ML20235V629

Special Rept:On 870614,15 & 16,diesel Generator 2B Experienced Six Low Lube Oil Pressure Trips & on 870621,2B Exceeded 11 Max Start Time.Trips Caused by Sediment or Air Bubble in Sensing Line.Oil Line Cleaned & Returned to Svc

ML20235V629
Person / Time
Site:	Mcguire
Issue date:	07/16/1987
From:	Tucker H DUKE POWER CO.
To:	NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM)
References
NUDOCS 8707230251
Download: ML20235V629 (5)

Text

,_ _ ___ -__-__

e l

'DuxE POWER GOMPAhT P.O. DOx 3318D l CHARLOTTE, N.C. 26242 HAL 11. 'IUCKER rzzzrnowz ,

vuscosasemat (704) 073-4531 mum san enoscorion July 16, 1987 l U.S. Nuclear Regulatory Commission Document Control Desk j Washington, D.C. 20555 Subjec t: McGuire Nuclear Station Docket No. 50-370 Technical Specification 3.8.1.1 and 6.9.2 Gentlemen:

Please find attached a special report submitted pursuant to Technical Speci-fication (T.S.) 6.9.2 as specified by T.S. 3.8.1.1. The report concerns six invalid failures on Diesel Generator 2B due to air and/or sediment in an oil line and a valid failure due to an unknown cause.

This event is considered to be of no significance with respect to the health and safety of the public.

Very truly yours, W

Hal B. Tucker SEL/89/jgc Attachment xc: Dr. J. Nelson Grace INPO Records Center Regional Administrator, Region II 1820 Water Place U.S. Nuclear Regulatory Commission Atlanta, Ga 30339 101 Marietta St., NW, Suite 2900 Atlanta, GA 30323 Mr. Darl Hood Mr. W.T. Orders U.S. Nuclear Regulatory Commission NRC Resident Inspector Office of Nuclear Reactor Regulation McGuire Nuclear Station Washington, D.C. 20555

~ '\/

8707230251 870716 ft PDR S

ADOCK 05000370 PDR p \

Diesel Generator 2B Experienced Six Invalid Failures Due To Air And/Or Sediment In An Oil Line And A Valid Failure Due To An Unknown Cause

Introduction:

On June 14, 15, and 16, 1987, Diesel Generator 2B experienced six Low Lube Oil Pressure trips during start-ups for post maintenance break-in and troubleshooting runs. The problem was traced to sediment or an air bubble in a section of the lubricating (lube) oil pressure sensing line. The line was cleaned and vented, and Diesel Generator 2B was successfully started.

On June 21, 1987, Diesel Generator 2B exceeded the eleven second maximum start time during a routine start-up. The cause for the slow start could not be determined. Four more start-ups were made to troubleshoot the diesel and all ,

were successful. Diesel Generator 2B was tested with the routine operability test and returned to s vice at 0712 on June 22, 1987.

Unit 2 was in Mode 5, ,old Shutdown, at the time of these incidents.

The first 6 incidents have been attributed to a Defective Procedure. There were no instructions in the procedures on how to properly vent the lube oil pressure sensing lines. The root cause of the slow start could not be determined.

EVALUATION:

Background

Each unit at McGuire Nuclear Station (MNS) has two independent Diesel Generators (D/Gs). As part of the Essential Power System they provide standby AC power to the equipment required to safely shut down the reactor in the event of a loss of normal power. The D/Gs also supply power to the safeguard equipment as required during a major accident coincident with a loss of normal power (blackout).

The automatic start /run circuits for the diesel have several interrupt circuits, one of which is for low lube oil pressure. This interrupt circuit shuts the diesel down when the lube oil pressure is too low for continued operation. The signal itself is routed through a start delay timer which requires 20 seconds to elapse during an engine start before the low pressure trip signal can trip the diesel. This is one of four trip signals which are not bypassed during an emergency automatic start initiated by the Engineered Safety Features System.

Technical Specification (TS) 3.8.1.1 requires that the D/Gs achieve 488 RPM, 57 Hz, and 4160 VAC within 11 seconds during a start to be considered operable. TS 3.8.1.2 applies in Mode 5 and requires one D/G to be operable in order to conduct any fuel movement or reactivity changes. At the time of these events no fuel movement or reactivity changes were in progress and D/G 2A was operable; therefore, no TS action statements applied.

L _ ______-_-__-____-_-

Description of Event Operations and Maintenance started D/G 2B on June 13, 1987 in accordance with HP/0/A/7400/65, Diesel Generator Break-in Runs, at the conclusion of the refueling outage periodic maintenance on the diesel engine. D/G 2B then experienced six intermittent Low Lube 011 trips beginning with the sixth start (start no. 537, Invalid Test Failure) of the break-in run set and ending with a trip during an adjustment troubleshooting run (start no. 546, Invalid Test Failure).

The symptoms indicated on the instrumentation were not consistent. The high differential pressure across the in-line filter was indicating the filter was clogged, but other local gauges indicated the lube oil flow to the engine was adequate. The oil pump discharge pressure was indicating within the normal range, but the pressure switches which trip the diesel had actuated on low pressure. Maintenance inspected the filter and found it clean and fully functional. The problem was eventually traced to a short length of lube oil pressure sensing tubing near where the line penetrates the engine crankcase.

