ML20057D714
| ML20057D714 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 08/03/1993 |
| From: | Cooper R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | Calvo J Office of Nuclear Reactor Regulation |
| References | |
| NUDOCS 9310050248 | |
| Download: ML20057D714 (12) | |
Text
'
gy g.
[
)
+
pa racg y
k fi k
UNITED STATES
/.
j i
y 5
p, NUCLEAR REGULAT'JRY COMMISSION i
3-
.. j REGION i
[
8,,
g J75 ALLENDALE ROAD l
%V,8 KING OF PRUSSIA, PENNSYLVANIA 19406-1415 ale 3 593
[
Ul M
9/~
'l MEMORANDUM FOR:
Jose Calvo, Assistant Director For Region I Reactors
{
Division of Reactor Projects - I/II Office of Nuclear Reactor Regulation FROM:
Richard W. Cooper, Director i
Division of Reactor Projects, Region I
SUBJECT:
PROPOSED TASK INTERFACE AGREEMENT REGARDING THE ADEQUACY OF THE OYSTER CREEK METHODOLOGY FOR DETERMINING THE REPORTABILITY OF PLANT OPERATION ABOVE THS LICENSED CORE THERMAL POWER LIMIT Your assistance is requested in determining the adequacy of the Oyster Creek methodology-for determining the reportabiliry of plant operation above the licensed core thermal power (CTP) limit. This ccncern was identified as an unresolved item in resident Inspection Report 50-219/93-04, dated April 22, 1993.
Following startup from a recent Oyster Creek refueling outage, the licensee implemented a new program for monitoring CTP while operating the plant at or near the licensed CTP limit.
The CTP monitoring program was developed as part of corrective actions proposed in response to three past occurrences that resulted in operation above the licensed CTP limit for an extended period of time. A summary of these occurrences is provided in the enclosed excerpt from Inspection Report 50-219/93-04.
The new CTP raonitoring program is controlled by an operating procedure and involves control room operator observation of instantaneous and averaged values of CTP. Software changes were made to the plant computer to include a 15-minute,1-hour, and 8-hour moving average of CTP on the heat balance information display that is routinely observed by the operators. General Public Utilities (GPU) had previously performed a statistical analysis of averaged CTP values and had concluded that the 15-minute average provided an appropriately
" smoothed" indication of CTP for compliance monitoring that eliminated data va2iation due to feedwater flow signal noise and deadbands in feedwater control system esponse and also provided adequate response to changes in power. The longer (1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />) CIP average was intended to be a value that, if appropriately maintaining power level with the 15-miaute average, would rarely, if ever, be excaMM. Operators are directed by Procedure 202,
" Power Operation," to reduce core thermal power if the instantaneous CTP value exceeds the licensed limit by 5 MWth or if any of the averaged CIT values exceeds the licensed limit.
In addition to these actions, operators are directed to immediately notify the Manager, Plant if Operations, if instantaneous CTP exceeds the licensed limit by 10 MWth or if the 1-hour or j
9(/&\\
9310050248 930803 PDR ADOCK 05000219 P
PDR tb AUtl ll)//jj_
- l}/bl6 f!!,1
~
f
A i
ALS 3 1993 1
Jose Calvo 2
8-hour averaged value exceeds the licensed limit. The Manager, Plant Operations, if notified, determines whether a plant deviation report should be written based on a particular occurrence. If a plant deviation report is written, it will then be reviewed to determine potential reponability. Other restrictions are included in the operating procedure involving control rod movement and recirculation flow changes with power level above 95% to reduce the possibility of overshootiag the licensed limit.
The inspectors questioned licensee management in operations, engineering, and licensing, as well as the site director as to what would constitute steady state operation above the licensed CTP limit warranting a reportable condition. The licensee consensus was that an occurrence of the 8-hour CTP average being above the licensed limit (1930 MWth) would obviously require submittal of a licensee event report (LER). The licensee acknowledged that there could be occurrences of the other indicated CTP values above 1930 MWth that would warrant reportability, but that these would have to be assessed on a case-by-case basis. In fact, there have been numerous occasions since startup from the refueling outage when the 15-minute average CTP value was above 1930 MWth for short periods of time and two occasions when the 1-hour average CTP value was slightly above 1930 MWth. The license has umW these occurrences and concluded that they did not represent a reportable violation of the licensed CTP limit. The primary reasons for concluding that the conditions were not reportable were: (1) that the conditions were not intentional or due to a lack of operator attention; (2) that none of the occurrences involved operation above 102% of the licensed limit; and, (3) that other core thermal limits were not threatened. The licensee's assessment was based primarily on conclusions made in a GPUN corporate position paper developed in September 1992, regarding compliance with the CTP licensed limit. A copy of this GPUN position paper is attached to this memorandum.
The new monitoring program appears to provide a good method to restrict intentional operation above the licensed CTP limit and readily identify periods of unintentional operation above tl,c licensed CTP limit. We ask that you review this request along with its enclosures and assess the acceptability of the licensee's program. In particulu, we ask that you assess (1) whether this program is an appropriate method for monitoring CTP near the licensed limit, and (2) whether this program appropriately considers reportability for all instances of steady state operation above the licensed limit. We would appreciate completion of the that this long-standing Oyster Creek issue can be brought to a close. The Region I point of contact is David Vito, Senior Resident Inspector, Oyster Creek, Division of Reactor Projects (609-693-0702). This TIA proposal has been discussed with Alex Dromerick of NRR.
