ML20209H648
| ML20209H648 | |
| Person / Time | |
|---|---|
| Site: | Davis Besse  |
| Issue date: | 10/15/1985 |
| From: | Gradomski R TOLEDO EDISON CO. |
| To: | |
| Shared Package | |
| ML20209H568 | List: |
| References | |
| NUDOCS 8511110182 | |
| Download: ML20209H648 (46) | |
Text
__
.- ?-
FINDINGS, CORRECTIVE ACTIONS, AND GENERIC IMPLICATIONS REPORT TITLE: AFPT OVERSPEED TRIP THROTTLE VALVE PROBLEM PREPARED BY: R. J. Gradomski PLAN NO. 1D .
DATE PREPARED: August 9, 1985 PAGE L of M CHAIRMAN REV. DATE REASON FOR REVISION BY TASK FORCE O 8/9/85 Initial Issue R. Gradomski B. Beyer 1 8/19/85 Add Corrective Actions R. Gradomski L. Grime 2 8/30/85 General Corrections R. Gradomski Ex GrM 3 10/15/85 Trip Throttle Valve R. Gradomsk 4
e 8511110182 851031 PDR ADOCK 05000346 S pm
r.
TABLE OF CONTENTS Page .
I. ISSUE / CONCERN 3 II. BASIC PRINCIPAL OF OPERATION 3 III'.
SUMMARY
OF TROUBLESHOOTING AND INVESTIGATION 6 IV. RESULTS/ CONCLUSIONS OF FINDINGS 10 A. Direct Cause of Problems 10 B. Root Cause of Problems 12 C. Disproved Hypotheses 13 V. PLANNED ADDITIONAL ACTIONS 17 VI. TECHNICAL JUSTIFICATION OF FINDINGS 18 VII. SPECIFIC CORRECTIVE ACTIONS 19 VIII. GENERIC IMPLICATIONS 25 ATTACHMENTS:
- 1. Chronology of Events Related to Restart After Overspeed Trip 28
- 2. Inspection of Condition and Operation of Overspeed Trip and Valve Latch Mechanisms 33
- 3. Results of Operator Interviews, Procedural and Training Reviews 37
- 4. Affected Procedures 40
- 5. Testing and Inspection of Trip Throttle Valves Related to l3 Difficulties in Opening 41 FIGURES:
- 1. Latching A W Turbine Trip Valve 44
- 2. Auxiliary Feed Pump Turbine Overspeed Trip Linkage 45
- 3. Auxiliary Feed Pump Turbine Trip Throttle (T&T) Valve 46 l3 I
v I. ISSUE / CONCERN During the reactor transient of June 9,1985, the Auxiliary Feedwater Pump Turbines (AFPT 1-1 and 1-2) were started and then automatically tripped on overspeed. Equipment Operators (EOs) were dispatched to determine the problem and restart the AFPTs.
1 While attempting the restart, the operators experienced difficulty in relatching and opening the trip throttle (T&T) valves. The EOs were successful in their attempts to relatch and open the T&T valves after several minutes and the AFPTs were restarted. This findings report addresses two problems: difficulty in relatching the T&T valve and 3
difficulty of opening the T&T valve.
II. BASIC PRINCIPAL OF OPERATION The following is a description of the operation of the overspeed trip
^
mechanism (OTM) in tripping the T&T valve:
(Refer to Figure 1 for the location of each component.)
- 1. A spring loaded weight is attached to the turbine shaft. As the speed of the rotating shaft increases, centrifugal force on the weight overcomes the force of the spring causing the weight to extend out from the shaft.
- 2. At 4500 RPM, sufficient outward movement of the weight causes the weight to strike the leaf spring.
l
P
- 3. The leaf spring and connected tappet move upward. (the manual trip lever provides this upward movement when manually tripping the OTM.) .
- 4. As the tappet moves upward, it removes the interference and allows the head lever to move toward the turbine.
- 5. The head lever and attached connecting rod are pulled by spring force in the trip direction. The limit switch operated by the head lever changes state to show " trip" on the control room annunciator and alarm printer.
- 6. As the connecting rod moves in the trip direction, it applies a
" hammer blow" through the slotted link connection in the trip crank to the trip hook.
- 7. The trip hook disengages from the T&T valve latca-up lever allowing the T&T valve closure spring to drive the valve stem and disc to the closed position. The limit switch on the T&T valve changes state to show "not fully open" on the computer alarm printer.
From Figure 1 the relatching of the valve trip mechanism involves two actions: first, the trip tappet must be reset, and second, the trip hook must be engaged with the latch-up lever so that the valve stem can be raised.
,a-- , --. . _-, - ,,. - - - - - ,
t The resetting of the tappet can be achieved by pulling the connecting rod toward the valve. This moves the head lever so that the tappet spring can force the tappet down into the reset position. When the ,
tappet is down, the head lever and connecting rod are positioned toward the valve. The head lever limit switch changes state to show
" norm" on the computer alarm.
The valve can be latched by turning the valve handwheel in a direction which would close the valve. Although the valve is already closed, it is necessary to rotate the handwheel in the CLOSE direction in order to raise the nut attached to the latch-up lever to a position where the trip hook and latch-up lever can be engaged. When the latch-up lever reaches the top of its travel, the trip hook engages.
If the rotation is continued, the latch-up lever contacts a mechanical 3 stop on the yoke of the valve. After the hook is engaged, the valve is latched and rotation of the valve handwheel in the OPEN direction raises the valve stem to open the T&T valve. The outward motion of the stem first takes up the clearance in the coupling (about 010 inch). After this freeplay is removed, the stem starts to open the pilot valve. The main disc does not move until the pilot has been opened (raised) 1/8 inch. The opening of the pilot allows the region above the main disc to be vented to the downstream pressure, thereby 3
reducing the pressure difference across the main disc so that it can be more easily opened. When the valve is fully open, the T&T valve limit switch changes state to show " full open" on the computer alarm printer. (Refer to Figure 3 for a cross section of the valve showing the main disc and the pilot.)
't III.
SUMMARY
OF TROUBIISH00 TING AND INVESTIGATION The troubleshooting and investigation involved the following major .;
elements- !
i i
I The plant records, specifically the station alarm printout, the Data Acquisition and Display System (DADS) output, and the security alarm printouts, were evaluated to determine a chronology of events associated with the trip and restart of the AFPTs.
i This detailed chronology is presented as Attachment 1 to this l report.
The physical condition of the OTMs and T&T valve latching mechanisms was inspected, and the operation of the linkages was checked. The results of these inspections are sumwarized in Attachment 2 to this report.
The operators were interviewed as to_their actions relative to relatching the valves by both Toledo Edison and NRC personnel.
3 The guidance in existing plant procedures and the training i
received by the operators on the relatching operation were also reviewed. The specific items of significance from these interviews and reviews are presented in Attachment 3 to this report.
Attachment 1 (the chronology) also reflects the results of the operator interviews.
l 1
l l
The T&T valves were opened both with and without a differential pressure of about 200 psig (using auxiliary steam). Attachment 5 summarizes the results of this testing. Some increase in .
torque required to open the valves can be attributed to the differential presotre. The valve for AFPT 1-2 was found to require more torque to open than the valve for AFPT 1-1 and was disassembled for inspection.
