ML20129D635

Rev 0 to Action Plan 26, Inadvertent Auxiliary Feedwater Pump 1 Suction Supply Transfer from Condensate Storage Tank to Svc Water Supply

ML20129D635
Person / Time
Site:	Davis Besse
Issue date:	06/26/1985
From:	Czuba T TOLEDO EDISON CO.
To:
References
NUDOCS 8507300119
Download: ML20129D635 (7)

Text

_ . _ . . - _ . ._=

,4 N

c-G 4

T l

ACTION PLAN # 26 TITLE: Inadvertent Auxiliary Feedwater Pump #1 Suction Sunniv Transfer from condanante Stormee Tank to Service Water Suvolv i

1 L

l APPR.

CHAIRMAN FOR REV DATE REASON FOR REVISION BY TASK FORCE IMPL.

0 6/26/85 Initial Issue

~u

[F 8507300119 850626 PDR P

ADOCK 05000346 PDR 9

J e

--s.-- w y - p ,- y-n , . 6-, ,w-y + -y ..,----v- vmer ms r -y-w er---, ,. ms 3 --e

. v- . - -

' ,, TITLE: Intdvartrat Auxilicry Fosdwater Pump #1 Suction Supply Trcasfer from Condznaate Stcrege Tenk to Sarvice Wster Supply REPORT BY: Timothy Czuba PLAN NO: 26 DATE PREPARED: June 26, 1985 PAGE: 1 of 4 INTRODUCTION:

The following is the analysis and evaluation to support the action plan for determining the root cause for the auxiliary feedwater pump #1 suction supply transfer from the condensate storage tank to service water. This report has been prepared in accordance with the " Guidelines to Follow when Troubleshooting or Performing Investigative Actions into the Root Causes Surrounding the June 9, 1985 Reactor Trip," Rev. 4.

1

SUMMARY

OF DATA:

This plant has experienced inadvertent auxiliary feedwater suction supply transfers on auxiliary feedpump #1 during pump startups. An analysis of this situation had been initiated prior to the June 9,1985 reactor trip.

The analysis had indicated a need for a time delay in initiating suction transfer due to transients in the auxiliary feedwater supply header on auxiliary feedpump start-up.

A. June 9, 1985 Sequence of Events:

i 1. The #1 auxiliary feedpump turbine speed was approximately 3625 rpm. The Auxiliary Feedpump #1 low suction pressure alarm i actuated. This alarm will actuate on 11 psi decreasing.

i (See time 1:57:53 in Section B.)

2. The auxiliary feedpump #1 low suction pressure alarm cleared.

(See time 1:58:27 in Section B.)

3. The auxiliary feedpump turbine #1 stop valve closed. (See time 1:58:34 in Section B.)

4. The auxiliary feedpump #1 suction supply transferred from the condensate storage tank to service water. (See time 1:58:40 in Section B.)

5. The auxiliary feedpump #1 suction supply was manually reset which returns suction to the condensate storage tanks. (See i time 2:00:31 in Section B.)

r .

B. Computer Alarm Printout Data:

TIME EVENT REF. PUMP SPEED (RPM) 1:57:53 APF 1 Suction Pressure Low P006 3623 1:58:27 AFP 1 Suction Pressure Norm P006 3471 1:58:34 AFPT 1 Stop Valve No 2001 2707

_ ,, _ - . - - -- -.,3. -- -.. ~ - . - - , ,,m

, ', TIME EVENT REF. PUMP SPEED (RPM) i 1:58:40 AFPT Suct Xfer to SW or PSL P008 2706 2:00:31 AFPT Suct Xfer to SW or PSL P008 1767 Norm C. Maintenance and Survei? lance / Testing History of Transfer Schemes:

3/25/85 ST 5071.03, Section 6.6 (Calibration of the Auxiliary Feedpump 1-2 l Suction Pressure Switches) was performed. The calibration was '

within tolerance. No adjustments were made.

4/8/85 ST 5071.03 Section 6.2 (Calibration of the Auxiliary Feedpump 1-1 7 Suction Pressure Switches) was performed. The calibration was within tolerance. No adjustments were made.

4/22/85 ST 5071.03, Section 6.6 (Calibration of the Auxiliary Feedpump 1-2 l Suction Pressure Switches) was performed. The calibration was l within tolerance. No adjustments were made. I 5/6/85 ST 5071.03 Section 6.2 (Calibration of the Auxiliary Feedpump 1-1 Suction Pressure Switches) was performed. The calibration was within tolerance. No adjustments were made.

5/20/85 ST 5071.03, Section'6.6 (Calibration of the Auxiliary Feedpump 1-2 Suction Pressure Switches) was performed. The calibration was within tolerance. No adjustments were made.

6/4/85 ST 5071.03 Section 6.2 (Calibration of the Auxiliary Feedpump 1-1 Suction Pressure Switches) was performed. The calibration was within tolerance. No adjustments were made.

CHANGE ANALYSIS:

According to the maintenance history taken from instrumentation and controls equipment records and from Davis-Besse Maintenance Management System, modifications of the system are as follows:

9/1/81 Auxiliary Feedpump Suction Line Strainers S-201 and S-206 were modified to prevent plugging per FCR.79-0215.

HYPOTHESES:

Hypotheses . for the auxiliary feedwater pump #1 inadvertent suction trans-fer are based on infonmation received during the transfer, information obtained from equipment records, DBHMS, and the computer alarm printout.

A. Hypothesis #1:

Suction header pressure switeb's 26s ,928A and PSL 4928B setpoints are out of specification.

