ML20154B979

Summary of 850701-02 Meetings W/Util Onsite to Audit Design Review/Quality Revalidation Program for Tdi Diesel Generators

ML20154B979
Person / Time
Site:	Comanche Peak
Issue date:	02/25/1986
From:	Burwell S Office of Nuclear Reactor Regulation
To:	Office of Nuclear Reactor Regulation
References
NUDOCS 8603040527
Download: ML20154B979 (7)

Text

'

2 f 8 o UNITED STATES

[' ], NUCLEAR REGULATORY COMMISSION 37 E WASHINGTON, D. C. 20555

%.....}1 re,,# %

Docket Nos.: 50-445 and 50-446 APPLICANT: Texas Utilities Generating Company (TUGCO)

FACILITY: Comanche Peak Steam Electric Station, Units 1 and 2 (CPSES)

SUMMARY

OF MEETING BETWEEN NRC STAFF AND TUGC0 TO AUDIT THE COMANCHE PEAK, TDI DIESEL GENERATOR RELIABILTIY PROGRAM A meeting open to the public between the NRC staff and TUGC0 was held on Monday afternoon and Tuesday, July 1 and 2, 1985, in the Visitor's Center /

Nuclear Operations Support Facility at the Comanche Peak Steam Electric Station. The purpose of the meeting was to audit the design review / quality revalidation (DR/QR) program conducted for the Comanche Peak diesel generators manufactured by Transamerica Delaval, Inc (TDI), a part of Phase 2 of the TDI Owners Group Program.

The NRC staff's earlier review of the Comanche Peak Diesel Generator Reliability Program was documented in Section 9.5.9 of SSER #6, issued in November, 1984. That review identified numerous items to be completed before licensing or before NRC final approval of the diesel generator reliability program. Since SSER #6 was issued the TDI Owners Group and the appifcant had submitted additional information to the NRC staff, and conducted additional inspections and tests. At the meeting, the applicant and staff discussed the numerous items and the efforts, status and results of bringing these items to an acceptable conclusion. On a majority of the items, the staff's review of the referenced material resulted in a conclusion that the guidance action item was resolved. On those items remaining, the staff gave additional guidance on actions required for resolution.

At the start of the meeting the staff stated that this audit was generic to the TDI V-16 diesels; i.e. that staff conclusions would be appropriate for all TOI V-16 diesels and that staff conclusions of a generic nature would be given in a forthcoming Safety Evaluation Report specific to this diesel model. In that respect, this meeting was similar to an audit meeting held at Shoreham for the TDI eight-cylinder diesels.

The meeting started with an audit of the " Questions Raised in CPSES DR/QR Audit Review," included in the meeting notice (Enclosure 1). The applicant responded with a 17 page handout entitled "CPSES DR/QR Audit Review' (Enclosure 2). All of the questions and responses were discussed. Of these, more in depth discussions were held on the following points:

I 8603040527 e60225 PDR ADOCK 05000445 S PDR

J

1) The staff questioned the use of Glyptol lacquer to monitor fastener moverent rather than the used of a device which provides a locking grip; i.e., double lock nuts or a bonding agent such as LOCTITE 217. The applicant committed to provide surveillance of the Glyptol for movement. Should the Glyptol st.ow movement, the applicant will examine whether the , junction should be modified.

2) The staff questioned the need to provide forced ventilation in the control cabinet. The applicant advised that it intends to implement forced ventilation at the first refueling outage. That decision is based on a belief that it would enhance reliability, but that it is not necessary. The applicant committed to monitor temperature within the cabinet with temperature indicator tape or other temperature measuring devices on resumption of testing.

3) The staff also questioned why 15 incidents related to fuse and relay failure which had been designated random failures by the Owners Group, were not evaluated by the applicant. The applicant stated that those incidents, were reviewerf by the CPSES startup group and other applicant groups which concluded they were unique and unrelated. The evaluation of the failures in search of common causes was described.

The staff indicated tentative acceptance of the above responses and commitments. The remaining items and responses on Enclosures 1 end 2 were given tentative approval after brief discussions.

The second part of the meeting consisted of a discussion o' maintenance and surveillance activities, in particular to any differences between the CPSES recommendations, and those given in the Owners Group DR/QR report. A portion of the exceptions taken to the Appendix 2 maintenance and surveillance program in the applicant's letter (TXX-4501 dated June 27,1985) served as the agenda for this pot tion of the meeting and is provided as Enclosure 3 for ease of reference. The applicant stated that it was in basic agreement with the guidance on maintenance and surveillance activities as documented in Appendix .

2 of the Owners Grcup DR/QR reports as modified by its letter TXX-4501 dated June 27, 1985. The staff acknowledged that the list of maintenance and surveillance reouirements given on page 9-8 and in Appendix ! of SSER 6 was intended as preliminary guidance, and that the final maintenance and surveillance requirements have been evolving since Appendix I was written.

It should be noted that the staff and the applicant did obtain the Owners Group concurrence during the meeting regarding only those modifications proposed by the applicant to the original Owners Group Appendix 2 maintenance and surveillance recommendations which were discussed at the meetino. This is important because a large number of future changes, including those r.ot discussed during this meetino which will be proposed by any utility to the Owners Group maintenance and surveillance program will be reviewed and approved by the Owners Group.

N 8s Wes Of the items listed on attachment TXX-4501 which were reviewed and discussed at the meeting, the staff indicated tentative acceptance of the proposed changes to the Owners Group recommendations for which owners group concurrence was given.

NOTE: The applicant submitted a revised description of its modifications to the maintenance and surveillance activities, (i.e. updating Enclosure 3) in a letter (TXX-4556) dated October 4, 1985. This latest letter included modifications in response to staff comments on July 2, 1985.

This portion of the meeting closed with a discussion on how the applicant could make changes to the maintenance procedures in the future as part of their continuing diesel reliability program. The applicant proposed using the same procedures described in the technical specifications for changing its procedures for the diesel maintenance. The staff was not ready to address this proposal, but advised that it would have a definite policy in the near future which would apply to all TDI OG utilities wishing to make a small number of future modifications to their naintenance and surveillance requirements.

The meeting attendance is listed in Enclosure 4.

Spottswood B. Burwell, Project Manager PWR Project Directorate #5 Division of PWR Licensing-A

Enclosure:

1. Meeting Notice, dated June 25, 1985

2. CPSES DR/0R Audit Plan

3. Letter to V.S. Noonan from J.W. Reck re: Design Review / Quality Revalida-tion (DR/QR) Report on the CPSES TDI Diesel Generators, dated June 27, 1985.

4. Meeting Attendance cc:

1 Seenextpaeb 0 o . ,

d :PD# :P 05/  : . R/PD#5  :  :

OFC :PD#"[

_____:_y .5...:_

__4

.:_3 g_______ __

j/_______:___ _

NAM *

, rl nger :A -Cook C ranmell VSNo n  :  : : :

____.:__ i... 1..:____.j..____:.y ,

86 DATE : N/14/R6 I

,d/f,(/86:  :-A /14786.._________h:_

/ /86  :  :

0FFICIAL RE' CORD COPY

D 25 g Of the items listed on attachment TXX-4501 which were reviewed and discussed at the meeting, the staff indicated tentative acceptance of the proposed changes to the Owners Group recommendations for which owners group concurrence was given.

NOTE: The applicant submitted a revised description of its modifications to the maintenance and surveillance activities, (i.e. updating Enclosure 3) in a letter (TXX-4556) dated.0ctober 4,1985. This latest letter included modifications in response to staff comments on July 2, 1985.

This portion of the meeting closed with a discussion on how the applicant could make changes to the maintenance procedures in the future as part of their continuing diesel reliability program. The applicant proposed using the same procedures described in the technical specifications for changing its procedures for the diesel maintenance. The staff was not ready to address this proposal, but advised that it would have a definite policy in the near future which would apply to all TDI OG utilities wishing to make a small number of future modifications to their maintenance and surveillance requirements.

The meeting attendance is listed in Enclosure 4..

/

ii \ g- [ '

Up , \lI{ y % \ ,' f\

Spottswood B. Burwell, Pro.iect Manager PWR Project Directorate #5 Division of PWR Licensing-A

Enclosure:

1. Meeting Notice, dated June 25, 1985

2. CPSES DR/QR Audit Plan

3. Letter to V.S. Noonan from J.W. Beck re: Design Review / Quality Revalida-tion (DR/QR) Report on the CPSES TDI Diesel Generators, dated June 27, 1985.

