ML20062B350
| ML20062B350 | |
| Person / Time | |
|---|---|
| Site: | Oconee  |
| Issue date: | 10/15/1990 |
| From: | Tucker H DUKE POWER CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NUDOCS 9010240276 | |
| Download: ML20062B350 (4) | |
Text
?f; yqi.' y jj $
'yll-l-
- f
'i
- Duke 1%wer Company IIAL 11 Take Y,i PO Bcx33198 l' ice President Charlotte, N C 28242 Nuclear Production.
2-(704)373 4531 Y\\
- . DUKE POWER
.u f
/
' October 15, 1990 U. S. Nuclear Regulatory Conunission Document Control Desk a '*
Washington, D. C. -20555-I 1
- o. <
Subject:
Oconee Nuclear Station, Unit 1 l
Docket No. 50-269 j_
Unit 1 Cycle 13 Startup Testing Report i.
i
[I Gentlemen:
-By latter dated August-20, 1990 I provided the Oconec Unit 1 Cycle 13 j
i Startup Testing Report pursuant to Oconee Nuclear Station Technical Specification 6.6.1.1.
For your information, attached is an internal
' letter documenting the evaluation of control rod group 5 worth.-
Very Truly Yours,
.\\
y Hal B. Tucker
-PJN/9 q
'xc S. D. Ebneter, Regional Administrator U. S. Nuclear Regulatory Commission Region II 101 Marietta Street, NW Suite 2900 Atlanta,:GA 30323 L. A. Wiens,' Project Manager
. L Office of Nuclear Reactor Regulation
~
U.'S. Nuclear. Regulatory. Commission Washington, D.
C.-
20555 P. H.. Skinner, Resident Inspector e
.0conee Nuclear Station
-3 i
". {gg2ggggoggg,T f
I
y 4
i;
~V t
?
4 August 6,=1990 D. M. Hubbard Oconee Performance
Subject:
Oconee 1 Cycle 13 Zero Power Physics Tests Control Rod Worth Heasurements During the Oconee 1 cycle 13 (01C13)- zero power physics tests (ZPPT) control rod groups (CRGP) 5 and 7 were measured to be abnormally high relative to predictions by core 3
physics models.
The core model used for the Physics Tests Manual-(PTM) was reviewed for errors and none were found.
All ZPPT-predictions were repeated. with a completely independent set of core physics models and the results were essentially -the same.
.Therefore it-was.unlikely that a inputL error contributed to the discrepency. All core shuffle procedures were reviewed at Oconee and within. Nuclear Design-to insure that the core was loaded as designed. No tests are performed at HZP which provide information on core. symmetry
~
so a direct indication of broken or otherwise' failed control rods was not available-at this time. However, core reactivity was. verified by the all rods out:
boron measurement and since the measured CRGP worths were.high there was additional conservatism in all: shutdown margin calculations. Also, as required by tests procedures, extra bank worths were measured until the ' average error in all measured CRCP's was less than'10%. Since the: measured rod-
-worths:were conservative with respect to. shutdown margin and there were 'no other reactivity concerns, it was judged acceptable to escalate power 'in order to utilize power distribution data from the incore detector system.
-l At 75 percent of full power with near all rods out a near I
.l equilibrium power distribution was carefully examined in an effort to explain.the rod worth discrepency. The axial power shapes were-plotted for rodded locations in an effort to
~l identify possible brcken control rod fingers. The radial power, distribution was examined for assymetries that would indicate a dropped control rod bank.
There was no j
information to suggest a mechanical failure of any control rod.
The radial power distribution measured by the incore D
r'~
- - - detector system was in general lower than-prediction in the IfK'j.y;M".j,' ((+1isur$ral portion of the core and higher than prediction in fjjF' S yf outer half of-the core radius where CRGP 5 and 7 are t
F
' located.
This shift in radial power relative to prediction
()gi }g jggg waslslightlymorethaninpastcores and was opposite in sign relative to 01Cl2. At this point it seemed likely that
. lu.
~:0.
}www Cv ~ mCE
i 9/'
~f
- p i
m a deficiency in'the~ core model's prediction of the radial power distribution had lead to the discrepency in' control
,/
rod. worths. Further investigations therefore focusedJ on trying to explain the radial shift in~the power distribution relative to prediction. Again a completely independent. core model was-executed which verified the design model used in the:PTM.' Beginning of-cycle (B00) assembly burnup estimates made by the operator aid computer and the design models were compared to determine if small errors were being carried along and.
magnified from cycle-to cycle.
The errors f
observed were demonstrated to have an insignificant impact on the power distributions predicted by the, core models.
Historical-BOC, power and burnup data from unit' 1 was compared-to.past ' experience on units 2 and 3.
Unit l's.
recent past history demonstrated some unique behavior.
Therefore an examination was conducted of the fuel shuffling-strategies employed in unit 'l relative to units 2 and 3. The o
characteristics of the fuel assemblies placed in the center of'the~ core was of particular-interest.-
As postulated,-
recent cycles of unit I had been shuffled with'a lesser degree of inside to outside mixing of assemblies than units 2 and 3.
m There are also a number of uncertainctes associated with the measurement of control rod worths. Acceptable reactimeter Lperformance, appropriate and systematic interpretation of reactimeter. traces, stable system flow and temperture conditions, proper detector performance, consistent and systematic execution of test procedures are-all required for
- w accurate measurement results. Reactor engineers reviewed all i
measurement hardware, procedures, and results. Nothing was found that would significantly compromise the precision of these measurements.
The exact cause of the rod worth errors in the 01C13 ZPPT
- )
has not been ' identified. Evidence indicates that the L
discrepancy between the predicted ant measured radial power distributions is the root.cause of' the rod worth errors.
It is possible, though not confirmed, that the fuel shuffling-techniques used in-recent cycles of unit I have tended.to carry along.small inaccuracies over several cycles.
If this is. true, one must postulate that the sum of these errors o
over several cycles has caused a radial power shift in the real core,that is not predicted by the current core design models. The effect of this inaccuracy may be diluted by practicing shuffling techniques that more thoroughly mix the p,
type of burned fuel that is placed' in the center of the l+
core.
Continued operation of 01C13 at full power is judged acceptable for the following reasons.
l I:
l l
r
- r
,c..-
- 4q q
y,
- 1.. There is no evidence-of mechanical failure of control rods..
1
- 2.,The imeasured' rod worths are higher than~ prediction.
Thus the current shutdown margin-and shutdown boron ll calculations are conservative.
3.
There' is no evidence-of a reactivity problem of any
.,1 kind.
- Reactivity letdown to date is well within k
acceptance criteria.and better than most other cores.
4 The prediction of the hot assembly has always been and-continues to be well within the uncertainty " assumed in t
the development of the operating limits.
5.
The-shift in the measured radial power distribution relative to prediction has been self-correcting.
-The core. is depleting in a. manner consistent with design i
predictions.
4_-
-l J. L. Eller-Engineering Supervisor Oconee Nuclear-Design cc:
E. D. Price K. S. Canady-G. A.-Lareau-1
't L
r t
i i