ML20132G756

Rev 2 to Procedure 12, Auxiliary Feedwater Sys Valve Problem Analysis (AF 599 & AF 608)

ML20132G756
Person / Time
Site:	Davis Besse
Issue date:	06/26/1985
From:	Long J TOLEDO EDISON CO.
To:
Shared Package
ML20132G743	List: ML20132G738 ML20132G745 ML20132G750 ML20132G756
References
12, NUDOCS 8508050133
Download: ML20132G756 (6)

Text

,

r. .

ACTION PLAN i 12 TITLE: Auxiliary Feedwater System Valve Problem Analysis (AF 599 & AF 608)

APPR.

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

0 6/14/85 Initial Issue See Rev. O for Approval General tormat enanges. claritications as result of discussion with the See Rev. 1 for Approval 1 6/16/85 c Revise Hypothesis #7 and add step 12 2 6/26/85 to the action plan. J. Long [Mp' v t

8508050133 850626 PDR ADOCK 05000346 p PM

a .

TITLE: Auxiliary Feedwater System Valve Problem Analysis (AF 599 & AF 608)

REPORT BY: James W. Long PLAN NO.: 12 DATE PREPARED: June 16, 1985 PAGE 1 of 2 This report has been prepared in accordance with the " Guidelines to follow when troubleshooting or performing investigative actions into the root cause surrounding the June 9, 1985, reactor trip", Rev. 2.

INTRODUCTION The following report is the analysis and evaluation to support the action plan for determining the root cause for the failure of AF 599 and AF 608 (AFW to SG isolation valves) to open during the June 9, 1985 reactor trip.

SUMMARY

OF DATA AF 599 and AF 608 are normally locked open valves and were open prior to the transient. During the transient, both valves went shut automatically because of the improper initiation of SFRCS. After the SFRCS was reset,'

both valves failed to open automatically. Operators were then sent to open the valves manually. According to the operator, the valves were placed in manual and the handwheel turned in the open direction. The handwheel was hard to turn and was only moved 1/2 turn in the open direction. The handwheel was then turned in the close direction 1/2 turn to try and'get a hammerblow in the open direction. This was repeated a second time, and when turned in the close direction the second time a rattling noise came from the operator and the valves were opened electri-cally. This rattling noise was probably the tripper fingers being kicked out by the motor and is to be expected.

The actual differential pressure (DP) seen by these valves at the time they were attempting to open is unknown but they are designed to open against a 1050 PSID. At 1515 on 6/9/85, both valves were cycled satisfac-torily within their required stroke time. At that time S/G pressure was 850 PSIG.

A review of maintenance and surveillance testing history shows that the torque switch settings were changed in March 1984 per FCR 84-0039 as a result of the Limitorque motor operated valves study and both valves were satisfactorily tested per ST5064.01 (CTMT Isolation Valve Post Maintenance

. Testing). During the 1984 Refueling Outage, the motors and magnetic brakes were replaced on both valves per FCR 83-0067. The brakes were replaced as part of the environmental qualification of safety related electrical equipment program (10CFR50-49 Rule Requirements). In addition.

AF 599 was disassembled. 3 bearings replaced, relubricated, and reassem-bled. During the testing of AF 599 following this maintenance, a loose spacer was found in the spring pack. This was corrected and both valves were tested satisfactorily. The only normal testing for these valves is a

_ _ stroke time per ST 5071.02 (AFW System 18 month refueling test), which was performed on 12/31/84, with satisfactory results. The brakes were replaced during the outage as noted above.

=

p .

AFS Analysis Con'd. 2 CHANGE ANALYSIS The only changes identified from the testing performed on 12/31/84 and the 6/9/85 reactor trip is the plant condition when the valves were cycled.

On 12/31/84, the plant was in Mode 5 t'.,eefore, the plant was cold and at i low pressure so the valves did not see a high DP across them. During the i 6/9/85 reactor trip, the plant was in Mode 1, therefore, at normal operat-ing temperature and pressure. Because of their location in the Mechanical Penetration Rooms, they would probably have been close to the ambient temperature of the rooms. However, they would have seen full S/G pressure when shutting. If the upstream check valve leaked, any pressure trapped between the check valve and AF 599 (AF 608) would have bled off. This would have caused a high DP across the valves when attempting to open.

HYPOTHESES Based on the information collected before, during and after the transient, it appears that both valves torqued out when opening. The following is a list of the hypothesis that could cause a valve to torque out. The Action Plan Item that will prove or disprove each hypothesis is listed.

. 1. Improperly adjusted torque switch bypass contact (this hypothesis covered by Action Items 3 and 5).