! 4. Meeting Attendance cc: See next page t

l l

l

\

I

s' l i

i W. G. Counsil '

Comanche Peak Stean Electric Station Texas Utilities Generating Company Units 1 and 2 cc:

Nicholas S. Reynolds, Esq. Resident Inspecr.or/Cofranche Peak Bishop, t.iberman, Cook. Nuclear Power Station Purcell & Reynolds c/c U. S. Nuclear Pegulatroy Comission 1200 Seventeenth Street, N. W. P. O. Box S3 Washington, D. C. 20036 Glen Rose, Texas 7 M 43 Robert A. Wooldridge, Esq. Pegional Adninistrator, Fegfon IV Worsham, Forsythe, Sampels & U. S. Nuclear Regulatory Ccmission Wooldridge 611 Pyan Plaza Drive Suite 1000 2001 Bryan Tower, Suite 2500 Arlington, Texas 76021 Dallas, Texas Mr. Robert E. Ballard, Jr. 4.anny A. Sinkin Director of Projects Christic Institute Gibbs and Hill, Inc. 1324 North Capitol Street 11 Pen Plaza Washington, D. C. 20002 New York, New York 10001 Mr. R. S. Howard Ms. Billie Pirner Garde Westinghouse Electire Corporation Citizens Clinic Director P.O. Box 355 Government Accountability Project Pittsburgh, Pennsylvania 15230 1901 Que Street, N. W.

Washington, D. C. 20009 Renea Hicks, Esq. David R. Pigott, Esq.

Assistant Attorney General Orrick, Ferrington, & Sutclitfe Environmental Protection Division 600 Montgomery Street P.O. Box 12548, Capitol Statien San Francisco, California 94111 Austin, Texas 78711 Mrs. Juanita Ellis, President Anthony 7. Roisman, Esq.

Citizens Association for Sound Energy Trial lawyers for Public Justice 1426 South Polk 2000 P Street, N. W.

Dallas, Texas 75224 Suite 611 Washington, D. C. 20036 Ms. Nancy H. Williams Nancy E. Wiegers CYGNA Spiegel & Mc0 farmed 101 California Street 1350 New York Avenue, N. W.

San Franc!sao, California 94111 Washington, D. C. 20005-4798 Robert P. I.essy, Jr.

Morgan, Lewis & Boktus 1800 M Street, N. W.

Suite 700, North Tower Washington, D. C. 20036

Texas Utilities Electric Company Comanche Peak Electric Station Units 1 and 2 cc:

Resident Inspector - Comanche Peak c/o U. S. Nuclear Regulatory Comission P.O. Box 1029 Granbury, Texas 76048 Mr. John W.' Beck Vice President Texas Utilities Electric Company Skyway Tower 400 North Olive Street, L. B. 81 Dallas, Texas 75201 Mr. Jack Redding licensing .

Texas Utilities Generating Company 4901 Fairmor.t Avenue Bethesda, Maryland 20814 William A. Burchette, Esq.

Peron, Burchette, Ruckert & Rothwell Suite 700 1025 Thomas Jefferson Street, N. W.

Washington, D. C. 20007 Mr. James McGaughy Southern Engineering Company of Georgia 1800 Peachtree Street, N. W.

Atlanta, Georgia 30367-8301 Adcinistrative Judge Peter Bloch U. S. Nuclear Regulatory Commission Washington, D. C. 20555 Elizabeth B. Johnson Administrative Judge Oak Ridge ifational I.aboratory P.O. Box X. Building 3500 Dat Ridge, Tennesse 37830 Dr, Kenneth A. McCollom, Dean Divisicn of Engineering, Architecture and Technology Cklahoca State University Stillwater, Oklahoma 74074 Dr. Waltar H. Jorden 881 Outer Drive Oak Ridge, Tennesse 378?O

F-

• E Rg Meetino Summery Distribution etDestetWCentral File NRC Participants NRC PDR C. H. Berlinger local PDR D. Persinko PD#5 Reading File S. Burwell

, J. Partlow (Emergency Preparedness only) E. l. Murphy V. Noonan Project Manager S. Burwell OELD E. Jordan

8. Grimes ACRS (10)

M. Rushbrook cc: Licensee and Plant Service list t_._

, ,,,,, ENCLOSURE 1 ff

%, UNITED STATES y g NUCLEAR REGULATORY COMMISSION

,  ; j wasHWGTON, D. C. 20655

/

JUN 2 51985 Docket Nos.: 50-445 and 50-446 MEMORANDUM FOR: Vincent S. Noonan, Director for Comanche Peak Project, DL FROM: S. B. Burwell, Project Manager Licensing Branch No. 1 DL

SUBJECT:

FORTHCOMING MEETING WITH TEXAS UTILITIES TO AUDIT THE COMANCHE PEAK DIESEL GENERATOR RELIABILITY PROGRAM DATE & TIME: Monday, July 1. 1985 Tuesday, July 2, 1985 1:00 p.m. - 5:00 p.m. 8:30 a.m. - 4:00 p.m.

LOCATION: Visitors Center / Nuclear Operations Support Facility Comanche Peak Steam Electric Staticn, F. M. 201 Glen Rose, Texas PURPOSE: Audit the design review / quality revalidation (DR/QR) program for the Comanche Peak diesel generators manufactured by Trans-america Delaval, Inc. (TDI), a part of Phase 2 of the TDI Owners Group Prgram. The enclosed questions will be used as t an agenda. Open issues identified in SSER No. 6 Section 9.5.9 may be discussed.

PARTICIPANTS: NRC APPLICANT S. Burwell D. Woodlan C. Berlinger R. Haskovec D. Persinko C. Ray, et. al .

NRC Consultants _.

! D. Dingee H. Hardy P. Louzecky A. Wendell r

b WT.-

S. B. Burwell, Project Manager Licensing Branch No. 1 Division of Licensing -

Enclosure:

As stated cc: See next page d"$[lh9h LS(y k: N ,

3 .

i j ENCLOSURE I

1 QUESTIONS ftAISED IN.Cpst$ 08/QR A00!T A1 VIEW Cp102 - Generater Centre 1_s _

i in %e opinion of %e PWL reviewers, the analysis reported for the gener4ter

! centrols does not meet Se objective of %e 04/QR offert, primartly because j

the methodology followed was too limited in scope, It addrssses only four of

%e generators control components and leaves many out that are critical to system reliability. -

Desten/ Quality Cancerns .

. Sonera11y nest voltage regulater and esciter failure medes are considered

! crittcal to the enrall reliability of the diesel generator, especially these

. tavelving commen mode effects. Ceasents pertinent to the results end.

) -

eenclusions of the Comanche Peak OR/qt report aret e Diedes and SCRs - States %at everent ratings ats adequate . subject to l .

preter eeeling. Does met address voltage ratings. At least fear fattore

! incidents listed in W E00 CTS are unen>1ained as to probable failure i inedea. One failure related to set-of-puse parelleling, it may have been l caused by the sustafaed everveltage conditions prevalent in such events.

i The temperature meniterfng prov1siens ef Attechneet 4 wegId not preclude i the scentrence of e failure derfog en emergency operating conditien. It is

! recesmonded that teste be implemented to posittwly identify over- .

teeperature condittens and personent secrective acttens tenen as required, i This effort should be seesup11shed by TUGC0 as seen se practical. .

l e Field flashing relay. voltage specifications in Attachment 1 may not be  !

adequate. A nominal voltege rating equal te er lower than the lowest l

sapected battery voltage should be specifted to fasure relay pick-vp. A l Jrepping resister by-pessed by a norpelly closed relay contact sheeld be use to prevent cetI dens e when energized at higher volta This is standard practice and is needed to insure relay pick ge levels.up under few v conditions and to preclude rgley soil everheating under mentaum voltage - -

sendittens.  !

e Rather than reising only on a device such se Glyptel 14c eer to monitor feetener tightniss, positive settona should be taken te insure such tightness. Fosioner teckine devices'soch as the use of double avts or a L bending egent such as LOCTI1t til (Lactito Corporation, Newington. ,

Conneettsut 06111) are more posittve. -

e . Substitution of lockable ntionneters may be preferred to monitoring by  ;

-use of Olyptel leequer. -

J e Cabinet ventilation - forced ventilation een create new reliability  ;

j .

Problems by setting-in large quantities of Ifnt and dirt if the cabinet is [

i en %e swetten side of Se fan (s), .The presence of vent filters weeld not '

i help if e %er opentage such as enclosure joints..enused fastener holes, er openings erevnd table penetratfens bypeas W filters. These problems rewire such attention to detati and can be avoided by insta11tns the fans i

to Oey diseherte inte % e sabinet G roogh filtere er with filters en the I-

- - - - - , . - c - - -- _ -,, _ n

i suction esbtnet through filters or with filters on the suction side. Or perhaps forced venttiation could be avoided by relocating tempsrature-critica) components to the bottom of W enclosure where air temperatures ers 1ower. ~

o Pooer Supply Sypassing -- The recosseendations for power supply bypassing were made without an apparent knowledge of the energy levels of trans f ant overvoltages to be suppressed. The EDGC13 shows facidents involving lar

. voltage spikes, An investigation should be made to assess such energy ge

~

Tevels to insum adequate soniconductor protection. TUGC0 should give this  :

4 high priority for resolution.

