ML20236N404
| ML20236N404 | |
| Person / Time | |
|---|---|
| Site: | Braidwood  |
| Issue date: | 11/03/1987 |
| From: | Hunsader S COMMONWEALTH EDISON CO. |
| To: | Davis A, Murley T NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM), NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III), Office of Nuclear Reactor Regulation |
| References | |
| 3533K, NUDOCS 8711160147 | |
| Download: ML20236N404 (6) | |
Text
- _ -.
L
.f
-V i
s.
% T y.
j m] Commonwealth 5dison S
]
(
~ ) : One Fir;l National Plazi, CNca00. Illittis
-(
C 'T Address Reply to: Post Offic: Box 767 -
l
, (j :..CNeago, Illinois 60690 0767 '
.. l November 3, 1987l
,d i
Mr. Thomas E. Murley, Director Office of Nuclear Reactor Regulation U.S., Nuclear Regulatory Commission' l
Washington, DC 20555
-l Mr. A. Bert Davis Regional Administrator l
.U.S.; Nuclear Regulatory Commission Region III 799 Roosevelt Road Glen Ellyn,'IL 60137 Attn: Document Control Desk
Subject:
Braidwood Station Unit-l Operability of an Emergency Core Cooling System (ECCS) Subsystem
(
NRC Docket No. 50-456
. References (a): -September 25, 1987, C. E. Norelius letter
~to Cordell Reed.
(b): October 23,.1987, L. D. Butterfield letter to A. B. Davis.
Dear Mr. Murley and Mr.. Davis:
Reference (a) provided the NRC Notice of Violation of NRC requirements
' involving the temporary alignment of the Residual Heat Removal (RHR). System i
during post maintenance testing of an RHR train. Ref9rence (b) provided
-Commonwealth Edison's (Edison) response and included a request that a meeting be. held with representatives from NRC Region III and the Office of Nuclear Reactor Regulation and ourselves to discuss this issue. The purpose of this letter is to provide, in preparation for this meeting, Edison's position with respect to.the closure of the RH8716A valve during Mode 1 operation and the resulting Technical Specification interpretation of operability for the RHR
. system. This position is being submitted for early NRC review and subsequent
)
discussion.
{
8711160147 871103 l
PDR ADOCK 05000456 i
k PDR
\\ \\
j s
i l
l l
US NRC November 3, 1987 c
't hs
}
\\
,I L
p.,
On August 8, 1987, during a review of the Unit 1 Operating Log, the atatin control. Room Engineer (SCRE) noted that the RH8716A valve had been closed during the previous shift for 64 minutes.
Edison personnel discussed this' event ori August 9,1987 'and initially determined that this event could be conservatively interpreted as a situation resulting in the inoperability of both trains of the Unit 1 RHR System for this 64 minute period, while the unit
]
was in Mode 1 operation. Accordingly, this event was reported to the NRC f
Operations Center pursuant to 10 CFR 50.72(b)(2)(iii). Subsequently, after a more extensive review, Edison concluded that the event was not reportable under 10CFR 50.72 and, therefore, the August 9, 1987 notification was retracted on August 12, 1987. That review showed that one train of the Unit 1 RHR system was in fact operable and the other train of the Unit 1 RHR System was in a recirculation mode. Under this configuration, the plant was operating under the required 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> time clock of Technical Specification 3.5.2.
'tne. fellowing presents Edison's views on why the event of August 8, j
1987 did not constitute the inoperability of both RHR Syutem trains of the Emergency Core Cooling System (BCCS).
d Figure 1 provides a piping line diagram of the ECCS System that' f
includes the RHR System. On August 8, 1987, the A train RHR System was in an operable mode but in standby. The station was recirculating the IB RHR pump to the Refueling Water Storage Tank in order to perform a leak test on a j
flushing connection isolation valve in the IB RHR pump discharge piping. This i
evolution resulted in the IB RHR heat exchanger being bypassed, the RH8716A valve being closed, and the RH8735 valve being opened.
It should be noted that approximately 3000 gpm was being provided to the discharge header from the 1B RHR pump; the RH8716A valve was capable of being opened from the control room; and an operator with radio communication to the control room was stationed at the RH8735 valve to close it, if necessary. Because this configuration reduced the overall capabilities of the RHR System, the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> j
timeclock was being observed. Had a safety injection (SI) signal been
{
received in this configuration, each RHR pump (lA and 1B) was capable of 1
injecting into two RCS cold legs, separately, resulting in four RCS cold leg
)
injection, without any additional operator action.