I
\\
,]l hac d
. Cooper, Di Div ion of Reactor Proj
, Region I m
AUG 3 1993 Jose Calvo 3
Enclosures:
1.
Excerpts from NRC Inspection Report 50-219/93-04 2.
GPUN Internal Memorandum dated September 23,1991 - GPUN Position on Compliance to License Limit on Thermal Power 3.
GPUN Internal Memorandum dated March 8,1993 - Core Thermal Power Averages 4
Oyster Creek Nuclear Generating Station Procedure 202.1, " Power Operations,"
Revision 36, dated May 15, 1993, Sections 5.4,5.5 and 5.6 5.
Sample Plant Computer Heat Balance Information Display cc w/ encl:
W.12nning, DRP R. Blough, DRP J. Rogge, DRP J. Durr DRS E. Jordan, AEOD DRP Directors (RII, RIII, RIV, RV) t
ENCLOSURE 1 ENCLOSURE 1 - Excerpts from NRC Inspection Report 50-219/93-04 l
1.2 Core Thermal Power Monitoring and Reportability Considerations Following startup from the 14R outage, the licensee implemented a new program for monitoring core thermal power (CTP) while operating the plant at or near t'te licensed CTP limit. The program is procedurally controlled by Operating Procedure 202.1, " Power Operation." The monitoring program involves control room operator observation of instantaneous and averaged values of CTP. The averaged CTI values (15-minute average,1-hour average, and 8-hour average) were added to the plant competer heat balance information display that is routinely observed by the operators. The way the averaged CTP values are calculated, the color coding of the displayed values and the operator actions directed by Procedure 202.1, if the CTP limit is exceeded, are noted in the discussion of Unresolved Item 50-219/90-12-01 provided in Section 7.0 of this report, Through observation of and discussion with the control room operators during the inspection period, the inspectors noted that the monitoring program was working well and that operator sensitivity to plant operation at or near the licensed CTP limit had obviously increased. De inspectors conduded that the monitoring program would keep the plar.t from intentionally being operated above the licensed CTP limit. However, the inspectors questioned how the, licensee was going to interpret tne indicated steady state CTP values in terms of reportability due to operation above the licensed limit (1930 MWth).
Procedure 202.1 directs the control room operators to reduce reactor power if the 15-minute, 1-hour, or 8-hour averaged CTP value exceeds 1930 MWth or if the instantaneous CTP value exceeds 1935 MWth. The procedure also directs the operators to notify the Manager, Plant Operations, if instantaneous CTP exceeds 1940 MWth or if the 1-hour or 8-hour averaged CTP value exceeds 1930 MWth. The Manager, Plant Operations, if notified, determines whether a deviation report should be written based on a particular occurrence. If a deviation report is written, it will then be reviewed to determine potential reportability.
The inspectors questioned licensee management in operations, engineering, and licensing, as well as the site director as to what would constitute steady state operation above the licensed CTP limit warranting a reportable condition. The licensee consensus was that an occurrence of the 8-hour CTP average being above 1930 MWth would obviously require submittal of a licensee event report (LER). The licensee acknowledged that there could be occurrences of the other indicated CFP values above 1930 MWth that would warrant reportability, but that j
these would have to be assessed on a case-by-case basis. In fact, there have been numerous occasions since startup from the 14R outage when the 15-minute average CTP value was above 1930 MWth for short periods of time and two occasions when the 1-hour average CTP value was slightly above 1930 MWth. The licensee has assessed these occurrences and concluded that they did not represent a reportable violation of the licensed CTP limit. De primary reasons for concluding that the conditions were not reportable were: (1) that the 1
l
conditions were not intentional or due to a lack of operator attention; (2) that none of the occurrences involved operation above 102% of the licensed limit; and, (3) that other core thermal limits were not threatened.
The licensee provided the inspectors with a GPUN corporate position paper developed in September 1992, that presented the licensee position on compliance with the CTP licensed limit. The position paper made four basic statements with regard to CTP limit compliance:
1.
Average power for any steady state period may not exceed the licensed limit. The defined steady state period shall not be longer than 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />.
2.
Instantaneous heat balance power shall not exceed 102%.
3.
Immediate corrective actions must be taken if item 1 or 2 is violated.
4.
Power shall not be intentionally increased above the licensed limit to increase the average power level over the steady state period.
This company position was based in part on two past NRC internal memoranda relating enforcement aspects of operation above licensed power level and a 1981 General Electric internal memorandum interpreting the earlier of the two NRC memoranda. The licensee's position contended that normal fluctuations above rated power are neither license violations nor outside the design basis and that a requirement to never allow indicated power above the limit would effectively constitute a power derating. The licensee position noted that small overshoots of the licensed limit may be caused due to limitations or deadbands in the control system and that operation in this manner is acceptable, provided the condition is corrected and power level remains below 102%. Power level fluctuations above 100% and less than 102% are considered to be within assumed design margins. Reportability was not specifically discussed in the licensee position paper. However, the licensee stated that operation beyond the position paper guidance would be representative of a reportable condition; i.e., 8-hour CTP average greater than 1930 MWth, any indicated CTP level above 102% power, or intentional operation above the CTP limit to increase power average.
