ML031560736

E-mail from Merschoff to Sheron, Regarding Food for Thought - Davis-Besse

ML031560736
Person / Time
Site:	Davis Besse
Issue date:	08/20/2002
From:	Merschoff E NRC/OCIO
To:	Collins S, Sheron B Office of Nuclear Reactor Regulation
References
FOIA/PA-2003-0018
Download: ML031560736 (1)

Text

Caidweli--ood for Thuh- DaisBese age1 Fies !rnes Caldwell - Food'for Thought- DavisBe'sse~ 'Page 1 From: Ellis Merschoff To: Brian Sheron; Samuel Collins Date: 8/20/02 7:29AM

Subject:

Food for Thought- Davis Besse Sam/Brian, Here is a first order approximation of a risk informed assessment of Davis Besse, that might be useful to you as a datum when reviewing the staff and licensee's more rigorous assessments.

Use our Davis Besse SPAR model with MLOCA set to 1.0. Result is 2.6 E-3. Seems reasonable given an MLOCA.

Now, for probability of MLOCA. Licensees ANSYS analysis page 16 of 29 plotting Von Mises strains shows failure criteria is met at the membrane midplane at 2500 psi vessel pressure. Uncertainty for this result is very high. Consider ANSYS modeled a flaw free membrane when it more likely includes stress risers and discontinuities from its method of application. Thus, a probability of failure of the membrane of 0.04 doesn't seem completely unreasonable. (If you believe 2500 psi, a trip with loss of heat sink gets you pretty close to the predicted failure criteria).

Now, MLOCA= 0.04 MLOCA CDF = 2.6 E -3 (0.04)(2.6 E -3)= 1.0 E -4 Which is RED.

I recognize this is a very simplified approach, but I thought you might find it an interesting point of comparision.

Ellis CC: Jim Dyer; Jon Johnson