ML20072S403
| ML20072S403 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 03/31/1983 |
| From: | Hukill H GENERAL PUBLIC UTILITIES CORP. |
| To: | Eisenhut D Office of Nuclear Reactor Regulation |
| References | |
| 5211-83-017, 5211-83-17, NUDOCS 8304070167 | |
| Download: ML20072S403 (11) | |
Text
f GPU Nuclear Corporation su Nucioar
- =rSs48o 8
Middletown. Pennsylvania 17057 717 944-7621 TELEX 84-2386 Writer"s Direct Dial Number:
March 31, 1983 5211-83-017 Office of Nuclear Reactor Regulation Attn:
D. G. Eisenhut, Director Division of Licensing U. S. Nuclear Regulatory Commission Washington, D.C.
20555
Dear Sir:
Three Mile Island Nuclear Station, Unit 1 (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 RCS Trip on 25 Subcooling Margin In response to your letter of March 4, 1983, GPUN intends to revise RCP trip criterion from 1600 psig ESAS to 25 F indicated subcooling margin by June 1, 1983. discusses the basis for a reduction in subcooling from 500F to 25 F developed from plant specific error analyses. Enclosure 2 provides the basis for RC pump trip on subcooling margin. addresses RC pump operation criterion for normal, transient and accident conditions.
These changes are in compliance with the requirements of 10 CFR 50.46 and improve the plant safety margins for certain non-LOCA events.
- Further, these changes do not involve a change to Technical Specifications or an unreviewed safety question and are, therefore, being implemented under 10 CFR 50.59.
Sincerely, r
9
!.' D. Sukill H
Director, TMI-l HDH:LWH:vjf Enclosures cc:
R. C. Haynes J. F. Stolz J. Van Vliet 8304070167 830331 PDR ADOCK 05000289 P
PDR GPU Nuclear Corporation is a subsidiary of the General Public Utilities Corporation
m Encicsure 1 25*F Subcooling Margin I.
Background
Since' the accident at TMI-2 the NRC and utilities recognized the need to maintain adequate subcooling margin.
The NRC, through butletins and NUREGs required each PWR Licensee to assure that adequate subcooling margin was maintained and in the long term to install a saturation margin monitor. Through coordl* nation with the utilities a minimum margin of. 50*F was established.
For B&W units, the 50*F was 'oased on an actual margin of 5 F (which ailows for the dif ferences in loop temperature of the highest elevation and the point where the hot leg temperature is measured) with a 45'F instrument error.
The 50*F subcooling margin became the determining criterion for allowing throttling of HPl.
11.
Analysis GPUN has performed plant specific calculations of instrument string error using
.the Root Mean Square (RMS) method f or subcooling margin at pressures greater than 300 psig. The calculations show that during normal and accident conditions (i.e., SB LOCA's), the maximum instrument string error is -18.7*F (+ 21.7*F).
With the addition of the 5*F physical configuration margin, a 25*F subcooling margin is justified.
During a Large Break LOCA the subcooling margin monitor is not required for RC pump trip.
Further, HPI operation criteria is not dependent on the saturation margin monitor but rather LPI operation.
In order to provide specific' guidance to the operator, GPUN proposes that the 25"F subcooling margin monitor criteria be used and alarmed.
'llt.
Discussion Since the fall of 1979 analyses performed by GPt5 and B&W have demonstrated that HPI initiation and throttling based on subcooling cargin is adequate to ensure subcooling during the three main events of interest, a.
Small Break LOCA Events The reduction in subcooling margin to 25*F during SS LOCA events allows better plant control during system recovery by allowing HPl throttling sooner. The lower subcooling margin allows a broader control band which permits the operator to maintain conditions with greater margins to possible overcooling conditions.
The SB LOCA events are bounded by the analysis discussed in Sectiom II (Containment Temperature - 245'F, 4 R).
Containment Pressure -<30 psig, RH - 100% and dose - 5 x 10 b.
Steam Generator Tube Rupture Events
-The primary to secondary leak rate during alngle and multiple tube ruptures is a function of primary to secondary dif f erential pressure.
The dif ferential pressure is minimized with reduced subcooling margin and by primary depressurization operating the RC pumps. Figure 1 illustrates that by changing from a 50*F subcooling margin (with pumps of f) to a 25'F subcoolIng margin (with pumps on) a 50% reduction in. the amount of RCS leakage is attainable.