3 The valve for AFPT 1-2 was disassembled and inspected. The results of the inspection are summarized in Attachment 5.
Although some minor scoring of bushings and bearing surfaces and some minor dimensional differences were found, the representative of the valve manufacturer (Gimpel Corp.) did not consider they were significant enough to prevent opening.
The major conclusions from these investigations are summarized below:
The chronology shows that the OTM for AFPT 1-2 was reset very soon after the EOs entered the room -- within about a minute.
Evidently, there were no serious problems with reset of this turbine. .However, there were problems with opening the T&T valve on AFPT 1-2, and that problem prevented immediate operation of that turbine. However, another E0 successfully opened the 3
T&T valve for AFPT 1-2 using a valve wrench.
The chronology shows that the T&T valve for AFPT 1-1 was evidently opened without resetting the OTM. Turbine AFPT 1-1 was started; however, it tripped about twelve minutes later while a manual .
speed change was in process, but not as a result of overspeed.
The OTM was reset very shortly after the mechanism's spurious trip (less than a minute) and the turbine successfully restarted.
The physical inspections and operational checks of the OTM linkage and the T&T valve latch mechanisms showed some abnormal conditions. These conditions may have contributed to the difficulty in resetting the OTM or to the spurious trip of AFPT 1-1. However, the mechanisms on both turbines were shown to be i capable of being reset when properly operated. Therefore, these conditions did not prevent resetting the OTM and relatching the valve. It was also demonstrated that it was possible to latch the valve and open it without resetting the OTM. In that condition the valve is not positively latched -- it is prevented from tripping only by friction on the latch faces.
- The procedures provided to the operators did not clearly identify that the valve relatching also required the OTM tappet to be reset. The training on the relatching operation included little hands-on operation of the mechanisms.
- -- e
1 Both T&T valves were difficult to open with the pressure differential !
L of about 900 psi which was present during the June 9 incident.
Neither of the operators who initially attempted to open the .
valves had ever opened the valves under the high differential ,
pressure conditions. Both valves were successfully opened later using a valve wrench; however, the operators were initially s
reluctant to apply what they considered were large forces to the s --
valves. When other operators who were more familiar with the operation arrived, the valves were successfully opened. The m operators were also uncertain initially whether they had, in fact, opened the valves.
The chronology shows that the T&T valve for AFPT 1-1 was opened without reset of the overspeed trip mechanism. The turbine 3 .:
started to spin very soon after the door to the room was opened 7 (less than one minute). The turbine speed was low, however, and the operators were not aware that they had it spinning. The.T&T valve for this turbine required considerable force to open. It was finally opened fully by the operator who had previously used a wrench to open the T&T valve for AFPT 1-2. This was about 6 minutes after AFFT 1-1 initially started and about a half a ^^3 minute after AFPT 1-2 was started.
e 9
,,,,-------.e,--- w r -- -- , m, - -
,'/ -
4 -
10-x IV. RESULTS/ CONCLUSIONS OF FINDINGS
(
,, A. Direct Causes of Problems .
c.
j.,
- 1. Difficulties in Resetting the OTM and Relatching
~, 3 the-T&T Valves s'; ..
e The most probable cause of the spurious trip of AFPT 1-1 was the operator's failure to reset its OTM prior to latching and opening its T&T valve. The investigations.
~~
showed that'it was possible to open a T&T valve without resetting of the GIM and that this would result in the
' valve latch' mechanism being in a metastable condition from which it could spuriously trip.
The reported difficulty of relatching the T&T valve was e
probably a direct result of the operator's attempts to latch the valve without resetting the OTM. Investigations 7
showed that opening the T&T valve without resetting the OTM requires the latch parts (trip hook and latch-up lever) to be held engaged while the valve is opened and until they are loaded sufficiently for friction to hold the latch parts engaged.
~
s.
g .*
v t
h
-- v
-The investigations shcw that the OTM could have been reset and the T&T valve relatched, had the operators involved been more familiar with the required manipulations of the ..
linkage. That is, the trip and latching mechanisms, even though they are not simple and straight forward to operate, were apparently not defective and were capable of being reset as was demonstrated by the successful relatching of AFPT 1-2. Note, however, that if the OTM on AFPT 1-2 had not been reset, AFPT 1-2 would have probably had problems in latching similar to those experienced on AFPT 1-1.
- 2. Difficulties in Opening the T&T Valves
, The T&T valves were not promptly opened because the particular EOs who were first involved did not know that the force they had to apply to open the valves would be large. Even though the forces were large, the T&T valves were capable 3 of being opened, as was demonstrated by the actions of another operator who subsequently opened the valves using a valve wrench. The operators who initially attempted to open the T&T valves could have opened them if they had used the valve wrench as was done later.
12-l B. Root Causes of Problems
- 1. Difficulties in Resetting the OTM and Relatching the .
T&T Valves Three significant root causes for the apparent failure of 3 the operator to reset the OTM before latching the T&T valve were identified:
- a. The reset or tripped condition of the OTM is difficult for the operators to determine. The design of the linkage does not include any labels, operator aides, or easily identified position indication.
- b. The procedural instructions for relatching the T&T valves are incomplete and if strictly followed would not assure a properly reset mechanism.
- c. The training given to the operators did not provide for adequate hands-o. experience for a specialized operation which would have to be performed under stress.
- 2. Difficulties in Opening the T&T Valves Two root causes for the difficulty in opening the T&T valves were identified:
e v w - "
- a. The training given the operators did not provide for adequate hands-on experience for an operation which was performed under stress and which required signifi- .
cantly different forces than they .had previously experienced.
3
- b. The open or closed condition of the T&T valve is difficult for the operators to determine. There is no s easily identifiable position indicator.
C. Disproved Hypotheses The investigations evaluated previously identified hypotheses and other potential causes for the difficulties in resetting the 3
OTM'and in latching and opening the T&T valves. These are discussed below: '
Overspeed Trip Mechanism
- l. Hypothesis 1. The tappet of the turbine trip mechanism did not J
return to its normal position while attempting the relatch evolution. It was shown in the investigation that the tappet.
would freely return to its reset position when the connecting rod was pulled to its reset position. The leaf spring that 1.
provides the return force to the tappet was functional. The tappet was not bound and moved freely. Since the control room alarm printout did not indicate that the head lever actuated the i
1
- , _ .m._ .- - ,,. , . _ m- ..- , . .
I I
J limit switch, the linkage of AFPT 1-1 was evidently not moved
.far enough to allow the tappet to drop to the reset position.
Hypothesis 2. The spring that provides the relatching force for the trip hook is defective or inadequate. It was determined that the spring provided adequate force to latch the trip hook when the OTM was properly reset. The spring condition was satisfactory.
Hypothesis 3. The pivot point of the trip hook is not sufficiently free to assure proper engagement of the trip hook to latch-up lever. It was demonstrated that the trip hook was sufficiently free to engage the lever when a proper reset and latching' sequence was used; however, some misalignment of the parts was noted and is discussed in Attachment 2. The pivot point of the trip hook was well lubricated.
4 Hypothesis 4. The linkage mechanism may not be adjusted
. correctly. The Terry Turbine Co. representative verified the proper adjustment of the linkage. Additionally, when the proper i
reset sequence was followed, the OTM reset and operated correctly, i It should be noted that the linkage operated as expected to trip '
I the turbine during the initial overspeed conditions.