, 3 B. Hypothesis #2:

The low suction pressure alarm pressure switch PSL 503 is out of specification and failed to alarm on an actual low suction pressure condition.

C. Hypothesis #3:

Pressure switches PSL 4928A and PSL 4928B were inadvertently actuated by vibration.

t D. Hypothesis #4:

Momentary loss of power to motor operated valves AF 786 and SW 1382 and their control circuits.

This hypothesis is considered infeasible for the following reason:

This system is fail safe and upon loss of power the valves will not transfer. The computer alarm point (P008) indicated that the valves did transfer. Therefore, there could not have been any loss of power to motor operated valves AF786 and SW 1382 and their control circuits.

E. Hypothesis #5:

Operators may have manually transferred suction supply to service water after seeing the low suction pressure alarm.

The scenario was discussed with the operators who were on duty during the transfer and no one recalled manually transferring auxiliary feed-pump suction supply to service water. No los entries were made con-cerning a manual transfer. Due to the nature of this hypothesis, no work can be performed to prove or disprove this hypothesis.

F. Hypothesis #6:-

1 Auxiliary feedpump suction lina strainer S-201 was clogged during the transient and caused incdvertent transfer of auxiliary feedpump #1 '

suction transfer.

This hypothesis is considered infeasible for the following reason: i The pressure switches PSL 4928A and 4928B are located upstream of the strainer and would not see the pressure drop of a clogged strainer.

The computer point (P002) for the differential pressure across the strainer was not indicating an alarm condition during the transient.

G. Hypothesis #7:

A' low suction pressure transient was induced into the system causing suction supply to be transferred to service water supply. In my discussion with the operator on shift on June 9, 1985, he stated that  !

he saw the station annunciator alarm 'for low suction pressure for j auxiliary feedpump #1. Then he saw that valve transfer to service water had taken place.

l

, -4 If the time response of the interlock control circuit is faster than the computer scan rate, it is possible that a low suction pressure transient could have caused the transfer without the computer seeing the transient, but the station annunciator would.

H. Hypothesis #8:

A low suction pressure was caused by strainer S-257 being clogged.

The low suction pressure was properly sensed in the auxiliary feedpump

1. 1 piping causing suction transfer to occur. S-257 is located in the condensate auction header which is common to both feedpumps during normal suction lineup. A suction transfer did not occur in the auxiliary feedpump #2 supply because PSL 4929A and PSL 4929B did not actuate due to their setpoints being out of specification.

i se a/12'

ACTION PLAN PLAN NUMBEH PAGE ,

b?

_ _ _ _ . . _ _ _ -Raymo 26 i

' TITLE DATE PREPARED PREPARED BY INADVERTENT AUXILIARY FEEDWATER PUMP #1 SUCTlo)LSl!PPLLTRANSEER SPECIFIC OBJECTIVE 6/26/85 T- 'Czuba _

To determine root cause of the Auxiliary Feedwater Pump #1 transfer to service water.

STEP PRIME ASSIGNED START TARGET DATE NUM8ER A TION STEPS RESPONSIBILITY TO DATE DATE COMPLETED ALL STEPS OF THIS ACTION PLAN ARE T0_BE PERFORMED IN ACCORDANCE WITH THE LATEST REVISION OF " GUIDELINES TO FOLLOW WHEN TROUBLE-SHOOTING OR PERFORMING INVESTIGATIVE ACTIONS INTO THE ROOT CAUSES SURROUNDING THE JUNE 9, 1985, REACTOR TRIP."

1. Independently verify by a calibration check the setpoint at T. Czuba T. Czuba which the pressure switches (PSL 4928A and PSL 4928B) actuate. K. Yarger

2. Simulate a low suction pressure on both pressure switches T. Czuba T. Czuba PSL 4928A and PSL 4928B simultaneously, and verify proper valve K. Yarger actuation occurs on AF 786 and SW 1382. Monitor PSL4928A and PSL4928B for actuation during auxiliary feedpump testing, which is being performed by action plans IA and IB. Steps 1 and 2 address Hypothesis #1. -

3. Check the calibration of PSL 503 and verify actuation occurs T. Czuba T. Czuba at the required setpoint. This step addresses Hypothesis #2. K. Yarger Kanut

4. Perform a seismic analysis on PSL 4928A and PSL 4928B mounting. Sagooleim T. Czuba This step addresses Hypothesis #3. K. Yarger

ACTION PLAN PLAN NUMHE R PAGE .. ,

Rev__.0 26 2 *' . 2 TITLE DATE PREPARED PREPARED SY INADVERTENT AUXILIARY FEEDWATER PUMP #1 SUCTION SUPPLY TRANSEER 6/26/85 T Czuba _

SPECIFIC oSJECTIVE To determine root cause of the Auxiliary Feedwater Pump #1 transfer to service water.

STEP PRIME ASSIGNED START TARGET DATE A '

NUM8ER RESPONSIBILITY TO DATE DATE COMPLETED

5. Perform response time on the interlock control circuit. T. Czuba T. Czuba This step addresses Hypothesis #7. K. Yarger  !

< 6. Independently verify by a calibration check the setpoint T. Czuba T. Czuba q at which the pressure switches (PSL 4929A and PSL 4929B) K. Yarger actuate. Monitor PSL 4929A and PSL 4929B for actuation during auxiliary feedpump testing, which is being performed by action plans IA and IB.

7. Check the calibration of PSL 509 and verify the actuation T. Czuba T. Czuba j occurs at the required setpoint. Steps 6 and 7 address K. Yarger Hypothesis #8.

NOTE: The action steps may be performed in any sequence.

i i