupQuestionsonMNPSRAiresponsesU2andU3.wpd Page 21

9. (U2) What is meant by the last sentence in the description of the Rev. 0 and Rev. 1 PRAs?

10. (U2) Describe the sequences identified as COOL in the response to RAI 1e.

11. (U3) Regarding the 10/02 revision of the PRA, the response to RAI le indicates a total CDF (excluding internal flooding) of 2.57E-5 with a truncation value of 1E-1 1. The response to RAI 1d gives a value of 2.04E-5 with a truncation value of 1E-9. The ER provides a value of 2.88E-5 and states that a truncation value of 1E-1 1 was used.

Please explain.

12. (U2) The second paragraph of the response to RAI 1h states "The Level 2 portion of the IPE PRA for Millstone Unit 2 has not been updated but there has been some modifications of the individual bin definitions for consistency between the Unit 2 and Unit 3 PRAs." However, page E-F-23 of the ER states 'Recent experimental results have shown that certain outcomes on the containment event tree are much less likely than previously thought. These changes were incorporated into the Level 2 model." These statements appear inconsistent. Please clarify and describe in more detail what was done.

13. (U2) An example of how the RC and PDS conversions were made, and how Table F.2-4 was generated would help explain some remaining confusion regarding the conversion process. Take new RC M6, for example. According to Table 1h-3 in the RAI responses, RC M6 is composed of IPE RCs E-LM-R and E-MH-R. In the IPE (Table 4.9-5 of the IPE), TLCH contributes 0.04% and 37.7% to these two RCs. However, in the revised PRA (Table F.2-4 of the ER) TLCH contributes 73.8% of RC M6. It is noted that a number of IPE PDSs are not Included in Tables 1h-1 and F.2-4 (for example, TEHA, TEHB, and TEHC, which are the dominant PDS in the IPE. Where are they assigned and is this the source of the difference noted above?

14. (U3) Given that the original Table G.2-4 is incorrect (according to the response to RAI 2.c) and results in incorrect (but high) frequencies in several release categories, is the increase in CDF used in the cost benefit analysis also in error?

2