ML20084G164
| ML20084G164 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 09/09/1971 |
| From: | Finfrock I JERSEY CENTRAL POWER & LIGHT CO. |
| To: | Morris P US ATOMIC ENERGY COMMISSION (AEC) |
| Shared Package | |
| ML20084G158 | List: |
| References | |
| NUDOCS 8304210508 | |
| Download: ML20084G164 (5) | |
Text
-[%
Q Q.
m J
Jersey Central Po,we?d r& Light Company s
MADISON AVENUC AT PUNCH DoWL Ro AD e MoRRISToWN, N.J. 07960 e 539 6111 September 9, 1971 O
A 9
[{
k Dr. Peter A. Morris, Director 9
Division of Reactor Licensing g
SEp j United States Atomic Energy Commission Washington D.C. 20545 u.s g, IO7/g
~
Q
- c. t y.;;
Ency f
Dear Dr. Morris:
ggf.p.ca
/7 q
Subject:
Oyster Creek Station Docket No. 50-219 Surveillance Testing Failures The purpose of this letter is to report to you several failures that were observed during surveillance testing at the Oyster Creek. Station.
Radvaste Storace Tank Execeding 0 7 Curies On August 12, 1971, at approximately 4:00 u.m., the results of the chemical analysis for La day,ver.: corpletca; ant they showed an y
c
/ k.,
activity of 1.1 euries in be combined v..ste sur6 and vaste x
sanple tank "B".
The remaining i.anks had not beet cumpled as the fff.1 72-hour tank analysis was not due until August 13.
"T
- f. n n; *.-
When.the activity in the outside tank became known, i'=ediate steps g
vere t3 ken to reduce the invenLe y by returnicg va.er to the condenser 4;p.#1 la
.S %
hotvc31.3 and ret.ycIlvs water back inte, the raavasta facility at fast as spce could be msde aveliable, g
Samples vera t+n nt appruntantely 8:30 a.:u. on August 13, the iic -f iaich wea :-" 6.noan until 4 :00 p.m. thit afternoon, (0 72 curirs).
In the interim, 2 waste sruple tanks were returned tc, the condenser hotwells and 60% of flooc drain sample tank "h" vas recycled back into a vaste neutralizer tank indicating that the outside tank activity was (.7 curies sometime during the day.
Samples taken on August 14 showed the activity to be 0.61 curies.
On August 22, 1971, the 72-hour outside tank activity analysis of samples taken at 8:30 a.m. shoved a total cutside tank activity of 0.87 curies.
Again, when the activity in the outside tanks became known, immediate steps were taken to recycle the water inventory.
8304210508 710930 PDR ADOCK 05000219 9
S PDR
)
S.w.qm ggggg ::ge w m.g.,
yQ Aa%.o e m
4
$!S O 5 W $ N Y N O $gy.y;y y& $ 0A $ $ $z.- <! ? ?% O y.. a p 39
.r.q. x x..
M
Dr.PeterA.Mor(}
(])
Page II September 9, 1971 Water in the rioor urain sample Lanxs was recycled to che vaste neutralizer tanks for processing through the vaste corcentrator.
Water from the vaste sample tanks was transferred to ' he hotvells.
c The volume in the surge tank was processed throu6d the concentrator as fast as space was available.
Samples taken from the outside tanks on Au6ust 23 at 8:30 a.m. showed the total outside tank activity to be 0.63 curies.
The safety significance of these events has been analyzed and appears in FD3AR Amendment No. 11.
The cause of these events was the vaste concentrator being out of service for maintenance and subsequent difficulties that were encountered in regenerating the vaste demineralizer, both of which contributed to a backlog of water in the outside storage tanks, principally the vaste surge tank.
In order to reduce the releases to the environment during the period the vaste concentrator was out of service, the water normally processed through the concentrator was transferred to the ve;t* surge tank _to.be processed after the concentrator was, returned to service During the l
time the concentrator was not available, the regeneration of 'he l
condensate demineralizers was postponed to decrease the load on the radvaste facility.
Additional spare par + are being ordered for the vaste concentrator; and in additicn, consideracion is being given to the inst ation of a second waste concentrator as a part of our overall evaluation of the performance of the redanste facili ty.
Generator Load Rejection and Turbine Trip Anticipctary Scram Pypsos Switch, PSif-C, Fai_ lure While conducting a surveillance test oa August 2,1971 to check the trip set points of the generator load rejection anticipatory scram and the scram byr-er. P.
+d e n a l o v v a".,
cne generator load raj
.s L..
ou.
r"rtine trip anti:tpatory scram bypass switch, PSH-C,
,as iound closed (bypass position).
At reactor thermal power less than 45% of rated 1690 MWt, the generator load rejection and turbine stop valve closure anticipatory scrams are bypassed since below that point, the anticipetory scram is not required to limit reactor pressure for a turbine trip transient without bypass valve action.
HovcVer, at power levels greater than 45% of rated thermal output, both anticipatory scram functions are required to be in service.
The 45% set point corresponds to a specific turbipe third stage pressure belov which it is bypassed.
9
Dr. Peter A. Morris Page III O
~
September 9, 197 Upon discovery that the generator load rejection acram, as sensed by PSH-C, was bypassed, closer investigation revealed that a bad packing leak existed on the pressure sensing root valve for PSH-C.
A check of the pressure being sensed by that switch was made and found to be approximately 115 psig, which was less than its set l
point of 200 psig. At 115 psig, PSH-C would be closed, bypassing i
the anticipatory scram functions provided by PSH-C and the combina-l tion of main stop valve 3 and 4 2190% open.
