ML20077R217
| ML20077R217 | |
| Person / Time | |
|---|---|
| Site: | Grand Gulf  |
| Issue date: | 08/15/1991 |
| From: | Cottle W ENTERGY OPERATIONS, INC. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| GNRO-91-00149, GNRO-91-149, NUDOCS 9108220183 | |
| Download: ML20077R217 (8) | |
Text
_ - - _ _ - _ _ - _ _ _ _
Ent rgy Operations,ine.
Operations W. T. Cottle August 13, 1991 U.S. Nuclear Regulatory Commission Mail Station PI-137 Washington, D.C.
20555 Attention:
Document Control Desk
SUBJECT:
Grand Gulf Nuclear Station Unit 1 Docxet No. 50-416 License No. NPF-29 Technical Appeal of the Denial of the Amendment Request For Reactor Protection System Instrumentation GNRD:
91/00149 Gentlemen:
By letter and SER dated July 10, 1991 the NRC denied a proposed amendment to the Grand Gulf Technical Specifications which would have modified a surveillance requirement for the flow biased simulated thermal power (STP) trip instrumentation.
Entergy Operations wishes to appeal this technical decision based on the enclosed information.
We would be happy to arrange an appeal meeting at your convenience to discuss this matter further.
Yours truly, w
7~~' &w J0F/be Attachment cc:
(See Next Page)
G9108151/SNLICFLR gg/
9108220183 910815
'* I PDR ADOCK 05000416
// f P
PDR f
J
~.-.-_
. August 15,-1991 GNRO-91/00149.
Page-2 of 3 cc:
Mr. D, C,- Hintz, w/a Mr. J. Mathis, w/a Mr. R. B.-McGehee, w/a Mr. N. S. Reynolds, w/a Mr. H. L. Thomas, w/a Mr.-Stewart D. Ebneter, w/a Regional Adininistrator U.S. Nuclear Regulatory Commission Region II 101 Marietta,St., N.W., Suite 2900 Atlanta,-Georgia 30323 Mr. L. L. Kintner, Project Manager, w/a Of fice of Nuclear Reactor R 39ulation U.S. Nuclear Regulatory Commission Mail Stop 11021 Washington, D.C.
20555
[.
l L
l l<
C9108151/SNLICFLR
Attachment to GNRO-91/00149 Page 1 of 6 In t reduct.J on Reference 1, dated April 26, 1991, submitted a request by Grand Gul f t o modi fy the daily channel check for the flow biased simulated thermal power (STP) trip instrumentation in TS Table 4.3.1.1-1.
The change was requested to reduce unnecessary operator burden in t he cont rol room and was r odeled upon the NRC SER which approved the same change for the Clinton Power Station (Reference 2).
The request was denied by the NRC on July 10, 1991 vin Reference 3.
In the SER accompanying the denial of the Grand Gulf request, the Sta f f noted:
"The STP trip signal is used by the reactor protection system to scram the reactor to limit maximum rector power so as to maint.In the minimum critical power ratio above the appropriate safety limit for events such as loss of foodwater heating."
On this basis, the NRC concluded:
"The requested changes to the TS are denied.cause the deletion of the surveillance r equ ir emen t for t he flow biased simulated thermal power scram would permit an undetected and unnnalyzed increase in the scram setpoint."
As will be discussed in det ail below, the STP trip signal is not currently used by the Grand Gulf reactor protection system to mati tain ttCPR above the safety limit for any event, in fact., any increase in the STP scram setpoint is well analyzed for the applicable Grand Gulf event (i.e., no scram is assumed), and has been repeatedly reviewed and approved by the NRC.
Grand. Gulf FSAR_
Before reviewing the analytic basis for our position, it is worthwhile to briefly discuss the arrangement of t he Grand Gul f FSAR.
Chapter 15 of the FSAR contains a descript io i of the analyses conducted by the NSSS supplier (GE) for Grand Gul f's first cycle of operation.
A separate volume of the FSAR, entitled the Current Cycle Safety Analysis (CCSA), containa n description of those Chapter 15 events which were found to be bounding for the current cycle of operation.