~

2. Improper torque switch setting (this hypothes1s covered by Action Item 2).

3. Wrong or improperly adjusted spring pack (this hypothesis covered by Action Item 7).

4. Failure of motor brake to release when energized or engage when deenergized (this hypothesis covered by Action Items 3 and 4).

5. Improper torque switch setting calculations (this hypothesis covered  !

by Action Items 8-11).

6. Improper torque switch installation (this hypothesis covered by  !

Action Item 6).

2 7. High DP across valve (this hypothesis covered by Action Item 12).

O i

t l

l

b ACTION PLAN PLAN NUMBER PALE Rev. 2 12 1 a' 3 sec..a faitt DATE PREPARED PREPARED BY ADI SYSTEM VALVE PROBLEM ANALYSIS (AF 599 and 608) 6/26/85 M. Bajestani SPECIFIC OBJECTIVE Ta Determine the Root Cause of Motor Operated Valve AF 599 and 608 Failure to Open STEP PRIME ASSIGNED START TARGET DATE ACTION STEPS RESPONSIBILITY TO DATE DATE COMPLETED quuaE7:

ALL STEPS 0F THIS ACTION PLAN ARE T0 BE PERFORMED IN ACCORDANCE WITH THE LATEST REVISION OF " GUIDELINES T0 FOLLOW WHEN TROUBLE-SHOOTING OR PERFORMING INVESTIGATIVE ACTIONS INTO THE ROOT CAUSES SURROU DING THE JUNE 9,1985 REACTOR TRIP".

1 Before beginning troubleshooting work, document the as-found J. Long condition of the valves (limit to those conditions which can ,

I be recorded without changing conditions - i.e., valve position, general condition, environmental conditions) .

2 The torque switch settings were changed for MV 599 and 608 J. Long under FCR 84-039 (1.5 open and 1.0 closed). These settings chould be verified.

3 The stem thrust load should be measured to verify the thrust J. Long calculation. MOVATS (Motor Operated Valve Analysis & Test System) should be used to measure valve stem thrust, time of  ;

i control switch actuation, and dynamic motor current).

i I

_ w __

an l'

I TION PLAN PLAN NUMBE R PAGE Rev. 2 12 2 a' 3 un LE DATE PREPARED PREPARED BY AFW SYSTEM VALVE PROBLEM ANALYSIS (AF 599 and 608) 6/26/85 M. Bajestani SPECIFIC OBJECTIVE to D2tcraine the Root Cause of Motor Operated Valve AF 599 and 608 Failure to Open IEP PRIME ASSIGNED START TARGET DATE

^ '"$ RF.SPONSIBILITY TO DATE DATE COMPLETED M8ER re MV 599 and 608 are fast speed operators. A magnetic brake is J. Long previded to oppose the motor inertia after the power is removed from the motor. The brake and motors were replaced last refuel-ing outage. These brakes should be checked for . proper operation.

h Verify number of turns on the handwheel of the valve from fully J. Long closed position, the limit switch contact 33/AC bypass; the torque i

cwitch contact 33/TO.

$ With valve in midposition (spring pack relaxed) verify that the J. Long torque switch is not preloaded.

(OTE STEP 12 should be performed before STEP 7.

7 Vorify by visual inspection the spring pack model number. J. Long If the heavy spring number 60-600-0068-1 is used - no problem.

However, if light spring number 60-600-0062-1 is used, the terque switch should prevent valve opening.  ?

R8 Motor horse power calculations should be performed in order to J. Long -

datermine if the motor is capable of providing enough~ torque.

7  :

3p 17

b)

PLAN NUM;ER PAGE ACTION PLAN n oos Rev. 2 12 3 3 DATE PREPARED PREPAREo BY TITLE AFW SYSTEM VALVE PROBLEM ANALYSIS (AF 599 and 608) 6/26/85 M. Bajestai SPECtFic OBJECTIVE To Determine the Root Cause of Motor Operated Valve AF 599 and 608 Failure to Open PRIME ASSIGNED START TARGET DATE STEP A Tl N STEPS DATE COMPLETO R ESPONSIBILITY TO DATE NUM ER E9 Actuator size should be checked to determine if it is capable of J. Long operating against a 1050 psi differential pressure.

010 Tortional stem stress and tensile stress should be checked to J. Long verify that these stresses do not exceed the ASME design allow-able values, c11 Torque dial settings should be established by opening and J. Long ,

I closing positions based on the extreme stem operation loads expected during the hot and pressurized condition.

12 Test operate the valves individually with up to a 1050 PSIG J. Long 2-pressure differential across the seat.-

=

l

• Steps 8-11 are not dependent on Steps 1-7 and can be performed in any order j a

, I i

s

..e

}% s>