IR 05000237/1990005
| ML17202G815 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 02/23/1990 |
| From: | Danielson D, Huber M, James Smith NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III) |
| To: | |
| Shared Package | |
| ML17202G814 | List: |
| References | |
| 50-237-90-05, 50-237-90-5, 50-249-90-04, 50-249-90-4, IEB-85-003, IEB-85-3, NUDOCS 9003120241 | |
| Download: ML17202G815 (10) | |
Text
u~s~ NUCLEAR REGULATORY COMMISSION
~EGION II I
. Reports No. 50-237/90005(DRS); 50-249/90004(DRS)
Docket Nos. 50-237; 50-249 Lic~nses No. DPR-19; DPR-25 Licensee:
Conunonwealth Edison Company P. O. Box 767.
.
Chicago, IL 60690_
Facility Name:
Dresden Station - Units 2 and 3 Inspection At:
Dresden Site, Norris, IL 60450 Inspection Conducted: January 29 through
. obJ"ftt.11~~~
Inspectorsr. F. smi~h
.. hJ~t~UJ~
M. P. Huber *
~}~~;/~-"-
February 8, 1990 Approved By:
D. H. Danielson, Chief Materials and: Processes Section Inspection Sununary
. Date
~/:;3/~
Date Date Ins~ection on January 29 through February 8, 1990 (Report No. 50-237/90005(DRS);
so-49/9ooo4(DRS)) *
Area~ Inspected:
Routine~ ~nnounced safety inspection of maintenance and
.inservice testing (IST) of pumps and valve The areas covered a review of the implementation of IST including a review of procedures, performance of testing and recording of the trends (73756), and a review of previously identified
. items (92701).
.
Results:
.No violations or deviations were identifie Of the areas inspected, th~ inspectors noted.the following:
0
Review of the IST prog~am indicated that a knowledgeable staff was dedicated to ensuring effective implementation o,f the. IST progra The aggress 1 veness of the 1 icensee' s sta*fr in addressing techn i ca 1 issues involved with the IST program was goo Problems were noted during the-inspection with regard to previous testing methods and administrative details; however, the licens~e's staff appears to have adequate support to complete the necessary reviews and implement a complete progra *
9003120241 900223
~0~
ADOCK b50002~7 Q
PD~
- DETAILS Persons Contacted ' Commonwealth Edison Company (CECo}
- E. D. Eenigenburg, Dresden Station Manager
- L. Gerner,.Technical Superintendent
- M. Strait, Technical Staff Supervisor
- K. Kociuba, Quality Assurance Superintendent
- J. Coonan, C.O.M. Coordinator
- G. Bergan, Onsite Nuclear Safety
- R. Falbo, Regulatory Assurance Assistant
- M. Horbaczewski,=Technical Staff
- F. bundek, !ST Coordinator u; S. Nuclear Reg~latory Commiss'ion. (NRC}
- D. E. Hills, Resident Inspector
- Denotes those present for Exit Meeting on.February 8, 199 The NRC inspectors*.also contact~d other license~ persbnnel during the
- course of the inspectio *
L~censee Action on Pre~ious Inspection Findirigs (92701} (Closed) Unresolved Item ( 50-237 /87033-02; 50-249/87033-02}:
Testing, test criteri~ and test evaluation were inadequate to assure that thermal overloads will perform properl The licensee generated a new procedure to resolve this proble Proc~dure DMP 7300-5, Revision 11, dated January 1989 "Unit 2(3) Inspection and Maintenance of 480V MCC Breakers/Contactors and 208V Contactors" provides a basis for scheduling tests and prescribes test parameters and test acceptance criteri In addition, the licensee uses Station Nuclear Engineering Department (SNED} Design Guide No. 26, 11Setting of Thermal Overload Devices for Mota~ Operated Valv~s" fer selecting thermal overload sites. This document provides guidance for determining the size of the thermal overload relay elements based on carrying locked rotor current for the maximum time without damaging the moto The combination of these documents provides assurance that the proper size thermal overloads were selected for motor.operated valves (MOV~)
and that they will be properly tested at* appropriate intervals. This item is close '
'
. (Closed) Unresolved Item (50~237/87033-03; 50-249/87032-03):
Operational controls to prevent exc~ssive valve cycling!