! o Current Feedback i from available perfSi!nal nnance - The need for this recommendation is not apperant date. -

o The status reserks of the EDGCTS designated fifteen l'acidents ss random failures having no relevance to Cosenche peak reliability, N se random failures include four fuseholder failures and seven N1ay or relay contact failures representing discernible fetture patterns ehich should be

.- investigated. No effort appears to have been made to identify and document all probable fatture. modes on other Itsted incidents. Most of these *

. Incidents caused outa of the emer ncy generator, therefore their fapact

.should have been cons dered crittee to reliebt11ty.

F-068 - Intercooler The report did not confira Wt W heat rejection ca ncity of W cooler is edequate,for the application. In the case of the tur >ocharger this was done via test legs. A related concern tar whether the cooler v111 function adequately if it were 101 (nample) fouled. .

04 420 - Lube 011 Pressure Reautator.v Valve The QR inspection reports apparently were not available to the Owners' Groop Review Team. Does the conclusion regarding adequacy of the component still remain on*epen question? ,

04 621A - rue 1 011 orta Tank Can TUGC0 conf,tre that fuel return entry is not by means of an open funnel which also can ohit dirt? .

02-3105 - Crankshaft Beerinn She1Ts The minimum oil film thickness of *147 microtaches is ebout 10% below the espected -

ainimum value for W Tevel of ff1tration pertainino so CP5E3. .Noreover, if a ,

10 micron filter is used it will pass dirt Wt is twice the size of the oil ffle thickness.

Normally, bearing shell crusti (radfel interference) and side locetton, and details of W design of oil holes and grooves are considered in a design review because Wee are fundamental to strength.1mbefcation and cooling e.f the l bearing. Why was this nos done?

w~r- ~ e,m--,,~~vem--, , . ,. ,--,,-.,w,,,-n-n,--,-.-, --

i The QR makes no direct reference to hardware checks (inspections). Wes anything done in this regard?

l 02-310C - Crankshaft Thrust Seerine Rinot l

The DR did not address amiel vibretton as a source of thrust taring loading.

This een be e significant force. Please expisin.

The QR makes no reference % inspections. Were any done?

~

02-3358 - Front Geer' Gasket /Seltian The QR does not confirm W t bolts were toreved to 60 ft-lbs. Ws this conffrued7 l

02-341C - Piston Pin Assembly I The QR fndicates a comp 1ste inspection of ptston pins on EDG-01 with

! unsatisfactory results. In50G-02 view of' wee inspected on all right bank pins with -

the unsatisfactory findinos en 100-01 and -

settsfactory results.

similar unsettsfactory results on other 70! engines (e.g.. Alver Send)f Can -

7 TOGCO, explain why all pine were not inspected en (06-027 The app 1teetion of chroes pieting can reduce the fatigue strength of steels.

Was this considered in the DR7 OfMS - Exhaust knifold Boltind and Saskets There have been a number of instances of ffres caused by impingement,of off on hot engine parts, The enheust manifsid elbow is not insulated free the cylindei It is constdered

' head to a point inside of the exhaust sentfold shield.

important that TUSCO confire Wt lube all or fuel oil eennot sprey on the hot i In pertievlar all piping Joints shovid be checked to insure en wust elbows.

that spray could not impinge on hot surfaces if a Ieek developed. Possible use of shrouded joints should be considered.02-420 - Eneine Driven Lube 011 puun It is noted that the pump body and neraleDid arethe sede OR ofeenfirm cast iront presumably these ports sen the sounting fienges are also of cast iron.

withstand seismic load?

02-540A.8.C - Lube 011 $ & Tank - Strate.ar Aas 4 19 The OR did not address accessibility for maintenance to verify that the strafne I and the lube oil heater con be reenved end refaste17ed with high probability of J

maintaintag syntes elesnitness end. tightness. Can TU0C0 comment en this? ,02-689 - Off-Encine A1erm 5ensors - Wrinn The QR did not confirm that wiring (items 2. 3. and 4) aset IEEE 383 requirements.

i 02-7170 H.I - Auxiliary Subbene Gaskets Bottina, Fittinas_

i The QR did not include e walkdown to confirm the absence of low spots that could trap sludge or foretgn matter. Can TUGC0 confirm this? -

02-410A - Misc. teuipment - Aux. Jacket l(ater Pum_o Th[e OR makas no sention of the capacit oftheauxiliary(orenginedriven)

Jacket water pumps. Are they edequate 02-410C - Jacket hter East Exchenaar ..

Wre 'ts no indication that the heat exchanger performance will be adequete

! under W extremes of tesperature that might prevail et TU0CO,. The review does I not correlate the performance of W various components of the system such as the jacket water pump, strainers and thennostats. Are they all compatible?

02-410E . Jacket Water $tandoine Nester PhblemswiththecomponentwereencounteredatShorehen. Can TUGC0 sonfirm 4

that there is no similarity between the Shorehen and Cp5E3 instellations?

02-8208 - Aux 111ary tube 011 puse The OR does not state that the cepecity and pressure of'this pump is equal to that of the engine driven pump. TUGC0 should verify this.

. 02-8200.t - Lube 011 Weenwaru Strainer / Filter The DR did not vertfy that the strainer er ft1ter flow espeoity is equal te or

• greater than that of the prelvbe off pump (90 GPM). 100C0 should confirm this.

' Note also that a ploi, of filter de versus time would be e useful meintenance tooI. .

gg-820F 6 Fuel Flow tube 011 Filters _

TUCCO should verify that the filter capacity is sufficient for de engine detven -

j lube oil pump (600 SPM). Also, there is no indication es to whether a relief 1 valve and bypass around the filter is provided. If it is the velve shovid be i installed at the top of the filter to prevent the passage.ef dirt and sludge l Into the edgtne if the valve lifts. If the engine is required to start when the J

lobe oil heater 1s inoperative. it may be necessary to have a byteas in order to -

l i insure ht the engine will be lubricated adequately under this sold start condition. In any case, the pressure drop should be chectred and recorded at regular f atervals so es to entistpate the necesstty for a filter change.

Appendix C indicates a number of instances of excessive pressure drop across j

filters which indicates W t ffiter elements are not always changed when they should be.-

02-8200 - Lube 011 Heat Escheneer If the cooling water is too celd. sludge in the att will deposts eh the fins and reduce W best exehanger sepacity. W s this considered in the OR/Q47 l

l 3-820H-Lube 011FuelPressureSLPainer

, TUGC0 should confirm ht the capacity is greater then or equel to that of the

• engine driven tube oli pump (600 GPM)., .

02-825A - Fuel 011 Day Tank There is no indication in the OR/QR that there is adequate means of removing water from the fuel oil to issu N that W fuel injection pumps are protected.

A drein Ifne at the bottoe of the day tank wovid h useful in view of W intermittent operation of the enetne. However even if the enetne runs sentinuously for several hours at fuli food, t$ere will be insufficient ifme for' water to settle in the de tank; TUGco should verify W t there is means for

- stripping or filtering va%r from the fuel oil. '

M-839A - Startine Air Skid Base. Tank Relief Velve. Ffeat Trap and Tank

.700C0 should verify that there to adequate eccessiblitty to' W eir valves and fTest traps.for matntenance. In addition. it is noted that W starging etr relief valve ca mity is slightly greater than the compressor seposity. TUGC0 should verify tiet it is not possible to connect more then one compressor to each air tank.

6 D

d 6

- - . - . , _ , -- . - - -. , , - - - , - n,-

J

=. .'

ENCLOSURE 2 CPSES DR/OR AUDfT REVfEW Page 1 CP102 - Generator Controls NRC Comment: In the opinion of the PNL reviewers, the analysis reported for the generator controls does not meet the objective of the DR/OR effort, primarily because the methodology followed was too limited in scope, it addresses only four of the generator control components and leaves many out that are critical to system reliability.

CPSES Response: PNL indicates that only four components were reviewed. The review considered six components: (1) the diode bridge assembly and diverter.SCRs: (2) the voltage regulators: (3) the SCR firing circuits: (4) the pts, cts and linear reactors: (5) the field flashing circuits and (6) the min-max excitation circuit.

These components were selected by reviewing failure patterns reported in the EDGCTS. The assessment of the critical components is inconsistent with the scope of work and ,

charter assigned by the Owner's Group.

