ML20107B769
Text
..
_ ~.
A g;
2.
9-4 8
+
u...
,. s u 4
Jersey Central Power & Light Co.
Docket No. 50-219 Attention?
Mr. T. J. McCluskey Station Superintendent Oyster Creek Nuclear Station.
4 Forked River, New Jersey 08731 Gentienent
{
A letter to the Directorste of Licensing,-AEC, from the Northern States g
Power Company, dated August 4, 1972, was recently placed in the Public Doctaient Room. The subject matter discussed in that letter is felt to have possible applicability to your plant and, accordingly, I-have fl' enclosed a copy for your information, i
l Sincerely,
,i 1
Robert T. Carlson Chief
{
Pacility Operations Branch 1
f
Enclosure:
}
Letter, Northern States Power Co.
to the Directorate of Licensing, dtd 8/4/72 e
i omer >
...,CRE S S,.,,
pit Carlson:keb.
sunwmc >
f 1/. 2. 4../.73 un>
Posa ABC Ste (Rev,9 53) AECM 0240 9604170068 960213
)
a-s i
NORYHERN SYATES POWER COMPANY M I N N E A PO L.I G. M IN N E S OTA 5 5 40%
August 4, 1972
- g. Q7 %.
~
'.1
- /!,
h 48p Mr. A Giambusco Deputy Director for Reactor Projects m
S, l
Directorate of Licensing c.', j7 *, g 4.
United States Atomic Energy Commission Washington, D.C.
20545
'4 g
/
Dear Mr. Giambusso:
M MONTICELLO NUCIEAR GENERATING PIAF2 1
Docket No. 50-263 License No. DPR-22 1
Supplemental Letter to the May 25, 1972 Report of Lov APRM Gain Setting i
Our hhy 25, 1972. letter reported that all APRM channels were found indicating a
lower than the core thermal power; however, the effective safety system set-tings remained less than the safety limit.. 'This situation was reported as an abnormal occurrence in compliance with Tecalical Specification 6.6.3 3 It
- 1 should be noteu that this was not a violation of Section 2 based on the dis-cussion, contained on page 18 of the Technical Specifications.
Tnis letter 1
provides a further description of analytical studies of the APRM performance i
capability.
Based on this work, the present Technical Specification concernirs APG1 calibration requirements, along with prudent calibration techniques, assures safe operation of the plant under all conditions.
Tne hhy 25, 1972 letter did not discuss the tracking accuracy of APICI channels, Each channel receives signals from 24 in-core detectors.
Tne average of these
' inputs provides a good representation of the average core power.
However, the fact that a finite nu;ber of discrete detectors cannot give an exact represen-i tation of average power at all times was acknowledged in the accident ant. lysis.
Section VII. h.5 2 3 of the FSAR discusses the performance analysis of the APFM System.
In part it states, "Tnat the APRM provides valid average power reasure-ments during typical rod or flow induced power level change has been shown by three dimensional analysis.
These analyses indicate tracking accuracy of I
approximately 5% of a vide range of power levels. " Further discussion of the tracking accuracy studies is contained in the General Electric Topical Tseport APED-5706, In-core Neutron bbnitoring System For General Electric Loiling Water Reactors, November, 1968.
In recent discussions and correspondence General Electric has confirmed that the accident analysis allows for an error in APTO: trackingofbulkthermalpowerof5%ratedpower.
1 i
- f. -
- ri S to 1 ;l : c ? g.
V-o
,i-A v.ORTHERN - OTATES POWE. ' COMPANY
- r. A'Giambusso August 4, 197;
/
The occurrence took place 'during a power ascension from a shutdown condition.
Under equilibrium conditions at 50% of rated power, as indicated by a heat-balance, the APK4 channels were calibrated.
For vach of the six channels the gain was decreased.
Twelve hours later a. heat balance indicated the reactor power to be 68% of rated.
At this time the APM:s all read lov; the average of the six channels being 64%.. Five channels were within the 5% criteria.
.One channel indicated low by 7 5% of rated power.
If the gain had not.been decreased while at 50% power, only one channel vould have indicated less than
/
68%; that one indicating 6L.L% Which was well within the 5% tracking criteria.
A review of the situation showed that while at.50% power, the power shape was such that a peak occurred near one of the four detector elevations. At'68%
power, having experienced changes in core flov, xenon concentration, and (most imporcant) in the control rod pattern, the power shape was such that the peak had moved away from the detector elevation.
In' retrospect, it appears that during changes in the flux shape the APR4 gain should not have been decreased.
k'ith the exception of this event, experience has shown that a SCmargin for APF0 tracking capability is conservative. Tne reason for exceeding the 5%
mar 61n as reported above is that we doubly perturbed the situation; first, by calibrating the system at a time when it appeared the gain was too high and second, due to the normal tracking deviation expected for a change in power shape.
Power shape transients are most pronounced at lower power levels, where control rods are being withdrawn from the core while operating with minimum-recirculatic:
Under these conditions, the flow-biased scram feature of the APD: syste=
flov.
maintains an additional margin of safety below the 120% scram trip setting, whici the FSAT demonstrates to be sufficient to proclude fuel damage during ' postulated transients.
From this experience we have strengthened our operating procedures in the fol-First, we believe that the best calibration is done during loving manner.
steady state operation at elevated power levels where the power shape is rela-tively uniform at all elevations;,unless the calibration check shows that the bc increased during power shape transients or unless some other justi:
gain must cause for recalibration exists, a change in gain is not recommended. The APFC:
channels are 1cft with a conservative bias.
Second, operating instructicnc make the operator aware that a certain tracking deviation can be expected, and that during any reactor operation that might potentially change the power shape, the operator should frequently compare the APM: readings to the reactor power level If the calculated by the plant process computer or other heat balancc methods.
deviation exceeds a prospecified threshold, he is instructed to inform adminis-trative personnel who in turn vill execute the necessary steps to have the situation analysed and corrected.
Tne primary purpose for frequent AP?l: calibr<
tions is to continually compensate for the loss of sensitivity of in-core detect
.due to neutron exposure.
Tne intention is not to calibrate the /.?E: systemHavin frequently enough to ecmfensate for transient changes in the power shapc.
sufficient margin for tracking accuracy allowed in our accident analysis and recognizing this in our operation procedures, we belicyc that the frequency of FPi?: en11bration required by the Technien1 Specifications is cufficient to TN:1 tic any unsafe conditions.
Eurs verv truly KO.A y,
......,,u.
.