Technical Specification 3.5.2 defines an operab?e BCCS subsystem to be comprised of:
(a) one operable centrifugal charging pump, (b) one operable j
safety injection pump, (c) one operable RHR heat exchanger, (d) one operable RHR pump, and (e) an operable flow path capable of taking suction from the Refueling Water Storage Tank on a safety injection signal and automatic 1
opening of the containment sump suction valves.
The Technical Specifications also define a subsystem to be operable if it is " capable of performing its specified function".
Edison's analysis shows that a subsystem of the ECCS is operable with the RHR flowpath limited to injection into two of the four RCS cold legs.
j
~
]
gg
- ~ = -.-
a.
4 gq,
$y
^'
- O W
p f
l y
4
[jD '
lij 1
t'Nd
,US NRC<~
t-3: '
' November 3, 1937
?
~
Mp j
,iJ f
t i i
q-k ;The primary, function of an ECCS subsystem is specified.in Chapter 6.3 1
~
1~ of:the' Byron /Braidwood.FSAR. ~"The primary. function of the ECCS isito remove
- the stored'.and fission product: decay heat fran the ' reactor core during -
z
.accidsnt conditions" (FSAR page 6.3-1).
The FSAR further states'"The design
!J'K : bases' for selecting the ' functional' requirements of the" BCCS are derived from -
, Appendix K limits for Fuel Cladding Teinperature, etc., following any of the above; accidents as: delineated in 10 CFR 50.46.'1The subsystem functional.
I iparameters are selected to. integrate so'that the Appendix K requirements are imet..."
(FBAR page 6.3-1)'.
- "The Emergency Core Cooling System (ECCS)'
-components'are designed in order that a minimum of three accumulators,'one D
charging' pump, one' safety injection pump, and'one residual heat removal pump.
together with their associated valven'and piping'will ensure adequate core cooling in'the even't'of.a. design-basis LOCA". :(PSAR.page.6.3-2; Section 6.3.2)
.. Previous analyses have shown'that these temperature limits can be L
maintained with the RHR flowpathilimited to one' RHR pump capable of injecting into two!of the.four reactor coolant system cold legs. Operability of the 1
~
y
' ECCS.. system under these' con'dit ions' has been demonstrated in a Westinghouse evaluation previously provided'to the NRC with regard to Byron Unit 1.
That evaluation showed;that closure of oneLSI8809 valve in the RHR System, while
. operating in this configuration in Mode 1, did not result in'a violation of 10:
CFR 50.46 acceptance criteria for ECCS performance. This evaluation assumed ECCS flow; availability. from at least_ one' centrifugal charging puanp;, one safety j
injecton pump;;one RHR' pump injecting i m two reactor:coolantisystem. cold l
legs; and availability.of the' accumulators.. Although-the August 8 event did
'not involverthe SI8809A and B: valves,.but:rather valve RH8716A, Edison has
' l determined'that closing of a RH8716 valve is; bounded by the closure of either of the SI8809 valves and:is less ;1imiting than the failure of an RHR pump as j
far as flow delivery to the core.during a LOCA is concerned. Thus, the.
l
~
condition which resulted from closing valve RH6716 is included in the_ previous analysis for the closure of valve SI8809. The previous Byron analysis is equally applicable to Braidwood-for operation in Mode 1.
1 The FSAR.LOCA analysis considers both large and small break events.
Since the small break Is0CA analysis is not dependent upon RHR flow, the small j
break LOCA results are not. dependent upon RHR system valve alignment and it does not'need to be considered here. The:large break LOCA event relies on RHR l
cystem alignment. During the August.8' event, the 1A RHR pump was in a stand-by l;
operable. status and was available to inject into two of the RCS cold legs upon lL initiation of a safety injection signal.
For the large break LOCA analysis with minimum. safeguards actuation (one ECCS subsystem actuation), Westinghouse conservatively assumed that the broken RCS cold leg is one of the two aligned to' receive'RHR injection flow. With this configuration, the RHR pumi. can L
L
)
g
l' h
US NRC November 3, 1987 deliver 190 lbs/sec. during the core reflood phase rather than the 390 lbs/sec. flow currently modeled. Early in the large break LOCA scenario, the accumulators inject to fill the downcomer of the reactor completely. Then the
- ECCS pumps must supply sufficient flow to maintain the downcomer level during core reficod. The centrifugal charging and safety injection pumps together can supply 110 lbs/sec. and this flow combined with the RHR pump flow of 190 lbs/sec. results in flow to the RCS of 300 lbs/sec. Westinghouse's review of the analysis indicates that 300 lbs/sec. is adequate flow to maintain the downcomer water level. Since this flow was as<ailable, very little or no I
penalty in calculated peak clad temperature can occur in the minimum safeguards I
large break LOCA FSAR analysis due to the inoperability of one RHR train and the simultaneous closure of one RH8716 valve. The increase in peak clad temperature is estimated to be less than 10'F and therefore the predicted peak clad temperature remairts well below the limit established in 10 CFR 50.46.
