ML20054J441
| ML20054J441 | |
| Person / Time | |
|---|---|
| Site: | Beaver Valley |
| Issue date: | 06/14/1982 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20054J439 | List: |
| References | |
| NUDOCS 8206280685 | |
| Download: ML20054J441 (9) | |
Text
- __
p urg'o UNITED STATES
. ["
NUCLEAR REGULATORY COMMISSION WASHINGTON, D C. 20555 a.,
a
(
/
l SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATED TO AMENDMENT NO. 51 TO FACILITY OPERATING LICENSE NO. DPR-66 DUQUESNE LIGHT COMPANY OHIO EDISON COMPANY PENNSYLVANIA POWER COMPANY BEAVER VALLEY POWER STATION, UNIT N0. 1 DOCKET N0. 50-334 Introduction By letter dated February 23, 1982, Duquesne Light Company (the licensee) informed NRC that it had completed its review of the safety and operational aspects of the design for Cycle 3 of the Beaver Valley Power Station Unit No. 1.
Also, by letter on the same date the licensee submitted a request for amendment to the Technical Specifications for Cycle 3 and beyond.
Since that time, the licensee has amended the February 23 request by letters dated May 5, May 26, June 1, and June 10, 1982.
In addition, the staff has had several conversations with the licensee and received additional infor,
mation submitted by letters dated April 21 and April 26, 1982.
This evalu-ation addresses the proposed Technical Specification changes.
Power Range Neutron Flux - High Positive Rate and High Negative Rate Trip Setpoints The licensee has requested changes in the Power Range, Neutron Flux, High Positive Rate and High Negative Rate values in Table 2.2-1 Reactor Trip System Instrumentation Trip Setpoints, to the original values. The present values were an interim solution when a potential problem following dropped rod events was first discovered in early 1979.
This interim solution was reevaluated later in 1979 and an interi.m procedural position limiting rod insertion when the plant is above 90% power in automatic control was adopted.
Westinghouse has reevaluated the setpoints of the flux rate trip setpoints considering the 1979 interim procedural position and has found that the original setpoints are appropriate.
On this basis, we find the requested change to Table 2.2-1 of the Beaver Valley Technical Specifications acceptable.
' 82062806s6 s20614
~'
~
PDR ADOCK 05000334 P
FAH and partial power Multiplier The 1icensee requested the FaH value of 1.5355 be changed to 1.55.
The 1.5355 value included a penalty for rod bow.
Rod bow is now accounted
^
for by a burnup dependent rod bow penalty which has been included in the proposed Technical Specification change. We find this change to account for rod bow acceptable.
The licensee also requested that the partial power multiplier be changed from 0.2 to 0.3.
However, insufficient justi-cation has been submitted by the licensee and we cannot accept the change of value of the partial power multiplier, which thus remains at 0.2.
Sufficient justification has been-submitted for the FAH value of 1.55 and we find this acceptable.
Fxy F
A change to the Technical Specifications on xy was requested to remove the pcle-dependentvaluesof Fxy as a function of core height and provide these xy values by means of a Peaking Factor Report.
It is anticipated that Fxy will change from cycle to cycle and this Technical Specification change would eliminate the necessity of making a Technical Specification change for each reload. A similar change has been approved for Farley Units 1 and 2.
The Radial Peaking Factor Limit Report will~be submitted at least 60 days prior to initial criticality for each cycle.
This report was submitted on March 9,1982 for the Beaver Valley Cycle 3, meeting, the 60-day require-ment.
Based on the information submitted by the licensee, we find the proposed change involving Fxy acceptable.
The licensee requested that the partial power multiplier be changed from 0.2 to 0.3.. for Fxy al so. We consider this change to be inappropriate and since the licensee has not submitted adequate justification, this change has not been found acceptable.
Control Rod Position Indication System
Background
The. characteristics of interest are of two general types.
First, the channels have non-linearity in the steady-state response. Second, the channels display a time-dependent (transient) response due to thennal effects in the detector assembly.
A typical steady-state calibration curve is an arc-shaped curve, with the indicated position low at the near full-in and near full-out extremi-ties and the indicated position high in the mid-travel region. For. most rods, but not all, the Zero and Span adjustments allow the steady-state calibration curve to be fitted within the + 12 steps acceptance band under hot ::ero-power conditions.
Prior to the start of operating cycle #2, the licensee reported some difficulties in attaining the required perfomance from the analog system of rod position indication. The fiRC approved (Reference 9) certain actions as an acceptable solution for that operating cycle. Those actions were interim; long-term solutions were to be arrived at by the licensee during the 12-18 months duration of cycle #2.
The rod position detector is a linear variable transformer consisting of primary and secondary coils alternately stacked on a stainless steel cylindrical tube. An extension shaft from the rod drive mechanism extends up into the tube and serves as the variable " core" for the transformer.
