ML20135D986
| ML20135D986 | |
| Person / Time | |
|---|---|
| Site: | McGuire  |
| Issue date: | 09/10/1985 |
| From: | Tucker H DUKE POWER CO. |
| To: | Adensam E, Harold Denton Office of Nuclear Reactor Regulation |
| References | |
| TAC-62981, TAC-62982, NUDOCS 8509160246 | |
| Download: ML20135D986 (6) | |
Text
l Duxu Powim GoMmNY l'.O. IBOX (5:111111 CitARIMt'rt*, N.C. Mil 2 lO HALH.Tt*CKEH tra.mruonn two ratesmaat (704) Il7 )-4fWlt aaname PatMMTT9tDe September 10, 1985 Mr. Harold R. Denton, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C.
20555 Attention:
Ms. E. G. Adensam, Chief l
Licensing Branch No. 4 f
Subject:
McGuire Nuclear Station, Unit i Docket No. 50-369
Dear Mr. Denton:
During the Cycle 3 startup physics tests performed at McGuire Unit 1 rod worths were measured using the method described in " Rod Bank Worth Measurement Utilizing Bank Exchange", (WCAp-9863-A). This test process incorporates a two tiered set of criteria for test result evaluation, j
These are " review" criteria, which have no direct safety significance, and " acceptance" criteria, which is based on meeting safety analysis i
l input assumptions. The spectfic lists of these criteria are defined in WCAP-9863-A, Section 5.0 and also listed in the NRC Safety Evaluation Report (SER) as attached in Appendix A.
The SER also requires that, in the event that a review or acceptance criterion has not been met, a
report be submitted to NRC.
On June 26, 1985, during the McGuire Unit 1 Cycle 3 Rod Exchange test, the reference bank, Shutdown B, worth was measured to be 12.5% below I
the predicted value. This deviation exceeded review criterion (1) on the reference bank worth of 10%.
Subsequent measurement of Control C was 17% below its predicted value. This exceeded review criterion (2a) on all banks other than the reference bank of 15%. All other measure-l ments met their review criteria. All rod exchange acceptance criteria l
were met.
Therefore, no safety analysis input assumptions are affected.
l This report sunnarizes review criteria evaluations an directed in the NRC l
SER.
If there are any questions regarding this report, please contact Duke through the normal licensing channels.
Very truly youru, A
. put t
1 Ital B. Tucker 0509160246 050910 yDR ADOCM 05000369 RI,G i nmh l*DM Attachment 4
l k
Mr. Harold R. Denton, Director September 10, 1985 Page 2 cc:
Dr. J. Nelson Grace, Regional Administrator U.S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW, Suite 2900 Atlanta, Georgia 30323 Mr. W. T. Orders NRC Resident Inspector McGuire Nuclear Station w
L I
INTRODUCTION /SACKGROUND l
l During the Cycle 3 startup physics tests performed at McGuire Unit 1, rod worths were measured using the method described in " Rod Bank Worth l
Measurement Utilizing Bank Exchange", (WCAP-9863-A). This test process i
incorporates a two tiered set of criteria for test result evaluation, i
These are " review" criteria, which have no direct safety significance, and
" acceptance" criteria, which is based on meeting safety analysis input assumptions. The specific lists of these criteria are defined in
(
WCAP-9863-A, Section 5.0 and also listed in the NRC Safety Evaluation Report (SER) as attached in Appendix A.
On June 26, 1985, during the McGuire Unit 1 Cycle 3 Rod Exchange test, f
the reference bank Shutdown B, worth was measured to be 12.5%
i below the predicted value. This deviation exceeded review criterion (1) on the reference bank worth of 10%.
Subsequent measurement of Control C i
was 171 below its predicted value. This exceeded review criterion (2a) on all banks other than the reference bank of 15%. All other measurements met their review ci aeria. All rod exchange acceptance criteria were met, i
Therefore, no safety analysis input assumptions are affected.
This report summarizes review criteria evaluations as directed in the NRC SER.
i EVALUATION t
The evaluation centered on the following three areas:
l
- 1) Review of analytical predictions f
- 2) Comparisons of rod exchange data to subsequent boron endpoint data
- 3) Review of reactivity computer traces and assumptions
[
The models and calculations on which the predictions were based were reviewed.