However, in addition to the configuration presented in this analysis on August 8, 1987, the IB RHR pump (though in an administrative 1y declared inoperable status) was running and providing flow to the remaining two RCS cold legs.
If one postulates no single failure, as is appropriate while the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> clock is in effect, four injection points were available; approximately 3
3000 gpm was available from the IB RHR pump; and full output was available from the 1A RHR pump for injection.
In this configuration, the IB RHR pump's contribution would have been reduced, at the time the safety injection demand signal would have been initiated, until the time the RH8735 valve would have been closed. However, upon closure of the RH8735 valve, full four loop 1
injection would have been established.
Operation in Mode 4 requires one ECCS subsystem to be operable.
However, while aligned for shutdown cooling, one RH8716 valve is required to l:
be closed but energized. This is done to preclude the possibility of l
pressurizing the RWST to RCS presrure (350 psi) if the RH suction check valve (SI8959 A and B) leaks. The only difference in the LCO for an ECCS subsystem l
is that, in Mode 4, the capability to manually realign the suction of the RHF l
pump to the RWST is permitted.
If closing an RH8716 valve were intended to l
render both BCCS subsystems inoperable, shutdown cooling could not be l
established with the required valve lineup.
Also, under restrictions of the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> time clock, an EHR pump train may be declared inoperable to perform maintenance activities. Such activities including pump teardown can be performed within the constraints of the 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> I
time clock in operating modes 1 through 5.
The valve alignment described above is less limiting than a condition that would be seen during maintenance l-activities.
V l
i
,y _
y[
,n g
gi yg m
- af "
'"At
'~
US NRC '
' November 3,-1987-
.g1 i.
4
,e It.is Commonwealth Edison's. position.that manipulation of valves, yb~.such as the RH8716:or RH8735, is acceptable'during operation in-Modes;1 through:
w 5 while' observing the' applicable 72. hour Technical-Specification action' state-k (ment time clock... Fulfillment of the' single active failure criterion is not'
.y required'while observing a Technical Specification time clock. For the event
. described above, Staidwood Station'was operating under.the required.72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> j
'?
.timeclock of Technical Specification 3.5.2.
Additionally,.the configuration
'f of tne plant was such-that.any. single active. failure'could be withstood without
,!cc_aplete;1oss of RHR cooling. At no time was this 72 hour8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br />' clock exceeded.
~: consequently, Technical Specifications 3.0.2,-3.0.3, and13.5.2'were not violated.
.For the'above. reasons, an ECCS subsystem was capable of. performing' 1
(
its,specified function during the surveillance testing period'and was therefore. operable.
4 The above'information is be'ing submitted for.your review. -As stated d
. in our. respons,J'to NRC Inspection Report 50-456/87-029 we would like to meetr with representatives from NRC' Region ~III and'the Office'of Nuclear Reactor.
Regulation to further discuss this. subject. We will~ contact'you in'the near1
-future to set up a convenient ' day and. time to meet' at Braidwood Station, j
~
Please. address any questions concerning this matter.to this office.
1 1
Very truly yours,-
q i.
.C.
S. C. Hunsader Nuclear Licensing Administrator
- j cs cc
- 15. Sands - NRR W. Forney - RIII
.Braidwood P.esident Inspector m
l p,
i l
l
-3533K-I 1
i
R lA
- 8 8
ij p,'5 8
5 9
e 9
8 l
p 8
c u y d k cl n eo a RHT B
A T
'r 3
3 S
hB 3
A.
i I
R W
3 /. R 7k H
.g S
H R
- R R
1A B
l 5
X 3<
X 8X 2X 1
1 2
7 H
H 8>
8 8
8 l
l
^
8 6
6 1
7 7
8 8
X
]
B A
9 X8 9X 0
0
~
8
_ s 8
8 ge L
t l
o
~
H o
T
~
X8 M'.-
M M+
5 3
8 u
h o
g g
e e
h L
L t
t o
o o
h H
H
}5 4
3 2
g g
g Se Se e
Se CL CL L
CL ld Rd d
Rd R
lo l
o o
C C
C
!