The crankcase penetration for this line is located immediately under the crankshaft and the line has several bends which form a seal loop type of configuration before going to the instruments. An air bubble in this section of the line can cause unpredictable instrumentation readings and sometimes acts like a restricting orifice. This section of tubing was removed, cleaned, reinstalled, bled, and returned to service. Start no. 547 (Valid Success) was then performed for PT/2/A/4350/15B, Diesel Generator 23 Operability Test, and D/G 2B was returned to service.

At 0255 on June 21,1987, D/G 2B was started (start no. 553, Valid Failure) to facilitate changing the 4160 VAC system alignment. The diesel required 14.8 seconds to reach 95% rated speed. Operations declared D/G 2B inoperable and wrote a work request to investigate and repair the diesel. Four troubleshooting starts were performed with numerous technical personnel in attendance and all four starts were within the valid start criteria. A review of all the data available failed to reveal a cause for the slow start. D/G 2B was successfully started again (start no. 558, Valid Success) for PT/2/A/4350/15B, Diesel Generator 2B Operability Test, and Operations declared it operable at 0712 on June 22, 1987.

Conclusion These incidents resulted in six invalid test failures and one valid test failure of D/G 2B. Previously, Unit 2 D/Gs had 4 valid failures in the last 100 valid attempts and 1 valid failure in the last 20 valid attempts resulting in a monthly testing interval. As a result of the valid failure, the ratios changed to 5 valid failures in the last 100 and 2 in the last 20. This l changed the testing interval to a weekly basis. ,

{

In the case of the first six trips the root cause was the incomplete bleeding of the oil pressure instrumentation sensing lines. Historically, the mechanical maintenance crew working on the diesel has bled the lines during the initial start-up of the engine af ter an overhaul or other major periodic maintenance, as they did in this case. This practice has proven satisfactory in the past. There are no instructions in the maintenance procedures on how j to perform an oil line bleeding operation. These six incidents have been I attributed to a defective procedure.

L - - - -------- ------___ _ ________ __ _ ___--___

1 The path the instrumentation sensing line follows af ter leaving the engine crankcase forms a loop seal where sediment can collect and several points in the line form air bubble traps. An air bubble in an oil filled line does not act the same as an air bubble in a water line. Sometimes the bubble acts as a restricting orifice and at other times like a dash-pot. The air bubble also does not bleed out as quickly as in a water system. Therefore, a certain amount of experience is required before a technician becomes sufficiently skilled at bleeding an oil line to do a complete job on the first attempt.

Also, the lube oil system must remain pressurized for a period of time before the bleeding can be completed and the first few runs of the diesel engine in the break-in run series are too short to adequately purge the pressure sensing lines. Because of these complications an oil pressure trip is not totally unexpected during the first operations of the diesel engine after the lube oil syatem has been drained and refilled.

After the sixth trip, Instrumentation and Electrical (IAE) personnel were

. asked to assist in resolving the problem, and IAE recommended cleaning a particular section of the tubing which was vulnerable to blockage. After cleaning and reinstalling that portion of the tubing the problem was resolved and the instrumentation resumed normal indication. Because of the location of the line, the technician could not see whether just air or air and sediment came out of the tube as it was being removed and cleaned.

In the case of the slow start there were no indications of the cause or any of the instrumentation. The symptoms did not exactly fit any of the common cause symptoms with which station personnel were familiar. The four troubleshooting starts did not reveal any discrepancies at all and all the acceptance criteria for the D/G were met in each case. An after the fact discussion between Maintenance and a Woodward Co. factory representative brought out a problem with the effects of air in the internal lines of the Governor / Actuator (G/A).

Since the symptoms and conditions were so similar to what they saw, Maintenance suspect that air bubbles in the G/A were the cause. During outage preventative maintenance, the oil in the G/A was changed. There are no provisions on the G/A to bleed air out of the internal lines; rather, the manufacturer suggests operating the engine and allowing the G/A to purge itself. If there is air in the lines in the G/A, the effect is to cause a slow start. The slow start will not necessarily be the first one, but is usually one of the first few. The root cause of the slow start could not be clearly identified.

A review of past McGuire reports indicates there have been two previous reports of a similar nature. One reported a problem with sediment in the lube oil sensing lines causing a trip. The symptoms were very similar to those seen in this incident, even though the cause in that case was clearly sediment. The corrective action was to replace some small tubing with larger tubing and would not have prevented this incident. The second reported a slow start with no indication of a cause.

This incident is not Nuclear Plant Reliability Data System (NPRDS) reportable.

~

CORRECTIVE ACTIONS:

Immediate: None Subsequent: The oil instrumentation line was cleaned, bled, and returned to service.

Planned: None SAFETY ANALYSIS:

In the case of the six Low Lube 011 trips, the D/G was out of service for periodic maintenance. The D/G operations were to break in the engine and prepare it for normal operation. Since this was a planned condition and no TS related actions were required, there are no safety concerns.

In the case of the slow start, the unit was in a condition where the D/G was not required to be operable. Had this event occurred during operations in a higher mode TS action would have been required. All available information indicates that D/G 2B would have started and loaded with the emergency loads had it been required for an incident or accident.

There were no personnel injuries, personnel overexposure, or releases of radioactive material as a result of this incident.

The health and safety of the public were not affected by this incident.

{

_ _ _ _ _ - - _ - __ - _ -