I No errors were found and all cuality assurance requirements were met.
Since Shutdown B was the reference bank for the Cycle 3 Rod Exchange test, its I
worth would impact the measured worth of the other banks. Measurements of changes to critical soluble boron concentrations for the dilutions associated t
with the independent insertion of the reference bank for the rod exchange, Shutdown B, and the lead control bank, Control D, matched the predictions within a few ppn. Also boron worth determined from the reactivity computer calculated rod warth (pcm) divided by critical boron change for rod insertion 1
(ppm) was inconsistent with current cycle prediction and previous cycle data.
These discrepancies imply a greater rod worth for Shutdown B than what was i
measured by the reactivity conputer.
This is supported by predicted values l
of boron worth which also imply a greater rod worth when combined with measured changes in critical baron concentrations.
In reviewing the reactivity trace from the recorder strip chart for the measurement of Shutdown B, the trace was found to contain no noticeable anomalies. Although the change from unrodded to the Shutdown B in condition results in a radial power redistribution into the peripheral assemblies of about 15%, this was not observed in the trace.
Specifically, the predicted radial power redistribution caused by Shutdown B insertion would result in reactivity "undershoot" following the rod motion. This spatial effect is typical in rod worth reactivity traces and would normally be accounted for in the trace analysis. However, no spatial effects were visible in this trace.
This could have contributed to the under measurement of Shutdown B rodworth.
The neutron flux seen by the excore detector in Cycle 3 is lower than in previous cycles due to the low leakage loading pattern used in this cycle.
Since the measurements were performed at a lower neutron flux level, the gamma source is a significant contributor to the detector signal. This can cause the measured reactivity to be lower than expected with the chang'. in boron concentration, but there is no other supporting evidence that substantiates this as a possible cause for the review criteria exception.
Radial redis-tribution "undershoot" and gamma source effects are estimated to be as much as a 5% reactivity effect.
The delayed neutron fraction (81) can be affected by rodded configurations by as much as 3% in some cores.
For McGuire 1 Cycle 3, the effect is small(0.5%).
Increases in si can also be obtained from refined 8 calculations. Work by R. J. Tuttle (1) results in a 2% increase in the delayed neutron fraction.
For McGuire, a 2.5% increase in measured reactivity of Shutdown B would have resulted from utilizing Tuttle's method and rodded s's.
CONCLUSIONS Westinghouse has reviewed the results of the McGuire Unit 1 Cycle 3 Rod Exchange test and has found no deficiency in either the rod worth measurement and procedures or the calculational models used for the predictions. The likely cause of the measured versus predicted differences in reference bank reactivity is a combination of small variances in input to the reactivity computer and measurement environment effects. There is also no effect on the safety analysis input assumptions due to these measurements.
(1)
R. J. Tuttle, " Delayed-Neutron Data for Reactor-Physics Analysis",
Nuclear Science and Engineering, Volume 56, pages 37 71, 1975.
APPENDIX A TEST REVIEW AND ACCEPTANCE CRITERIA WCAP-9863-A (From NRC Safety Evaluation Report) l
}
The review criteria are:
(1) The absolute value of the percent difference between measured and predicted integral worth for the refer'ence bank is 1 10 percent.
(2)
For all banks other than the reference bank; either a) the absolute value of the percent difference between inferred and predicted integral worths is 1 15 percent or b) the absolute value of the reactivity difference between inferred and predicted integral worths is 1 100 pcm, whichever is greater.
(3) The measured / inferred worth of all the rods must be 1 110 percent of the predicted worth.
The acceptance criteria are:
(1) The measured / inferred worth of all the rods nust be 190 percent of the predicted rod worth.
(2)
For all banks other than the reference bank; either a) the absolute value of the percent difference between inferred and predicted integral worth is 1 30 percent or b) the absolute value of the reactivity difference between inferred and predicted integral worths is 1 200 pcm whichever is greater.
(3)
The absolute value of the percent difference between measured and predicted integral worth for the reference bank is 1 15 percent.
a