ML23209A767

From kanterella
Jump to navigation Jump to search

Share File Observations on Response to Question 06 and 22 TSTF-505 and 50.69 (E-mail Dated 6/13/2023) (EPIDs L-2022-LLA-0185 and L-2022-LLA-0186)
ML23209A767
Person / Time
Site: Nine Mile Point Constellation icon.png
Issue date: 06/13/2023
From: Richard Guzman
NRC/NRR/DORL/LPL1
To: Reynolds R
Constellation Energy Generation
References
EPID L-2022-LLA-0185, EPID L-2022-LLA-0186
Download: ML23209A767 (1)


Text

From:

Richard Guzman To:

Reynolds, Ronnie J:(Constellation Nuclear)

Subject:

FW: SHARE file Observations on Response to Question 06 and 22.pdf Date:

Tuesday, June 13, 2023 12:15:00 PM Attachments:

SHARE file Observations on Response to Question 06.pdf Ron - see notes attached per the high-level observation discussion from Todd Hilsmeier.

Thanks, Rich

Staff Observations on Response to Questions 06 & 22 (FLEX credit)

[6.a, 22.a - describe and justify approach]

Response did not justify NMP1s approach for FLEX failure rates (i.e., 2x permanently installed failure rates):

Response did not provide a basis for the 2x factor (value seems arbitrary)

PWROG-18042 (where FLEX failure rates are based on industry experience and vetted by industry and approved by NRC) suggests the 2x factor is too low.

Response primarily focused on questioning PWROG-18042 (i.e., its too conservative), which doesnt validate NMP1s approach.

Response seems ok with FTS failure rates in PWROG-18042, and these FTS failure rates are 10-16x larger than the permanently installed SSCs. This suggests the 2x factor used by licensee is too low.

NMP1s approach (or justification of approach) does not consider available generic or plant-specific FLEX data.

Doesn't meet Conclusion 6 of NRC Memo dated May 6, 2022 (ML22014A084) which states, "should not use failure rates for permanently install equipment."

Response states that DG FTR probability in PWROG-18042 is biased to failing within the first hour (i.e., short FLEX run times). However, from PWROG-18042, early DG FTR events (before reaching stable conditions) are counted as FTS.

[6.c, 22.c - provide updated uncertainty analysis]

Response states FLEX failure probability uncertainty has a minor impact on RICT results (i.e., not a key source of uncertainty).

However, for LCOs 3.6.3.C and 3.6.3.H in Table 6c-3, the impact of FLEX failure probability uncertainty is significant (i.e., impacts RICT by 50% or more).

FLEX failure probability is a key source of uncertainty.

Response mentions one RMA to address FLEX uncertainty (i.e., dedicating the N+1 FLEX generator during AC or DC RICT). It is unclear what this RMA represents:

- Describe RMA in more detail?

- How does this RMA reduce risk (i.e., how credited in PRA sensitivity study)?

- Is there sufficient time to credit a spare FLEX DG?

- Since FLEX uncertainties primarily impact fire CDF sequences, what fire-specific RMAs can be implemented?