ML20084K458

AO 50-237/1974-51:on 741006,during Shutdown,Zero Flow Indicated on Chimney Monitoring Sys.Caused by Gasket Failure for Glass Jar Muffler on Discharge Side of a Sample Pump Due to Deterrioration.A Sample Pump Replaced

ML20084K458
Person / Time
Site:	Dresden
Issue date:	10/17/1974
From:	Stephenson B COMMONWEALTH EDISON CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
Shared Package
ML20084K461	List: ML20084K458 ML20084K463 ML20084K477
References
751-74, NUDOCS 8305190430
Download: ML20084K458 (3)

Text

- Commonwrith Edia::n One First nal Plaza Chicago. It!.nois

- Address to: Post Office Box 767 i Chicago, lilinois 60390 1 BBS Ltr.#751-74 Dresden Nuclear Power Station R. R. #1 Norris, Illinois 60450 m

. October 15, 1974

p. , 'N p-

/ d_ W I s

fx,K' %

r 0

f J .,

t._

Mr. James G. Keppler, Regional Director t

~~

00T211974 c Fa 4 Directorate of Regulatory Operations-Region III u "Gy (mn U. S. Atomic Energy Co:nnission 1 EIij 799 Roosevelt' Road W 8Mca 4 Glen Ellyn, Illinois 60137 OJ s;>

m SUBJECTr REPORT OF ABNORMAL OCCURRENCE PE'1 SECTION 6.6.A 0F THE TEENICAL SPECIFICATIONS.

LOW SAMPLE FLOW ON D2/3 CHIMNEY MONITOR.

References:

1) Regulatory Guide 1.16 Rev.1 Appendix A

2) Notification of Region III of AE Regulatory Operations Telephone: Mr. F. Maura, 1700 hours0.0197 days <br />0.472 hours <br />0.00281 weeks <br />6.4685e-4 months <br /> on October 8, 1974 Telegram: Mr. J. Keppler, 0830 hours0.00961 days <br />0.231 hours <br />0.00137 weeks <br />3.15815e-4 months <br /> on October 9, 1974 Report Number: 50-237/1974-51 Report Date: October 15, 1974 Occurrence Date: October 6, 1974 Facility: Dresden Nuclear Power Station, Morris, Illinois IDENTIFICATION OF OCCURRENCE A zero flow was indicated on the Unit 2/3 chimney monitoring system. At the time of the incident, this was thought to,be contrary to Section 3 8.A.1 of the Dresden Unit 2 and 3 Technical Specifications which requires that at least one chimney monitor be in service at all times. Further investigation indicated that no Technical Specification violation had occurred since the chimney monitoring systems were in operation during this period.

CONDITIONS PRIOR TO OCCURRENCE At the time of the occurrence, Unit 2 was'in the shutdown mode and Unit 3 was in the run mode at 530 s e and increasing power at 50 MWe/hr.

03d519o43o741015

$DRADOCK 05000237 D 6I eDR YQ m.:t=

47-a39.

a pel>N

.0CT 1 7 1974 COPY SENT REGION 7

n - '

a -

. . . . . . . . y _g +

Mr. James G.'Kipp r October 15, 1974 s

DESCRIPTION OF OCCURRENCE At approximately 1400 hours0.0162 days <br />0.389 hours <br />0.00231 weeks <br />5.327e-4 months <br /> on October 6, 1974, a radiation protectionman, during his routine filter change at the Unit 2/3 chimney monitoring system, noted a zero flow indication on the system flow meter. He also noted that "A" cample pump was operating, the pump vacuum was normal (11 in.) and the correct valves were open. After changing the filters, a zero flow was still indicated on the flow meter.

Operations was notified of the above condition at approximately 1410 hours0.0163 days <br />0.392 hours <br />0.00233 weeks <br />5.36505e-4 months <br />.

At this time, the power level increase on Unit 3 and the startup of Unit 2 were delayed until the chimney monitoring system could be repaired. A shift foreman, investigating the occurrence, secured the "A" sample pump and started "B" sample pump. When "B" sample pump would not operate properly, he started "A" sample pump and fully opened the "A" pump flow control valve. The sample flow then increased to 1.0 cfm.

Maintenance was notified at approximately 1450 hours0.0168 days <br />0.403 hours <br />0.0024 weeks <br />5.51725e-4 months <br /> and while "A" cample pump remained operating, maintenance replaced "B" sample pump. The system was switched to the new "B" sample pump and the flow meter then indicated a nomal system flow of 2 5 cfm. These repairs were completed on October 6 at 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br />. "A" sample pump was replaced on October 8.

DESIGNATICU OF APPARENT CAUSE OF OCCURRENCE (Component Failure)

Subsequent investigation indicated that the gasket for the glass jar muffler on the discharge side of "A" pump had failed due to deterioration. The pump was pulling a nomal sample through the iodine and particulate filters and through the chimney monitors. However, due to the deteriorated gasket, the sample. flow was leaking from the muffler into the chimney monitor building (and eventually to the chimney through the max-recycle building ventilation) and was not passing through the flow meter on the discharge side of the pump.

Follow-up investigation to determine the cause of the "B" sample pump failure showed that the pump failed due to worn pump bearings.

ANALYSIS OF OCCURRENCE At no time during this event was the function of the chimney monitoring system impaired. This statement is substantiated by the following:

~

1. The vacuum gauge on the pump suction indicated normal operating vacuum levels.

2. -The chimney monitor readings did not decrease. Normally, upon loss of sample flow, the. monitor will indicate a definite decrease due to the loss of noble gas flow and the decay of noble gas daughter products.

There was no such inciication during the period of this incident.

. James G. Kipp -3 October 15, 1974 3.- Analysis of iodine and particulate filters from the period in question indicated that the total amount of cetivity collected was nomal for the ex' i sting operating conditions.

Since the chimney discharge was monitored continuously, and the leaking gases exhausted back to the chimney, there was no violation of the Technical Speci-fications, nor any safety implications.

CORRECTIVE ACTION The corrective action at the time of the event was to replace the "B" sample pump in order to obtain the necessary sample flow. Further action involved replacing the "A" sample pump. During this action the muffler Jar gasket problem was discovered and the gasket was replaced.

Further action is not deemed necessary due to the fact that the sample flow-rate is monitored on a daily basis which in light of this event will allow for rapid detection of this problem should it recur.

FAILURE DATA Although there have been sample pump failures in past history, the problem with a deteriorated gasket is a unique event with no precedent. All previous history in this area involves bearing failure or pump impeller failure.

Sincerely,

b. B. B. Stephenson Superintendent BBS DAAndo en 9

+

x.