Deslan/Ouality Concerns _

NRC Comment: Generally most voltage regulator and exciter failure modes are considered critical to the overall reliability of the diesel generator, especially those involving common mode effects. Comments pertinent to the results and conclusions of the Comanche Peak DR/OR

- report are:

Diodes and SCRs - States that current ratings are adequate subject to proper cooling. Does

{ not address voltage ratings. At least four failure incidents listed in the EDGCTS are unexplained as to probable failure modes.

One failure related to out-of-phase paralleling, it may have been caused by the sustained overvoltage conditions prevalent in auch events. The temperature monitoring provisions of Attachment I would not preclude the occurence of a failure during an emergency operating condition. It is recommended that tests De implemented to positively identify over-temperature conditions and permanent corrective actions taken as required. This effort should be accom-plished by TUGCO as soon as practical.

l CPSES DR/OR AUDIT REVIEW Page 2 CPSES Response: The voltage ratings of the diodes and SCRs were verified and found to be adequate. The DR/OR report discussion of voltage rating was omitted for that reason. .

PNL suggests that sustained overvoltages due to out-of-phase paralleling is a cause of diode failure. Out-of-phase paralleling is a result of an operator error or failure in other equipment. Protection from such events is provided by transient. suppressor CR-8 which is rated at 950 volts. The isolated incident mentioned by the consultant did not involve equiment manufacturered by TDl/PORTEC.

Thermal analysis showed that the diode mounting represented a potential diode over-temperature problem if the diode case to heat sink thermal resistance increased due to loosening of the diode. The loosening problem is one that is expected to develop gradually over a period of time, rather than on any one start. Thus detection of the onset of a problem would be possible with the method suggested. The concern about the mounting method is further supported by the diode manufacturer's installation recommendations.

NRC Comment: Field flashing relay voltage specificatiors in Attachment ? may not be adequate. A nominal voltage rating equal to or lower than the lowest expected battery voltage should be specified to insure relay pick-up. A dropping resistor by-passed by,a normally closed relay contact should be used to prevent coil damage when energized at higher voltage levels. This is standard practice and is needed to insure relay pick-up under low voltage conditions and to preclude relay coil overheating under maximum voltage conditions.

CPSES Response: Apparently the consultant has misinterpreted the relay specification as stated in the DR/OR report. The relay is to be selected so that. the coil operates over the full supply range of 90 through 140V. Discussions with the manufacturer of the existing relay suggest that such a relay is available. The consultant's solution would require substantial requalification of the design.

CPSES OR/OR AUDIT REVIEW Page 3 NRC Comment: Rather than relying only on a device such as Glyptol lacquer to monitor fastener tightness, positive actions should be taken to insure such tightness. Fastener locking devices such as the use of doube nuts or a bonding agent such as LOCTITE 271 (Loctite Corporation. Newington, Connecticut 06111) are more positive.

CPSES Response: The intent of the recommendation in the DR/OR report is to use Glyptol as an agent for monitoring the tightness of bolted electrical connections, Use of a bonding agent which may flow on mating conducting surfaces would adversely affect the electrical performance of the connections, and no visual indication of loosening would be present.

NRC. Comment: Substitution of lockable potentiometers may be preferred to monitoring by use of Glyptol lacquer.

CPSES Response: The Glyptol Lacquer is recommended in Attachment 1 of the DR/OR report for imme-diste use. The consultant's recommendation would require requalification of the PC board because of the mechanical design changes required to accommodate the lockable poten-tiometers. Further, the use of Glyptol would atill be recommended to provide positive indication of any adjustment change.

Long term recommendations in Attachment 3 of the DR/OR report suggest Mit style, sealed, multiturn potentiometers. The intent of the recommendation is to improve the long term stability of the adjustments. Glyptol monitoring is recommended with these l

potentiometers as well i

l NRC Comment: Cabinet ventilation -

forced ventilation can create new reliability problems by sucking-in large quantitles of lint and dirt if the cabinet is on the suction side of the fan (s). The presence of vent filters would not help if other openings such as enclosure joints, unused fasteners holes, or cpenings around cable penetrations bypass the filters.

These problems require much attention to detail and can be avoided by installing the fans so they discharge into the cabinet i

through filters or with filters on the l suction side. Or perhaps forced ventilation l could be avoided by relocating temperature-critical components t o* t h e bottom of the L -

CPSES DR/OR AUDIT REVIEW Page 4 _ __ _

enclosure where air temperatures are lower.

CPSES Response: Positive pressure ventilation of the generator controls is recommended by PNL.

The need for adequate cooling is the focus of the DR/QR recommendation for ventilation, rather than the details of the implementation. However, since positive pressure ventilation may be simpler to implement while still assuring cabinet cleanliness, it may be preferred.

NRC Comment:

~

Power Supply Bypassing - The recommendations for power supply bypassing were made without an apparent knowledge of the energy levels of transient overvoltages to be suppressed. The EDGCTS shows incidents involving large voltage spikes. An investigation should be made to assess such energy levels to insure adequate semiconductor protection. TUGCO should give this a high priority for resolution.

CPSES Response: The purpose of the power supply bypassing is to prevent AC noise voltage generated outside '

the voltage regulator circuit from inter-fering with its operation and to prevent voltages generated by the circuit itself from propagating in the circuit. There are no recorded cases of transient voltages within the generator control equipment, since no instrumentation is provided for monitoring short-term transients. This was presented as a long term recommendation to be implemented at the discretion of TUGCO NRC Comment: Current Feedback Signal - The need for this recommendation is not apparent from available performance data. -

CPSES Response: It is well known that voltage and current sensing circuits should utilize 3-phase or 6-phase rectification so that the smoothing filter may have a small time constant rela-tive to that of the rest of the system. By so doing, the adjustment of the gain and damping is less critical to the desired level of stability. This was presented as a long l term recommendation to be implemented at the discretion of TUGCO.

l NRC Comment: The status remarks of the EDGCTS designated i

fifteen incidents as random failures having no televance to Comanche Peak reliability.

i

--,.__..m. ...._,_,,.,_,,,y,,-,,,.,,,,,y. m,,,,m._,_._,m_____..,.

=

M CPSES DR/OR AUDIT REVIEW Page 5 __

These random failures include four fuseholder failures and seven relay or relay contact failures representing discernible failure patterns which should be investigated. No effort appears to have been made to identify and document all probable failure modes on other listed incidents. Most of these inci-dents caused outage of the emergency genera-tot, therefore their impact should have been considered critical to reliability.

CPSES Response: The four fuseholder failures were not in TDl/PORTEC equipment. Three failures were due to loose fuse clips or holders, but because they did not involve TDl/PORTEC equipment they were not studied further. The fourth fuse failure was due to a blown fuse in the measuring PT circuit, which was an isolated occurance.

Seven relay failures have been mentioned by the consultant. None of these were in TDl/PORTEC equipment. Two failures were in the K-1 relay of the Basler voltage regulators at Zion Units 1 and 2. These two failures have no relevance to the TDl/PORTEC equipment because the K-1 relay in the Basler ,

voltage regulatot is used in an entirely '

different way than the K-1 relay in the TDl/PORTEC equipment.

One failure was due to sticky and damaged auxiliary contacts Trojan Unit 1. This was considered a random event because no similar failure was identified in TDl/PORTEC equipment.

Three failures (two at Arkansas NUC Unit 1 a r.d one at Brunswick Unit 2) were related to the malfunction of protective relays which were not considered part of the generator controls for the review.

The CTS had insufficient information about the relay failure at Kuwanee and therefore was not analyzed.

F-068 - intercooler NRC Comment: The report did not confirm that the heat rejection capacity of the cooler is adequate for the application. In the case of the turbocharger this was done via test logs. A related concern is whether the cooler will

CPSES DR/OR AUDIT REVIEW Page 6 function adequately if it were 10% (example) fouled.

CPSES Response: The TDI Owners Group found the intercooler to be adequately sized based on:

- successful TDI tests

- successful site check-out successful present operation the intercooler cooling water is treated engine jacket water.

• Furthermore, there is a lack of identified problem experience with shell side fouling, and CPSES has agreed to inspect the tube side after every refueling outage.00-420 - Lube Oil Pressure Reaulatina Valve

~

NRC Comment: The OR inspection reports apparently were not aval'able to the Owner's Group Review Team.

Does the conclusion regarding adequacy of the component atill remain on open question?

,CPSES Response: The adequacy of the component is not an open question. The TDI Owners Group found the component adequate based on:

- No reported incidents of TDI related failure due to valve design.

The valve is identical to the SNPS valve which was found adequate.

- Successful continuous operation.

CPSES has agreed to periodically disassemble and clean the valve.

00-621A -

Fuel 01I Drlo Tank _ . , _ _ _ _ _ _ _ _ . . _ _ _ _

NRC Comment: Can TUGCO confirm that fuel return entty is not by means of an open funnel which also can admit dirt?