To prevent thermal degradation of motor insulation and shortening of MOV motor life, the license~ has instituted Operating Order No. 5-90. This order specifies the maximum number of times that a MOV may be started
\\\\
..........................................
,~-.,..........................
.,,.
within a* given.pe/1od~ The maintenance dep~rtment has instituted a similar order ~or ~se when local operation of the valve is require during maintenance~. Th~ licensee has provided both operations an, maintenance ord~is;to prevent excessive cycli~g of MOV This item is considered closed *. *
- *
t._. (Clb~e~j Unresol~ed Item (50-2j7/8703§-01~*50-249/87032-0i): Inadequate controls* *are 1n place to. assure open limit switch settings. The setting of the open limit swi_tch is of concern and*
control ~f the setting*needi to.be est~blished to ensure that possible.
damage *from backseating, is prevented.*
The licensee uses procedure-DEP 040-9, "Limitorque Valve Operator; Maintenance and Surveillance"*
when the open 1imtt switch is ~djusted. By the ptoc~dure,. the_ open limit switch is set at.a point wher~ the valve should not.backsea The open limit swt~ch also.is verified ~o be set properly by operating the valve electrit~lly to the open prisition and then manually oper~ting the. valve to ensure that the valve does not back~eat. * This is an acceptable practice and therefor_e *this item is considered close. *
- . '..
'
.
(Closed) Op~n Item {50-237/87033-08~ 50:249/87032-08): * Evaluation of inaccurate o~en indication after setting open torque switch bypass; Normall~ wi~h a two rotor control switch configuration in L imitorque actuators, a conflict b.~tween the setting.of the open
- torque switch bypass *and the*o~e~ iriditation exists. The licensee
. has evaluated this condition and implemented procedures and modifications to -ddress the:conflict:of the two switch setting The licensee has*i~plemented Operating Order No. 6-90 "Motor Operated
.
Valves with Torque Switches Set pe~ NRC Bulletin 85~03". This order
identifies valves that do not ~av~ a seal-in and ~pecifies ~ct~ons to be taken when operating these valves; AdditiQhally, the licensee is implementing modification number M12*-2(3)-88-063, "Dresden *station Two to Four_Rotor(Limit Switch Replacement" to allow valve open position indjcation and torque switch open bypass to be set
Lindependently: Either the two rotor limit switch configuration ~ill
- be chang~d to a fciur rotor conftg~ration or.the torque switch open bypass functton w'll.be jumpered out and the position indicating lights set at a pciin~ that will more accurately reflect the actual valve ~Q~itio The modification 1s* ongoing; but in the foterim, the *licensee performs stroke time testing by t_iming, the valves at the.motor control center JMCC) by observing the contactors at th_e MC No problems were note~ with the licehse~*~-actiVities to address this item. and therefore it is considered_ close '(Closed) UnfesolvJd Item (50~23~/8~633~09~ 50~249/87032~09): The setting technique'for the limitorque open torque switch_ bypass can be expected to lead. to improper.settings. The use of current traces to identify the point: of unseat~ng ~f ~ valve could. lead to mischaracteriz~ti9n*of the traces. This could result~in mistakin~ stem motion for disc motion, and therefore the bypass ~ould be set, based
- 3
';
'*
.,,
,l
. ~*
on the faulty reading of. tH_e current trace, at the point where it was needed most and,could no longer serve its.functio The li~ensee ~ses current traces duri~g the setting of the open bypas~ switch to.set the switch at least an additional 253.of full stroke time beyond*: the initial motor current spike or 20% of full.
stroke time beyond: the dis~ unseating spike, both. ~~ikes are*
identified on the current trace.* The 20% of full stroke time margin.