In developing the average CTP values for the plant computer, the licensee performed statistical analyses to determine the appropriate time intervals for the CTP averages. The purpose was to determine two time periods that effectively reduced the instrumentation system noise yet provided adequate response to actual changes in power. The shorter of the time periods (15-minute average) was determined to be the best value for the operators to observe for normal power control. The licensee expected that the 15-minute average could occasionally exceed the licensed CTP limit during minor power changes, such as recirculation flow bumps. The longer (1-hour) CTP average was intended to be a value that, if appropriately maintaining power level with the 15-minute average, would rarely, if ever, be exceeded. As noted previously, there have been two occasions since startup from the 14R outage when the 1-hour average briefly exceeded 1930 MWth. In both cases, the CTP 2
i
i 1-hour average was about 1930.05 MWth for less than a minute.
Based on the above observations, the inspectors concluded that the new CTP monitoring program provides the control room operators with a good method to ra. strict intentional operation above the licensed CTP limit and readily identify periods 01 Laintentional operation above the licensed CTP limit. However, the inspectors questioned whether the licensee was appropriately considering the reportability of all instances of indicated steady state operation above the licerised CTP limit for the reasons noted below.
The inspectors acknowledged that use of the instantaneous CTP indication for monitoring j
CTP level was difficult due to instrumentation system noise. The 15-minute and 1-hour average CTP values appeared to effectively reduce instrument noise to allow for CTP level control closer to the licensed limit, and for this reason, care should be taken by the operators to keep these values from going above the licensed limit. While the licensee has stated that an indicated 8-hour CTP average above the licensed CTP limit would be reportable, the criteria for determining reportability of other indicated values above the CTP limit (but less than 102% power) were much less clear. The inspectors contended that there could be -
potentially reportable situations with the 1-hour and/or the 15-minute average values over the limit. The inspectors also noted that just because a particular occurrence of operation above the licensed limit was unintentional did not necessarily mean that it was not reportable. The inspectors acknowledged that assessment of the effects on other core thermal limits was an appropriate way to assess the safety implications of a period of operation above the licensed CTP 'imit but that a conclusion of little or no safety impact did not, by itself, preclude i
reportability. Finally, the inspectors reminded the licensee that the NRC internal memoranda referenced in the licensee's position paper were in regard to enforcement issues, not necessarily reportability issues.
i The inspectors considered this issue to be unresolved (Unresolved Item 50 219/93-04 01) i pending further discussion with the li'ensee and further NRC assessment of the licensee's program for controlling power level at or near the licensed CTP limit and criteria for determining reportable conditions.
7.0 REVIEW OF PR.EVIOUSLY OPENED ITEMS (92701,92702)
(Closed) Unresolved Item 50-219/90-12-01. During reviews of plant thermal performance in 1987, GPUN identified an error in the calibration equation for calculating feedwater flow for subsequent input to the core thermal power heat balance. The calculational error was corrected by a procedure change. GPUN did not recognize until after a review done in 1990 that the calculational error resulted in a non-conservative value for feedwater flow and that prior to the 1987 procedure change, the plant had been operating above its licensed core thermal power (CTP) limit due to the error. The magnitude of the feedwater flow error was i
2.09%. This was reported by the licensee in LER 90-012, dated August 31,1990.
On October 21,1992, the licensee discussed with NRC the effects of this and two other prior 3
O non-conservative errors in the calculation of CTP on core operating limits See Section 4.1 of NRC Inspection Report 50-219/92-21 for more details. The licensee concluded that while the reasons for the three instances of non-conservative CTP calculations were not interrelated, their cumulative effect could have resulted in plant operation between 103% and 104% of the licensed CTP limit for extended periods of time since initial plant operation.
The licensee also concluded that no thermal limit violations had occurred. NRC representatives at the meeting concluded that GPUN had appropriately assessed the effects of i
these calculational errors and that the errors had been effectively resolved.
During the meeting, GPUN presented a list of corrective actions which had been or were to be taken to establish better control over CTP calculation and monitoring. The inspectors assessed the licensee's implementation of these corrective actions. The inspectors verified that the current versions of Procedure 106, " Conduct of Operations," and Procedure 1001.6,
" Core Heat Balance and Feedwater Flow Calibration-Power Range," require that reactor water cleanup (RWCU) input to the heat balance calculation cannot be removed without approval of the Manager, Plant Operations. RWCU input to the heat balance calculation has been properly accounted for since a May 11,1990 occurrence, involving a non-conservative CTP error of 0.08% when RWCU was returned to service but not included in the next heat balance calculation.
Calibration of the three feedwater nozzles (one for each feedwater train) in 1991, brought out a 1.19% non-conservative error in feedwater flow. The inspectors verified that the orifice discharge coefficients for the three feedwater nozzles were changed in the plant computer in December 1991. The inspectors also reviewed Procedure 1001.6 which provides for a manual check of the plant computer heat balance. The procedure appropriately accounted for the new feedwater flow orifice discharge coefficients. During power operation, the manual heat balance is performed weekly to verify the accuracy of the plant computer heat balance.
Also, GPUN has completed its evaluation of feedwater flow calculation using the total feedwater flow element installed during the 13R outage vice the three individual flow elements for each feedwater train. The total flow element was iutalled downstream of the feedwater heaters to more accurately measure flow to the reactor (the individual flow elements are located upstream of the high pressure feedwater heaters). The total feedwater flow element input to the heat balance calculation was made operational on March 24,1993.