CPSES Response: The return to the Fuel Oil Drip Tank is by hard pipe with no open funnel.

02-3108 - Crankshaft Bearina Shells NRC Comment: The minimum oil film thickness of 147 micro-inches is about 10% below the expected mini-

CpSES DR/OR AUDIT REVIEW Page 7 , _ _ . __

mum value for the level of filtration per-taining to CPSES. Moreover, if a 10 micron filter is used it will pass dirt that is twice the size of the oil film thickness.

Normally, bearing shell crush (radjal inter-forence) and side location, and details of the design of oil holes and grooves are con-sidered in a design review because these are fundamental to strength, lubrication and cooling of the bearing. Why was this not done?

The OR makes no direct reference to hardware checks (inspections). Was anything done in this regard?

CFSES Response: The successful operating history of the main bearing shells, indicated by the absence of any designed-related deficiencies in the component tracking system, demonstrates the acceptability of this component. In reviewing the specific geometric design features of the main bearing shells, the effect of crush on radial clearance, side location, 360 degree grooving and oil supply were considered (Reference 1 of the OR/OR report) and found to conform with accepted practice.

The analysis of crankshaft main bearing shells also considered the effects on bearing performance of particles which pass the 10 micron oil filter, and which are larger than the minimum oil film thickness (Reference 1 of the DR/OR report). The babbitt layer has sufficient thickness and embedability to allow such particles to pass between the journal and the bearing or to be rendered harmless by embedmont in the babbitt layer.

The recommended inspections at alternate fuel outages will disclose any harmful wear that could adversely affect bearing performance and reliability.

The OR report and TER 10-029 together with NRC 84-0080 were reviewed to evaluate the results of hardware inspections which were performed. The unsatisfactory results re-ported were due to scratches disclosed in the visual inspe. tion. These scratches were attributed to foreign particles in the engine w**ch passed through the bearings before they were trapped in the lube oil filter. The nature and severity of the scratches were e

s' n' CpSES DR/OR AUDIT REVIEW Page 8 evaluated and the bearings were judged acceptable for continued use.

02-310C - Crankshaft Thrust Bearino Rinas NRC Comment: The OR did not address arial vibration as a source of thrust bearing loading. This can be a significant force. Please explain.

The OR makes no reference to inspections.

Were any done?

CPSES Response: Axial vibration of the c r,a n k s,h a f t is driven

• primarily by variations with time in the axial component of the gear force. The OR considered the peak axial gear force, and thus, in a conservative way, accounted for the peak thrust bearing ring load. In the absence of resonant axial vibration of the crankshaft, the peak t h r u s,t bearing ring load will be less than or equal to the sum of the peak axial gear force and the other axial loads. There is no evidence of resonant vibration of the crankshaft on TDI diesel engines, and the EDG component tracking system had not reported any problems with

, thrust bearing rings on TDI engines.

OR inspections of the thrust rings were not required and none were performed.

02-3358 - Front Gear Gasket /Bottino NRC Comment: The OR does not confirm that bolts were tor-qued to 60 ft-lbs. Was this confirmed? .

CPSES Response: All external bolts on the front gear case of DG-01 and 00-02 were verified to be torqued to the proper valge per MAR 85-0095 and 85-0097. No loose bolts were found. Internal bolts, which require equipment disassembly, will be inspected at the next scheduled main-tenance period per OGTP-672-0-404 02-341C - oiston Pin Assambly NRC Comment: The OR indicates a compteto inspection of piston pins of EDG-01 with unsatisfactory results. EDG-02 was inspected on all right bank pins with satisfactory results. In view of the unsatisfactory findings on EDG-01 and similar unsatisfactory results on other TDI

CPSES DR/OR AUDIT REVIEW Page 9 enginees (e.g.. River Bend)? Can TUGC0 ex-plain why all pins were not inspected on EDG-027 The application of chrome plating can reduce the fatigue strength of steels. Was this considered in the DR?

CPSES Responses wrist Elg insceetion. Inspection results in _

TER 10-080 noted in the DR/OR Report for Comoonent 02-341-C shows that visual inspec-tion was performed on all wrist pins in DG CPI-MEDEGEE-02. Results were satisfactory except for pin 4L which "shows some slight evidence of scoring. Scoring does not pene-trate the chrome plating". The Maintenance Engineering Evaluation MEE #84-037 determined that pin 4L was acceptable for use because the damage noted did not penetrate the chrome plating.

The information in TER 10-078 for component 02-3408, which includes inspection results for connecting rod bushings ,from ,EDG,-01 and EDG-02, was received. The results showed that the adverse conditions.found on pins from EDG-01 did not cause detectable damage on the bushings. The results were also satisfactory for all bushings from EDG-02.

Fatioue Strenath -

Effect of_ Chrome _ Platina.

It is known that the cracks normally present in electroplated hard chromium can act as stress raisers and thus reduce the fatigue strength of the underlying steel. A compres-sive residual stress in the surface is often used to offset this effect. Such surface compressive stress can be produced by shot peening or appropriate heat treatment (1).

The piston pins are case carburized and hardened, which produces a residual compres-sive stress in the case hardened surface and near sub-surface material. The beneficial effect of the case hardened surface was con-sidered in the analysis of piston pin

fatigue strength. No adjustment was made for the effect of chrome plating. The very con-servative fatigue analysis indicates a suffi-cient margin to tolerate a reduction of fatigue strength by as much as 40% because of the hard chromium plating.

The component tracking system disclosed no evidence of fatigue damage or failures for piston pins in nuclear or non-nuclear

CPSES DR/OR AUDIT REVIEW Page 10 service. The successful operating history of the piston pins provides confirmation of ,

component acceptability.

i f

02-3808 - Exhaust Manifold Boltina and Gaskets NRC Comment: There have been a number of instances of fires caused by impingement of oil on hot engine parts. The exhaust. manifold elbow is not insulated from the cylinder head to a

. point inside of the exhaust mainfold shield.

~

It is considered important that TUGCO confirm that lube oil or fuel oil cannot spray on the hot exhaust elbows, in particular all p6 ping

, joints should be checked to insure that spray l could not impinge on hot surfaces if a leak developed. Possible use of shrouded joints should be considered.

CPSES Response: The 701 Owners Group does not believe that fire caused by impingement of oil on hot engine parts is a concern at CPSES. The i

basis for this belief is given below:

- There is a drip collection system In-stalled on the diesel.

- The fuel oil lines have been seismically qualified.

1

- The high pressure fuel oil lines are shrouded and are the only oil lines in the area of the unshrouded exhaust manifold elbow.

- Shrouding the exhaust manifold elbow is physically impossible because of ther-mal growth and physical constraints.

- Engine configuration prevents oil build-up in the area (as described in 1

the DR/OR).

,02-420 - Enoine Driven tube Oif pume NRC Comment: It is noted that the pump body and nozzle are made of cast irons presumably the mounting flanges are also of cast iron. Did the DR j confirm these parts can withstand seismic Ioad?

CPSES Response The mounting flanges for this pump have been seismically quellfled, and as delineated in

j CPGES DR/OR AUDIT REVIEW Page 11 the DR/OR, its qualification may be found in Stone & Webster Engineering Corp. calculation No. 11600.60NM(B)-001-C2C-040.

02-540A. B. C Lube Oil Sumo Tank-Strainer Assembly NRC Comment: The DR did not address accessibility for maintenance to verify that the strainer and the lube oil heater can be removed and reinstalled with high probability of

. maintaining system cleanliness and tightness.

Can TUGCO comment on this?

CPSES Response: Using the appropriate procedural precautions to ensure cleanliness control. TUGCO has demonstrated the ability to enter,the lube oil sump without introducing contaminates to the system. The sump was entered several times during initial system flush and once when replacing a foot valve with no adverse effects.02-689 -

Off-Encine Alarm Sensors - Wirina NRC Comment: The OR did not confirm that wiring (items 2, 3, and 43 meet IEEE 383 requirements.

CPSES Responses it was not the intent of the DR/OR to deter-mine if the m. ring complied with IEEE-383.

The intent was to dotarmine if the wiring system was adequate for its application.

Item 2 (of the DR/OR) was found acceptable based on the following:

The wire was run in conduit. The con-dult will make it difficult for a fire to be initiated, and it will limit and l contain combustion of the insulation should it occur, and prevent fire from spreading to other components.

- This wire (a thermocouple extension wire) is not safety related. The en-gine will continue to operated satisfactorily during a LOOP /LOCA with-out these wires being intact.

CPSES DR/OR AUDfT REVIEW Page 12 _ _,

items 3 and 4 Cof the DR/OR) were found .

. acceptable based on the following:

- The insulation of this wire is composed primarily of non-organic materials which should not support combustion.

