ML20040A616
Text
,
tS t,E o
UNITED STATES
.f
~
NUCLEAR REGULATORY COMMisslON g
7.
t WASHING TON, D. C. 20555
[/ Y k *... e,o V\\
MAY 2 31978
/
MEMORANDUM FOR:
D. B. Vassallo, Assistant Director for Light Water Reactors, DPM FROM:
D. F. Ross, Jr., Assistant Director for Reactor Safety, DSS
SUBJECT:
EVALUATION OF THE EFFECT OF A CHANGE IN THE FLOW MALDISTRIBUTION FACTOR ON DNB FOR BABC0CK AND WILC0X 205 FUEL ASSEMBLY PLANTS (REISSUED)
References:
1.
B&W 1etter dated December 20, 1977, J. H. Taylor to S. A. Varga, "VMFT Experimental Results" 2.
B&W letter datcd February 6,1978, J. H. Taylor to S. A. Varga, "The Effect of a Change in the Flow Maldistribution Factor on DNB for Babcock and Wilcox 205 Fuel Assembly Plants" 3.
B&W letter dated March 28, 1978, J. H. Taylor to S. A. Varga, "VMFT Experimental Results" 4.
B&W letter dated May 15, 1978, J. H. Taylor to Olan
/_^ '
D. Parr' l
L' My April 27, 1978 memorandum on the same subject contained an error in that it assumed a margin of.08 in DNBR existed for the most limiting transient whereas the actual margin is.01 in DNBR. This memorandum corrects that error.
6 Reference (1) reported to the NRC that the minimum flow maldistribution factor, as obtained from vessel model flow tests, should be reduced from 0.99 to 0.965. The 0.99 value was used in the BSAR-205;PSAR.
The effect of the change in flow maldistribution in a CHATA/ TEMP analysis is a reduction in minimum DNBR of 0.08.
CHATA/ TEMP was used for the BSAR-205 PSAR.
B&W has developed the LYNX 1/ LYNX 2 codes which allow cross-flow among adjacent assemblies. The effect of the change in flow mal-distribution in a LYNX 1/ LYNX 2 analysis is a reduction in minimum DNBR of 0.01.
Analyses which show the effects of the change are given in reference (2).
Reference (2) demonstrates that although the BSAR-205 core thennal hydraulics analysis was performed with the wrong flow fraction for the hot channel, the non-conservatism is very nearly offset by the conservatism in using closed channel codes (CHATA/ TEMP).
Reanalysis with the LYNX 1/ LYNX 2 codes at the FSAR stage will therefore give approximately the same results as the CHATA/ TEMP analysis for the PSAR.
Reference (3) discusses the effect of the upper plenum pressure gradient 8201210298 810403 PDR FOIA MADDEN 80-515 PDR
. t%Y Z 31978 on the LYNX 1/ LYNX 2 analysis. Upper plenum pressure gradients were measured in vessel model flow tests (VMFT) and in the Oconee 1 reactor; a pressure profile which bounds the data was then used in the LYNX analysis to determine the effect of the pressure gradient on minimum DNBR. The minimum DNBR was reduced by 0.04 using either the B&W-2 correlation or -the interim BXC-2 CHF correlation.
The LYNX 1 and LYNX 2 codes are under review by the staff. There may be a need for additional margin to offset the effects of code uncertainties and any penalties which may arise from the review, but the extent of that margin is not known as yet. The effect on plant operation can be bounded by considering the effect of the radial upper plenum pressure gradient with a closed channel code; staff calculations show the closed channel effect to be approximately 10% in DNBR. Actual code penalties will probably be 5% or less (.06 in DNBR). This penalty of.06 in DNBR will be applied by the staff until the LYNX codes are approved.
The-BXC-2 correlation is an interim CHF correlation derived from data on test bundles with rod pitch and diameter typical of the 17 x 17 fuel design. The CHF testing is in progress and a final CHF correlation will be derived at the end of the test series.
Should the BXC-2 correlation prove to be non-conservative, adjustment in plant parameters may be required to accomodate full power operation. However, the 1.25 limit used by B&W exceeds the 95/95 limit calculated with available data by more than 5% and is acceptable as an interim limit.
g.
C/
The BSAR-205 design calculations had a margin of 0.01 between the minimum DNBR and the DNBR limit for the most limiting A00.
This margin is less than the' combined effect of the inlet flow error correction and the upper plenum pressure gradient effect by.04 in DNBR. Therefore, there is
~
need for identification of other margins which are available to offset this deficit.
Adjustments in plant parameters required to accomodate LYNX 1/ LYNX 2 uncertainties, possible BXC-2 non-conservatisms, flow maldistribution.
corrections, and upper plenum pressure gradient effects will be addressed in the FSAR review.- Additional margin could possibly be obtained prior to FSAR stage through rod bow testing which is in progress and thr,ough the statistical core design which is under review.
Should these margins not materialize, the operating parameters of BSAR-205 plants will be changed to meet the thermal-hydraulic design criteria. The applicant indicated that sufficient margin exists in the present BSAR-205 design to operate the core at nominal power with a radial peaking factor of 1.43, an inlet temperature of 567"and with the axial power shaping rods removed.
These changes from the initial design are equivalent to a DNBR margin of 0.26; the required margin for flow maldistribution, pressure gradient, rod bow and code uncer cainties is 0.24.
Therefore, the available margin is sufficient to assure full power operation.
l l
P
1
~
/
$ 2 3 I3IO
(~
The FSAR analysis, when submitted, must use the correct inlet flow
~
fraction, the upper plenum pressure gradient and an approved thennal-hydraulic method.
It is particularly important that the pressure gradient effect be included in the analysis with other than center peaked axial power distributions.
)
[I
't
(.v14 D. F. Ross, Jr., Assistant Director s
for Reactor Safety Division of Systems Safety
Contact:
W. Hodges, NRR, X27588 cc:
S. Hanauer R. Mattson D. Ross
- 0. Parr S. Varga T. Cox J. Wilson W. Pike Z. Rosztoczy T. Novak P. Check
/..
L. Philli s G. Mazeti
g G. Kelly W. Hodges 3
)
W.
0; j
I
_ _. _ _ _. _ _ _ _. _. _ _ _ _ _