With a constant a.c. current source applied to the primary windings, the position of the rod drive extension shaft changes the primary to secondary coupling and produces a secondary voltage that is directly related to rod position The transient response for the RPI's is typically of the "over shoot" type. That is, if the rod is being pulled out, the RPI indication will show a greater withdrawal and later settle (at themal equilibrium) back to the steady-state value; if the rod is.being inserted, the initial indication is greater insertion than actual. The magnitude of this thermal transient response appears empirically to be insignificant in the region of the lowest one-third of rod travel. However, near the fully withdrawn positions, this transient response at some plants can be as great as 25 steps. The time constant of the themal recovery toward the steady-state value varies with rod location radially across the core and has values between 10 and 15 minutes.
" Settling Times" (or " soak times") of 20 to 45 minutes have been observed before steady-state is reached.
Evaluation The February 1982 submittal reported that a core physics analysis had been performed for cperating cycle #3 which indicated that a wider rod misalignment tolerance may be adequate (i.e., + 32 steps vs.124 steps).
Staff review of this approach indicated technical difficulties.
The April 1982 submittal provided additional technical information and modified the approach to provide a three-step tolerance, as had been suggested by W.
The modified approach proposed a wide instrument error tolerance (+ T6 steps) near the full-in position (0-30 steps), a
" reference curve" for the intemediate positions, and the standard 1,12 steps tolerance for the 150-228 step range. The staff considered this approach and concluded that it would be awkward and could introduce human factors engineering questions that could outweigh the benefits of the approach.
The May 5,1982 submittal withdrew both expanded tolerance approaches and proposed certain design and operational changes, within the original 1 12 step instrument error allowance. Our review of this proposal indicated
. ~ -... _.. ~. - ~ ~.
. ~
.~
general acceptability, but three questions remained. The licensee's May 26, 1982 submittal addressed these concerns. The licensee provided additional infomation on June 1,1982, and revised Technical Specifications (and bases) on June 10, 1982.
To address the rod position indicators (RpI) concerns, the licensee has proposed a multi-faceted program. Each aspect is described and discussed below.
1.
The group demand counters will be the primary indicators for precise rod position infomation, with the individual analog RPI's display-ing general infomation regarding rod motion, especially during the first hour. We have previously approved, for another W PWR facility, this shift of which rod position indicator is to be considered the primary indicator (see References 11,12). The bases for the approval are (a) that a potentially misaligned rod that is undetected for an hour or so is not an unacceptable risk, (b) the operating experience over the years with the control rod drive system and the demand counter has indicated a very high reliability, while the experience with the analog indicators has been plagued with less-than-desirable performance, and (c) the demand counters can be checked periodically to confim that the rods did in fact go to the position requested by the demand counters. We find this approach acceptable for this facility also.
2.
The licensee has proposed to develop " custom made" meter scales for the analog indicators that exactly correspond to the individual cali-brations of the analog position channels. These custom made scales will, in effect, eliminate most of the steady state errors that are due to the non-linearities of the detector response. We believe this approach is sound and desirable, and should help restore operator reliance in the analog indicators.
3.
After the individual rod position channels are calibrated at hot, zero-power conditions, the steady-state calibration shifts slightly as full power is approached. These shifts are predictable and reproducible, and have values of three to six steps (i.e.,1 to 2-1/2%) depending upon the radial position of the rod in the core.
The exact values of these shifts must be determined empirically for each rod.. We are allowing the licensee to make minor corrections to the channel calibrations after reaching full power to accomodate the steady state shifts due to power ascension. We are requiring, since the calibration adjustments (zero and span) are interdependent, that following any such correction, the channel will be tested at least one intemediate position and at least one low position t~o confim that the overall calibration has not been adversely affected by the correction.
l 4.
The deviation between the position indicated by the demand counter and by the individual analog position channels will be periodically
m..._._
.._m.._ m.
m..
m m
-_.m 5-monitored to detect any potentially misaligned rod. When the plant process computer is available, the Rod Position Deviation Monitor will provide continuous and automatic monitoring. Additionally, the deviation will be monitored manually by the. reactor operator on at least a daily basis, as a backup cross-check. When the automatic monitor is not available, the formally-required manual surveillance will be increased to a minimum of every four hours. As a matter of practice, the operators will be monitoring the rod positions much more frequently.
If this deviation monitoring should indicate that a rod is possibly misaligned, the primary detector' voltage of.the position channel'will be measured immediately (less than 15 minutes) to confim the validity of the indication.
If the primary voltage indicates a rod position that is different from the demand counter by 12 steps or more, the rod will be declared to be misaligned and the appropriate Action Statement of the Technical Specification will be invoked. This level of surveillance on the potential deviation of the rods and planned actions are the maximum practical and provide timely protection for the core physics parameters.