- These wires (thermocouple wires) are not safety related. Engine will operate sat 6sfactorily during a LOOP /LOCA without these wires being intact.

02-717G. H. t - Auxillary Subhase Gaskets. Botfina. Fittinan NRC Comment: The OR did not include a walkdown to confirm the absence of low spots that could trap sludge or foreign matter. Can TUGCO confirm -

this?

CPSES Responses it is stated correctly that the OR did not perform a walkdown to confirm the absence of

" low spots" that could possibly trap sludge or foreign matter. However. " low spots" are not a problem because the diesels are equipped with sufficient filters and stralners (all of which have been requall-fled) to preclude engine damage or the entrapment of sludge or foreign matter into any possible " low spots." Therefore, a walkdown was not required nor performed.

In addition the lube oil is sampled monthly to check for chemical and particulate contamination.

._ _ 02-810A-Mine. Eauinment -

Aur. Jacketwater Puma _ ._ _______ .,

NRC Comment: The DR makes no mention of the capacity of the auxillary (or engine driven) Jacket water pumps. Are they adequate?

CPSES Response The auxillary Jacket water pump capacity is adequate for its intended design function.

Its adequacy is based on the following

- The pumps have operated successfully at CPSES and have been circulating adequate quantitles of water to achieve the desired results throughout the diesels' operational history.

- There have not been reported incidents related to TDI diesels, regarding the

cpSES OR/OR AUDIT REVIEW Page 13 leadequacy of the auxiliary Jacket water cooling pump capacity.

02-810c - Jacket Water Heat Exchancer ,

NRC Comment: There is no indication that the heat exchanger performance will be adequate under the extremes of temperature that might pre-

~

vail at TUGCO. The review does not correlate the performance of the various components of the system such as the jacket water pump, strainers and thermostats. Are they all compatible?

CPSES Response: The jacket water heat exchanger performance is adequate for the temperature condition,s that exist at CPSES. This assessment is made on the basis of the following:

- The jacket water heat exchanger found at CPSES is the same manufacturer as that for SNPS. (The heat exchanger is a non-TDI fabricated component). The ,

manufacturer's heat exchange design at SNPS was given an exhaustive review, and revealed nothing but acceptable design practices. Therefore, if a manufacturer can properly design a heat exchanger for one site (SNPS), it is sound reasoning that the manufacturer can properly design a heat exchanger for another site (CPSES).

- The jacket water heat exchanger has

, not been an operational concern at l SNPS or CPSES. Also, the only i -

concerns that could occur would be poor water chemistry which cannot be used to discredit the jacket water j

heat exchanger design.

- The jacket water heat exchanger performance has been acceptable through its operating history.

! 02-810E - Jacket. Water Standoine Heater NRC Comments Problems with the component were encountered at Shoreham. Can TUGC0 confirm that there is i

l

CpSES OR/OR AUDIT REVIEW Page 14 _ _

no similarity between the Shoreham and CPSES installations?

CPSES Response: The CPSES jacketwater standpipe heater is unlike the one found at SNPS. The CPSES unit has one immersio.n heater and has proper circu-lation to prevent stagnant water around the element.

02-820B - Auxiliarv Lube Oii Pumo _ _ _ . _ . _ __ ___ . _ _ . ,

NRC Comment: The DR does not s t'a t e that the capacity and pressure of this pump is equal to that of the engine driven pump. TUGCO should verify this.

CPSES Responses it was not the intent of the OR/OR to deter-mine if the auxiliary tube oil pump was equal in capacity and pressure to the engine driven lube oil pump. Rather, it was to assess the adequacy of the auxillary pump for its intended design function.

.The acceptability of the auxiliary lube oi!

. pump 91s established based on the pumps proven ability to satisfactorily maintain the re-quired system pressure. Furthermore, the auxillary tube oil pump maintains a lube oil system-pressure that exceeds that of the engine driven tube oil pump.

Finally, many of the TDI diesels operating in nuclear stand-by service do not have these pumps. As stated, the pump is an

" auxiliary", which is not required, and is not " standard equipment" on the engine.

02-820D E -

Lube _OfI Meenwarm Streiner/FIIter . . _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ __,

NRC Comment: The OR did not verify that the strainer or

, filter flow capacity is equal to or greater than that of the prelube oil pump (go GPM).

TUGC0 should confirm this.

Note also that a plot of filter dp versus time would be a useful maintenance tool.

CPSES Response A review of the 701 Manual, Volume lil, con-firms the flow capacity of the strainer /fil-ter as being equal to the prelube oil pump (90 GPM).

e ,.

CPSEE DR/OR AUDIT REVIEW Page 15 02-820F - Full Flow Lube Oil Filters NRC Comment: TUGCO should verify that the filter capacity is sufficient for the engine driven tube oil pump (600 GPM). Also, there is no indication as to whether a relief valve and bypass around the filter is provided. If it is, the valve should be installed at the top of the filter to prevent the passage of dirt and sludge into the engine if the valve lifts, if the engine is required to start when the tube oil heater is inoperative, it may be necessary to have a bypass in order to insure that the engine will be lubricated adequately under this cold start condition. in any case, the pressure drop should be checked and recorded at regular intervals so as to anti-cipate the necessity for a filter change.

Appendix C indicates a number of instances of excessive pressure drop across filters which Indicates that filter elements are not always changed when they should be.

CPSES Response: Verification of equipment capacity is not required. The engine driven positive displacement lube oil pump, maintains system pressure at 50 psi by incorporating a pres-sure regulating valve and recirculation line back to the pump. The performance curve of a positive displacement curve is relatively flat and will maintain the required system pressure over a range of flow rates.

It should be noted that lube oil system pres-sure control feedback is from a point downstream of the subject equipment. This being the case, plus successful system / engine operation, the pump and downstream equipment of the system, correlate the performance of the system component as being compatible and adequate.

There is no relief valve and bypass around the subject equipment. The concerns raised  !

regarding this component are not related to ,

, the attributes to be verified. l The pressure boundary, nozzle loads, pressure drop, and particle retention size were the [

attributes to be verified and all were found acceptable.

I

e CPRES_04/0p AUDIT REVIEW Page 16 02-82C3 - tube 06 Heat Exchancar NRC Comme.nts if the cooling water is too cold, sludge in the oil will depos6t on the fins and reduce the heat exchanger capacity. Was the: con

• sidered in the OR/OR?

CPSES Response: Excessively cold cooling water causing oil sludge to deposit on tube fins was not a n'o should not be considered in the CR/OR based on the folicaing:

- During engine operation, warm Jacket

, water is used as cooling water. '

During engine standby, jacket water la

. heated and assists in maintaining proper Ivbe oil temperature.

- The diesels are equipped with suffi-cient filtars end stra,iners to preclude oil sludge.

02-820H -

Lute _Olf Full Pressuet Strainer NPC Comment: TUGCO should confirm that the capactty is greater then or equel to that of the engine drlwen tube col pu.nr (600 GPM).

CPGCS Aesponser Reter to 02 920f'-

02 128A.;_fuef Oli Day Tank NRC Comment: There in no &ndlestion in the OR/OR that there is adequate means of removing water from the fuel all to insure that the fuel injectica pumps are protected. A drain line at the 30ttom of the day tank would be useful in view of the intermittent operation of the engine. However, even if the engine runs continuously for several hours ei full load, there will be i n e u f f i c iis n t time for water to settle in the day tank. TUGCO shoule verify that there is means for stripping or filtering water from the fuel oil.

CPSES Responsen At least once per g2 days, the fuel oil storenge tanks ere checked for any accumulated water. At least once per 31 days and after each operation equal to or greater than 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />, the day tank is also checked for accumulated water. This is done per procedure OPT-214A by tahAng a semple at the fuel oil pump

CPSES Onton a00!T REVIEW Page 17 suction strainer This prevenis any water from entering the Fuet Oil System a n d , 1,f any water is found it is removed.

02-935A -

S _t a r t i n a Air Skid Bese. Tank Aelief vgfve. Float Tran ned Tsnk NRC Cowie n t : TUGCO should verify tnat there is adequate accessibility to the air valvos and fJost traps for maintenance, in addition, it is noted that the starging air rel6ef valve capacity is sfightly greater then the compressor capecity. TUGC0 should verify that it is not possible to connect more than one compressor to each avr tank.

CPSES Response: Accessibility to the air valves and float traps is adequate for the required maintenance, it is not poseible to connect more than one compressor to an air tank.