In the process of investigating the operation of the OTM linkage, it was found that a force on the connecting rod in the trip direction would sometimes dislodge the tappet from the reset to i
1
)
i.
the tripped position. Although there is some possibility that this could have been the cause of the spurious trip of AFPT 1-1, the force required is large and no mechanism to get such a force .
has been identified during the June 9, 1985 event.
It is also possible that an operator, by forcing the connecting rod in the trip direction after resetting the OTM mechanism (as
~
an attempt to assure that the OTM had reset), could inadvertently trip the OTM. This apparently did not occur since it would have led to multiple indications of operation of the head lever limit switch on the alarm printout. These were not present.
Trip Throttle Valve The design of the T&T valve is such that the force to open it will increase as the pressure differential, i.e., the pressure above the main disc, increases. The action of the pilot valve to vent the main disc is intended to reduce this effort; however, it cannot be eliminated. Analyses were performed which confirmed that the presence of large amounts of saturated water upstream 3 of the valve would reduce the rate at which the pressure above the main disc is reduced by the venting through the pilot. This would not affect the force to open the pilot; however, it could delay the bleed-off of pressure above the main disc and thereby make it more difficult to open the main disc. From a practical standpoint, even venting water through the pilot would result in enough steam to start and run the turbine at significant speed.
5 Consequently, this was not a factor for AFPT 1-2, since the turbine did not spin until the third E0 opened the T&T valve.
However, it could have been a factor for AFPT 1-1, since the .
pilotwasapdarentlyopenedverysoonbutthevalvewasnot opened further until another E0 used a valve wrench.
For AFPT 1-1 a large opening force was still observed after the turbine had been spinning for about six minutes. It is considered unlikely that significant water could be present after that long a time with flow out through the pilot. Consequently, although water in the valve could have made the forces higher, it is not considered to be the direct cause of the difficulties in opening the T&T valves on June 9.
3 The operational testing of the T&T valve for AFPT 1-2 showed large forces could be required to open the T&T valve irrespective of the pressure difference across the valve. This was apparently the result of bringing the valve latching mechanism hard against the mechanical stop during the relatching evolution. Under ;
those conditions, large axial valve stem (spindle) loads will be developed. Since it is necessary for the valve spindle to rotate in the threads and in the coupling, high axial loads will <
1 lead to high handwheel torques to overcome this loading and permit the valve to open.
Since the T&T valve for AFPT 1-1 did open slightly (probably only the pilot) these forces could not have been a cause of its t
high opening resistance. The T&T valve for AFPT 1-2, on the -
other hand, did not open until a valve wrench was used and, therefore, the forces at the mechanical stop could have contributed a to the difficulty experienced by the operators in opening that valve. The NRC interview of the operator who initially attempted to open the AFPT 1-2 valve is not clear as to whether there was any rotation of the handwheel in the open direction. The operator 3
responded "not much" when asked whether he had opened the valve and in addition stated that he could not remember whether he "didn't open it far enough". Even though it cannot be proven to have been a factor on June 9th, the potential to make T&T valve opening difficult by running the valve into the mechanical stop with excessive force should be avoided in future valve operations.
V. PLANNED ADDITIONAL ACTIONS No further investigations of the difficulties of resetting the OTM and 3
relatching the T&T valves are planned. The direct and root causes of the spurious trip of AFPT 1-1 T&T valve have been identified.
No further investigations of the difficulties in opening the T&T valves are planned. The direct and root causes of the reported 3 difficulties have been identified.
, . _ . . _ _ _ + , _ _ _ - , _ _ _ . . _ , , , -
1
., l VI. TECHNICAL JUSTIFICATION OF FINDINGS The direct cause of the difficulty in latching and the spurious trip .l of AFPT 1-1 has been established as the failure of the operator to properly reset the OTM prior to latching and opening the T&T valve.
The root causes have been identified as: 1) lack of consideration of human factors in the linkage design so that a clear indication of the OTM position was not available to the operator, 2) procedures which did not adequately define the actual actions required to reset the OTM and relatch the T&T valve, and 3) training which did not adequately prepare the operators for performing the relatching activity under stress. Accordingly, the overspeed trip mechanism linkage should be removed from the freeze list so that corrective action may begin.
The reported difficulties in opening the T&T valves were the direct result of the operators not applying sufficient force to the valve.
The root causes have been identified as a lack of knowledge or experience an the part of some operators that large forces and a valve wrench could be required to open the valves under full pressure conditions. Another factor which contributed to the operator's 3 unwillingness to apply larger forces to the valve was the lack of an easily understood indication of valve position. Consequently, they could not be sure that the valve was not already open and, therefore, that more force would risk damage to the valve. The investigation concluded that the T&T valves were not defective and that, as demon-strated later in the incident, where more experienced personnel were involved, the T&T valves were capable of being opened.
VII. SPECIFIC CORRECTIVE ACTIONS A. Required Corrective Action: .
The following actions have been identified to correct existing deficiencies related to the identified root causes and prevent similar problems associated with the AFPT OTM and T&T valves. 3 Human Factors Considerations
- 1. Design and install local indication (trip / reset) of the OTM. Additionally, supply simplified local operating instructions. This item is to be completed prior to plant restart.
- 2. Paint the yoke of the T&T valve, the latch-up lever, trip hook, and connecting rod (unthreaded portion), for both AFPTs, a yellow color to distinguish this equipment as important in the operation of the overspeed trip. Addi-tionally, the manual trip lever should be painted red.
This item is to be completed prior to plant restart.
- 3. Install local position indication on both T&T valves. This indicator has been requested by the operators and will be installed prior to plant restart.
i l
I
0 l
l I
- 4. Provide enhanced communication for EOs in the AFPT rooms.
i EOs in both pump rooms should be able to communicate ;
simultaneously with each other and the control room while -
operating the AFPTs. This item is to be completed prior to plant restart.
Procedural Considerations
- 5. Modify the affected procedures (reference Attachment 4) to reflect the proper reset sequence for the OTM and to also check that alarm points S007 and S017 reflect the reset condition on the control room alarm printer and annunciator.
Also modify the affected procedures to provide cautions 3
against running the valve hard against the mechanical stop.
This modification is to be completed prior to plant restart.
- 6. Modify ST 5071.04 to ensure that the T&T valve and OTM are properly reset after the completion of testing. This modification is to be completed prior to plant restart.
Training Considerations
- 7. Operators will receive instruction on the theory of operation for the OTM and T&T valve. This instruction is to be completed prior to plant restart and will include all physical and procedure changes made as a result of this
report. This instruction will also be incorporated in operator annual requalification.
- 8. Operators to complete " hands-on" training in the proper resetting of.the OTM and opening of the T&T valve with steam pressure > 800 psi. This training is to be completed with the plant in Mode 3 during restart. This training will also become a requirement for operator qualification.
g ..
B. Additional Planned Action:
The following are corrective actions to discrepancies noted during the course of these investigations that are not related to the direct and root causes but are included for completeness.
- 1. Install "no step" indication on the handwheel of both T&T valves. These are to be installed prior to plant restart.
- 2. Install "no step" indication on the connecting rod of each OTM. These are to be installed prior to plant restart.