Reduced integrated leekage will, as a consequence, reduce the dose to individuals on and off site.
c.
Overcool ing _ Transients During overcooling events a' reduced subcooling margin provides an increased operating band making-it easier for the operator to stay within the presssure temperature limits.
For all other transient and accident conditions, there is no reduction in the safety margin or consequences of an accident as described in the FSAR.
Mitigation of LOCA, tube rupture and overcooling events is not dependent on subcooling margin as a signal to-initiate autoinatic plant protection.
Since HPi is'not automatically initiated for any events analyzed in the FSAR besides LOCA, tube rupture and. overcooling events; HPI throttling does not affect the consequences as analyzed.in the FSAR-(See Table 1).
IV.
Conclusion Maintaining a minimum of 25'F indicated subcooling margin assures that the RCS is subcooled during normal, transient and accident conditions. Therefore, HPI initiation and throttling may be based on 25 F subccoling margin without a reduction _ in plant safety.
Furthermore, a reduction in subcooling margin from 50*F to 25'F reduces the primary to secondary leak rate, integrated leakage, dose rate and integrated dose resulting from a tube rupture event and, therefore, increases plant safety.
V.
References:
1.
_ NRC letter dated 23,1979 " Meeting Minutes for October 12, 1979".
2.
Letter from Warren J. Hall to Darrell G. Eisenhut dated March 31, 1982, Question 1.13.
e r,-
RCP Trip i.
Background
Based on initial analyses received from PWR vendors, NRC concluded in-NUREG 0623 that delayed trip of reactor coolants pumps during-a small break LOCA can lead to predicted fuel cladding temperatures in excess of current licensing limits.
B&W examined what would happen if the reactor coolant pumps were tripped at some time into the accident when the system void fraction was high. They arbitrarily assumed that the pumps were tripped when-the system void fraction was 90%.
At the time of pump trip the liquid that was previously dispersed around the primary system through pumping action now collapsed down to low points of the primary system such as the bottom of the vessel and steam generatcrs. This resulted in significant uncovery of the reactor core, due to an insufficient amount of liquid being available to provide acceptable core cooling.
Due to design features as well as temporal behavior of system void fraction, B&W concluded that unacceptable consequences would result f rom delayed reactcr coolant pump trip only for a range of small breaks (.025 to 0.25 ftM and a range of trip delay times af ter accident initiation.
Based on these findings, a meeting of utility vendors and owners was held with NRC in September 1979.
At this meeting it was agreed that the 16C0 psig ESAS signal provided timely Control Room Indication for manual action to prevent possible voiding scenarios.
II.
Anal ysis in a recent B&W's, anai ysi s, various RCP tripping schemes were investigated.
They included saturation margin, coi nci dence ESAS/Tsat, low low pressure and void fraction.
The saturation margin scheme was chosen by GPUN since this signal results in fewer pump trips during overcooling events and if the saturation margin is sultabiy low, in no pump trips during design basis tube rupture events or very'small break LOCA's.
The results of the analysis centered around a review of void fraction for pumps on/off conditiens.
/
/fs
@9 g with I
with pu-.;
f p m.o
.\\
\\
without c
b without h
%I
'c 3
/
\\
mo
/
00 Pum?
8 N s
% ~ ~ ~
>c
,/
p k
Time 9
gg A
The above schematics IIiustrate that the system wIII initieiIy be less voided es well as 'at a. lower pressure with pumps running.
These results occur because the continued operation of the RC pumps keeps the coolant circulating throughout the loops providing better-steam generator heat transfer and lower hot leg temperatures. -The lower system pressure with the pumps running during the initial phases of the transient results in a decreased Icak flow and thus a lower void fraction. However, the continuous operation of the RC pumps results in' lower quality discharge through the break whlch eventually of f sets the decreased pressure ef fect on the leak flows.
Thus, the fluid in the primary system ultimately evolves to a high void fraction for certain break sizes as a result of the continued RC pump operation. The studies done by B&W show that the crossover point between the pumps on and of f cases occurs at 40% system vold fraction independent of the break size.