For a particu.lar Chapter 15 event which has been reannlyzed and included in the CCSA, the appropriate Chapter 15 subsection int roduct ion includes instructions to the render to refer to the CCSI, for the current analysis.
Chapter 15 also contains appendicos which may update the Cycle 1 event descriptions.
Appendices include r er.n a l y s e s for applicable events to support such initiati.ves as the tiaximum Extended Operat ing Doma in (t1EOD).
Again, the Cycle 1 event description in the approprinte Chapter 15 subsection will refer the render to any applicable appendices for more current information.
Grand Gul f has employed th is practico (i.e., CCSA and Chapter 15 appendices) since the first FSAR update following Cycle 1.
G9108151/SNL1CFLR
Attachment _to GNRO-91/00149 Page 2 of 6
- Loss of Foodwater 11eatingjLFWill ArJ ysg 1
- The LFWil analysis is contral to 'the issues surrounding the requested chango to the STP trip-surveillance. requirement, and the subsequent NRC denial because historically it is thn only safety _ analysis for which STP trip credit has bnen taken.
- The analysis-mothodology for this: event changed significantly during Cycle 1.
As originally licensed for Cycle 1
'3n LFWil analysis credited the STP trip to minimizo the calculated severity r o transinnt.
An improved analytic methodology, first employed during cno MEOD analysos during.Cycin 1 and then repnoted -for each reload, assumes no-STP t. rip (or any trip for that matter).
-1.
Initial Cycle 1 17411 Analysis The initial Cycle 1 LFWil analysis is discussml in Snctlan _15.1.1 of the Grand _
Gulf FSAR.
With - respect to.t he STP trip (also called the tharmal power monitor t.rlp),
Subrection 15.1.1.2.2 of the FSAR notes:
"The thermul power monitor (TPM) is the primary protection system trip in mitigating the connnquences of this event.,
If there was no high. thermal power trip neram design availablo in the Grand Gulf plant. design, ' reactor scram during the losa of feedwater heating transient would ' occur when the r.eutron flux exceeds the high APRM flux
= scram set point.
Usually, the high APRM flux scram set point is higher than the high thermal power scram set. point by.approximately 6 to_a l-percent.
Thornforo, the loss of feedwater heating-transient would be morn ll severe without thn high t.hermal power trip scram design. -Tiils would lead t.o 'a higher operating CPR limit nud reduce the flexibility of plant L
operation."
In other words, rather than revising the CPR limit which would have unduly restricted plant operation, Grand Gulf chosn to credit the STP trip for Cycle 1.
In some sense, tho 'nood to credit tne STP t.rlp was an artificini situation i
neceSHltated by the lim.ltations lmposed by the GE computer Codes and analys15 methodology at that timo.
' Subsection 15.1.1.3.1 of the FSAR and its' associated references discuss the mathematical mcxleling (including a' point kinetics core model) employed for the initial-Cycle 1-analysis of thn LFWil event.
As discussed in the next section,
- point kinetics modnis y told extremnly conservativo results for the LFWil event.
l 2.
Change in Analysis Methodology
't Subsequent. to-the initial Cycin 1 analyses for Grand Gulf, GE (and the industry)
. changed Its approach to modeling and analyzing LFWil event.s.
Rather than a l
j transient analysis of the event, a steady state code was employed which examined l
steady state power levnt and cther core conditions beforn and af ter the LFWil.
l This approoch, which was used oc Grand Gulf for the MEOD and all subsequent
(
analyses discussed below, climinated the need to credit any plant trip (includlag the STP trip) In the coursn of demonstrating acceptahin CPR results.
G9108151/SNLICFLP l
m
+
' Attachment to-GNRO-91/00149 Page 3 of 6 Soma additional background conentning the change in LFWil analysis methodology may be hnipful.
A loss of feedwater heating event results in a decrease in feedwater temperature entering the reactor vessel.
This lowers tho temperature of the water entoring the core (i.e.,
increases subcooling), reduces the average core void fraction and thereby causes the corn average power to incrensn.
The change in subcooling enuses a void redistribution and a corresponding power redistribution. While the net. change is a total core power increase, not. all areas of the corn increase at the same rate.
This tends to flatten the radial power distribution.