- 3. A semi-annual preventative maintenance (PM) ites specifically fr cleaning and lubrication of the T&T valve and OTM of each AFPT will be instituted. This PM is to include:
1) the removal of old, built-up grease and grime from the T&T valve and OTM, 2) lubrication at all grease fittings, 3)
- lubrication of the sliding nut, screw spindle, and external
l guides of the-T&T valve, and 4) lubrication of all other pivot points and sliding surfaces of the OTM excluding the mating surfaces of the latch-up lever to trip hook and ..
tappet to head lever. This semi-annual PM will also check the tightness of the connecting rod adjustment lock nut, overspeed trip device hold down screws, and the proper adjustment and alignment of the OTH. Also, the PM will l3 determine the need for touch-up painting on identified surfaces. This ites is to implemented prior to plant restart.
- 4. A refueling outage preventative maintenance (PM) ites will be institued to: 1) perform NDE on the tappet leaf spring,
- 2) confirm proper adjustment of the tappet, 3) perform dimensional inspection of the tappet nut and head lever, connecting rod to trip hook lever, and trip hook to latch-up lever mating surfaces. This item to be completed prior to the next refueling outage.
3
- 5. Institute a new surveillance test, or equivalent, to exercise the T&T valves weekly while in plant modes 1,2 or 3 by turning the handwheel in the closing direction through the distance equal to k of the total lift of the valve.
This ites will fulfill a manufacturers recommendation.
1
~ _ .- -~ .. . ..
- 6. Straighten the bowed connecting rod and ensure that the connecting rod adjustment lock nut is tight on the AFPT 1-2 OTM. Reference Non Conformance Report (NCR) 85-0114. This .
correction must be completed prior to plant restart.
- 7. Replace the connecting rod socket on the head lever for the OTM on AFPT 1-2. Reference NCR 85-0114. This item must be replaced prior to plant restart.
i
~
- 8. Retighten the hold down screws for the OTM on AFPT 1-1.
Reference NCR 85-0113. This action must be completed prior to Action Plan IA/1B testing.
f f
- 9. Adjust the trip hook crank to obtain perpendicular alignment between the connecting rod and the trip hook pivot shaft for both AFPT OTMs. Reference NCR 85-0114. This alignment must be completed prior to plant restart.
- 10. Replace the head lever and tappet nut for both AFPT OTMs.
Refrence NCR 84-0114. This item is to be completed prior to plant restart.
The following are actions that are being considered to enhance performance with respect to the operation of the AFPTs.
- 11. Investigate means to enhance the local manual control of the AFPTs. This will include, but will not be limited to, local indication of steam generator level and manual .
control of the auxiliary feedwater pumps using the T&T valve. Appropriate training in the manual operation of the AFPTs will have to be provided.
- 12. Investigate means to provide for remote reset of the AFPTs, if they trip on overspeed. This would include:
- a. The addition of an electric trip solenoid and
- b. A motor operator on the T&T valve.
Consideration will be given to whether the additional complexity created by the above modifications will, in fact, affect the auxiliary feedwater system reliability.
- 13. Check operation and installation of the local speed indicator for AFFT 1-1.
The transcript of the interviews of the EOs involved in the restarting of the AFPTs implies that the local speed 3 indicator for AFPT 1-1 may not have been functioning at some times on June 9. The operators apparently were unaware that AFPT 1-1 had started prior starting AFPT 1-2.
i Later on in the event, the local speed indicator was evidently functioning.
1 .
- 14. Perform testing to confirm that breakaway forces developed by the normal expected contact of the latch-up lever to the mechanical stop are. consistently within the operator's capability when using a valve wrench.
- 15. Since a valve wrench is essential to reliable opening of the T&T valves, a valve wrench will be provided at each turbine.
3
- 16. The bore in the yoke cip of the T&T valve for AFPT 1-2 will be reased to the correct drawing dimension.
- 17. The pilot seat of the T&T valve for AFPT 1-2 will be machined to remove the indentations observed in the inspections.
VIII. GENF.RIC IMPLICATIONS A. Sianificance There are no other Terry Turbines like the AFPT at Davis-Besse and consequently there are no generic implications for many of the identified problems. There are, however, some broad impli-cations of the root causes of the problems with the reset of the AFPTs. These are:
, .e
, g-c 1. There may be other crucial operations which the operators are called upon to perform locally under conditions of stress which are not straightforward and for which the -
equipment human factors could be significantly improved. i.
- 2. There may be other in-plant operations for which the exist-ing procedures are inadequate or incorrect.
- 3. There may be other operations for which specific hands-on-l training is needed to assure that the operations will be correctly performed in times of stress. This issue is specifically addressed in the final report for Action Plan l.
3, Operator Actions and Procedural Adequacy, Item III. A.6. .
In addition to the generic implications of the root causes discussed previously, there may be generic implications cf two other conditions which were observed in the investigations.
Specifically:
l
- 1. Equipment was found to have been damaged by being used to support personnel (by being stepped on).
. 2. Temporary scaffolding was placed in locations which reduced the access to critical equipment and communications.
e w - - - - - -,,- y-. - , . , - - - v - , y ...- , . - ,, , , - -m_,. ,,,m_.__ . , , , - , _ - . - - , , , . -. . . , . _ _ _ _ . - . _ - , , . - ., ,-
B. Planned Action The potential generic problems described above may also be :
factors in some of the other events on June 9, 1985. Deterai- !
nation of whether these potential problems are present elsewhere at Davis-Besse is beyond the scope of the investigation of the AFPT overspeed trip and reset problems. As a result, selection of specific corrective action is not appropriate in this report.
However, to assure that the observations are addressed, action plans for each of the identified items will be prepared.
1 RJG REPORT 2
1
. 1 I
ATTACHMENT 1 {
CHRONOLOGY OF EVENTS RELATED TO RESTART AFTER OVERSPEED TRIP l
.I The following chronology describes the time sequence of events !
related to the tripping and reset of the Auxiliary Feed Pump Turbines ;
(AFPTs) as determined from the plant records of June 9,1985. The records utilized were the station alare printer output, the output of DADS, and the security computer records. Thi,s chronology also includes comments indicating the most probable interpretation of each recorded event. These comments are based on the investigations conducted to date and interviews of the operators. (See also Attachments 2 and 3.)
COMPUTER TIME
- POINT ID DESCRIPTION 1:41:31 S007/Z001 AFPT-1 overspeed trip /stop (T&T) valve
~
started to close.
Comment: The AFPT tripped at its correct setpoint and the T&T valve closure is instantaneous. These events are confirmed by the AFPT speed data.
1:41:44 S017/Z002 AFPT-2 overspeed trip /stop (T&T) valve started to close.
Comment: The AFPT tripped at its correct setpoint and the T&T valve closure is instantaneous. These events are confirmed by the AFPT speed data.
1:45 Security Hatch to AFPT 1 and 2 room open.
Computer Comment: Two EOs entered the room.
One went to AFPT-2 and the other 3 went to the door to the room for AFPT-1.
1:45:50 S017 AFPT-2 overspeed trip reset.