Prior to startup from the 14R refueling outage, GPUN established more specific controls for monitoring CTP at full power operating conditions. These controls were included in Procedure 202.1, " Power Operation." To support the enhanced monitoring effort, software changes were made to the plant computer to include moving averages of CTP and add these values to the heat balance information display that is routinely observed by the control room operators. The instantaneous computer value for core thermal power, which has always been on the heat balance display, is updated every two seconds and uses a rolling one minute average of feedwater flow. The new 15-minute average CTP value is an average of the previous 15 minutes of instantaneous CTP values. The 15-minute average CTP is calculated every 15 seconds. The new 1-hour average CTP value is an average of the previous 60 4
O
,-y-
D I
l-minute blocks of CTP values and is updated once a minute. The new 8-hour average CTP vahe is an average of the previous 4801-minute blocks of CTP values and is updated once a minute. The new average CTP values were installed on the plant computer just prior to shutdown for the 14R outage so that empirical data could be taken by core engineering. A statistical analysis was performed on this data to determine the most appropriate means of CTP monitoring. GPUN concluded that the 15-minute average provided an appropriately
" smoothed" indication of CTP for compliance monitoring that eliminated data variation due to feedwater flow signal noise and deadbands in feedwater control system response and also provided adequate response to changes in power.
Procedure 202.1,
- Power Operation," was changed to indicate specific operator actions related to monitoring CTP near the licensed limit. Opemtors are directed by the procedure to reduce core thermal power if the instantaneous CTP value exceeds the licensed limit by 5 MWth or if any of the averaged CTP values exceeds the licensed limit. In addition to these actions, operators are directed to immediately notify the Manager, Plant Operations, if i
instantaneous CTP exceeds the licensed limit by 10 MWth or if the 1-hour or 8-hour averaged value exceeds the licensed limit. Other restrictions are included involving control rod movement and recirculation flow changes with power level above 95% to reduce the possibility of overshooting the licensed CTP limit.
The inspectors observed and interviewed the control room operators on several occasions during this inspection period to assess their attention to the new heat balance display and compliance with the changes to Procedure 202.1. The operators were aware of the new computer points and the procedure changes, as well as the color coding applied to the displayed values. Specifically, the instantaneous CTP value will turn yellow at 1933 MWth and red at 1935 MWth. The 15-minute and 1-hour average CTP values turn yellow at 1929.5 MWth and red at 1930 MWth.
l The control room operators have developed an obvious increase in sensitivity to power operation at or near the CTP limit as a result of the new monitoring program. The inspectors also concluded that the monitoring program, if consistently enforced, will keep the i
plant from being intentionally operated above the licensed CTP limit. Based on these f'mdings, the inspectors concludM that the licensee had completed the corrective actions which had been committed to dunag the October 12, 1992 management meeting, and that Unresolved Item 50-219/90-12-01 is closed.
The inspectors also questioned how the licensee will use the new average CTP values for determination of reportability due to steady state operation above the licensed CTP limit.
This issue is discussed in Section 1.2 of this report.
5 i
1
a
- 33bwe43 9*. elbd DCLOSURE 2 I d SSF Memorandum 3"W GPUN PosrrION ON COMPLIANCE TO Sept. 23, 1992 LICENSE LTwrT ON THE3tNAL POWER Dat s
R. V.
from:
Furia - Manager, OC Fuel Proj.
Matris Ocmp. Ctr.""
I Locaton:'
5411-9240034 Distribution i
It was recognized within GPUN that a consistant and taiform position for compliance with the licensed limit on thermal power should be established for its Nuclear Units. Over the past several months a-position was developed with input by appropriate oyster Crealt and TMI personnel.
'Ita ateme%t to this memorandum contains the established position and a discussion section for clarification.
It is intended that this document provid, the guidelines for establishing proceduras for operation at or nnar the licensed limit for thermal power.
F.ach plant la requested to review existing procedures and to revise them as necessary to insure that they provide guidance consistent with the position established.
If there are any questions or additional clarification is neoc..ad, please contact me.
. V.
R. V.
Furia Extension 7244 f
concurred By:
KnubEl censing & Reg. Affair Director
/and Attachment
l m.a c..u GPUN Position on Compliance with the License Limit for Tternal Powar at Steady-State Operation.
1.
For any steady state period (as defined by each plant) the average power level may not exceed the licensed limit.
The steady state period defined by each plant shall not be longer than eight hours.___
2.
An instantaneous heat balance power level limit shall be established for each plant that may not be intentionally exceeded. This limit shall not be greater than 1024 of the license limit.
3.
Immediate corrective actions must be taken when the power level violates eiths' of the above.
The reactor shall never be intentionally maneuvered above 4.
its licensed limit for the purpose of increasing the average power level over the steady state period.
DISCUSSION It is common for reactors to experience small variations in calculated and/or actual power level when operating at staady state.
Normal fluctuations above rated thermal power are neither license violations nor outside the design basis.
If operators are expected to control any indication of power level to less than or equal to the license limit, then normal fluctuations above rated power would not be permitted during steady state operation.
This would impose a de f acto derate in average reactor power.
- Howevar, it is acceptable to average the calculated power to minimize fluctuations and provide a suitably noise smoothed indication for in monitoring for compliance to the licensed thermal power use limit.
The period of tine chosen for averaging should be based on the frequency of heat balance calculations, control system limitations, etc. that will define a plant specific sinimum steady state period for evaluation against the licensed power. Each plant should determine this period of time and strive to maintain the steady stata power level at or below the licensed limit based on the smoothed power calculation.
A maximum of eight hours for this period is established based on NRC enforcement guidance to their inspectors.
During the approach to the license power limit, or adjustments to maintain power at ths licensed limit, small overshoots of the licanned limit may result due to limitations or dead bands in the control system.