Loss of subcooling margin in.the hot leg occurs well before the RCS void fraction can become large enough to threaten core uncovery. Therefore, the use
.of the subcooling margin is an ecceptable alternative to 1600 psig ESAS.
til.
Discussion Since the f all of 1979, analyses performed by GPUN and B&W have demonstrated that Reactor Coolant Pump trip on subcooling margin accomp1ishes the originaI
. objective of-tripping RCP's without decreasing safety margin for the three main events of interest, a.
Small Break LOCA The change in parameters from low pressure ESAS actuation to loss of
. subcool Ing margin accomplishes the original objective of tripping the RCPs f or smal l break LOCAs.
For the break sizes of concern, a loss of 25'F subcooling, margin will occur slightly b> fore low RCS pressure actuation of ESAS.
The operator has essentially the same time to trip the pumps at the onset of saturation as witn the current ESAS actuation criteria.
A review of the SB LOCA analysis indicates that the basis fcr RCP trip was 20% veld fraction.
b.
Steam Generator Tube Rupture Events During Steam Generator Tube Ruptures in which minir.u. subcooling margin is maintained, continuous RC pump operation assures expeditious cooldown with a minimum primary to secondary differential pressure.
This change in criteria for RCP trip will allow RCP's to :s cpercied for a greater
. spectrum of tube ruptures (including ruptures beycnd the design basis) and to reduce the offsite doses for those events.
c.
Overcooling Transients For overcooling events in which the pressurizer does not empty, the subcooling margin does not drop below 25'F even though RCS pressure is below the ESAS initation setpoint (1600 psig).
When combined with our proposed RCP trip criteria, the RCP's'will remain operational, thereby precluding void formation in the hot _ lecs anc minimizing formation and duration of voids in the reactor vesse! head.
Furthermore, RCP operation allows continued use of Main Feedwater in lieu of Emergency Feedwater and
the use of pressurizer spray to control RCS pressure during recovery from the overcool _i ng-transient.
Table 1 lustrates that the 25'F subcooling margin is not lost for any event in wb'.. f orced RC flow is required for event mitigation. Therefore, the operator would not be forced to terminate forced flow when it had previously been taken credit for in FSAR analysis.
lV.
Alarm in addition, a Control Room alarm will be adjusted to annunciate if either subcooling margin monitor indicates less than 25 F.
The plant computer also independently computes pressure and temperature saturation margin for logging, trending and alarm.
V.
Conclusion The change in RC pump trip from low. pressure ESAS actuation to loss of subcooling margin:
ensures'RC' pumps are turned off when required which o
assures no decrease in safety.
permits RC pump operation f or some overecoling events o
which increases plant control.
o permits RC pump operation for more tube rupture events (including those beyond design basis) which increases safety by decreasing leakage and expediting plant cocidown.
VI.
Reference 1.
" Evaluation of Transient Behavior for Small Reactor Coolant System Breaks in
-177 FA Plants", dated May 7, 1979.
Z f
l TABLE 1 HP I-Forced Flow 25'F SCI'.
JAccident or.
For Event Assumed in Reached During Transient:
Mitication FSAR Event Rod Withdrawai '
-Power No Yes No Startup No Yes No Boron Dilution No Yes No-Cold Water Addition No Yes N/A Loss of Flow No No N/A Dropped Rod No Yes No Loss of Elec. Load No No-N/A fBlackout Yes No N/A Steamline Failure Yes See Discuss.
(Enc.1,Ill.c)
Tube ~ Rupture Yes-Yes No Fuel Handling No N/A N/A
. Rod E,jection--
Yes-Yes Yes Feedwater Line.
Break-No.
No No Waste Gas Decay'~
Tank. Rupture No N/A No SS LOCA-Yes See Discuss.
See Ciscuss.
Large Break LOCA.
No-No Yes
- Analyzed with and' without RCP cperation.
L
_______________________j
FIGURE 1
Effect of EC Purnp Tri; G:f asrion on Entegrated Systs:n Leakage for Single Ruptured Tube
- 180 t
i i
i i
i 1: 25of SC, f dMPS ON 160 i
II: 50 F SC, PUMPS ON
~
tr
/
c3 111: 50"F SC PUMPS OFF
=
ka s
E S=
i Ill
- u. 2 120r-
/,
/-
m m W'
cu
!i
==
100
/
_f b$
'80
.'. j
=5 e-60
/
D 40 i
1 20 L
0 20 40 60 80 100 12u TIME (MINUTES)
SC - Subcooling
RC Pump Operating Criteria 1
I.-
RCP~Operattons - Trip A.