Comment: The EO had successfully reset the OTM. On July 5, 1985, it was noted that the OTM must be in its full reset position before the switch associated with the alare point changes state. There were no additional trip alarms throughout the event which means that the OTM was properly holding in the reset position. It is not clear from the interviews con- 3
..___.~,m,__,-,-.,,,_...m-..,..-,.,.-. . , - - . , _ _ , . . . . , , , , , -
o, 5
- ducted by the NRC whether or not the valve mechanism was forced against the mechanical stop or whether the handwheel was rotated t
until the " free play" was removed. 3 -
In any case, the E0 did not open the pilot since the turbine did not start.
1:46 Security 'The door to AFPT-1 was opened.
Computer 1:46:33 S008 AFPT-1 showing speed increase. The speed leveled out and remained rela-
- tively constant at 1800 to 2100 rpm 3 until 1
- 52:47, about six minutes later.
Comment: The E0 stated that he had tried to reset the OTM; the computer alare printer, by absence of reset printout, indicates that the OTM was not moved a sufficient distance to cause the switch for alarm point S007 to change state and did not allow the tappet to reset. On 3 July 5, 1985, this E0 demonstrated the amount of movement applied on June 9, 1985. His understanding of the operation of the OTM had his pull the connecting rod until the trip hook met with the latch-up lever. It is noted that the connect-ing rod to trip hook is a slotted connection that allows additional movement of the connecting rod to reset the OTM. The appearance of the trip hook meeting with the latch-up lever does not mean that the OTM was pulled sufficiently to allow reset. It was demonstrated on July 5, 1985 that the movement on June 9, 1985 would cause neither: 1) the tappet to reset, nor 2) the switch to change state.
The E0 at this time during the June 9 event, physically held the trip hook to the latch-up lever and cracked open the T&T valve. l The OTM was not reset. This evolution was shown possible on June 29, 1985 and July 5, 1985.
During both investigations, the
~
e hook could be held on the lever, the T&T valve opened, the OTM 1
,---n. . ,, ,,..-.,-,,.,_...,,,,.,.m___,., ,_,.,_,.___,,_,~,,.,_,_,,m_,__,.__~-,.,__-_-,,..-.4 _ , . _ , , _ , - _ . _ _ . .
.?t not reset and the switch not change state. The OTM for AFPT 1-1 was not reset until 1:58:57.
Although the DADS output shows both speed and outlet pressure -
for AFPT-1 were achieved, the transcript of the operators interview by the NRC indicates 3 that the operator was not aware that AFPT-1 had started.
1:52 Security A third E0 logged into the AFPT rooms.
Computer He was accompanied by an Assistant 3 <
Shift Supervisor. )
1:52:21 S018 AFPT-2 showing speed increase.
Comment: The third EO, after viewing that the OTM was reset, realized that the T&T valve needed additional force to move the disc from its seat. At that time, he applied a valve wrench to the hand-wheel and began opening the T&T valve. The third E0 stated that 1 after moving the T&T valve disc from 3 its seat, the T&T valve became easy to open. The speed peaked at about 2850 rpm at 1:52:32 and then dropped back to about 2450 rps.
1:52:53 2002 AFPT-2 T&T valve fully open.
Comment: The pump speed was still about 2,450 RPM at this time. About 1:53:12 the speed started to increase steadily and at about 1:53:44 leveled out at about 3740 RPM. This 3 increase appears to be the result of speed increase from the control room operator.
1:53:51 2002 AFPT-2 T&T valve not fully open.
Comment: Believing that the AFPT governor was not controlling properly, the EO took control of pump speed via the T&T valve to reduce pump speed. At about 1:54:07 ;
the speed dropped rapidly from about 3740 RPM to about 3400 RPM where it remained relatively 3
constant for more than six minutes.
This mode of operation continued until about 2:02 where contol was transferred to the control room.
4
. , - - - , . . - , , ,me--- - -m---,
O 1:56:08 Z001 AFPT-1 T&T valve fully open.
Comment: According to the transcript of the interview of the EOs by the NRC, the T&T valve for AFPT-1 was opeaed by the same operator who .
opened the T&T valve for AFPT-2.
In that interview, the E0 stated that it was difficult to open and that he used a valve wrench. The speed of AFPT-1 started to increase rapidly above 2100 RPM at about 1:52:47. At 1:56:08 the pump speed was approximately 3600 RPM and remained essentially constant for more than two minutes. 3 The EO was apparently controlling pump speed locally using the T&T valve prior to this time. AW flow to the steam generator started about 1:54:33.
1:58:39 Z001 AFPT-1 T&T valve not fully open and turbine speed dropping rapidly.
Comment: The third EO, contralling the pump locally, had attempted to reduce the pump speed by placing the valve wrench on the handwheel and closing the valve. A rapid reduction in speed of AFPT-1 from the 3600 RPM level had started at 1:58:27 and at 1:58:38 the speed was 3188 RPM. During 3 this process, the trip hook dis-engaged from the latch-up lever and the T&T valve slammed shut.
It is noted that there was no indication of OTM reset. It was shown that engagement without the OTM being reset is unstable due to the connecting rod / springs providing a constatt force in the disengagement direction. A review of ptap speed data confirms the EOs statement that the T&T valve
" tripped" shut.
1:58:57 S007 AFPT-1 overspeed trip reset.
Comment: After the T&T valve slammed shut, the third E0 properly reset the OTM.
1:59:02 S008 AFPT-1 showing epeed increase.
Comment: The third E0, after resetting the OTM, began to open the T&T valve. AFPT-1 was con-trolled locally via the T&T valve -
for the remainder of the event with no further problems associated with the T&T valve or OTM.
2:01:58 2002 AFPT-2 T&T valve fully open.
Comment: Previous to this, attempts were made to place control of the pump in automatic from the control room (2:01:11). At 2:01:24, the pump control was placed in manual from the control room and it was realized that the T&T valve must be fully open so that the governor throttle valve could function. The pump was then manually controlled from the control room for the re-mainder of the event with no further problems.
- Times are Modcomp time. 6 seconds have been added to DADS time.
Access times are only to the nearest minute.
-.v v-r wr e t-- 4 *- -- -% c--w m---T -w----mT- -e-tm*-- ST-'mF- - - ' -----m-+ e77 --Y e-
u 3 >
ATTACHMENT 2 INSPECTION OF CONDITION AND OPERATION OF OVERSPEED TRIP AND VALVE LATCH MECHANISMS 1 This attachment suusarizes the results of the investigations performed to verify the physical condition of the mechansias and confirm their operability.
These investigations were performed without steam and without disassembly.
j' 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. 4", Maintenance Work Orders (MW0s) were I prepared. Activities associated with MWO 1-85-2063-00 (AFFI 1-1) and NWO 1-85-2065-00 (AFPT 1-2) were conducted on June 29, 1985.
Both MW0s required the performance of Steps 1, 2, and 3 of Action Plan
, ID. Representatives of the Nuclear Regulatory Commission (Messrs. W.
i Rogers and N. Choules), Terry Turbine Co. (Mr. J. Kregel), and MPR Associates, Inc. (Mr. T. Clark) were present during investigative actions conducted on June 29, 1985.- Additional investigations associated with Rev. 01 of the aforementioned NW0s were conducted on July 5, 1985.
. The specific observations considered to be potentially important either in this problem or to be potential future problems are discussed below:
i 1. Exercise of the OTM Linkane All checks associated with Step 3 of Action Plan ID were completed.