Due to the stored energy and mixing of feedwater in t.he downcomer and lower plenum with thn recirculat,lon flow, the temperaturo reduction at thn core inlot (and, thernfore, the power increaso) occur relatively slowly - on the order of seconds to minutes. The relative slowness of this event allows the core to maintain a quasi-equilibr.ium in which the water temperaturo, core power and neutron flux distribution maintain their steady state rnlationships.
The LFWH analysis presented in (SAR Section 15.1.1 for Cycle 1 ovaluated this l -
event with a t.ransient thermal hydraulics ccxle using a point kinetics model to represent the core feedback mechanisms. The point kinetics model is a simplistic representation of the act.ual phenomena - it cannot represent. thn flux redistribution that occurs in this-event. and aust assume kinct.ics parameters that bound all the statepoints in the event.
cansequently, the point kinetics model results in a large overprodletion of the hFWil corn power incronse.
Because the LFWil event is relatively slow and thn transient is smooth with no sudden increases or decreases of important parameters, the conditions at the beginning and end of the event bound those throughout the transient. Thereforo, performing t he LFWil analysis with a three-dimensiona l quasi-steady state code is appropriate.
Essentially, a new steady state corn power level (and other steady stat.e core conditions) are determined based on the positive reactivity addition of the colder feedwater, and the of fect on CPR is calculated.
Since thn radial power red ist ribut ion o f f ects a re accounted for 1.n the methodology, the excessive core power overprediction and associated need to credit a trip is eliminated, 3.
Maximum Extended Operating Domain (MEOD). - Cycle 1 During Cycle 1 GE conducted and NRC approved analyses to support Grand Gulf operation in the HEOD.
A bounding analysis was performed using a etandard BWR/6 plant.
All abnormal operat.ing transients, including t.he LFWH event were examined.
In analyzing the LFWH event, GE employed the revised steady state approach discussed above.
l l
At the request of the NRC staf f, a plant specific LFWil event was performed using
-tha GE thren dimensional liWR Core Simulator (Reference 4) documented in the NRC approved GESTAR Amendment (Reference 5).
This analysis was submitted to the NRC by Reformco 6 and approved by the NRC by letter dated August 15, 1986 (Referenco 7),
j G9108151/SNLICFLR l
l t
. ~ - -...
,mm._7-,.
w Attachment to-GNRO-91/00149 Page 4~of 6
- The analysis was performed for the GE P8X8R fueled Grand Gulf Cyclo 1 core and'
-was shown to be bounded.by the generic tlys is.. Nolther the generic BWR/6 LFWH 4
analysis nor the Grand Gulf-specific Lh - analysis assum<d credit for the STP, or any.other, trip..
Thn results of these analyses were incorporated into Appendix 15D of the Grand Gulf FSAR (Rovision 2) in Decembor, 1987.
41 Cycle 2 Reload and LFWii Evout Analysis Grand Gulf contracted with Advanced Nuclear Fucis (ANF) to provide reload fuel 1
for Cycle 2 and to conduct the reload analyses necessary to obtain NRC approval 1
for the fusi cycle.
The. Cycle 2' reload analyses.for LFWil utilized the ANP 3-D XTGBWR steady stato coro simulator codo which is similar to t he GE code used in the HEOD analysis of the LFWil event. The ANF generic analysis of the LFWil event is documented in Reference 8.
At the -roquest of the Staf f, Grand Gulf submitted by Reference 9 a plant specific analysis of.thn -LFWil event utilizing the XTGBWR code.
This analysis provided-the steady state conditions of the core before and af ter the LFWil event and the resulting delta CPR. No credit was taken for the STP trip.
The Staf f's review and approval of thn Cycle 2 analysis is documented in
. Reference-10.
5.
Subsequent Roloads Tho.LFWH event has -been consistently analyzed for each Grand Gulf reload using a steady stato=approcch and NRC approved methodology.
In no instance was the STP trip (or any trip) credited for achieving acceptable CPR results.
In particular, the LFWil analysis for the current cycle of operatleu (Cyc16 5) employed a relatively new-ANF computer code (CASM0/MICRODURN), albeit the same steady state approach described-above which did not involve credit. for a plant
'. t. r l p.