for setting the open'torque switch bypass is generally accepted as adequate, and ~his item is considered.close * (Closed) Open Item_, (237/87033-10; 249/87032-10):
Close torque switch bypass may be opened before valve is free of backseat. The concern was that the toi~ue switih might become challenged by the clpsing force required to disengage the valve: stem from a backseated conditio and thereby interrupt ~he closing action. The principal cause for challenging the close to~que switch when coming off the open position was eliminated by procedure DEP 040-9, Revision 1 dated October 1988,
"Limitorque Valve Operators Maintenance* and'Surveillance. 0 This procedure requires: not only setting valve switches to avoid backseating but also manually confirming the absence of backseating. Without backseating, ther~ is no routine operating conditton which would
. challenge the torque ~witch. The licensee also has a procedure DEP 040-11, Revisipn o. dated August 1988, ~Electrical Backseating of
. Limitorque Operated Valves" which could challenge the torque switch, but this is a spec~a1 application which requires special precaution For normal usage of motor operated valves, this jtem is close (Closed) Open It~m (50-249J87029-02):. Followup on Event Number 11482,
"LPCI Inop due to LPCI Minimum Flow Valve Failure."
Ori February 11, 1988, Dresden notified NRC that the Unit 3 Low Pressure Coolant Injection (LPCI) ~ystem was inoperable due to an inoperable minimum flow valv The.valve was.declared inoperable because the valve yoke was found to be broke NRC Inspection Report Hb. 50-249/87029(DRS)
documented the licensee's actions to repair the valve and *noted that further eva luat iori' of the* root.cause of* the failure and the comp let ion of corrective acti~ris taken was necessar I The licensee completed the analysis of the failure and concluded that an improper *weld was used on the va-lve yoke. The* yoke bonnet to motor operator "adapter p 1 ate was rewe l ded ut ili z fog a f u 11
..
penetration weld rather than the improper fillet wel As further corr*ective action,, the licensee initiated action to inspect other operators and reweld any connections where necessar Additional commitments were also generated when it was noted in the everit investigation that the torque switch ~etting was set too high*
- in the close dire~tiorr (higher than specified by the~vendor). It was determined by the licensee that the higher setti~g would ~ot
.result in loads o~ the valve o~ operator in excess of their capabilities *. The possible cause for the torque switch setting error was confusiqnwith the Limitorque SMB-000 torque switch. The
- *
licensee has revised their torque switcfi setting procedure to clarify the i 11 ustrat ions *ind '\\cat i ng the torqu*e sw_i tch adjustment screw Training sessions were also given to maintenance technicians to address this issu No problems were noted and the licensee's actio~s are considered acceptable. This item is considered close Pump and Va.lve IST Program Implementation (73756)
The licensee was implementing Revision 3 of their IST program, which is currently under review by NRC and due to receive a Safety Evaluation Report (SER).
This inspection wa~ conducted to verify that the implementation of the licensee's IST program was conducted. in accordance with their current IST program submittal and to verify compliance with Appendix B of 10 CFR SQ; 10 CFR 50.55a(g); and Subsection IWP and IWV of Section XI of the ASME Cod The inspection consisted of a review of administrative controls, selected surveillance procedures, test results and documentation. *
- Completed Surveillance Review The NRC inspec:t;ors reviewed completed surveillance procedures to verify implementation of the licensee's IST progra The surveillance$ reviewed are listed belo ' "Quarterly Valve Timing"; DOS 1600-1~ Revision 15, dated April 1989 with Temporary ChJnge Number 89-162, work performed August 12-20, 1989
"Quarterly LPCI_ System Pump Operability Test with Torus Available for the In-Sjrvice Test (IST) Program," DOS 1500-10, Revision 9, dated December 1988, work*perfo~med February 27, 1989, Revision 5, January 1988, work performed March 28, 1989
"H~gh Pressure Coolant Injection (HPCI) System Operability Verification, "DOS 2300-3, Revision 17, dated May 1989 with
- Temporary Change {TC)89-173, work performed June 18, 1989
"Quarterly Diesel Generator Cooling Hater Pump Test for the In-Service Test {IST) Program," Revision 7, dated April 1989, work performed February 2, 1990
"Core Spray Sysfem Pump Test with Torus Aviilable," DOS 1400-1, Revisipn 15, !August 1989, work performed September 11, 1989
'"HPCI MO Valves and Pump Operability Test Monthly," DOS-2300-l,
. Revision 17, Septem_ber.1987, work performed October 11, 1988 *
"Monthly Suppression Chamber to Drywell Vacuum Breaker Fu.il.