All items performed their respective functions. The " exercises" were performed using the sequence for manually tripping and resetting the OTM, as stated in Section II of this report, were successful.
l However, incomplete engagement of the trip hook to latch-up lever was i noted on AFPT 1-1 when latching of the valve was attempted without
- j. resetting the tappet in the OTM.
The E0 involved with AFPI 1-1 on June 9, 1985 demonstrated how he had attempted to reset the OTM. His understanding of the operation of the OTM had his pull the connecting rod until the trip hook met with the latch-up lever. The movement of the OTM so that the trip hook
!. met with the latch-up lever was not sufficient to allow either the turbine end of the OTM to reset or the limit switch to change state.
i
- 2. Enassement of Latch Without Tappet Reset j-
- During the performance of Action Plan ID, Step 3, it was noted that Section 6.11 of Surveillance Test (ST) 5071.01.26 (Attachment 3) did
{ not provide adequate direction for resetting the OTH. The purpose of the surveillance test is to exercise the OTM and is not primarily intended to train operators. The wording of the ST places secondary
. importance in resetting the OTM with the primary emphasis on engagement of the trip hook to latch-up lever. During the exercising, it was
- i. shown possible to engage the trip hook with the latch-up lever without resetting the OIM. By holding the hook in the engaged
- t position and opening the valve, it was possible to load the latching faces and thereby hold the hook in place with friction. In this condition, the spring forces would be acting to disengage the hook.
Note that in this condition:
- a. The valve was metastable and could trip without any action of the overspeed mechanism.
- b. The turbine would have had no overspeed trip protection even though it could have run.
- 3. 'AFPT 1-1 and 1-2 OTM Reset Stability It was demonstrated that the OTM could be tripped when the connecting rod was pulled hard in the trip direction, particularly when done repeatedly. This should not occur. One experienced EO stated that in order to determine if the OTM has reset, he pulls in the connecting rod in the trip direction. This action could result in the retripping of the OTM. The retripping is attributed to a worn tappet and head lever that would not allow the proper interference between parts when the OTM was reset.
- 4. Alaru Indications On July 5, 1985 additional investigations were conducted under revision 1 to the previously listed MW0s. The purpose of the additional investigations was to correlate the computer alarm printer data from 1:41:31 to 2:01:58 on June 9, 1985 with reported E0 actions.
It was observed, from the June 9, 1985 computer alarm printer data, that there was one trip and one reset for each of the AFPT OTMs.
The following summarizes these investigations:
- a. It was noted that OTM must be in its full reset position before the limit switch, which indicates head lever position, changes state. This shows that if the OTM were not moved far enough to allow the tappet to return to its normal position, then the switch would not change state. An associated test showed.that if the OTM were repeatedly moved from the tripped to reset position, with the tappet manually prevented from resetting, then multiple trips and resets were recorded on the alare printer. This test demonstrates that the computer alare printer would indicate if the OTM were being moved sufficiently but failed to reset.
- b. It was demonstrated that the trip hook could be manually held engaged with the latch-up lever and that T&T valve fully opened without resetting the OTH. This action would not result in an overspeed trip " norm" (reset) on the computer alars printer.
The " norm" indication would only be received when the OTH was moved far enough to allow reset of the mechanism.
.s 4
- 5. AFPT 1-2 Connectina Rod AFPT 1-2 OTM was noted to have a " bowed" connecting rod and loose 4
lock nuts (Figure 1). Although these conditions existed, the OTM was within adjustment. The connecting rod provides the mechanical .)
connection between the turbine trip mechanism (device) and the T&T l 4 valve. Proper adjustment is verified by assuring that the gap i between the connecting rod link pin and the front edge of the slot is not less than 1/4" (see Figure 1, Detail A) in the reset position.
r Improper adjustment of the connecting rod could cause either 1) improper engagement of the trip hook to latch-up lever, or 2) failure of trip hook to disengage with the latch-up lever. Operational i i
checks did not show that the bowed rod or the loose connections caused the problems with the OTN.
4
- 6. 'AFPT 1-1 Hold Down Screws Two of the AFPT 1-1 turbine trip device hold down screws were noted to be loose. There are three screws that hold the device to the turbine shaft casing. Although two screws were loose,.the third was ,
tight preventing rotational or vertical movement and maintaining the
! position of the device.
- 7. AFPT 1-1-and 1-2 Leaf Sprinas Attempts were made to view the leaf springs for both AFPTs from the overspeed trip adjustment port. Direct and fiberscopic (AFFT l-2 only) examinations, to the extent possible without disassembly, did not detect any unusual conditions. As discussed, spring integrity
- was verified by exercising and confirming the free motion of the tappet.
- 8. AFPT 1-1 Trip Device Cleanliness There was some build-up of dirt on the mechanisms. However, based
, upon the exercising of the mechanism, particularly the tappet and leaf spring, it was concluded that the observed grime build-up on the overspeed trip device would not, by itself, cause the tappet not to return to its normal position during proper relatching of the OTM.
4 The connected leaf spring appears to provide positive return force to
, the tappet as discussed in Section IV.
, 9. AFPT 1-2 Connectina Rod to Head Lever Ball Joint A ball and socket is provided as the connection of the connecting rod to the turbine trip device head lever. The socket on the AFPT 1-2 linkage appeared to be oblong versus a normally round geometry. This was shown not to cause binding of the OTN and the ball moved freely in the socket.
j l
-,,__r__,__,_,_--..-,.m.,,__. . . .
t
- 10. AFPT 1-2 Pump Room Environment The auxiliary feed pump rooms are congested, particularly the room for AFPT l-1 which contained scaffolding. It was noted by both EOs that some scaffolding had been removed from the AFPT 1-2 room. The .
scaffolding had caused problems with access and operator mobility in the pump rooms while attempting to restart the pumps on June 9, 1985.
It was found that the communications were not adequate to permit the E0 to communicate with'the control room while he was in a position to operate the OTM linkage or the T&T valve.
The pump rooms are not particularly well lighted in the area of the T&T valves cad the OTM linkage. The EOs indicated that some light bulbs may have been burned out on June 9, 1985.
- 11. Lateral Alignment of Linkage Inspection of the linkage showed that the connecting rod is slightly non perpendicular to the axis of the trip hook pivot shaft in the horizontal plane. This results in a slight misalignment between the clevis at the end of the connecting rod and the trip crank. This may lead to some resistance to movement at this joint in the linkage.
This appeared to have contributed to less than complete engagement of the trip hook with the latch-up lever on some occasions. However, a small force manually applied to the trip hook was sufficient to bring the hook and the latch-up lever into full engagement in all cases.
.) r a
+. --- .----e-~ .-.*-m.-+,, -.--..e - - - - - ,+-e--'--- ---'--=---'------c'-~-*-**7
s 4
ATTACHMENT 3
. RESULTS OF OPERATOR INTERVIEWS, PROCEDURAL AND TRtINING REVIEWS i
s .,
4 A. Operator Interviews i
The operators involved in the June 9, 1985 attempts to restart the AFPTs were interviewed and asked to demonstrate and recall their actions at that time. The significant items from those interviews are as follows:
i
- 1. The E0 involved with AFPT 1-1 on June 9, 1985, demonstrated how he had attempted to reset the OTH. His understanding of the operation of the OTM had his pull the connecting rod until the trip hook met with the latch-up lever. The movement of the OTM so that the trip hook met with the latch-up lever was not sufficient to allow either the turbine end of the OTM to reset ,,
or the limit switch to change state.