4 e

=

7 a.~7 l,  :" -

E , i:3 ENCLOSURE 3 Log # TXX-4501 TEXAS UTILITIES GENERATING COMPANY File i 10010 setYwAY TOWSs .400 WORT,8 OLIYE STREST L.B.et

• DALLAS,TEXA513801 June 27, 1985 "A"." ". 1*S" Director of Nuclear Reactor Regulation Attention: Mr. Vince S. Noonan, Director Comanche Peak Project Division of Licensing U. S. Nuclear Regulatory Commission Washington, D.C. 20555

SUBJECT:

COMANCHE PEAX STEAM ELECTRIC STATION (CPSES)

DOCKET NO. 50-445 DESIGN REVIEW / QUALITY REVALIDATION (DR/QR)

REPORT ON THE CPSES TDI DIESEL GENERATORS REF: 1) J. B. George letter TXX-4377 to H. R. Denton dated December 17, 1984

2) J. W. Beck letter TXX-4430 to H. R. Denton dated March 1, 1985

Dear Mr. Noonan:

Reference 1 provided the NRC with the CPSES evaluation results of the TOI Diesel Generator Owners' Group DR/QR report. Enclosure 2 to the letter specifically provided the CPSES position on the OR/QR report maintenance and surveillance activities. Based on Revision 1 to the OR/QR report (Reference 2) and on plant specific implementation of the report recommendations at CPSES, attacned is an update to Enclosure 2.

Sincerely, Q.1%L lJo'hnW. Beck RWH/gre Attachment

~

A Ott'inlHN UP ttJrAn L*TRLittEn ELACTMit' COMPANT

\

e; -, .. . ,

Attachment to TXX-4501 dated June 27, 1985 Enclosure 2 Results of Texas Utilities' Evaluation of the CPSES OR/QR Recommendations Regarcing Maintenance and Surveillance Activities Texas Utilities concurs with all Owners' Group recommended maintenance and surveillance activities as stated in tne revised CPSES OR/QR component reports and revised maintenance matrix, with the following proposed modifications and/or clarifications:

1) Component F-068, Intercoolers Item 2 in the revised maintenance matrix recomends cleaning and inspection of intercooler shell and tube sides after every refueling outage. Texas Utilities concurs with the inspection interval for the tube side, however, it is proposed that the shell side be comparably inspected at a five (5) year interval, based on a lack of identified problem experience for the shell sloe.

2) Component PP-022/23 Turbochargers Item 2 in the revised maintenance matrix recomends cleaning of the turbocharger impeller and diffuser at every refueling outage.

Texas Utilities proposes as a more practical maintenance approacn

, th'at the impellers and diffusers be inspected every refueling outage and cleaned if necessary.

A supplementary Phase I report on turbocharger nozzle ring assemalies has also been issued by the Owners' Group (Reference 7) which contains the following additional Utility recommendations:

a. At any turbocharger disassemoly tnere should be a visusi inspection of nozzle ring components for any apparent damage, failure or apparent mispositioning of vanes.

Replace all affected nozzle ring components. During reassemoly ensure that capscrews are properly installed with recommended pretorque,

b. Monitor engine operation to ensure exhaust gas temperatures do not exceed those specified, Texas Utilities will comply with the utility reconnendation a) above during all turbocharger disassemblies and will perform pre-turbine exhaust temperature nonttoring on a monthly b3 sis.

3) Component 00-420, Luce Oil pressure Regulating V_alve Item 1 in the revised maintenance matrix recem.nends valve disassem31y and cleaning at every outage. Texas Utilities concurs, but proposes that the interval be chJnged to alternate refueling outages based on satisfactory industry experier.ce with this valve.

1

.. - , , ~.

~

4) Component 02-307B, Lube Oil Fittings, Internal Texas Utilities concurs with the recommendation in Item 1 to eneck tubing for dents or crimps at every refueling outage, however, this will be performed on accessible tubing only.

5) Component 02-310A, Crankshaft Item 2 in the revised maintenance matrix recoecends that all crank journal diameters be measured at alternate outages. Te.r as Utilities proposes that a sample of main journals be measured at alternate refueling outages in order to be compatiole with the established sampling frequency for the main bearing shells (see component 02-3108).

6) Component 02-3108, Main Bearing Shells Item 1 in the revised maintenance matrix recomends visual and dimensional inspection of all main bearing shells for evidecce of wear or misalignment at the first refueling outage and at alternate outages thereafter. Texas Ottlities believes that the minimal problems encountered with the CPSES Unit 1 OG rain oearing shells do not warrant inspection to the degree recomended by the Owners' Group. Tnis position is in agreement with the assatssment by NRC and PNL of C.PSES main bearing shell experience, as stated -

on pages 5.12 and 5.13 of FNL-5234 (enclosure to Reference 5), As recommended by hRC and PNL, Texas Utilities will perform a sanple inspection of two (2) nignly loaded bearings per engine (nearings 5 and 6) at alternate refuelin) outages. Associated caps and saddles will also be checked. The need for additional inspection will de datermined by the results of tne initial sample inspection. Due to accessibifity restrictions, irspection of bearings 1, 9, and.10 are to be performed only during an engine teardown (every 5 years).

7) Ccmponent 02-31CC, Thrust Bearing cling item 2 in the revised maintenance matri.x reconcends a visual inspection of the thrust bearing for signs of wear or degradstion concurrently with visual and dimensional train bearing snell inspections as recomended above. However, the main bearing stell inspection frequency has been modified as stated sonve. Texas Utilities proposes that the recoemended visual intpectfon would be adequate if performed at a five (5) year interval (i.e., at overnauls) since a " bump check" for thrust b&aring clearance will be performed at every refueling outage.

8) Component 02-3154, Cylinder Bloces Item 1 in the revised maintenance matrix recomends a visual

. inspection of the cylinder b1cck and v?ddy. current inspection for stud-to-stud cracks beteeen cylinder heads and for cracks between the block edge and studs at the blocr einds. The rec ynnended frequency for the inspection is prior to returning the engine to standby service af ter any perico of operation above 50% load.

2-

l Texas Utilities agrees that surveillance of the block top surfaces to ensure the absence of stud-to-stud cracks and stud-to-eoga cracks is appropriate for the first cycle of operation at CPSES Unit 1, until such ti:se as the indications found on Train A, '

cylinders 4R and SR and on Train B, cylinder: 1R anc 4R are reinspected .st tne first refueling outage. Future inscection L requirenents should be based en the results of the above inspections and on any additional information provided ty the Owners' Gr.oup. This position is in agreement with PNL's position in PL-5234 Texas Utilities proposes at this time, however, that a visual and boroscopic examination be substituted for tne eddy-current technique at the same frequency of performance. Texas Utilities believes that the use of visual /ocroscopic examination will result in greater engine availability and is justifiable based on the stated conservatism in the Cwners' Grcap assessment of the CPSES bicek indicaticas as aligament cracks" and on the absence cf Widmanstaetten graphite in the CPSES cylinder blocks.

The indications in the Train A right bank cylinder clock are the largest found at CPSES, but tne Owners' Grcup has cor.cluded that .

they were castirig induced not service indu:ed. The smaller indications in the Train 6 right aank cylinder block were also determined to be casting defects (by TUSCG) and are less then 0.050 inches in depth. All of these inoications are in lower stress regions of the block top than lig went craco and are all smaller than ligitent cracks. ,

'diomanstaetten graohite is a . degenerate reicrostructure tnat ,

reduces the strength and fatigue resistance of 9'e/ cast iron.

The Owners' Group nas determined that there is no Wiccanstaatten ,

graphite present in the CPSES Unit I cylinder 01ceks.

i The Oer,ers' Group has stated in the latest Phase i recort (refererce 5) that blocks whicn have cc are ess eea to have ligament cracxs and which have no kidranstaetten grapqite are capable of withstanding a LOOP /LCCA event with sufficiert . margin, provided no stud-to-stud or stud-to-edge cracks exist.

Furtnernore, engines with such cracks can still oe returred to l standoy service provided these cracks are less than 11 inctes in '

depth from the block top. Because initiation of stud to-stud or stud-to-end cracks occurs at the block tcp surface ard because they pro; agate downward, visual surface inspection of the block '

top will ne adequate to assure the absence of sucn cracks.

l As required in Reference 5. Texas Utilities aill also perfor.n routine daily visual inspections of the block and external surfaces d'. ring operating periods, with a more thorough monthly irssectioS under streng ifghtirg (also with the engir.e caerating).

As stated asode, cloc< indications are to be reintpected for propagation at tne first refueling outage. l To ensare ecgine availability after surveillance testtrg aoove 5:n load, Texas Lt:11 ties will maintain the engine in stancoy service and perform the required surf ace inspection of tne t,1ccr top witnin 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.

9 *.

9) Component 02-340A, Connecting Roos CPSES is in agreement with the revised DR/QR report.

