ML110200594
| ML110200594 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 01/20/2011 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| Download: ML110200594 (5) | |
Text
1 DiabloCanyonNPEm Resource From:
Stuyvenberg, Andrew Sent:
Thursday, January 20, 2011 3:24 PM To:
'Grebel, Terence' Cc:
'Tan, Miranda'
Subject:
Jan. 4 Telecon Summary for Review Attachments:
Summary of 1-4 DCPP SAMA Call for Review.DOCX Terry - Attached is our draft of the January 4 SAMA teleconference for your review. Please contact me by COB on Monday, January 24, if you have any concerns with it.
Best regards, Drew DrewStuyvenberg U.S.NuclearRegulatoryCommission 3014154006 Andrew.Stuyvenberg@nrc.gov
Hearing Identifier:
DiabloCanyon_LicenseRenewal_NonPublic Email Number:
2436 Mail Envelope Properties (AF843158D8D87443918BD3AA953ABF781C89D41001)
Subject:
Jan. 4 Telecon Summary for Review Sent Date:
1/20/2011 3:23:59 PM Received Date:
1/20/2011 3:23:00 PM From:
Stuyvenberg, Andrew Created By:
Andrew.Stuyvenberg@nrc.gov Recipients:
"'Tan, Miranda'" <M1TF@pge.com>
Tracking Status: None
"'Grebel, Terence'" <TLG1@pge.com>
Tracking Status: None Post Office:
HQCLSTR02.nrc.gov Files Size Date & Time MESSAGE 390 1/20/2011 3:23:00 PM Summary of 1-4 DCPP SAMA Call for Review.DOCX 21091 Options Priority:
Standard Return Notification:
No Reply Requested:
No Sensitivity:
Normal Expiration Date:
Recipients Received:
DRAFT - FOR STAFF REVIEW ONLY TELEPHONE CONFERENCE CALL DIABLO CANYON NUCLEAR POWER PLANT UNITS 1 AND 2 SAMA ANALYSIS FOR PROPOSED LICENSE RENEWAL
SUMMARY
OF TELEPHONE CONFERENCE CALL JANUARY 4, 2011 1:00pm EDT NRC staff received PG&Es response to follow-up RAIs on December 8, 2010 (letter dated December 6). After reviewing PG&Es responses, NRC staff requested a conference call to confirm the staffs understanding of the responses before moving forward on the staffs SAMA review. NRC staff provided six informational requests to PG&E via e-mail on December 16, 2010.
The teleconference focused on the December 16 requests. These items are listed below, along with the explanations provided by PG&E during the conference call. All items were resolved during the call, though PG&E will submit to NRC a formal, written version of its response to request number 3.
- 1. The responses to a number of RAIs state that the DC01B model includes or does certain things or gives certain results. Presumably these statements are also true for the DC01A model since the only difference between the two models is the seismic hazard curve.
PG&E indicated that the only difference between DC01A and DC01B is the seismic hazard curve. The statement is true for both models.
- 2. The response to RAI 4 (or RAI 1.i) mentions that the DC00 (stage 1) model was modified to take credit for RCS depressurization following loss of charging in the seismic PRA and this resulted in a reduction in total CDF to 5.05E-5 per year. Was this change incorporated in the DCC0 model? The results for DCC0 do not show such a significant decrease in CDF.
PG&E indicated that it addressed this issue in its December 6 RAI response, beginning at the bottom of page 4. Excluding the Cable Spreading Room and Control Room fires reduced the DC00 CDF to 5.26E-5 per year, while taking credit for RCS depressurization following loss of charging accounted for the remainder of the reduction to 5.05E-5 per year. In the DCC0 model results shown on page 5 (total CDF: 5.38E-5), fires are once again included.