- 2. The Eos involved on June 9 were interviewed on June 15. In that interview, the EO involved with AFPT 1-1 used the palm of his hand to show how he attempted to keep the trip hook engaged to the latch-up lever. This action would be required if the OTM were not reset since manual force would have baen necessary to overcome the spring force holding the OTM in the tripped direction.
It was demonstrated that with sufficient force on the trip hook, the hook could be engaged with the latch-up lever without '
resetting the OTM. This engagement was unstable and could be ^
disengaged with slight disturbances of the mechanism such as T&T valve operation or mechanical shock. If engagement was maintained,'
the T&T valve could be opened.
In addition to the items noted above, the operator interviews were used in preparing the comments in the chronology of events presented in Attachment 1.
B. Procedures For Reset and Latchina Typical of plant procedures which include instructions for reset and relatching of the trip and throttle valve is. Section 6.11 of Surveillance Test (ST) 5071.01.26. This procedure states the following:
"6.11 Perform the following to exercise the overspeed trip mechanism:
6.11.1 Using the manual trip lever, manually trip the trip throttle valve. ~
4 l 6.11.2 Turn the trip throttle valve handwheel clockwise until
- the sliding nut rises and engages the latch up lever to the trip hook.
4 1
i
__, .. .- m. . . -. , _ _ ~ _ _ , . . . , . , , . . . . . _ . - . . _ _ _ _ , . . . _ . _ . . , , - _ _ _ - _ _ . , _ _ . , . _ _ _ _ _ _ . _ - _ . . , _.
l S
^
h
f NOTE: It may be necessary to pull on the trip
,, throttle valve linkage to fully energize the latch up lever to the trip hook.
6.11.3 Verify the latch up lever and the trip hook are fully ..
engaged.
6.11.4 Turn the trip throttle valve handwheel counterclockwise '
until the trip throttle valve is fully open.
a 6.11.5 ~ Turn the trip throttle valve handwheel 1/4 turn N
q ~, clockwise.
6.11.6 Seal the trip throttle valve handwheel.
Independently Verified v, .,
6.11.7 Verify computer point Z001 (Z002) AFFT 1 (2) Stop Valve reads "0 PEN".
1 6.11.8 Verify the red IL ICS 38E (38J) AFPT I (2) Governor Valve fully'open light is on."
P ; It is particularly important to note that the procedure steps ignore the trip tappet, head lever, connecting rod, and alarm indication of Q.
reset in the control room. As was verified in the inspections, unless the connecting rod is pulled in the reset direction, the
,t tappet will not reset. That is, simply engaging the latch hook with
^
. the latch-up lever will not move the head lever enough to allow the tappet to drop to the reset position. Consequently, this procedure
.. is technically incorrect.
p
' C.- Operator Trainina
' Discussions were held with the training staff to ascertain the depth of the information provided to the EOs. The discussions centered-around the hands-on training conducted on November 2, 1983 and actions taken by the Training Department as a result ef the Institute
- for Nuclear Power Operations (INPO) Significant Operating Experience Report (SOER) 82-8, dated August 4, 1982.
As a result of the normal Toledo Edison (TED) review process for INPO SOERs, SOER 82-08 was transmitted to the training department with the recommendation to include the SOER in Reactor Operator / Senior Reactor Operator required reading and non-licensed operator requalifica-i , tion training. The SOER did not indicate potential problems with.the operability of the OTM. The SOER was primarily intended to warn of a deficiency in the verification of OTH reset. It was noted at other plants utilizing Terry Turbines that the OTM could be in the tripped condition and without control room indication; these pumps would then not be available for operation. Toledo Edison reviewed this SOER and determined that Davis-Besse did in fact have control room indication of the condition of the OTM. According to training personnel, the 4
informatien of the SOER was given to the operators. It is note.d that 4
0 I
(
. - . _ - - , ._.__,_e..-_ _ _ . _ - - . _ _,__.. . - - - _ . - - , - ~ _ _ _ _ . . _ . - - -
s the SOER did not detail trip or reset operations and therefore, training personnel relied on the detail included in the existing plant procedures.
The E0 training conducted on November 2, 1983, was a 20 minute .
session conducted in the AFPT pump room. There were seven operators and one instructor. Of those seven operators, the three principal operators involved in the AFPT restart of June 9,1985 were present.
The training instructor stated that the training was largely a demonstration and that all operators were not given an opportunity to physically trip and reset the mechanism. There was no lesson plan for this training session.
D. NRC Interviews of Operators On June 17, 1985, the NRC interviewed the Equipment Operators (E0s) who were involved in the resetting of the OTM and the relatching and opening of the T&T valves. The statements in these interviews paralleled those obtained earlier by Toledo Edison interviews.
Several specific conclusions can be drawn from the statements by the operators in the NRC interviews which are particularly important in the investigations of the problems with the T&T valves.
- 1. Both T&T valves were difficult to open. The third E0 to enter the AFP room opened both T&T valves, and in both cases he used a valve wrench.
- 2. The two operators who initially attempted the opening of the T&T valves were unfamiliar with how much force would be required, how much hand wheel motion would result, and how to determine 3 when the valve was open.
- 3. The operator who initially attempted to relatch and open the T&T valve for AFFT l-1 was not aware that he had, in fact, succeeded in getting AFPT 1-1 started.
i I
s ATTACHMLNT 4 AFFECTED PROCEDURES LIST Surveillance Tests (ST) 5071.01 Auxiliary Feedwater System Monthly Test 5071.02 Auxiliary Feedwater System Refueling Test System Procedure (SP) 1106.06 Auxiliary Feedwater System Periodic Test (PT) 5150.01 Auxiliary Feed Pump Turbine Overspeed Test Alara Procedures (AP) 3010.15 Auxiliary Feedwater Pump 1 TRBL 3010.16 Auxiliary Feedwater Pump 2'TRBL 3010.47 AFPT 1 Over SPD Trip 3010.48 AFPT 2 Over SPD Trip
s ATTACEMENT 5 TESTING AND INSPECTIONS OF TRIP THROTTLE VALVES RELATED TO DIFFICULTIES IN OPENING This attachment summarizes the results of oper.tional testing of the trip throttle (T&T) valve opening and the results of the subsequent disassembly and detailed inspection of the T&T valve for AFPT 1-2.
A. Operational Testing Testing of the T&T valves for both AFPTs was conducted on September 11 and 13, 1985, using steam from the auxiliary boiler at about 200 psig.
The testing of T&T valve for AFPT 1-2 was done under NWO 1-85-2063-02.
In these tests the force to rotate the valve handwheel was measured using a spring scale acting on a valve wrench so that the lever arm was 20h inches. (The valve handwheel is 12 inches in diameter so that the force at the handwheel rim would be 3.4 times the measured force in these tests.)
- 1. T&T Valve for AFPT 1-2 The first tests were performed on the T&T valve for AFPT 1-2.
With no pressure across the valve, a force of 32 pounds was required to break the valve loose. After that, 5 pounds was required to turn the valve. With 200 psi across the valve, 17 3
pounds was required to break loose the valve. A force of 5 pounds was required for about 3/4 turn and then the force dropped to 3 pounds.