Stroke Exercise Test," DOS 600-9, Revision 6,* June 1988, performed January 25, 1989 *
-
0
"Q~arterly Containment Coolipg Service Water Pump Test for the In-Service Te~t (IST) *Program," DOS 1500-2, Revision 8, December 1988 performed February 27, 1989
"Standby Liqu~d Control System Pump Te~t, 11 DOS 1100-1, Revision 9, not dated, approved April 17, 1989, performed April 29, 1989 Of the procedures reviewed, discrepancies were noted with the High Pre~sure Coolant Injection (HPCI) s~stem pump testing and the method
'of determining vibration data acceptabilit *
.
During the review af the HPCI surveillance procedure~, the NRC inspector noted that the*surveillanc~ did not adequately test the HPCI pum The testin~ method outlined in the surveillance procedure was not in accordance with the testing method that is specified by the ASME Cod S~rveillance procedure, DOS 2300-6, Revision 4,
"Quarterly HPCI System Pump Test for the In-Service T~st (IST)
Program" dated April 1988, incorrectly adjusted both the discharge pressure and flow rate to specified values and the procedure was not conducted with the. pump operating at a reference spee IWP-3100
"Inservicc Test Procedure".requires that the pump tes~ shall be conducted with the pump operating at a speed adjusted to the reference speed for variable speed drives. Additionally, the resistance of the system (to flow) shall be varied until either the measured differential pressure or the me'asured flow rate equals its corresponding reference value. Unless th~ testing is performed as specified, a comparison between tests cann6t be made to evaluate pump degradatio The way the procedure was written, the licensee effectively set the pump conditions to an acceptable point of operation and t~en evaluated the parameters to determine the acceptability of the pump operation~
Subsequently, the inadequate procedure was revised t6 test the HPCI pump in accordance with the ASME Code specified testing procedur Subsequent reference valve testing showed that there were no problems noted with the pump operatio The licensee corrected the discrepancy prior to conduct 6f this NRC inspectio *
-
The second discrepancy noted was with the method of d~termining that vibration data was acceptable. The current revision of the IST program specif1e$ that pump vibration wfll be measured and recorded in velocity units. However, of the sample of procedures reviewed by the NRC inspector~, some of the current IST procedures still use displacement measurements to determine the acceptability of pump vibration readings. The use of displacement as a means to evaluate the pump vibration was used by the licensee prior to the current revision of the. licensee's IST progra Both means of measuring pump vibrations are acceptable, however, the use of velocity units is a better means of evaluating pump degradation and is used by the licensee as a part of their vibration analysis progra The licensee init~ated pr6cedure revisions to co~rect the discrepancy between the program and the implementing procedures pribr to the completion of the-NRC inspection. Additionally, it was stated by the licensee that p~ogram and procedure revision are continuirig as a result of ongoing discussions with the NRC prior to issuance of the SER for the IST prpgra No other problems were note Test Observations The NRC inspectors witn~ssed the performanc~ of DOS 6600-8, Revision 7,
"Quarterly Diesel Generator Cooling Water Pump Test for the In-Service
- Test (IST) *Program~" performed on February 2, 199 The licensee used operations pe~sonnel and a designated technical staff technician to perform the test~ Vibration data was obtained using calibrated equipment and the points used for measurement were clearly marked on the pump Reference conditions were established in accordance with the ASME Code to a~sess the pump condition. The work was done in a professional manner by a proficient staff. *No problems were note Declaration of Inoperability of Pump and Valves Generic letter 89-04, Position 8~ ts surmnarized to state that, 11 ***
as soon ~s data is' recognized as being within the required action range for pumps o~ exceeding the limiting valve of full stroke time for valves, the as$ociated component must be declared inoperable and the TS action time* must be started."
