ML20217C698

Requests Staff Comments on Ucs Assertion That NRC Made Nonconservative Assumptions in Safety Analysis Which Permitted Operation of Plant Through End of Current Fuel Cycle.Response to Request Encl

ML20217C698
Person / Time
Site:	Yankee Rowe
Issue date:	07/10/1991
From:	Russell W Office of Nuclear Reactor Regulation
To:	Blaha J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO)
References
NUDOCS 9107180079
Download: ML20217C698 (4)

Text

.- _ . _ - - .- -

%g 6# 80 C .

L t'4

.+  % UNITED STATES I

. ' ! 16 ' i NUCLEAR REGULATORY COMMISSION wAsaiNoton. o c. Posa

[c;'kMi;,f;f

• • • • • July 10, 1991

$$~

MEMORANDUM FOR: James L. Blaha Assistant for Operations Office of the Executive Director for Operations FROM: William T. Russell, Associate Director for Inspection and Technical Assessment Office of Nuclear Reactor Regulation

SUBJECT:

COMMISSIONER CURTISS' REQUEST FOR INFORMATION RELATED TO THE YANKEE R0WE REACTOR VESSEL Comissioner Curtiss requested the staff provide comments on the UCS assertion in their petition that the staff made nonconservative assumptions in its safety enalysis which permitted operation of Yankee Rowe through the end of the current fuel cycle. The enclosed responds to Comissioner Curtiss' request.

I request that you provide tnis information to Comissioner Curtiss and the other Comissioners. You should also note that the staff is working on a detailed reply to the UCS petition which will be forwarded to the Comission by July 23, 1991.

(

dunka William T. Russell, Associate Director for Inspection and Technical Assessment

Enclosure:

As stated cc: J. Taylor T. Hurley F. Miraglia J. Partlow B. Liaw S. Varge J. Calvo 1  %

9107190079 910710 1 PDR ADOCK 05000029 P PDR f{C

.;20027 I

ENCLOSURE Information Related to Yankee Rowe Reactor Vessel Introduction This informstion relates to the UCS/NECNP Petition submitted to NRC on June 4, 1991. The Petitioner in section IV.D.3, "Nonconservatism of and upper shelf energy assumptions (USE) calculations, underlying reference temperature d (RTofthepetitionallegedfouEbo)ncons staff's safety assessment. The items identified by the Petitioner were considered by the staff during tne preparation of the August 31, 1990, Safety Assessment and the staff's October 9, 1990, memorandum to the ACRS.

Alleged Nonconservatism No. 1 Page 16 of the Petition states that the revised neutron fluence reported in  !

Yankee Atomic Electric Company (YAEC) letter dated February 20, 1991, causes a significant increase in reference temperatures and a decrease in Charpy upper shelf Energy. Hence, the Safety Assessment significantly underestimates the degreetowhichYankeeRowe(YR)violatesNRCstandardsforreference temperature and Charpy a E.

Reference temperatures are used to calcul.ite the probability of vessel failure resultinf f rom pressurized thermal shock (PTS) events. The reference f temperatures were evaluated by the staff in an October 9, 1990, memorandum to i the ACRS. The revised reference temperatures resulted in a conditional failure probability of 3x10 for the limiting PTS event (small break LOCA). In its August 31, 1991, safetyassessment,thestaffdeterminedthecondj,tional2 probability of vessel failure for this event was in the range 10' to 10' Since tne revised neutron fluence resulted in a conditional probability of vessel failure within the range specified by the staff, the staff assessment did not underestimate the probability of vessel failure.

Appendix G of 10 CFR 50 requires that vessels have 50 ft-lb Charpy USE unless lower valves provide margins of safety against fracture equivalent to those of f Appendix G to the ASME Code. Only recently (within the last year) has ASME l Code experts formulated criteria which, if satisfied demonstrate that equivalent margins of safety exist. ThestaffidentlfiedintheirMay1,1990 letter to the lictnsee methods and criteria to be used in the analysis, which were based in part on criteria sent to the NRC on November 20, 1989, by the Chairman of the ASME Subgroup responsible for developing the code criteria.