Comment Based on observations made during these tests, the high Nreakaway forces were related to forcing the latching mechanism aga nst its mechanical stop. No effort was made to control how hard the valve latching mechanism was forced against the mechanical stop; consequently, the observed forces at breakaway should not be considered to be related to the pressure forces.
- 2. T&T Valve for AFPT 1-1 The second tests were performed on the T&T valve for AFPT 1-1.
In these tests, care was taken to assure that the valve was not forced in the closed direction, i.e., the valve was backed off the stop slightly before measuring the torque to open. With pressure across the valve, a force of 4 pounds on the valve wrench was sufficient to open the valve, and 3 pounds were adequate to keep it moving. With no pressure across the valve, the force was small - estimated at one pound.
l
s Comment There was, as would be expected, additional force required to open the valve under pressure. The valve for AFPT 1-1 appeared to operate with less torque under no pressure than the valve for -
AFPT 1-2.
- 3. Settina of Throttle Screws The throttle. screws which adjust the pressure in the region over the main disc were found to be set the some on both valves.
^
B. Disassembly and Inspection Because the test results indicated that the T&T valve for AFPT 1-2 was harder to operate than the valve for AFPT 1-1, it was considered prudent to disassemble and inspect the valve for AFPT 1-2 to determine whether it was damaged. The disassembly was done under MWO 1-85-2132-03
, on September 21 and 22, 1985. Representation of the manufacturer (Mr. R. Reed of Gimpel Corp.) and NRC (Mr. D. Kosloff) as well as Toledo Edison were present for the disassembly and inspection.
The following are the major observations made during the disassembly and inspection:
- 1. The steel valve spindle rotates in the bronze yoke cap at the 3 top of the valve yoke. There were some signs of scoring of the spindle 0.D. and cap I.D. Measurement showed the spindle was 0.873 inch 0.D. rather than the drawing dimension of 0.872,
+ .000, .002 inch. The yoke cap bore was measured as 0.8755 inch I.D. rather than the drawing dimension of 0.880, + .003,
.000 inch. The measured diametrical clearance was 0.0025 inch rather than the drawing minimum diaoetrical clearance of 0.008 inch. The spindle, however,.did turn in the cap without excessive torque.
- 2. The sliding nut was found to move freely along the threads on the spindle. There was no evidence of damage of the threads on either the nut or spindle.
- 3. There were some scars and gouges on some surfaces on the coupling between the spindle and the stem. These were not on any rubbing or bearing surfaces and did not appear to affect valve operation.
- 4. The area where the coupling guides on the bonnet showed only minimal rubbing. There were no signs of large forces or loads from misalignment.
- 5. The axial clearance between the spindle hub and coupling was slightly larger than the drawing value. The height of the cavity in the coupling was 0.760 inch compared to the drawing i dimension of 0.750, + .005, .000 inch. i
)
l I
\
c s
The spindle hub, which fits in the cavity of the coupling with a thrust wasber, had a measured length of 0.303 inch compared to a drawing dimension of 0.312, + .000, .005 inch. These dimensions would increase the axial free play by 0.009 inch. Consequently, the rotation of the handwheel from the closed stop until the pilot .
starts to open would have been slightly greater than the design value.
- 6. The stem and valve disk assembly moved easily within the valve body. The stem and assembly could be litted by hand - no .
binding or excessive friction was observed. .
- 7. The pilot disk moved freely within the main disk. The pin connecting the pilot valve to the main disk moved freely.
'8. There was a small amount of pitting on the valve stem at the location of the high and low pressure extraction points and at the location of the guide bushing. The overall finish of the stem was smooth.
- 9. There was some water in the valve and in the extraction piping.
- 10. With the exception of rubbing or seating surfaces, there was a smooth oxide layer on the internal valve surfaces.
- 11. The inside surface of the cylinder which guides the main disk, showed some marks at the location of the pin which connects the main and pilot disks. There was also a mark on the cylinder 3 wall under the throttle screw port which was about 1/16 inch deep and 2 inches long. However, no binding of the main disk was found.
- 12. The pilot valve seat contained six indentations. One was approximately 3/32. inch in diameter, the others were 1/64 to 1/32 inch in diameter. ,
- 13. There were 7 small pieces of what has been identified as weld slag on the exterior of the integral strainer.
The Gimpel Corporation representative concluded there were no dimensional variations or wear which could have directly caused the high handwheel forces.
s Figure 1 -
Latching AFW Turbine Trip Valve
~
Y TRIP HOOK REPOSITION SPRING
. Trip Hook !
I h- Um4 Switch
-p te.r if Valve N / Manual Trip cg '
'h M ism Umit Switch
\ Head p Lever o I l r"1 ,
b
\\ b \\ \l Trppet [ l TurbineTrip Mechanism
~
Steamin Steam 8
-J_ Trip Valve C'
J Ek .
. rc 1 k
s DETAI L A
e e~
DEAotz wa w Twytt .
, ,_x. i T H-- ,-h "m -m.
= --- - --
? \ w comecsmeace
,, / ; ' m*ve .
I massa tw aswa g [ PLAN VIEW w
\
l am
( - ) =a.--an n I I anse
~~ e . .
. e y '.
\' '/., . ,
['
l .
. /! '
- ek ~,P Eg .
a m
_
- wa _2%._
- 2=
[isaotzwn T w canau co m e cross nao -
~
urcie w tzwm
. TArrET w woms watw ELEVATION VIEW )!j e }Nk,N [1] .
FICURE 2 4 masamtwtzwn AUXILIARY FEED PUMP TURBINE OVERSPEED TRIP LINKAGE (VALVE LATCHED. TAPPET RESET)
s
-46_
.a u . .
-- (- . ..
3 I
J T },
>[. @ yh) >,
~ y$1: : :
li jV -
.. , <= i _
j
< = .__p
!? : i :
. < ;,= = x M
= = /
/c a -= ,
,g
~
<4ma: /s
,3
,t 4'g i p s, 3 ,' _g i:
' y nf"- ; -
'?4 At' ' l ;1l + ll
,, alt. J'
[l d!fl[r'c Tsq N-t e.J g 6
!lg!
- j=qv
- EI f* '
.e: . a
') ,
p(;H :l :f c[t ji
-l ,
r..
d,~ .
y ~1 .
It =A ~ 6b am bs [ s
) / '
g,9sgj(fi s
s e
f U;0 - z :-
sw( b _i s *W/ u m -*
1 ' --- -g /k
'/ /// . g 'j/Ax-// e/ >)nn - n- -,w hi 4
B., T lk~ ~$ [ 5 K
x y-.s
!.N
.s
~-f.f~
5, - e
- 9-
'e -
a
/,3 y, _,
, N u a( g , g-sN q'
'7ya g T
,g s I .
g, j I L IE .
s
-g 4 4'
--f -
- . _ _ .. _ . . s ,..- ___
' r ! .
T' g W' s s' >
j sN ' :: N - --
NN
.d_
s
,1 i .)-
LLI I. ___ L j
\s !j r.
di Yh a .. Mix Figure 3 Auxiliary Feed Pump Turbine Trip Throttle (T&T) Valve ,