ML20236R397
| ML20236R397 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 11/18/1987 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML20236R398 | List: |
| References | |
| TAC-65068, NUDOCS 8711230254 | |
| Download: ML20236R397 (4) | |
Text
_
/
o,,
UNITED STATES
[
p, NUCLEAR REGULATORY COMMISSION WASHINGTON, D. C. 20555 g
- j
\\*...+/
SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION RELATING TO THE CHANGE OF TECHNICAL SPECIFICATION BASES SECTION 3/4.7.1.2 THETOLED0EDISONCOMP,NJ DAVIS-BESSE NUCLEAR POWER STATION, UNIT 1 DOCKET N0. 50-346
1.0 INTRODUCTION
In a letter from Donald C. Shelton (Toledo Edison Company) to the U. S. Nuclear Regulatory Commission (NRC) dated March 23, 1987, a change in Technical Speci-fication (TS) Bases for Davis-Besse Nuclear Power Station, Unit I was request-ed. The proposed modification involves a decrease from 800 gpm'to 600 gpm for the Auxiliary Feedwater (AFW) flow requirement in the Davis-Besse TS Bases, Section 3/4.7.1.2. The reason for the proposed change is to assure equal flow requirements on all three AFW pumps now that a faulty variable flow valve has been removed from one of the three pump lines. According to information provided by the licensee, such a change would maintain reactor core parameters within safety limits and impose no additional risk to the public based on analyses of the applicable bounding case.
2.0 EVALUATION The tounding transient for a decrease in AFW flow is the loss of feedwater transient.
In Da.vis-Besse's Final Safety Analysis Report (FSAR) Accident Analysis, this transient may be initiated by any of three failures: the abnormal closure of a feedwater valve, pump failure, or a fcedwater line break (FWLB). Babcock & Wilcox has determined the most bounding of these to be the abnormal feedwater valve closure. The reactor coolant system (RCS) pressure surge due to the FWLB, occurs before AFW initiation takes place. Therefore the i
reduction in AFW flow is of less concern than for the valve malfunction case, 1
which relies upon AFW for the recovery of steam generator (SG) heat transfer l
8711230254 071110 i
PDR ADOCK 05000346 P
PDR q
b
l 1
2
~
capability. Also, the reactor trip for the FWLB occurs significantly sooner
)
than for the valve malfunction, making the feedwater valve malfunction case more limiting.
I Assumptions made in the reanalysis of the loss of feedwater transient include:
Initial reactor power at 102%
No credit for PORV, pressurizer sprays, or makeup flows, 1.2 times ANS 5.1(1979) decay heat curve, Availability of offsite power, and AFW feed to only one SG, due to single failure considerations, at a flowrate of 600 gpm.
In the reanalysis, the loss of feedwater results in a rise in RCS temperature and pressure leading to a reactor trip at the 2400 psia high pressure trip setpoint. The primary pressure peaks at approximately 2605 psig, which exceeds the system design limit of 2500 psig, but remains within the safety limit of
, 2750 psig (110% of design pressure) and is therefore acceptable.
1 Meanwhile, on the secondary side, the SG level drops. Ten inches is the SG low level setpoint which actuates AFW with a 40 second delay for flow delivery.
The analysis shows dryout of the SG at the same time as AFW flow initiatiun begins. However, the dryout condition exists for a relatively short duration (seeFigureA). The dryout condition is tolerable due to the Babcock & Wilcox SG design which includes AFW injection directly on the SG tubes for immediate J
recovery of heat transfer capability. The RCS hot leg reacts with a maximum temperature of approximately 615 F which lies below the 635' F Reactor Core 1
Safety Limit. The consequences of SG dryout in this Babcock & Wilcox designed plant is acceptable, mainly due to the once through SG design.
l l
Pressurizer level goes offscale high but maintains a bubble approximately four l
vertical feet from the top of the pressurizer. The pressurizer thereby avoids filling completely and is able to control primary pressure conditions.
l l
l l
l
-_-_____--J
i 3
The RELAP5 computer code was used to model the bounding loss of feedwater transient. RELAP5 has not been approved by the NRC staff for safety analyses for the Davis-Besse plant.
In this particular application however the code was found to give conservative results. The reactor trip time of 14 seconds calculated by the code is longer than the actual expected reactor trip time.
With the conservative reactor trip time, SG-inventory and heat transfer is underestimated resulting in a conservative RCS temperature rise and pressurizer insurge. Because the results of the use of RELAP5 are conservative in this i
application, we find it acceptable for use in this licensing action.
3.0 CONCLUS!0N
)
Because the licensee evaluated the bounding transient associated with the proposed AFW flow decrease using a conservative computer model, and because the results are within the safety limits of the TS, we find the proposed change acceptable, j
Dated:
November 18, 1987 h
e
-_-.-_._______m._m__._
F r2 CURE A b
~
DB.LOFW - AFW FLOW 0F 600 GPM j
STERM GENERATOR COLLRPSED LEVEL (Loop 1) l i
i i
i i
j I
i i
iffly e
sn l
l U
W i
td
- i Ba.
4 I
~
j i
I 8
i e
i M.
80.
160.
240.
320 400.
480.
j 87/01/07.
RELAPS TIME. SECONDS L
i L
<