-Set points for RCP Trip 1.- Forced Flow Tabie 1 of enciosure 2 provides an accident / transient sum,ary matrix for which forced flow is assumed.. Note, specifically, that RCS flow is assumed for design basis tube ruptures and that for a-25*F subcooling margin trip setpoint RCP trip does not occur._ Further, tripping RC pumps on 25'F subcooling margin reduces trips for non-LOCA events 'for which continued RC pump operation is desirable.
.2.
Volded System By tripping pumps _ on 25*F subcooling margin, operation of R^ pumps in a volded system is precl uded.
~3.
PORV Chailenges TMI-1 will use a manual operator RCP trip which avoics inedvertent trips.by.
automatic systems.
Experience in the utility industry since the TMI-2 accident
-has'shown that inadvertent operator initiation has a very low probabliity of
~
occurrence.
Tripping RCP's on 25'F subcooling margin reduces spurious RCP
' trips, thereby reducing. challenges to the PORY f or non-LOCA events since pressurizer sprays-is available to control RCS pressure.
4 Hot Stagnant - FI ui ds
-The addition 'of 5*F margin to the instrument error calculation sumarizec in enclosure 1-of this letter provides suf ficient m.argin to precl ude saturation-conditions in the RCS due to physical location of the senscrs. _ Current pl ant emergency-procedures discuss ' operation in a velded ccndition and removal of volds that resulti from flashing. Operator traini ng specif ically addresses _ these
~
procedures.
5.
RCP Services As described in section 2.1.1.5 of the TMI-1 Restart Repert-Inject ion f water to the - RC ' pumps is ' capable of - being restored following centainment isolation.
In the event that the cooling water to the pumps is lost, the injection. water
-provides. adequate cooling for the' seals anc the pamp can be operated-
- IndefiniteIy If.the seal injection is functioning ncrmal!y.
6.- ICC Instrumentation The saturation margin monitor which is part of the ICC instrumentation system wi l l be used f or Indication of 25'F subcooling margin for RCP trip.
' B.1 Justification 'for Manual RCP trip 1.
Conformance with 10CFR 50.46 A generic analysis has been performed by B&W which envelops TMi-1.
This analysis is sumaarized in Section 11 of enclosure 1.
2.
Most Probable Best Estimate Since there is no substantial change in time to trip RCP's for the worst cese SB LOCA under the 1600 psig criteria or the 25'F subcooling margin and in the time f or the operator to respond to the initiating signal, this analysis is not considered necessary.
C.
Other Considerations 1.
Instrumentation The saturation margin mcnitor and alarm are addressed in section 2.1.16 of the TMi-1 Restart Report. This. instrumentation will be upgraced to safety grade during the Cycle 6 refueling.
2.
Pump Restart Small break LOCA' and.other emergency procedures will be updated to reestablish the RCP restart criteria from 50*F to 25'F.
3.
Training Training'on RCP operation during transients and accidents is_integra! to the training program on emergency procedures which includes simulator training.
II.
Sunrnary of 10CFR 50.59 Evaluation 1.
The probability of occurrence or the consequences of an accident er malfunction of equipment important to saf ety previcusl y eval uated in the Safety Analysis Report has not increased.
Enclosures 1 and 2 provide a summary analysis of those events of concern and Table 1 address all acci dents eval uated in the FSAR Chapter 14.
In fact, by tripping RC Pumps on 25*F subcooling margin, greater control and reduced consequences occur.
No additional equipment is added by this change nor af f ected which would lead to a mal f unction.
-2.
The possibility for an accident o malfunction of a dif ferent type than any evaluated previously in the Saf ety Analysis Report is not
~
created.- All credible events analyzed are enveloped by existing FSAR events and no new unanalyzed events are created.
3.
The margin of saf ety as defined 'in the basis of any Technical Specification is not reduced.
RCP trip anc pressure temperature limits
l Tech Spec bases have been carefully evaluated as indicated in Enclosures 1 and 2 with no resulting reductic, in saf ety margin.
For certain events the saf ety margin has been increased.
l e
_ _, _ _ _