ML20004C271
| ML20004C271 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 05/26/1981 |
| From: | Hukill H METROPOLITAN EDISON CO. |
| To: | Stolz J Office of Nuclear Reactor Regulation |
| References | |
| L1L-159, NUDOCS 8106020466 | |
| Download: ML20004C271 (4) | |
Text
- -
Metropolitan Edison Company gj g
Post Office Box 480 Middletown, Pennsylvania 17057 Mter's Direct Cial Number May 25, 1981 A
LlL 159 C
N FH 4
Office of Nuclear Reactor Regulation dI 9
Aten: John F. Stolz, Chief b
HMJ6
[ j g
Operating Reactors Branch No. 4 U. S. Nuclear Regulatory Commission J N 011981 m ',
g Washington, D.C.
20555 u.a g s
Dear Sir:
,j
/
g Three Mile Island Nuclear Stari.t, Unit 1 (TMI-1)
Operating License No. DPR-50 Docket No. 50-289 Small Break LOCA The enclosed letter was provided to GPU by B&W for consideration in conjunction with ongoing activities associated with small break LOCA analysis. It is provided dere for your information. We have reviewed this letter against analysis
previottsly developed for the determination of system performance parameters to meet Appendix K to 10CFR50 and against analysis developed since the TMI-2 accident which define an expanded basia for operator ' guidelines. We have concluded that che information contained in this letter would not cause us to alter any previous conclusions with respect to compliance with Appendix K or modify any of the operator guidelines or systam actions that had resulted from previous analysis.
We have, as suggested by B&W, confirmed that our plans to upgrade the timeliness and availability of emergency feedwater will ensure EFW flow within one minute of an actuation signal. Further, the actuation signals will also be upgraded consistent with guidance based on earlier analyses. See enclosure 3 to Licensee letter dated January 23,1981 (TLL 680).
Sincerely, j
. D.
ukill Director, TMI-l HDH:CWS:1ma Enclosure cc:
D. Dilanni oOI H. Silver L. Barrett b
/!/
8106020f/ek l
Metropolitan Ecan Company is a Member of the General Public Utilities System
s Babcock & Wileox
- u.., p
, a. r.ii.o oi.ision a Mcoermott company 3315 Old Forest Road P.O. Sox 126o March 25. 1981 t.ynenburg. vagin,a 24 sos File-177/T1.2 m as4.sn1 ESC-634 GPU-81-042 Mr. D. G. Slear TMI-1 Project Engineering Manager GPU Service Corporation 100 Interpace Parkway Parsippany. New Jersey 07054
Reference:
R. W Ganthner to D. G. Slear. letter of October 3. 1980
Subject:
Reactor Coolant Pump Suction Small Break LOCA
Dear Mr. Slear:
Following the TMI-2 accident, the NR,C requested several small break accident scenarios be evaluated in order to develop operator guidelines for these events. These analyses included scenarios where auxiliary feedwater (AFW) was assumed not to be available at the stgrt of the event.
The assumed worst case small break LOCA (less than 0.01 ft.4) for these analyses is located at the reactor coolant pump discharce. This assumption was made because under normal circumstances a greater degree of HPI penetration into the reactor vessel is achieved during a suction line break. The purpose of this letter is to provide some information regarding this worst case assumption as it is affected in the scenario where HPI is not actuated and AFW is delayed. A brief summary of this situation was provided to you in the referenced letter.
Specific l
details are included in this letter.
For ptnp suction line breaks, under normal circumstances,100% of the HPI flow enters the reactor vessel. For pump discharge line breaks only about 70%
of the HPI flow enters the vessel. However, for the scenario where break size is such that HPI is not automatically initiated and AFW flow is delayed. the rate of system inventory loss before AFW actuation becomes important. During pump discharce breaks a two phase disci arge results due to the effect of the reactor vessel internal vent valves. This reduces the rate of system inventory loss. A pump suction break will result in the loss of lower quality fluid which will deplete system inventory at a higher rate. Thus at the time of AFW actuation the RCS inventory will be less for the pump suction line break than for the pump discharge line break, p b 22l; I c y,j, c--
hP APR 2 BEC'O
t Analyses which have been conducted for the pumo discharae break condition F
demonstrate that operator actiJns to start AFW f% in 20 minutes will result in acceptable conditions. However for the case where there is a pump suction break of a size such that boil dry of the steam generators occurs prior to an RCS pressure decrease to the HPI actuation setooint, the 20 minute delay in AFW actuation has not been demonstrated to result in an acceptable RCS inventory condition. That is, the 20 minutes of delay analyzed for in the pump discharge break case has not been analyzed for in the pump suction break case.
Thus.
although there is certainly a delay in AFW actuation which will result in acceptable conditions, the actual time has not been identified for the pump suction break. condition.
While the actual analyses for the pump suction break delayed AFW scenario has not been conducted, there is a significant amount of guidance for the operator regarding actuation of HPI and AFW. Following the TMI-2 accident, small break LOCA operating guidelines were developed.
These guidelines instruct the operator to ensure that AFW is being delivered to the steam generators and if it is not, to restore feedwater as soon as possible. Additionally. the guidelines require manual actuation of HPI should the system reach saturated conditions. These actions provide for mitigation of the delayed AFW scenario.
Additionally. uogrades of the AFW control system have been implemented which l
would ensure AFW flow in times on the ordqr of one minute. Thus. althouah the
(
specific analy',is has not been conducted. there is adeauate reason to believe that current procedures and system characteristics make the identification of the specific time delay superfluous. However, it is not clear what significance the demonstration of a 20 minute operator response time was to the NRC.
The licensing significance associated with the demonstration of a 20 minute operator l
response time is best determined by each utility. Additionally. the AFW upgrades are plant dependent and B&W cannot assess to what extent the probability of this event has been diminished. At present, the following positions appear to be possible resolution paths on this issue:
1.
Review the AFW systems and confirm that the small break LOCA with delayed feedwater is a highly unlikely scenario and need not be considered part of the desian basis for the plant.
Thus. while the analyses performed may not have considered the worst break location for demonstrating the minimum allowable operator response time the probability of this event along with the generation of the operator guidelines provide adequate assurance that this transient can be safely mitigated.
2.
Use the basic position outlined in Item i except report co the Commission the potential change in the previously submitted analyses.
3.
Perform detailed evaluations of the pump suction small break LOCA with l
a delay in the delivery of AFW and determine the time frame available l
to the operator to restore either feedwater or HPI.
.~
l l
l i i It is B&W's position that Item 3 is not necessary. We believe that either Items 1 or 2 are viable alternatives for the ultimate resolution of this issue.
l If you have any questions regarding the nature of this concern please call i
l me (804-384-5111. extension 3489) or R. W. Ganthner (804-384-5111. extension l
-2751) at our Lynchburg office.
Very truly yours.
- 5. H. Duerson Engineering Product Manager l
cc:
J. G. Herbein R. A. Knief E. G. Wallace
~
J. J. Colitz R. J. Toole J. F, Fritzen l
D. W. Demers R. W. Keaten H. D. Hukill F. R. Faist K. Menon L. R. Allen l
l l
[
1 i
.