10) Component 02-3408, Connecting Rod Bearing Shells Item 1 in the revised maintenance matrix recommends that visual and dimensional inspection of all bearing shells be performed at the cutsge which precedes 500 hours0.00579 days <br />0.139 hours <br />8.267196e-4 weeks <br />1.9025e-4 months <br /> of operation Dy at least the sum of hours of operation in a LOOP /LOCA plus the expected hoars of operation between outages, .

In place of the above scheme, Texas Utilities will perform a visual and liquid penetrant examination of connecting rod bearing shells for 2 sets of pistons per engine, during the first refueling outage. Lube oil analysis for contamination will be perfcened on a monthly basis, and bearing cicarance will be measured by " bur.p check" at every refueling outage.

11) Component 02-341A, Pistons .

CPSES is in agreement with the revised OR/QR report.

12) Component 02-350A, Cam Shaft Assemoly CPSES is in agreement with the revised OR/0R report.

13) Component 02-380A, Exhaust Manifold Item 1 in the revised maintenance matrix recommends magnetic particle examination of a sample of circumferential pipe welds and corresponding heat affected zones at the first refueling outage and at alternate refueling outages thereafter.

Texas Utilities proposes to visually examine all accessible welds at the frequency given above, instead of a sam 6Te of magnetic particle tests. It is felt that this procedure would provide a more practical approach with broader weld inspection coverage.

14) Component 02-387A, Crankcase Vacuum Fan _. _ __

Item 1 in the revised maintenance matrix recottends cleaning and inspection of the fan at alternate outages. Texas Utilities proposes, as a more practical approach, that the fan be inspected at alternate refueling outages with cleaning as necessary.

15) Component 02-390E, Rocker Arm Bushings Item I in the revised maintenance matrix recommends visual and dimensional inspection of intake rocker arm bushings at the outage which precedes 2300 hours0.0266 days <br />0.639 hours <br />0.0038 weeks <br />8.7515e-4 months <br /> of operation by at least the sum of expecttd hours of operation in a LOOP /LOCA plus the expected hours of operation netween outages. However, NRC has concurred in Reference 5 with Texas Utilities earlier proposal to visually inspect rocker arms at each refueling outage.

, m ,. . y. .., ._ ..

~

. . , ss ,

Based on the above, Texas Utilities will perform a visual inspection of the intake rocker arms at every refueling outage, and will measure them every 5 years during overhauls. +

16) Component 02-4104, Overspeed Trip Governor

CPSES is in agreement with the reviseo CR/0R report.

17) Component 02-410C, Overspeed Trip Drise Couplin g item 1 in the maintenance matrix recomends that the present L-110 Lovejoy ccuplings be replaced with new units at the next outage, per TDI :.ervic.e Information Memo (SIM) 3.53. Texas Utilities does not intend to replace these coupilngs at the first refueling outage because they were already replaced per SIM 363 instructions.

19) Compor.er,t 02-4134,_ Governor Linkage l

Texas Utilities will install lockwire on all fasteners designed for use with locKwire. Other locking hardware will be installed where specified. -

19) Comocnent 02-415A, Woodsa d Governor _

.t Recom1ertded settings will oe reviewed to ensure compatibility with the Teras Utilitics grid systent requirements, and may be ad, lusted accordingly.

20) Component 02 4418, Air Filter to Starting Air Distributor ,

item C in the revised maintenance matrix re:cmends tnat air start strainers ce cleaned and inspected monthly. Fouling of.tnese strainers (s not expected cecause this portion of the system o contains air which has been dried and filtered by the air fryer.

Texas Utilities proposes cleaning and inspecting the strainers every outage and will olowdown the strainers at the drip legs assembly daily.

21) Component 02 500C, Breakers & Contact Blocks , __

ltem 3 in the revised maintenance matrix recommends that circuit

! breakers be trip-checked at each outage. Texas Utilities believes that breakers of the molded case type should not be tested at frequent intervals and proposes to perform this test every 5 years for this type of breaker. Other types of breakers would bo tested at each refueling outage as recommended.

22) Components 02-5258, C, Barring Device Conten1s & Filter Because the carring device is not required during engine >

operation, the Owners' Group reccmendations (or tne associated components may not necessaril/ De incorporated b/ Texas fjtilities, but will oe taken into consideration.

i

? ** . j

~

l  !

i

!' 23) Component 02-717C, Jacket Water Piping, Couplings, Fittings, Oritrices & T-5 trainers f The DR/QR component report recomends that flanges be torqued per l the general torque tables in the TDI instruction Manual. Texas  !

1 Utilities flange torquing procedure provides for greater torque f than the general torque tables and will provide adequate load  !

, transfer to attaching supports.

l (

v l 24) Component 02-717G, K Lube Oil valves and Fuel Oil Valves  !

Item 1 in the original maintenance matrix recommended disassembly, I inspection and refurbishment of fuel oil valves on a 5 year l l interval, while the revised matrix specifies an interval of every i i -

outage. Based on satisfactory experience with these valves, Texas j Utilities will comply with the original recomendation for those valves which have identified deficiencies. This is consistent i with the TOI Maintenance Manual and the OR/OR Component report. j r

i 25) Component 02-810E, Jacket Water Heaters i

e The Owners' Group recommends in the revised maintenance matrix <

t that the following De perfortred at eacn cutage: 1) measure heater I insulation resistance, 2) inspect and clean heater elements, and l l 3) check calioration and inspect thermostat. However, this -

t j recommendation appears to be arbitrary in that it is not based on i l either a design review or on adverse component experience

}

?I

"' according to the component report. Texas utilities believes that  !

the activities recommended above would De adequate if performeo at i alternate refueling outages and intends to implement that i

! interval.

t j I

26) Component 02-820A, Lube Oil Sump Tank Heaters -

The Owners' Group recommends that heater insulation resistance be ,

measured and that thermostats be checked and calibrated at every

! outage. Based on satisfactory experience with these heaters, I l however, Texas Utilities proposes that these activities be "

performed at alternate refueling outages. The heater elements j i will be inspected at every refueling outage in conjunction with  ;

l the tank inspection, as recommended. j l

! 27) Cowiponent Cp-102, Generator Controls I

CPSES is in agreement with the revised 09/QR report.  !

1  !

i The DR/QR maintenance and surveillance activities recommended by the  ;

i Owners' Group are to be incorporated into the CPSES Unit 1 diesel  ;

1 generator preventive maintenance and surveillance program, as modified' i

} and/or clarified doove, as follows: l

} i

! Monthly and daily surveillance and maintenance items are to be  !

j incorporated into the surveillance or maintenance programs prior 1 l

)

1

{ <

6-l

-t .

to exceeding 5% power at CPSES Unit 1. Items based on number of engine starts or hours of operation, as well as 18 month surveillance items, are included in this category.

Surveillance and maintenance items which are to be accomplished on a refueling outage or longer basis will be incorporated into the appropriate programs prior to the first refueling outage. .

It should be noted, however, tnat evolution of the maintenance and surveillance program can be expected to occur based on implementation feedback, inspection results and on additional input from NRC, the Owners' Group, TDI or other applicable sources. Also, variations in the methods of achieving the intent of the recommendations may be necessary

, upon implementation in order. to provide a practical approach to maintenance. NRC is hereby requested to provide guidance on the subject of possible future changes (both major and minor) to the programs for the reasons stated above, particularly regarding NRC notification of prospectivit changes.

Finally, the TOI inspection and maintenance forms referenced in the

" Comments" section of the maintenance matrix may or may not be used in the performance of the recommended activities. Maintenance instructions and procedures will specify which forms are to be used, if any, for a specified activity.

e 7

ENCLOSURE 4 MEETING ATTENDANCE COMANCHE PEAK TDI DIESEL GENERATOR RELIABILITY PROGRAM AUDIT Monday and Tuesday, July I and 2, 1985 Texas Utilities . NRC Staff J. B. George C. H. Berlinger J. C. Kuykendall D. Persinko R. D. Caler S. B. Burwell G. Mullens K. B. Becker NRC Consultants D. A. London J. P. Shrewsbury D. Dingee J. R. Green A. H. Wendel D. Reimer H. M. Hardy F. W. Madden N. N. Rivera D. R. Woodlan P. S. Louzecky R. W. Haskoree R. G. Cockrel Oboserser Texas Utilities Consultants D. L. Dill, Carolina Power and Light Co.

K. T. Fitzpatrick, SWEC (Owners Grou J. C. Kammeyer, SWEC (Owners Group) p)

C. L. Ray, Duke Power Co. (Owners Group)

A. Kusko, FaAA W. Littman, FaAA r

9 1

- - . . .._m. _m, - . __ -..-_r, ,.m-,_-r-, ,__ __ ,-_. ,,_--__-_.-m.,,__,.. -- - _ - .m,,,. . , ,.., ,-_ . , , - - - . _ . -