Using RG 1.99 Revision 2 to predict the decrease in Charpy USE, the

• revised neutron fluence results in a predicted Charpy USE of 34 ft-lb for the l shell. plates. The welds have higher predicted Charpy USE than the shell l plates. The licensee performed an analysis at 35 ft-lbs that demonstrates the reactor vessel had fracture toughness substantially greater than the ASME Code l margins. In addition, A. Hiser, Jr. in a letter to C. Y. Cheng dated August i

90, 1990, determined that the YR reactor vessel could meet the equivalent margins in the ASME Code _with Charpy USE of 33 f t-lb. Based on the i

, F g

i

- 2-t 3 substantial margins in the licensees analysis and the A. Hiser analysis, the staff determined that the revised neutron fluence did not change the staff 1 assessment that the YR reactor vessel had fracture toughness equivalent to the  !

margins of the ASME Code.

Alleged Nonconservatism No. 2 f1 Pages 17 and IB of the retition state that the 50'r correction for the  ;

difference in YR irradiation terrperature (C00'f) and the norrinal value identified in RG 1.99, Rev. 2 (550'F) was not substantiated by reference to  ;

actual data. >

t s

The staff assumed a correction factor of I'F per l'T ch6nge in irradiation l temperature. Hence, the 50'T difference in irradiation temperature results in l a 50*F correction factor. The correction f actor of l'T per l'F change in i irradiation temperature is based on data reported in Reference 3 in the August j 31, 1990 letter. The author of Reference 3 evaluated test data irradiated and i

tested in the HSST Program. The materials irradiated and tested were Babcock &  !

Wilcox fabricated Linde 80 subrnerged - arc welds. All welds are representative i of the high copper content Linde 80 submerged arc welds that exhibit low  :

initial Charpy upper-self energy and high sensitivity to neutron irradiation  !

embrittlement. Since the Yankee Rowe reactor vessel beltline welds were j g fabricated by Babcock & Wilcox using Linde 80 flux, and the submerged are t s process snd are assumed to be highly sensitive to neutron irradiation f embrittlement the data frota the HS$1 irradiatiori series are applicable to  ;

Yankee Rowe. j Alleged Nonconservatism No. 3, ff Page 19 of the Petition states that the staff used 0.70 percent nickel instead '

of 1.0 percent nickel, as recommended by RG 1.99, Revision 2.  ;

The Yankee Rowe reactor vessel welds were fabricated by Babcock & Wilcox using [

Linde 80 flux and the submerged are process. The chemical composition of these i types of welds were evaluated by the B&W Owners Group. All welds had less than  !

0.70 percent nickel. In addition, the BR-3 reactor vessel was fabricated'by  :

Babcock & Wilcox at the same time as the Yankee Rowe reactor ve:sel. The amount of nickel in the BR-3 beltline weld was reported as 0.70 percent.

Based on the information from the B&W Owners Group and the ER-3 reactor vessel,  !

the Yankee Rowe reactor vessel was assumed to have 0.70 percent nickel.  ;

i Alleged Nonconservatism No. 4 Page 19 indicates that the safety assessment did not take into account the -

fact that the amount of embrittlement (chemistry factor) from the surveillance  !

data-substantially exceeds the embrittlement predicted by RG 1.99, Rev. 2. -

f I

i i

t 4 wy -

pywpr $www ps-i=*r- v-p'p- =+-+r-y- p' gv y - y- go ft--,y~- ----yewop yy- ,,- , -y Jr-9,gwsy'.,y,ap-ygye~- 9-w+~4,--vw---vg-yqis- p- es -g

e 3

i The surveillance data is from samples removed from the upper shell plate. Eh i 1.99 Pevision 2 predicts a chemistry f actor of 90.1 for the upper shell plate. i I

The staff estinate of the reference temperature for the YR upper and lower  :

shell plates is based on the analysis of the surveillance data performed by F Odette in Reference 4 in the August 31, 1990 letter. This resulted in a  :

chemistry factor of 164 The staff used this chemistry factor in determining ,

the amount of embrittlenent in the upper and lower shell plates. Hence, the staff assessrent took into account the f act that the amount of embrittlement -

from the surveillance data exceeded the values predicted by the RG. I i

t h

i i

i i

l l

!