- 3. RAI 8 (or RAI 2.e and 2.d). The RAI response does not support the conclusion that the higher cesium (Cs) release fraction for RC16U than for RC14 would not be expected to impact the consequences by a significant amount. Consider that the RC16U/RC14 Cs release fraction ratio, derived from information provided in ER Table F.2-8, is 2.9 (4.3E-02/1.5E-02) for the MAAP3 results and 1.7 (8.9E-02/5.1E-02) for the ZISOR results, while the release frequency ratio, derived from information provided in the response to RAI 2.b, is 0.7 (5.98E-07/8.55E-07). The NRC staff performed an analysis that scaled
DRAFT - FOR STAFF REVIEW ONLY the population dose-risk and OECR for the ST2 release category by the MAAP3 ratio.
The NRC staffs analysis showed that SAMA 3, and potentially SAMAs 10 and 18, were potentially cost-beneficial, in addition to the SAMAs previously determined by PG&E to be cost-beneficial. Provide further justification for the conclusion in the RAI 8 response that the difference in the Cs release fractions for RC16U and RC14 would not impact the results of the SAMA analysis.
PG&E will submit a written response to this question to provide the NRC with formal documentation. The following is the NRC staff synopsis of the teleconference discussion.
Diablo Canyon staff indicated that the SAMAs in question are not cost-beneficial, even when analyzed with the RC16U source term. PG&E noted that Table F.2-8 of the ER is a reproduction from the DCPP IPE, and is not SAMA-related. Instead, PG&E used MAAP4.0.7 for the RC16U source term in ST2. PG&E noted that cesium iodide (CsI) is an important contributor with impacts on costs, but costs are not directly proportional to Cs releases. Relying purely on Cs release would, PG&E asserts, overstate the impact of a given release.
The initial issue with RC14 is that, for estimating purposes, using one-third of the source term shown in Table F.2-8 results in an underestimate. PG&Es other option was to calculate the source term using two-thirds of that for RC14, which would results in an overstimate. PG&E noted that using either RC16U and RC14, even with values set at 95%, still results in the cost-benefit balance of the SAMAs in question being negative (not cost-beneficial).
In RC16U, PG&E developed a specific source term to account for a lower pressure core melt.
This modified source term changes the cost-benefit analysis for SAMA 3, though it remains, PG&E asserts, non-cost-beneficial. In general, PG&E asserts that the source-term differences have a minimal effect on results. The primary difference is that due to MACCS modeling assumptions the higher RC16U source term affects how local populations would respond to the event.. Given larger emergency-phase doses, more people would evacuate. This would lower the early-phase dose to population. In the base-case analysis, doses were driven by those received in the emergency phase. In the RC16U case, long-term doses drive the results.
PG&E noted that the CsI impact was a factor of 8 larger under RC16U assumptions (4% release versus 0.5% release).
- 4. The response to RAI 10 (or RAI 2.j) discusses some mitigation actions called for in the SAMGs. To what extent are the SAMGs incorporated into the DCPP Level 2 model?
PG&E indicated that no SAMG actions are included in Level 2.
- 5. RAI 11 (or RAi 3.c). SAMAs 5 and 18 are described as identical except that the alternate EDG is seismically qualified in SAMA 18. How does split fraction TD2 differ from TD1 and TDF?
Split fraction TD1 includes random failure of EDG with all external support available. TD2 introduces the unavailability of instrument A power. TDF includes total failure of all supporting systems.
DRAFT - FOR STAFF REVIEW ONLY
- 6. RAI 12 (or RAI 3.d). Are AFW seismic failures in the sequence therefore failures or unavailabilities of the AFW water sources?
PG&E noted that, as seismic events are likely to result in a loss of offsite power, the reactor would have to rely on natural circulation for cooling. The volume of the condensate-storage tank (below a non-seismically-qualified nozzle) is insufficient for cooling, so fire-water systems would have to be aligned - by operator actions - with the cooling system. PG&E attempted to account for the fact that operators, in an emergency, may fail to align the fire-water system to the cooling system.