In= approving the Cycle 5 reload, the NRC indicates-in Reference l' that:
"... a comp 1nte now analysis was run for the LFWii.
The LFWil was nualyzal with the newly approved MICROBURN-B/ANF following the approach previous.ly approved for Cycle 4, using an expanded GG1 data hasn."
NRC Conclusions in the Denia) SER In denying the requested TS change to modify the STP trip daily chanrol check the NRC's SER indicates-that the STP trip is necessary "... to limit maximum reactor power so as to maintain the minimum critical power ratio above the appropriate safety limit for events such as loss of foodwater heating."
-G9108151/SNLICFLR
s Attachment to GNRO-91/00149 Page 5 of 6
.This is a correct statement only for-thn initial _ portion of Cycle 1 at Grand L
-Gulf,.until operation _in-the MEOD was approved by the Staff.
As discussed in
' detail above, at all subsequent times the STP trip provided no, credited s9fet'y function for the LFWH or other events.
In other words, the new LFWil analysis methodology rnquired no limit on reactor power to maintain CPR above the safety L
limit.
L L
The SER concluded that the requested TS change was "... denied because the deletion of the surveillance requirement... would permit an undetected and unanalyzed increase in the scram setpoint."
L
- The analyses provided by Grand Gulf for MEOD and each cycle reload allow reactor power to reach a new, higher steady stata level and demonstrate ecceptable CPR results. This approach is equivalent to considering the _ complete failure of; the -
The analyses provided by Grand Gulf for HEOD and each cycle reload allow reactor power to reach a new, higher steady state level and demonstrat.e acceptable CPR results. This approach ia equivalent to considering the_ complete failure of the l
STP (ar.d any other) trip.
Far from being unanalyzed, the Grand Gulf and BWR l
Industry approach to analyzing the need for the'STP trip has been detailed, complete and. extensively reviewed by the NRC, Previous NRC Approvals for Similar Requests l'
.To our knowledge, at least two other BWRs have received NRC approval to delete the daily channel check for the STP trip - Fermi-2 (Reference 12) and Clinton Power Station (Reference 2).
In preparing requests.for changes to Technical Specifications, it is Grand Gulf's practice to determine if similar. requests have been previously made and approved by the NRC.
Infdoing so, our intent is to ensure that we have
' considered and addressed the sa fety issues r 'ised by other licensees or by the.
NRC in the approval SERs. We followed this practico for the STP request and-conformed our request _to include known issues raised on other dockets, as well e
as addressing Grand Gulf-specific concerns.
A i
l I
5 i
i
?
G9108151/SNLICFLR
J-Attachment to GNRO-91/00149 Page 6 of 6 References 1.
Entergy Letter, GNRO-91/00043 dated April 26, 1991; " Reactor Protection System Instrumentation Survel1 1ance Requirements."
2.
NRC 1.etter, John Ilickman to Dale L. Iloltzsher, dated January 31, 1990.
3.
NRC Letter Theodore R. Quay to William T. Cattle, dated July 10, 1991.
4.
J. A. Wooley, "Three-Dimensional BWR Core Simulator," NEDO-20953-A, January 1977.
5.
NEDE-240ll-P-A-US, Revision 7, " General Electric Standard Application for Reactor Fuel" (GESTAR) August 1985 (GE Proprietary 6.
Entergy Letter, AECM-86/0174 dated June 9, 1986; " Addendum to MEOD Submittal."
7.
NRC Letter, Inster L. Kintner to Oliver D. Kingsley dat d August 15, 1986.
8.
XN-NF-900(P), "A Generic Analysis of the Loss of Feedwatnr llenting Transient for Boiling Water Reactors", February 1986.
9.
Entergy Letter, AECH-86/0273 dated September 5, 1986; " Cycle 2 Reload Submittal Additional Information (LOFWii, LOCA, Fuel Li f tof f)."
10.
NRC Letter, Lester L. Kintner to Oliver D. Kingsley, Jr. dated October 24,
- 1986, 11.
NRC Letter, Lester L. Kintner to William T. Cottle dated November 15, 1990.
12.
NRC Letter, Theodore R. Quay to B. Ralph Sylvin, dated June 3, 1988.
G9108151/SNLICFLR
.