Although the words supporting this position in the 11censee 1s procedures are subject to misinterpretat1on,.the intent is clarified in OAP 9-11, "Procedure Usage and Adherence," Revision 1, dated August 198 P~ragtaph C.2.~.2.(a) states that the shift supervisor should be notified whenever test results fall outside ~cceptance criteria. The action is clearly prescribed in the pump and valve IST procedures;
~s soon as the shift supervisor is aware of the condition, the equipment-must be declared inoperable irmnediately~
Without this guid~nce.it is possible t6 interpret the procedure to allow delaying the reporting of ~quipment in the required action range to be deferred until the test i~ completed (up to eight days later) and the data is given to the shift petsonnel for analysi i Backseating of Valves Dresden does not normally backseat valves. Procedures are written and switches are set to preclude the impact of the rising stem with the backsea Normal operation of valves is not a problem in this respec Any backseating which might occut under normal cirtumstances
- would* be the result of individua_l *error or oversight and should be
- tre~ted on an individual basi There is, however, one condition under which deliberate backseating offers a benefit. * That is when the stem packing leaks and the licensee wishes to stop the leak without depressurizing the syste.+
Some valve.s are designed to accommodate this action.. They include an enlarged segment o~ stem which seals against a seating ba~kseat secti6n in the valve bonnet when the* stem is fully withdraw However, the seatin'g. area is small compared to the normal valve seating area, so the load applied to it must be limited to prevent damage to the back~eat:or the stem~
In order to electrically backseat ~ valve, the licensee must first establish what force can be ~afely applied to the backseat, establish a*
means to reproducibly control the force within the acceptable limit, and* then demonstrate the.effectiveness of that syste.
- .
Procedure DEP 040-11, Revision 0, dated August 1988, "Electfic~l.
Backseating of Limitorque Operated Valves", is intended for use to stop a packing leak in a Limitcrque operated valve by electrically backseating the valve. 'It is reported to be an emergency measure *
which has-not.been nec~ssary to apply t6 date~
The backseating of a v~lve to control leakage p~st the packing is a conventional application of one of the.valves' normal usage option Normally the valves are intended to be backseated manuall Backseating with a*motor operated valve adds numerous complications tb the operation. It is possible to do, btit it is a complex issue which requires special: consideration and usually, sp~cial equipmen The above procedur~ contain~ a number of areas with which the NRC inspectors have concerns. These include the following:
O*
0
0 There is no 1,mit *on fbrce applied to the backsea The control o~ force applied to the backseat is les~ than that available from setting the torque switch, and the torque switch is considered.inadequate to properly control backseatin There is no control of temperature reductions while the valve is backseated. These can increase tensile stress in the ste~
when the thermal expansion of the stem is greater than.that of the valve bod The test to demonstrate that the backseated valve will close, after the previous backseatfog,_ do.es not prove that it will close after the next on *
ihe same tech~ique used with different bperators on the same valves will yield diffeient result There is no evide.nce that an emergency analysis was performed by the licensee o'n the valve to justify the type of loads
. involved on the backse*ts and stem *
.The NRC inspectors'. recommended that this procedure not be used until there is objective* evidence that it will not damage the valve or operato '
I',
i Check Valve Program (73756)
The 1icens.ee's. chec'k valve program is based on the information provided in
86~03 and:!lectric Power Re~earch Institute (EPRI) NP-5479~
The principal considerations. in selection of valve$ for sampling inspection is*their me~bership in specific groups and the maintenance*history of similar valve types and applications in which there is a high probability
- of failure and lit'tle correlation has been found with th~ predicted results~
During the first app 1 icat ion of the program, the fina 1 n*umber of valves in~pected was 6 Of that number, 17 were found to contain worn, loose, or missing parts which ~ight have been overlooked w'thout the program~
Management is convinced that the program is effective in preventing
. failures and is providing s~pport so that it may cont~nue to do s..
