ML20084Q216
| ML20084Q216 | |
| Person / Time | |
|---|---|
| Site: | Quad Cities  |
| Issue date: | 01/26/1974 |
| From: | Stephenson B COMMONWEALTH EDISON CO. |
| To: | Oleary J US ATOMIC ENERGY COMMISSION (AEC) |
| References | |
| BBS-74-19, NUDOCS 8306130172 | |
| Download: ML20084Q216 (3) | |
Text
-
(
i o
q a=,
[7}h C:mm:nw:alth Edis:n A
t0 C y Quad-Cities Generating Station Post Office Box 216
- I h;[
Cordt.,ta, lihnois 61242 Telephone 309/654-2241 ici A
Q.
f].
f S.
/
'f BB3-7"-19 t>Q {
}
O k
January 26, 1974
~] (. -,ggl%,,4 p w
p W
Mr. John F. O' Leary, Director i !, gg' Directorate of Licensing D-Regulation United States Atomic Enercy Commission Washington, D. C.
20545
Reference:
Quad-Cities nuclear Power Station, Unit 2, Docket No. 50-265.
DPR-30, Appendix A, Sections 1.0. A.2, 1.0. A.4, 3 2.B and 6.6.B
Dear Mr. O' Leary:
The purpose of this letter is to inform you of the details of an intennitteat failure on January 17, 1974 of a reactor water level instrument and subsequent valving error following maintenance. This level switch is one of four similar sensors providing a core cooling system initiation sicnal thus through the redundant components, total system operability was not affected. This abnormal occurrence was previously reported to you by telegram on January 18, 1974 and discussed with Region III, Directorate of Regulatory Operations on January 18 !.: 24, 1974 by phone.
PROBIE1 AND I??!ESTIGATION On January 17, 1974 routine quarterly calibration of the Unit 2 reactor low low water level switches was performed. These switches are set to trip at a test pressure of 112 (-4,+0)" U20 increasing. Test pressures less than 115" H2O correspond to reactor water level > 83" above the fuel. When a pressure in excess of 120" H2O was applied to switch number LIS 2-263-72 it did not actuate the first time. The instrument mechanic inspected the switch internals and calibrated the switch to trip at 110.2" H20. In addition, the switch was cycled twelve times to demonstrate repeatability and the failure did not recur.
On January 18, 1974 the switch, a Yarway Model 44160, was inspected again and the mercoid switch involved was replaced. A closer examination of the switch in the shop detected traces of contamination in the mercury.
On January 24, 1974, LIS 2-263-72D was observed to be reading about 20"
'&en the switch was inspected again on that higher than instruments 72A,mC.
day it was discovered that the instrument isolation valves _in both sensing lines and the equalizing valve were closed. Apparently the instrument had not bcen properly returned to service in accordance with procedures following
[
maintenance on the 18th. The switch responded properly when it w halved in,,
g on the 24th.
COPY SENT REGION-b B306130172 740126 a
PDR ADOCK 05000265 i
%g PDR S
p
_ ~
U
(_)b R
N L
Mr. J. F. O' Leary January 26, 1974 EVAUJATION AND CORMETIVE ACTION SAFE W IMPLTCATIO!IS During the period of time that the 723 switch was valved out and the period during which its operability was questionable because of the mercoid switch, the renaining three sensors were fully operable. Since these sensors are combined in a one-of-two-twice logic, the capability of core cooling system initiation on Icw low level was available at all times. Also the initiation capability of the four high drywell pressure sensors was not affected by this occurrence. The implications of not returning a safety-related instrument to service and not detecting this condition for six days are of concern even though this was the first such occurrence and the effect of the error did not result in any system being disabled. It is believed that the corrective action described below will prevent a recurrence. This was also the first failure of this nature experienced with this type of instrument thus there are no safety implications in view of cumulative experience.
DEISC4TI ATIOIT OF CAUSE A"3 CORRECTIVE ACTION The Yarway level switch involved contains two mercoid switches set to actuate at a 112" H2O differential pressure. These internal switches are tripped by two permanent magnets. 'lhon one of the two switches failed to actuate, the mechanic renoved the cover and checked the alignment of the switch with respect to the uagnet before he attempted to ectuate the switch again. As stated previously, traces of foreign material were also observed in the mercury. In reviewing the failure we have concluded that there are three possible causes:
(1) the foreign material inhibited the switch move-ment; (2) the switch was slightly misaligned in its bracket such that the magnet's field wasn't strong enough to actuate it everytime; and (3) the strength of the mcnet's field has weakened and the switch needed to be moved closer to it.
Of these possibilities, nu-ber (2) scens the most probable at this time.
'le are continuing to investigate this problem with the supplier and will advice you by letter of the results. Functional testing of this switch will also be conducted weekly during this investigation period.
With regard to the error in not returning the instrument to service, the procedure used for the post-maintenance calibration was reviewed. The procedure was adequate in that it contained the proper controls for informing the reactor operator when the instrument was removed from service and returned to service.
The checklist also required the instrument mechanic's initials to indicate-completion of these steps. A review of the readings taken on the instrument from the 18th through the 24th, however, indicated that it was valved out during this period.
Although the indicating feature of these instruments is much less accurate than the switch setroints, the discrepancy between the 72D sensor reading and the others should have been detected before the 24th by the required daily instrtment check.. In this respect the station procedures will be improved by adding criteria which will specify an acceptable variation between seilsors and an action level requiring inmediate corrective action. This will be accomplished by March 1,1974 in conjunction uith our efforts to upgrade all station procedures to meet the requirements of Regulatory Guide 133 m
(
"N (b
(T,_
V Mr. J. P.'O' Leary January 26, 1974 This occurrence and its significance will also be reviewed in detail with all operatins personnel and instru~ent mechanics. The following points are being stressed to prevent a repetition: (1) the importance of following step-by-stop maintenance procedures and the potential hazards of being "tco familiar" with the procedure; (2) the importanco of checklists when used with a procedure to insure that critical steps are not omitted; (3) the importance of the reactor operator and his log as a means of detecting this type of error; and (4) the inportance of closely reviewing what may seem to be " routine readings" by cupervisory personnel.
Very Truly Yours, CO:TO TSALTII..EDISC?i CO:" PLY QUAD CITIEY IEUt ;,J.E:t STATICII
/
/> bh/h
. S,tep.ehson Station S'perintendent bBS/rhb cc: Regional Director Directorate of Regulator /' Operations - Region III l
l k