These overshoots may be self correcting or may require operator action to correct the power level.
In either case, the core thermal power may exceed the licensed limit ~during the maneuver.
Operation in this manner is acceptable provided the instantaneous indication of thermal power remains below the 102%
~
_ _ # ' & '. 4,gdiiL %
n c.
Licensing basis limit, and the smoothed indication of power does not exceed 100% When averaged over the plant specific time interval.
If either limit is exceeded, immediate action must be taken to rastora operation to within limits.
The licansed thermal power is a steady state limit such that small power overshoots are acceptable during power maneuvers. However, steady-state operation above rated power is not permissible, and requires corrective actions to reduce thermal povar..
While, from a licensina basis consideration, power overshoots up to 102% of rated power itr a short duration may be acceptable, the types of power maneuvers needed to maintain full power should not result in povar transients that approach 102% of rated povar.
Therefore, each plant should establish a movinum value for an acceptable instantaneous power overshoot based on the control systems and types of power maneuvers parformed to maintain full power.
Tnis value must be less than 102%
and should be procedurally establiabad as a value not to be intentionally exceeded and require immediate corrective actions if it is exceeded.
This position is intanded for steady-state operation and the minor power maneuvers associated with maintaining full power.
Unplanned or inadvertent power excursions over the licensed limit beyond the type discussed above should receive appropriate review and evaluation.
REFERENCFS 1.
August 22, 1980, NRC memorandum from E.L.
Jordan to RCENSB Region Chiefs, Discussion of Licensed power Level.
- 2. November 9, 1990, NRC Memorandum form J.A. Calvo to F. !!abdon, Enforcement of License Power Limit.
- 3. August 4, 1981, General Electric Memorandum from R. G. Saranka to J. L. Rash, S&I4 Position on Thermal Power Limit Compliance.
TUTR P.05
n' i
NUCLEAR POWER SYSTEMS DIVISION MEMO l
l TO:
R. G. Sere ka DATE:
8/4/81 JLR-813201 FROM:
J. L. Ras REQUIRED RESPONSE DATE:
SUBJECT:
S&LO POSITION ON THERMAL POWER FOR:
ACTION l
LIMIT COMPLIANCE
Reference:
NRC Memorandum SSINS 0200, E. L. Jordan to distribution, i
" Discussion of ' Licensing Power Level' (AITS F14580H2)",
August 22, 1980 S&LO POSITION This memo documents Safety & Licensing's position on compliance with license limits on thermal power.
This position is:
i 1)
The average power over an 8 hr. interval may not exceed the license limit,
)
2)
The instantaneous power may not exceed 102% of the iteense
- limit, 3) ta ur t on
". overshoot" may not exceed the equivalent 4/-
of 102% power for 15 min.' i.e.,101% for 30 min,100.5% for 4 60 min...,,
~'
% w.,,. ;
4)
That the best way to comply is to customize tne process computer filtering algorithis to minimize the fluctuations in calculated power (see discussion) and attempt to keep the filtered calculated power within the license limit.
A filtered calculated power greater than the license limit would require corrective action if:
a) 1, 2, or 3 above would be violated, or b)
There was reason to believe that the situation would not be self correcting.
DISCUSSION It is comon for BWR's to experience small variations in calculated and/or actual power level when operating at steady state.
These fluctuations lead to the question of how to monitor for compliance with
\\
the licensed limit thermal power, i.e., is 't necessary to dorate such i
that instantaneous power never exceeds rated or is it permissible to average over a period of time.
x
~
Reference 1_(copy attached) documents the NRC position on this sattar.
Following discussions with'NEED, S&LO has concluded that.from a safety viewpoint, the NRC position is adequate.
R. G. Serenka August 3, 1981 Page 2 m
While S&LO believes that the NRC position is adequate, we believe that the best means of compliance is to use the process computar digital filter en fesowater flow to minimize fluctuations in calculated power and to take corrective action, as requitec, when the calculated filtered power exceeds the license limit.
TECHNICAL BASES Standard process computer software for computing core thermal power is designed to offer customers the flexibility needed to deal with process noise, while minimizing the impact on efficient plant operation, and j
remaining within plant design bases.
When operating under steady con-J ditions core power is detemined by performing a reactor heat balance.
i The most important measured input to the heat balance is feedwater flow.
I Before undergoing conversion to a digital process computer input, the j
analog feedwater flow signal is put through a thres second time constant j
filter.
Further, after being digitized a digital filtering technique is j
also applied.
This technique gives a feedwater flow value that is the weighted average of the previous digitally filtered result and the last measured feedwater flow value. The weighting scheme is controlled by a process computer variable called EMAV.
ENAV is nominally set to a value of 12.0, <hich results in cigital'sTgnal"7TTtering with a 60 second effective time constant.
Customers should feel free to experiment with different values of ENAV to fit their particular plant operations.
Larger values of ENAV result in longer effective time constants and would be desirable curing steady operations.
During non-steady operations (e.g., increasing power) utilization of the above technique could result in non-conservatively low power estimates.
Therefore, when the last measured feedwater flow differs by more than 3%
from the previous digitally filtered result, a backup procedure to determine core thermal power is automatically used.
The backup procedure is to use the measured Aversge Power Range Monitor (APRM) core power signal.
The APRM analog signal input to the process computer is also put tnrough a 3-second time constant filter.
However, no digital filtering is parformed.
Therefore, the computer power determined from APJtM inputs represents a suitably noise smoothed measurement during non-steady core cperations.