During our review of the licensee's proposed program, we identified 3 additional questions.
First, while the custom made meter scales will improve the accuracy of the rod position.information presented by these meters, the voltage inputs from the position channels to the process computer which generates the Rod Deviation Alarm will continue to suffer from both steady-state non-linearities and thermal tr' nsient responses.
a This condition is expected to generate frequent spurious control-room alams, which are quite undesirable. The licensee has determined that the computer has the capacity to use a curve-fitting process in deter-mining rod position from the input voltage from the analog channels.
The licensee has therefore agreed to provide a curve fit for each rod of up to as cluch as a fifth-order equation, as may be necessary to achieve a curve that matches the calibration curve within + 2 steps error. The use of curve-fitting eliminates steady state error from causing spurious alarms. Alarms generated by thermal transients are discussed further below.
The use of the curve-fitting has previously been reviewed and approved for another }{, pWR (see Reference 11) and its acceptable for this facility also.
Second, are there methods and procedures to demonstrate the accuracy of the group demand step counters?
It has been generally presumed that the accuracy of these counters is probably no worse than + 2 steps. The basis for this value is the possibility that the mechinical counter could skip one step as a group is withdrawn and could skip another step as the group is re-inserted, resulting in a possible 2-step error at the end of the rod motion cycle. The licensee and Westinghouse have stated (Reference 13) that there are no physical provisions in the design of the rod drive control system that could be used on a practical basis to prove the 2-step accuracy presumption. However, the licensee has proposed two surveillance actions intended to provide assurance that the counters are operable and within reasonable accuracy. By comparing the twa counters within a bank and by comparing the mechanical counters to an electronic counter, differencies due to mechanical action of the counter
can be detected. We have detemined that this level of surveillance is the maximum practical available for this system design, and does provide reasonable assurance both that the most likely source of error will be detected promptly and that no gross errors are present.
Third, manual surveillance action of the analog rod position indicators is to take place daily after 1-hour soak time following rod motion, to allow thermal equilibrium to be established. Our question was that, realizing that with the automatic reactor control system in operation the rods are typically re-adjusted a few steps every few minutes, can the manual surveillance actually be performed? Upon reconsidering this point, the licensee has stated that, above 50% power, rod motions are usually small and are, hence, not expected to induce significant themal transients in the analog rod position channels. Therefore, above 50%
the allowance of a 1-hour soak time is not needed. Below 50%, a 6-step limit (" absolute" value) is placed on rod motion during the soak time.
This band of 6-steps is ample for typical automatic rod motion and still retains at least quasi-equilibrium conditions at the rod position detectors.
Additionally, the small rod motions that typically occur above 50% power are not expected to generate false Rod Position Deviation alarms, and hence spurious alanns due to transient effects should not have a signifi-cant impact on the operator. We believe these adjustments in the proposed Technical Specifications will make them more workable, provide reasonable operational flexibility and retain the original safety intent.
After resolving these three questions, we recognized that we may have made the computer-generated Rod Deviation alarm so quiet that it is no longer self-testing and could, in fact, be in a failed-silent condition for a long time. Therefore, periodic functional testing appears necessary.
Typically, alarms are tested on a monthly basis. The licensee has proposed to test this alarm on a weekly basis.
In consideration of typical plant process computer availability factors across the nation, we believe that weekly testing is appropriate for this feature.
The test will be performed by injecting a test signal either on the demand side or on the analog position instrumentation side to demonstrate that the plant computer remains capable of recognizing a deviation of 12 steps or more.
We are continuing to allow that detector primary voltage measurements be used in lieu of an operable analog rod position indicator channel.
However, in view of the system improvements that have been made, we are returning to the original limit of 1 channel per bank for which voltage measurements are acceptable. Further, we believe that the number of potential Licensee Event Reports will not be an onerous burden. The licensee has agreed with the reasonableness of this step.
Along the same line, we are requiring that, if the. manual surveillance of the analog indicators or the automatic deviation monitor indicates a misaligned rod, a confirmation measurement will be made by measuring the channel primary voltage prior to declaring the rod to be misaligned and invoking Action Statement 3.1.3.1.c.
. Our review also encompassed the r.eed for rod position indication during plant-shutdown modes of operation. We are requiring that, in addition to the demand counters, at least two RpI's per rod group be ope able. The purpose of rod position infomation during shutdown modes is to protect the shutdown margin and hence prevent an inadvertent criticality incident.
We recognize that with the reactor system substantially subcritical during the shutdown modes'(i.e., K 1.0), that given the noise level that typically exists on the source range (startup) nuclear instruments, the increase in the subcritical multiplication due to withdrawing a shutdown bank may not be readily discernible.