.
.
~he licensee is also searching ~Qr ne~ ways. in which t~ i~prov~ inspection methods by the use of non-intrusive inspection, such as Acoustic Emission (AE).
Toward this end, they have evaluated several new products in. the fiel The latest round qf inspections was performed prior to the current 9utag At least one of the valves inspected has since been discovered to contain a significant mechanical failure. The examination of the data
pre~iously prepared for;this valve and the problem predicted f~r it will be of considerable interest*to the.licensee in his evaluation of the testing device *. The NRC. inspectors requ~sted:that they.be informed of the results of this compa~ison~.
'
Check valves with press~re seal bonnet~ malfunctioned at Comanche Peak
.as ~result of improper cdntrol *of valve assembly iristructions. The first fasteners to be.tightened were subject~d to excessive torque before the
- seal ring in the bonnet.:was* bro-ught 1nto. uniform contact with the retaining ledge in the bod As*a result, the bonnet was cocked, the fasteners were bent, and ~he seal rings were not properly seate *
To explore the possibility that the e.xisting.procedures might permit the duplicatio~ of this problem at Dr~sden~ three assembly procedures were reviewed by the NRC )nspectprs~ These included the following:
0 DMP 0200-15, Revision 7, "MSIV Maintenance," January 19, 1990
.
'
. '
..
'
DMP 3200-01, Revision 3, "Feedpump Discharge Check Valve Disassembly, Seat and Disk Replacement and Reassembly," December 1989
'
.
.
DMP 220-1, Revisio~ J, "Feedwater th~ck Valve Disass~mbly, Seat and D\\sc Replacement and Reassembly, 11 April * 1989
..
.
.
The ffrst two procedures contain direction to cross tighten belts, but. *
with no specific tight.eriing order and"to no specific torque values. The last procedure included a ~ightening sequeh~e, but_did not specify torque value. i
.: r
- l I i
- , l
.* ~
'
I J
'
'..
,..
'*
Discussion with the licensee disclosed that the term used in the first two procedures ("cross-~.i.ghtening") to describe bolt tfghtehing was a.
standard technique taught in the training of technicians and was not
.considered an ambiguous 'instruction~ The absehce of torque values from the procedure was* not an ove.rsighL The licensee was aware of the numerous*
factors which can influenc* the bolting stress prodticed by~ given torque value and deliberately avoided reference to any, preferring to rely on the skill of the*~raft to* achieve a satisfactory r~sul Th~ licensee has n~~erous pressure~seal valves in the piant. If the craft was ~ot properly reassembling these valves, the poor practice ~ould be evident in the form cif leaks, damaged components and, particularly, bent studs or bolts. A*review of the recent history of Dresden's valve repairs indicates no instances in which such repairs were n~cessar Based on this.evidence~"
1 the. existing procedures appear to be achieving acceptable results even 'though they r~ly heavily on the skill of the craft. If the trending data should ever tndicate deterioration in this area, the addition of mo~e explicit guidance in assembly, tightening sequence end ti'ghtening *force would be 'recommende.
Exit Interview (30703}
The NRC inspector met ~1th licensee representatives {denoted in Paragraph 1)
- ,on February 8, 1990, t.o :discuss the sc.ope and findings of the *inspectio The licensee ackno~ledged the statements made by the inspectors with respect to items discussed in the report. *The licensee was also notified of the status of the 6p~n a~d unresolved items identified in Paragraph 2 and the results of the review of the IST progra The inspectors discussed the likely informational content of the inspection report with regard to documents or processes revie~ed by the inspectors during the inspection and the licensee did not.identify any such documents or processes as proprietar