As core operations approach steady conditions following a core power change the feedvater flow measurement will again become steady, and the process computer will automatically switch back to the heat balance method of determining core power.
The not result is that the parameter LNAV can be chosen to provide the degree of smoothing required for specific plant operations.
The software will assure process noise is appropriately considered in the power level calculation per plant design / licensing bases.
Action should be taken per the recommendations in this letter when a value exceeding rated power is calculated.
h The foregoing was prepared primarily for your use in responding to questions from Taiwan Power Co.; however, it is completely generic and a;ay be used to respond to similar inquiries from othar SWR owners.
JLR: hec /1776-77
PBAPS ADM 4TH FLOOR
,E3-12 '92 LED ;5;23 I;:FEAFS ;P JTH F 3 g;,,,, 7;7 156 'I22
- 757 324
%,\\
UNITED STATES 3
j NUCLEAR REGULATOAY CCMMISSION AWN 07eN,. D c.:0$$5
[f 1
e...
November 9, 1990 PEM0P.ANDUll TCP t Fred Heldon. Project Director (138 3)
Project Directorate 1Y-1 Division of P.eactor Projects. I!!, IV. y r.nd Special Projects TECM:
Jose A. Calvo, Chief Technical Specifications Branch Division of Operational Fyents Assesst.ent, NRR SUEJECT:
ENTORCEPIfiT OF LICENSE POWER L1}l1T KEFEFEECE:
August 22,1980 Memcrandun from E. L. Jordan to FC&h53 Regien Chiefs, Discussion of Licensed Power leYe1 Plant }:ece:
Sequcyah Units 1 and 2 Docket Hos:
50 307.50 328 TI.C 80.:
76679 Project Directorate:
Project Directorate Freject Managar J. Donohtw Review Branch 0TSB/00EA Review itatus:
Corp 1ete Inspection Reports 8912 and E915 for Sequoyah L' nits 1 and 2 documented a concern raised by the resident inspectors regarding the cer. trol of therszl power with respect te the licensed limit. The specific cer.cern raised in the
, inspection re>crts was intentienal operation by)the Sequcyth licensee at aus power level 6)ove 1005 rcted thema) power (RTF referenced cemorandum. The f acility license liriits the renctor core power level to "not in excess of 3411 regawatts thermal'. The question is whether routint operation according to the referenced memorandum is permitted. The referenced memorandum provides enforcement cuidance to inspectors dealing kith the degree to which licensed therrel power 1:ay be excences before enforcement action is appropriate and it reither condones nor authorizes routine operation
, of the facility in excess of the rated therral power Itvel specified in the license. The referenced mercrandum is not intended as operational guidance fer J
licensees.
We recognize that thermal pcwcr ray fluctuate without operator action.
Povertheltes, operators are expected to contro,1. steady state power to Ints than or equt.1 to 1001 Rtp Norr.a1 fluctuations are neither license violations nor.outside the design basis. Licensees should not intentionally operate aboYe 1001 RTP steady state and should taks. corrective action to reduce therril pcwer when it is discovered to be above the licanted limit.
Ai' l
4:5-;2 'E2 :.E: :s;<a ;;:esa s acn 4 - -,: g g,_.,;. ;-,,3,, ;
l 1
Fred Hebdon 2
This concludes cur work on TAC 7EC79.
~
Oil!GINAl. !!GED B710:E A. CALVO Jose A. Calvo, chief Technical Specifications tranch Division of Operational Events Assessment, nR cc: J. Eenchew s.
Cl5TEIBUT10h (Giralfile OTSB R/F COEA R/F A.Thtdani L. Phillips C.E. Rossi i
R. Jenes J.A. Calve R. H. Lebel R. L. Emch R. Wharten T.R.Tjader l
h 00 CUP.ENT NME: MEP 7 FRED HEE 00N I cT!r:DotA:t1B OTSB 00EA:ttRR SRI : !T C:0T18:LCES:HAR C 1RTjeder FMLobal.
RJcn JACalvo 11/7/90 -
11/r /90 11/ /9011/g/90
p.s c /, ad 1,
h
I-
~
~
AUG 21 ISSO
$s:H5 #0200 a
I
,~
pqy;rjn;p. FO?.:
E. J. Brunnar. Chief, RC&N58, RI R. C. Lewis, Acting Chief, RC&MSB, R13 s
i R. F. He{shman, Chief, ROLMSB, R111 G. L. Madsen, Chief, RO&N58 RIV J. L. Crews, Chief, R0&M58 _RV
. -E..L. Jordan..Assi:;. tant Direrter for Technical Pr.agrams.
_ _..__ r a :y.:
Division of Reactor Operations, Inspection, IE v
SUBJECT:
DISCUSSICH OF "t.1 CENSED POWER LEVEL" (AITS F14540H2) i Dating ba:t at leas 1, to 1974, there have been many lengthy " discussions" regercing the exact meaning of " full, steady-state licensed power level" (and simi,3arly werded power limits).
We do not believe the real safety banefits thet.might ce derived from an NRC-vide agreement would be worth the further expyditure of ranpower in meetings, etc. that would be required to achieve a consznsus.
Ve do realize that some common uniform casis for enforcing maximum licensed 1
powtr is noteed by I&E inJpectors.
Therefore, until and unless an WP.C-wide
)
position is put f orward and agreed upon (and as stated, I&! does not propose r
i to ripitiate proceedings to that end), ILE will use the following guidance.