Therefore a bank could be believed to have been withdrawn, given credit for shutdown margin calculation purposes, but not available for shutdown margin if i+ n not actually withdrawn. Therefore, analog indicators will serve to verify any real rod withdrawal. By requiring that RPI's in each rod group be operable, we assure that both groups in the bank actually received power and did in fact move. By requiring more than one RPI in each group to be operable, we assure that a single malfunctioning instrument will not give the operator incorrect and misleading infomation. By these requirements, we conclude that the rod position input to shutdown margin calculations will be valid.
The June 10, 1982 submittal provided revised technical specification changes to implement tha engineering solutions that had been reached between us and the licensee. The changes include a requirement that the plant computer (i.e., the automatic Rod Deviation Monitor) be operable.
In consideration of the availability factors for nuclear power plant process computers, we are granting an exemption to the standard specification 3.0.4 for this requirement. The purpose is to allow the plant to startup when the computer is not available, but anytime the computer is not available to provide its service, additional actions will be provided manually as compensatory measures.
The revised Technical Specifications also define the overall rod positicn indication system as' including the demand counters, the analog position
)
indicators, and the automatic rod position deviation monitor. This definition causes certain requirements to be shifted from specification 3.1.3.1 to specification 3.1.3.2, but results in a more logical arrange-ment.
The specific Technical Specification page changes co'vered by this evaluation are included as Attachment "A."
l Conclusion The licensee has proposed to retain the + 24 steps potential rod misalign-ment limit, which is consistent.with original licensing basis for this pl ant. The licensee agrees that the proper accuracy (potential uncertainty) to be associated with the rod position instrumentation is + 12 steps and hence remedial action will be taken when the indicated devTation is 12 steps or greater.
The licensee has proposed a comprehensive approach of design changes and operational changes to the rod position indication system. We conclude that this approach will provide a satisfactory method of determining the rod positions within the allocated accuracy of 12 steps. Further, the approach reduces the potential for misleading information and spurious alarms that could generate ambiguity and lead to erroneous actions by the reactor operator. Therefore, we conclude that the licensee's tech-i nical proposal and associated Technical Specification changes are acceptabl e.
Unless operational problems are encountered ~during this operating cycle,.
we consider the licensee's proposal to be an, acceptable long-term solution for this particular plant.
~
Environmental Consideration
-I We have determined that the amendment does not authorize a change in effluent types or total amounts nor an increase in power level and will not result in any significant environmental impact.
Having made,this determination, we have further concluded that the' amendment'involvas an action which is insignificant from the standpoint of environmentai, impact, and, pursuant ~to
~
10 CFR 551.5(d)(4), that an environmental impact statement or negative declaration and environmental impact appraisal need not be prepared in connection with the issuance of this amendment.
Conclusion We have concluded, based on the,cunsiderations discussed above,'that: (1) j because the amendment does not ' involve a significant increase in the proba-bility or consequences of accidents previously considered and does not involve a significant' decrease in93 safety margin, the amendment does not involve a significant hazards consideration, (2) there is reasonable assurance tAat the health and safety of the public will not be endangered by operation in the proposed manner, and (3) such activities will be conducted in 'ctmpliance with the i
Commission's regulations.and the issuance of this amendment will-not bef inimical to the common defense and. security or to the health and safety of the.public.
s a
Date:
June 14,1982
/
Principal Contributors:
/
J. T. Beard D. Hoffman M. Chatterton 6
k 1
^'**"m=
^ ' ^
9
, References 1.
Letter:
J. Carey (Duquesne Light) to S. Varga (NRC), dated February 23, 1982 2.
Letter:
J. Carey (Duquesne Light) to R. C. Haynes (NRC), dated March 9, 1982 3.
Letter:
J. Sieber (Duquesne Light) to S. Varga (NRC), dated April 21, 1982 4.
Letter:
J. Carey (Duquesne Light) to S. Varga (NRC), dated April 26, 1982 5.
Letter:
J. Carey (Duquesne Light) to S. Varga (NRC), dated May S, 1982 6.
Letter:
J. Carey (Duquesne Light) to S. Varga (NRC), dated May 28, 1982-7.
Letter:
J. Carey (Duquesne Light) to S. Varga (NRC), dated June 1, 1982
/
8.
Letter:
J. Carey (Duquesne Light) to S. Varga (NRC), dated June 10,1982 9.
Memo:
J. Olshinski-(NRC) to S. Varga (NRC), dated November 18, 1980, 11.
Memo:
G. Lainas (NRC) to T. Novak (NRC), dated April 1, 1982 12.
Memo-L. Rubenstein (NRC) to T. Novak (NRC), dated May 20, 1982 13,- Telephone Conference:
L. Sieber, K. Grada, (Duquesne Light);
J. Jenkins,(Westinghouse); J. T. Beard, D. Wigginton, P. Tam, D. Hoffman (NRC), May 28, 1982 Y
e 9