Several attempts were made to stop the packing leak in the root valve and thereby permit PSH-C to sense proper third stage pressure.
These attempts proved unsuccessful, and a lead was lifted from PSH-C in order to place the previously bypassed anticipatory scram function in service.
Single failure criteria was never violated in that other sensors on protection bus I were operable and that both sets of sensors on protection bus II did not fail.
The root valve for PSH-C will be repacked at the first outage, and the pressure switch returned to service. An investigation is being conducted to determine if the root valves should be replaced with ones of a different design.
j Stw.iby Gas Treatment Systed train No. 1, Minimum Flow Valve, Failure to Open While conductir.o a surveillaned test on July 6, 1971 to check the trip set point of ve.rious arec radiation monitore in the plant which initiate the standby gas treat..ent system and to check tor propel operation of that system, the tinimum flow salve for tne selected train (no. 1),(,' 28-24), failed to open on the laitial. start.
The solenoid valve.hich ports sir to and from the valve operating cylinder was dir.cssembled, and the plunger shich is fitted with e rubber disc was found to be jarmed on its metal = cat.
The caura f
appears to be a combination of heat generate 3 by t' e c'eroid, the
/
r number of cycles of the valve, and the force exerted on tha plungt * / :
am 2a3 operation.
The aolenoid plun$er was replaced and the valve g was cycled catisfactorily.
Reactor building ventilation isolation and standby gas treatment system initiation occurs whenever one of four selected area radiation monitors reaches its upscale trip set point. Under normal conditions, the minimum flow valves open upon initiation of the standby gas treatment system trains and reclose when the nonselected train is automatically secured.
If the selected train should fail in operation, the nonselected train would start, the minimum flow valve in the faile1 train opens, and the suction cross connect valve, opens so that cooling air could be drawn through the filters of the failed train. During this test, the minimum flow e
r Dr. Peter A. Mor Pa6e IV F.rptember 9, 197 valve for the calactaa trn.in failed to nran gnn bitintion of the standby gas treatment system.
If this system had been called upon to operate, it would have performed its intended function.
It vould require a second failure with the train in operati'on to have cause to use the minimum flow valve. Although train no. 2 was not the selected system, its valves operated satisfactorily during the test.
The plungers in each of the solenoid operators will be inspected and replaced if warranted. Consideration is being given to replacing these solenoids with those of a different manufacturer.
Failure of Scram Dunn Volume Level Switch, RD08-B While conducting a surveillance test on August 17, 1971 to check the 7
Ahl..
j.,[
trip set points and protection system response to high level in the scram dump volume, level switch RDOS-B failed to function properly 9: f at the predetermined trip set point of 37 gallons.
" ~
.5 y.j} _.'
The scram discharge volume is used to limit the loss of and contain p cJW. '
the reactor primary vessel vater from all the drives during a scram.
?? ^
d;'y %
I'uring normal plant -pom. tion, the discharge volume is empty and
.i. '
b 3th its vent valves and drf:' valve are open. Upon a scram signal,
.g yd a'1 three valves close.
During.a ceram, the discharge volume partly (p,~,,
f1'.ls with water from over the drive pistons. After a scram, the VM e f. _
i cc trol rod drive seal leakage continues i.,o flew to the discharge 7CD$ -
vo.ume until the pressure in the discharge volume equals reactor
.c./ S.;
venel pressure.
([g Level switches RDOS-A through ? connected to th'e diccharge volume MUkb guarJ against the volume being in9.dvertently full when'a scram is
%Q1 requi' red.
Level ::vitch RD08-F se onds an alarm at three gallons h,.N in the volu.nc.
RDOS-:: counds an clarm and initiatec a rod block OI.d at 18 gallons, aal RD'. d-A through D result in e. reactor scram ac
/ c 'y, x.
37 gallons in the velume.
49 T
jf '.. ?
During this surveillance test, level switch RD08-B was isolated and
["#Y filled to the predetermined tr'ip level; and although the acram was 5..k initiated, switch operation appeared to be sluggish.
The level switch 4g.' _. [
was drained to that the operation could again be checked. However, d.y,f{
it was discovered at this time that the scram signal could be reset
@%gcg?
from the reactor protection system; but the alarm did not clear t
indicating an additional problem with the alarm contact on relay 1K8.
The second time the level switch was filled, no protection system trip
.N e
signal was initiated. The switch was drained and refilled several more d
4" times, but a trip signal was not initiated. The level switch was O i rf manually actuated several times causing a trip signal to be initiated,
. kip substantiating,the fact that the relay contacts associated with the g.y {
protection system were satisfactory. The alarm function, however, T
p,yJ was not satisfactory in that at times it would " clear" and at other Ley times it would remain annunciated.
s j
y
=
s' Dr. Peter A. Mor" s Page V
)
O Septe=ber 9, 1971 The level switch civot shaft was cleaned and lubricated since it appeared to be binding and switch actuation checked satisfactorily three times. The alarm contacts on relay 1K8 appeared to be oxidized and were burnished.
Subsequent operation of the alarm proved satisfactory.
All other switches performed satisfactorily during the test includin6 alarm, rod block, and scram functions.
We are enclosing twenty-five copies of this report.
Very truly yours, j'On JB Ivan R. Finfr ek, J.
Manager, Nuclear Generating Stations IRF/pk cc: Mr. J. P. O'Reilly, Regional Director Division of Compliance, Region 1 e
O b
e w
5
/
O k
e