The-3verage power level aver any eight hour shift should not exceed the." full.
o i
The exact stesey-state licensed power level" (and similarly worded terms).
eight hour periods defined as " shifts" are up to the plant, but should not be varied from day to day (the easiest definition is a norsal shift sanned by a It is pemissible to briefly exceed the " full, staaey-q particular " crew").
L In no state licensed power level" by as much as 22 for as long as 15 minutes.
case should 102% power be exceeded, but lesser power " excursions" for longer periods should be allowed, with the above as vidance (i.e.,1% excess for 30 There are no limits on minutes, 1/2% for one hour, etc., should be e lowed).
the nucher of times these " excursions" may occur, or the time interval that nun noW that 1.ha steve misrL nt
-musCT@Yfa1.s sucn =excurnions, regarding the eight hour average power will prevent abuse of this allowance.....___
COWTACT: "M.-1t.-Woods IE 49-21120
~
- I*
@U 32 Ogg within the licensin basis and, therefore, d to ba captthekous d it is $1so f air to the util ties and the,lr rattptytri.
. J rdan Allistant Directer for Te
.icalfragtms Diy{ ion of Reacter Operattent Inspection Office of lospection and Inforcasent cet P., C. 0, Young IE S.,E. Bryan, IE gy. Eisenhut, HRR, HRR 70.Rossi 9
i i
=
e
]
9 a
e I
s g
0 O
e I
s e
e 6
=
3
ENCLOSURE 3 i
Nuclear memorandum subject:
CORE THERMAL POWER AVERAGES Date:
March 8,1993 From:
E. E. Bradley, Jr.
tu adon:
Oyster Creek Core Engineer 5414-93-0018 To:
R. J. Thompson, Jr.
Core Engineering Manager i
Prior to the implementation of the core thermal power averaging program at Oyster Creek, a detailed analysis was carried i
out to determine the most appropriate times for the averaging. The purpose of the analysis was to fm' d two time periods that would significantly smooth system noise yet also provide adequate response to actual changes in power. The first would have a faster reaction time to power changes. This average would only be expected to exceed the full power limit during changes in power such as recirculation flow bumps. The second would be a longer average that would never exceed the full power limit. This average would be long enough to permit operator action once the shorter average had exceeded the limit.
r Attached, please find a sample of the analysis. Over periods of steady state, a statistical analysis of several possible averages was performed. Some of the typical results were:
INSTANT 5 MIN.
10 MIN.
15 MIN.
30 MIN.
1 HOUR j
i Variance 0.64836 0.09766 0.04814 0.03169 0.02064 0.00720 l
1 Sigma 0.80521 0 31250 0.21941 0.17801 0.14365 0.08483 2 Sigma 1.61041 0.62501 0.43881 0.35603 0.28731 0.16 % 5 For the final selection, it was assumed that the operators could control the power to within 0.5 MWt. Therefore, the 15-minute aserage was selected as the shorter average. The 10-minute average wrs not consistently below 0.5 at two sigma.
The 1-hour average was selected as the longer average. The reasoning for this decision can best be seen on the graph of the instantaneous,15-minute, and 1-hour averages..At point 2701, when the 15-minute exceeds the limit momentarily, the 1-hour remains below the limit as long as power is reduced. This was not always true for the 30-minute average.
Longer averages of 2,4,6, or even 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> were considered. These averages did not provide any significant improvement in the variance or standard deviation values. Also, these averages include the negative aspect of longer response times.
Please contact me if you have any questions.
E. E. Bradley, r.
Extension 4097
/njm/CE93-018.MEM Attachment W. H. Behrte - Technical Functions Site Director (OC) cc:
G. R. Bond - Director, Nuc! car Analysis & Fuel R. V. Furia - Manager, Nuclear Fuel Projects (OC)
~
S. Kowkabany - Instrumentation Manager S. Levin - Director, Operations & Maintenance (OC)
.S. T. Mueller - Plant Analysis Manager (Acting)
J. Sedar, Jr. - Engineer III, Nuclear
Instantaneous, 15 Minute, and 1 Hour Averages.
1938 1936 -
1934 --
1932
'i l
I 1930 1
I i
l 4 i g
l r,:
1928 -
l es 1924
/
1922 1920
.l 1918 1916 1
301 601 901 1201 1501 1801 2101 2401 2701 3001 3301 Umit Instant.
15 Min Avg.
1 Hr. Avg.
.... a w,... w t
a 4
e 7
8 I
i l
n 8
e E
6
<W Le8>4 l
e Qu Q
Q J.
8
?
=Q U
o I
f/3 O
N J
l o
I f
I
/
4 6
N e
y o
o e'
g 6
ENCIOSUR N Nuder o'$'"" c"z'" "oc'=^^ 55"ra^'2"a
"==6 r STATION PROCEDURE 202.1 Title Revision No.
Power Operation 36
+
r 5.4 Perform power level changes as follows:
5.4.1 Immediately prior to making a power level change >10 percent, record the following initial conditions in the Control Room 3-Log:
Reactor power 2
2 Turbine generator output
- Recirculation flow Reactor vessel level and pressure Time of power level change 5.4.2 Notify the system dispatcher.
5.4.3 Make the power level change in accordance with this procedure and the instructions of the core Manager, or his designee.
EQ21 The " instantaneous" indication of core thermal power on the Plant Computer System uses a rolling one minute average of feedwater flow, and will lag behind the power change by a few minutes.
)
carm0N
~
w.
'n' The 15 minute and one hour averages of core thernal power
- have much longer response times. The use of these indications to monitor core thermal power during non-steady state conditions could result in operation above the Licensed Limit.
5.4.3.1 During the power change, monitor the following parameters to verify expected responses:
(0095P) 10.0
s OYSTER CPJ:EK NUCLEAR GENERATING Number g
STATION PROCEDURE 202.1 Title Revision No.
Power Operation 36 LPRM and/or APRM response Reactor pressure
- Steam line flow Turbine generator output Turbine control valve and/or bypass valve i
position
- Feedwater flow Core therrr.al power 5.4.3.2 IE the instantaneour value of core thermal power, as indicated by the Plant Computer System, exceeds the License Limit (1930 MWth) or any lower imposed limit by 5 Mwth, IggH reduce core thermal power.
5.4.4 Record the following final conditions in the Control Room Log for power level changes > 10 percent:
i Reactor power Turbine generator output
- Recirculation flow Reactor vessel level and pressure Time of completion 5.5 The following guidelines apply to prevent overshooting the License limit of 1930 MwTH, or the control valve capacity.
(0095P) 11.0 r
g gp OYSTER CREEK KUCLEAR GENERATING Nutsber STATION PROCEDURE 202.1 Title Revision No.
Power Operation 36 l
5.5.1 II the reactor power is greater than 95% of rated power QB the control valve position is greater than 90%,
THEN 1.
Withdraw control rods one notch at a time, i
i 2.
Limit recirculation flow changes to maximum step changes of 0.2 x 104 gpa.
3.
Allow power to stabilize in betweets i
i steps prior to proceeding further.
l
)
4.
Monitor Nuclear Instrumentation & Turbine Control System response closely during all reactivity manipulations.
5.6 The following operating rules shall be observed at all times during i
operation near rated power to ensure compliance with the Oyster Creek Operating License Limit of 1930 MWth.
H2E If indicated core thermal power, as calculated by the Plant Computer System, is limited to below the License Limit, then the i
following steps apply to the limit placed on the indicated PCS power.
5.6.1 II the 15 minute or one hour average of core thermal power is not available on the Plant Computer System, TER continue at Step 5.6.4.
(0095P) 12.0
1 OYSTER CREEK NUCLEAR GENERATING Number i
STATION PROCEDURE 202.1 Title Revision No.
Power Operation 36 l
NOTE Fluctuations in the instantaneous core thermal power values above and below the License Limit without operator action are acceptable and expected and shall be controlled as noted below; these fluctuations will not be seen in the averaged power levels.
5.6.2 R
any cf the following limits are exceeded,
)
The instantaneous value of core thermal power as
)
l calculated by the Plant Computer System system 1
exceeds the License Limit by 5 Mwth.
The 15 minute average value of core themal power as.
calculated by the Plant Computer System system exceeds the License Limit.
The one hour or longer value of core thermal power as calculated by the Plant Computer System exceeds the License Limit.
IH]l2[ reduce core thermal power.
5.6.3 R
the instantaneous core thermal power exceeds the License Limit by 10 MWth OR the one hour or longer averaged values of core thermal power exceed the License Limit, THEN the GSS will immediately notify the Manager, Plant Operations giving details of the highest power reached and the duration of this condition.
(0095P) 13.0
4 1
glgg OYSTER CREEK NUCLEAR GENERATING Number STATION PROCEDURE 202.1 Title Revision No.
Power Operation 36 i
5.6.4 II the moving average displays are not available, I
igg monitor core thermal power using the instantaneous value i
displayed by the PCS.
Maintain power at or below the l
License Limit.
5.6.5 II the moving averages and the instantaneous values are not available, Igm monitor core thermal power using a manual heat balance calculation completed every shift. Maintain core theznal power at or below the License Limit.
5.7 A manual heat balance shall be performed weekly in accordance with J
Procedure 1001.6, to verify the accuracy of the Plant Computer System heat balance. When completing the manual heat balance use method two.
5.7.1 II the reactor is less than 90% of rated power and the manual and PCS heat balance differ by more than 20 MWth i
or more, 1
IHH Perform another heat balance calculation when reactor power is above 90% of rated but prior to exceeding 95%
of rated power.
5.7.2 II the reactor is greater than 90% of rated and the manual heat balance exceedt. the PCS heat balance by more than 10 MWth, igg perform an additional heat balance to confirm that the difference exists.
(0095P) 14.0
^
ac wstu 5 g
CORE UfRI M 1867.75 E RX PESS 1828.12 PSIG H R0li PERCENT RATED M
%.77
'4 FEEDFLOW 6.94 ItMR til ELECTRIC 627.38 E FEEDTBP 308.75 DEG F REQ'D N SP 1938.80 M
STEfft FLOW 7.85 1!MR R 1004' SP 1938.00] E T0TRECIRCFLOW 15.92 104 GP!1 BJETED N
%.Tl
'4
@G RECIRC BP 527.29 DEG F HUCLEfR IHSTRltEHTATION CORE M P0ER WEREES N'4 Gff BYPRBSED R1 95.93 1.0061 W
15-111H 1867.97 M N2
% 79
.9968 M
IM 1867.97 M N3 96.26 1.8003 W
86 1867.44 M N4
%.04 1.0049 W
M5
%.64 1.9987 W
^
N6
% 68 1.9987 M
N7
%.53 1.9968 H0 N8
%.49 1.8003 W
CACKUP:$
W fiPSR1R7[ TDT ENCLOSURE 5 - Sample Plant Computer Heat Balance Information Display
.