ML20079P167

Provides Answers to NRC Questions Re Evaluation of Supplemental Info on PRA Assessment for Tornado & Hurricane Missile Hazard to Containment Isolation Valve Compartment Equipment

ML20079P167
Person / Time
Site:	South Texas
Issue date:	12/23/1983
From:	Simiu E NATIONAL INSTITUTE OF STANDARDS & TECHNOLOGY (FORMERL
To:	Singh B NRC
Shared Package
ML20079P130	List: ML20079P159 ML20079P167
References
NUDOCS 8401310108
Download: ML20079P167 (3)

Text

..-.:.. t.-. .~. ; . : . -w

-- -.. . :4 .. .. . . .-. -- ..

g **f UNITE] GTATE3 CEPARTMENT CF COMMEICE

" National Bureau of Standards weshington, D.C. 20234

\ , . .

December 23, 1983 Mr. B. K. Singh Mail Stop P-1022 Nuclear Regulatory Comission Washington, DC 20555

Subject:

Evaluation of Supplemental Information Pertaining to Proba-bilistic Risk Assessment (PRA) for Tornado and Burricane Missile Ha::ard to the Containment Isolation Valve Conpartment Equipment, South Texas Project

'Ihe report " Technical Evaluation of Probabilistic Risk Assessment for Tornado and Burricane Missile Bazard to the Containment Isolation Valve Conpartment Equipnent, South Texas Project," was based on documents mde available to the National Bureau of Standards before Decenber 12, 1983.

• Following that date, I have received from NRC: (1) a copy of questions addressed by NBC to Bechtel concerning the subject PRA, and of responses frcxn Bechtel to these questions (attached herewith), and (2) a copy of NBC

. memorandum Docket Nos. 50-498/499 dated Decenber 13, 1983, to George W.

,. Knighton and Olan D. Parr from William P. Gamill, all of NBC, on the l Meteorology Review of Probabilistic Evaluation of Isolation Valve Cubicled Roof Design for South Texas. -

NRC Ouestions to Bechtel i In my opinion, given the present state of the art, Bechtel's responses to the NBC questions satisfactorily answer concerns that may be raised with respect to:

1. the missile surface density assumed in the Bechtel South Texas Study -

l 2. the possibility that missiles lighter than those assumed in the

[ Bechtel South Texas Study might be available and danage the Isolation Valve Conpartment, and l 3. the possibility that Bechtel's estinates of injection probability E

may be scxnewhat low. In the event that the threshold hurricane l wind speed for missile injection is reduced to 165 uph, the l median probability of hitting the target will increase by about l one order of magnitude if 10 percent of the missiles are unrestrained and by about two orders of magnitude if all missiles are unrestrained.

[.

1310108 840106

' 4 ADOCK 05000490 l PDR l.

I~

e a -

l *c 2-NRC Meteorolcgy hwiew I With respect to the E Meteorology Review, I note that most of the dif-forence tor between the E and the Bechtel estimate o{ probability that a

,' 10~pper strikes yead tk assite dL=in to any one year between the difference (1.7 x10~ theper year and respective estimates 1.17 x the mean tornado path area (0.206 square miles by NRC; about 0.049 square mi16.:, to which there corresponds a == Mart of 0.022 square miles, by Bechtel).

N NBC mean tornado path area was estinated from a sanple of 250 tornadoes (period of occurrence 1954 - 1981) reported in the two-degree latitude-longitude square (excluding overwater area) containing the plant eite. On the other hand, the Bechtel estinate was based on a sanple of 2,730 tornadoes reported in 1953 - 1982 throughout the state of Texas. I believe that tornadoes in south Texas and in the other areas of Texas do indeed not constitute a meteorologica11y honogeneous set and, therefore, that the NBC estinate is more credible. This would increase the rate of occurrence of tornadoes at the site and, hence, the nominal probability of damage to Ivc from tornadMenerated missiles by about one order of magnitude. '!hus, the o M enerated estinated missiles would median be about of probability 3 x 10 9,f rather danage thanto 2 xivc 10fromfornad as indic ted in the Bechtel study. ,

The NRC memorandum mentioned above notes that the Bechtel es;inate of ,

hurricane wirxl speeds appears reasonc.ble. '!his is also my view. Thus, the NRC memorandum would not entail a modification of the estinated median value of the nominal probability of damage siles put forth by Bechtel (i.e.,1.2 x 10_g Ivcper byyear).

hurricane-borne However, themis- NRC memorandum points out that winds other than hurricanes and tornadoes were not considered in the Bechtel study.

. Calculations conducted at NBS and reported in "Elrricane Wind Speeds in the

! United States" by M.E. Batts, et al. (Journal of the Structural Division, l- ASCE, October 1980, pp. 2001-2016) indicate that the effect of nonhurricane and nontornadic winds on the probability of occurrence of winds of any j given speed is perceptible only for spoeds corresponding to mean recurrence intervals of about 25 to 50 years at most. '!herefore, it is my opinion that failure to consider winds other than hurricanes and tornadoes does not affect the calculations of the Bechtel study.

O .

M The estinated median value of the nominal probability of danage to the Ivc by grnado ,gr hyris.; borne missiles would then be about 3 x 10-9 + 2 x

= 3 x 10~ . Given the various uncertainties inherent in the Bechtel 10~

estinates (including uncertaint'ies with respect to the probability of in-jection as reflected in the Bechtel answer to question 93 by NRC), it is

,,nps. % a%. qp. . ..ws- -

, , - - , - - - - - - - - , , . - - , , - - , , - - - - - - - - , . , , ,--w.. .,,,,,,..-,,...---,.,---,,,w...--ww-v -

.-. q. .y a.. . . . . . .~ .- ~ . . . - -_..___

7j . - ,

7 J*

-i '

t. ,l 4

my opinion that the estinated probability of damage to the IVC of 3 x 10

~

is correct to within one or two orders of magnitude.

.i

sincerely, 1 Os .

Nl A Emil Simiu Research Structural Engineer Structural Engineering Group Structures Division Center for Building Technology, NE:L cc: S. Boyd L. Rubenstein B. Grenier I

l  %

e 1

i i-l 4

l'S

?

l

. + ~ . , .,

l ; er-#

. - . . _ _ . _ . _ , . _ _ _ . _ . . . _ , . _ . _ _ _ . _ . . _ . _ _ _ _ _ _ - - - , _ . . . . _ _ _ _ _ _ .. . - _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ . . _ .