Text

Review & Evaluation of Tdi Diesel Engine Reliability & Operability - San Onofre Nuclear Generating Station Unit 1, Technical Evaluation Rept
	ML20214E480
Person / Time
Site:	San Onofre
Issue date:	11/30/1984
From:	Laity W, Richmond W; Battelle Memorial Institute, PACIFIC NORTHWEST NATION
To:	Berlinger C; Office of Nuclear Reactor Regulation
Shared Package
ML20214D670	List: ML20214D667; ML20214E060; ML20214E265; ML20214E423; ML20214E480; ML20214E485;
References
	CON-FIN-B-2963, FOIA-86-656 PNL-5304, NUDOCS 8611240455
	Download: ML20214E480 (52)
	v • d • e

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _

9

. ~ ~ - ,,, - ,,,.w

, m eg,nr '1 . - - .. -

3rlA @i;> Gr... o

, e .o

,e , .

-?,. i -ny,;.:gr.tg;N4/gg;;g,.eynv 4 he g p;.6 - oc ^'

^

PNL-5304

. mmwm, w .

+

.q .m . ..l 4.g: yn '.h ; w a,: p.ym.mw h, 'C'. 6M.5My ; qi ,s y } lg% Q:w.lc.a,0.

. ., v. c. - .;.w s. ve c,e,.. - ,

p m. - mggg. -M,w.u, . .

g g,.h. g g.e.g . ,g ,w 3- .

np. ,. . w a w g,y, 9 w w gn.g y 4, w: a . 4,.$i.Q 3.g ,

. . n n un. .w wg : -....i es #we. m. s.n.h w. mey. e p. 4.. %.. ,,.- m .

. . . ' ..,6 - 6 y

.T4 .p.

y jp 1

a

'TechnicalEvaluathniReport%' 4%- * *%%. 'g', , y;f' , p#+i^

$ - . .J V ; : .s_f . ,* t, nfanggggggggn Review and Evaluation of wnde 9 Transamerica DhisiGLMEN$d.r. -bJ m .

Diesel Engine Rella c .

and N W 9 M -

. Operability-San y

wr 'wwwwrp... T

q. -~

. Nuclear Generg.ag~ w.an.nawgggg .

Unit 1 ~ . ...>

3.:m :, u .

.. ;.,y, 4* b ? g.

(fii . . " yiNNJ.%h J,$g y.*yNi y$;sg .+1#

+

s.s s i@memgur 9

' November 1995 v M'"

ep

WNT.:7ln@.

, ., .*?'. . N.WM.# -

. i:rgi w + +wcm.py;py.9%%g;gq.g;pp%i.up. ,

-g. mms.gnpgy,.yg,ppyy y m y

v if$ '

Prepared the U.S. Nuclear forRegulatory n'.AM;p;4hC.[qi9/jg$$f%s' Conunission' f "ljg' y)g b

under NRC FIN Contract B2963 DE-ACSG-76RLO J '," 'NGR$idf.$$$$. 1830. f@d ~ t W 3 "d"M ': ,

i . .

T Pacific OperatedNorthwest for the U.S.'Departme Laboratory ofEnseg 6;s;M6c%e[...MO. *7"~.

- . by Battelle Memoriallnstitu,te' '~,,. ,.~',r 4 gq s a' .; L 7,,t ' a a s v

~

r Id' .* ,(,s s h', ' i? f,* 's

......'*}N

"> % - A,Ws ' ,j r'

Y e-4

' e y ;.y M .6jP. % i @ i.3 ,.

4 A

- . -. .' , ' M. : L

' 's e d. . Y.O J* ins4.0 m- ' ns pyd v gl^ .1 Nr\'s l , L.

. . m. p p.,mw n q. . e w- .,

t *

,.._.19 w-b i g ,, % } /. , t g t ; y .. ;

k .j y ; ' , . 1

i. ,;

. *v i > > Q af%tQ :. e9...

, . Wsj}, hM, r ^*\ (.y: ,,

)' .. c;. 'g'W q' " q O r g" hipp U "' ' ";p Q ,y .,

8611240455 861110

. . NP% hf: PD" L '? 1 ' ' "".

L ' '. .'

PDR FOIA '" ~ "p D , . ** 'M MNN. . . .

ELLIS96-656 PDR - .

> . .~ .:- . :. . a. . ,u .wuuu ,

.i ._.m,,_ _

. . . . . ~

t a . i

. .. ~

+ m.+ yn v n

. - n any, z.my e ,

, PNL-5304 N + .$ . .* 4;1 . g' ;' s;3 e

.- A'

[* ' ,p s',,.', g

. ..:aya:; p , :p, 3:a u:-- a ;

. .~

a v 9,:p n.,.. w:q:f.w- ugm, u.. -a mp,.w. sus c u e - f. y ,

y swppggggpeggy.im ,

a J

./. m .; .: . m

'?' -s * ': ) " .,A

&h+. . . . ., .. . *;. . f * .

Technical Evaluation'Reporti~dk., N@$Nif8.rlM. .n M, . ' '

5

- g, . ya,gg~ggJ. _. , .

g ggg.. . p Review and Evaluation of o . _ :" . m. .

Transamerica Delasal,M$$PN Diesel Engine Reliability;andOM y .

Operability2 aS ( O. .ri d.f Cn " ?. W W*"~

a . u .. " w Nuclear Generatu,3.u.g igypmip&yg,ggg, .y ,

. - .g... y . ..

nl t'.

~ M' ' wA'? S h D[ri ? )'..,_,'+' , , g lh.hi'

- , p ;w.i ..' v:/

~ November 19R """" '

' 'l"Y. O W h

, ,>J

y k* 'l f* ,Y.;' ; ?% e{ 5 k 1 .. sh j } b,s,'_4 -

1 w ^= Oce4%s,yy y r. 35,' q < ~

o Prepared for ..e 9 4Jfi9/3;.y.$fiMfg;jp@^jpus '

the U.S. Nuclear Regulatory Commission ~ ' ' '

under Contract DE-AC06-76RLO 1830 . . . - -

NRC FIN B2963 s

'C- .h,9 M.H ..e W...,. . ;.iM.Pi

,.. . .j id. ~

Pacific Operated Northwest ior the U.S. Departme Laboratory <.4'rii of Energy' "'" '

f by Battelle MemorialInstitute ,.. +. -

' 5

.s., ( , , .{

s '- .' ,

, , . a. &

OBallelle = w w. . . M l

, %tA ., ,J a 2 'I

. :.,. , m. <

w

, c

'j,

> an:. . ,Wq ,

.i.

]'s f.

_ 4 } ~ ' * ' 0 Y;f l 'l ( !

v 8611240455 861110 ' '

O '

PDR FOIA e*

ELLIS96-656 PDR . .* ^^V.

s ,

. * . .. ' W ".. ;' T '"'

.. - - . . .c

.I 6

PNL-5304

?

Technical Evaluation Report REVIEW AND EVALUATION 0F TRANSAMERICA DELAVAL, INC., i DIESEL ENGINE RELIABILITY AND OPERABILITY - SAN ON0FRE NUCLEAR GENERATING STATION UNIT 1 i

November 1984 Prepared for the U.S. Nuclear Regulatory Commission Division of Licensing Office of Nuclear Reactor Regulation l

under Contract DE-AC06-76RL0 1830 l

NRC FIN B2963 Project

Title:

Assessment of Diesel Engine Reliability / Operability NRC Lead Engineer: C. H. Berlinger Pacific Northwest Laboratory Richland, Washington 99352

- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___ _J

PACIFIC NORTHWEST LABORATORY PROJECT APPROVALS Date // ~ 7 - Ef W. W. Laity, Project Manager Pacific Northwest Laboratory

/]>c, L-_- -- ;f gate l,

'l - 9 4 W. D. Richmond, Chainnan Senior Review Panel Pacific Northwest Laboratory e

iii

o' FOREWORD This report is supplied as part of the Technical Assistance Project, Assessment of Diesel Engine Reliability / Operability, being conductett for the U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, Division of Licensing, by the Pacific horthwest Laboratory. The U.S. Nuclear Regulatory Comission funded this work under authorization B&R 20-19-40-42-1 FIN No. B2963.

I e

I l

4 V

I

~

CONTENTS PACIF IC NORTHWEST LABORATORY PROJECT APPROVALS . . . . . . . . . . . . . . . . . . . . . . . . iii FOREWORD .............................................................. v ABBREVIATIONS ......................................................... ix

1.0 INTRODUCTION

..................................................... 1.1 1.1 ORGANIZATION OF REPORT ...................................... 1.2 1.2 LIMITED APPLICABILITY OF CONCLUSIONS ........................ 1.2 1.3 R E P ORT PR E P AR AT I ON . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.3 2.0 B A C K G R O U ND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1 2.1 OWNERS ' GROU P PROGR AM PLAN . . . . . . . . . . . . . . . . . ... . . . . . . . . . . . . . . . 2.1 2.2 SAN ON0FRE NUCLEAR GENER ATI NG STATION . . . . . . . . . . . . . . . . . . . . . . . 2.2 3.0 COMPONENT PROBLEM IDENTIFICATION AND RESOLUTION ... ..... . .. ... . .. . 3.1 3.1 SCE INSPECTION .............................................. 3.1 3.1.1 Inspection Procedures ...........,..................... 3.2 3.1.2 Results/SCE Conclusions .............................. 3.2 3.2 PNL EVALUATION .............................................. 3.3 3.2.1 Engi ne Base and Beari ng Caps . . . . . . . . . . . . . . . . . . . . . . . . . 3.3 3.2.2 Cylinder Block ....................................... 3.5 3.2.3 Crankshaft........................................... 3.8 3.2.4 Connecting Rods ...................................... 3.16 i

3.2.5 Connecti ng R od B ea ri ng Shell s . . . . . . . . . . . . . . . . . . . . . . . . 3.21 3.2.6 Piston Skirts ........................................ 3.24 i

3.2.7 Cy l i nd e r L i n e rs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.26 3.2.8 Cylinder Heads ....................................... 3.29

! 3.2.9 Cylinder Head Studs .................................. 3.32 vii

3.2.10 Push Rods ........................................... 3.33 3.2.11 Rocker Arm Capscrews ................................ 3.35 3.2.12 Turbochargers ....................................... 3.36 3.2.13 Jacket Water Pump ................................... 3.40 3.2.14 Hi gh-P ressu re Fuel Oi l Tubi ng . . . . . . . . . . . . . . . . . . . . . . . 3.42 e.

3.2.15 Air Starting Valve Capscrews ........................ 3.43 3.2.16 Engi ne-Mounted El ect ri cal Cabl e . . . . . . . . . . . . . . . . . . . . . 3.44

3.2.17 Components Possibly Affected by Crankshaft Torsional Vibration.................................. 3.45 4.0 PROPOSED MAINTENANCE AND SURVEILLANCE PROGRAM .................... 4.1 4.1 MAJOR MAINTENANCE ITEMS ..................................... 4.1 4.1.1 PNL Evaluation and Recommendations ................... 4.2 4

4.2 ADDITIONAL MAINTENANCE ITEMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.13 4.2.1 Rationale ............................................ 4.13 4.2.2 PNL Evaluation ....................................... 4.13 4.3 OPERATIONAL SURVEILLANCE PLAN ............................... 4.16

) 4.3.1 Rationale ............................................ 4.16 4.3.2 P NL E v a l u a t i on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.16 1

4.3.3 PNL Recommendation ................................... 4.18 4.4 STANDBY SURVEILLANCE PLAN ................................... 4.18 4.4.1 El ements and Rati onal e . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.18 4.4.2 P NL E v a l u a t i on . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.18 .

4.4.3 PNL Recommendation ................................... 4.19 5.0 ENGINE TESTING ................................................... 5.1 5.1 ENGINE TESTS AND RESULTS .................................... 5.1 5.1.1 Shop Qualification Tests ............................. 5.1 vili

5.1.2 Onsite Preoperational Tests .......................... 5.2 5.1.3 Post-0perational Testing ............................. 5.2 5.1.4 Special Tests Following Major Maintenance /

Inspection ....... ................................... 5.3 5.2 PNL EVALVATION .............................................. 5.5 5.3 PNL CONCLUSIONS ............................................. 5.6

, 6.0 OVERALL CONCLUSIONS .............................................. 6.1 6.1 GENERAL CONCLUSION .......................................... 6.1 6.2 BASIS FOR CONCLUSION ........................................ 6.1 6.3 LONG-TERM APPLICABILITY ..................................... 6.1 6.4 SONGS-1 RESTART CONSIDERATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6.2 6

IX

)

ABBREVIATIONS BMEP brake mean effective pressure CFR Code of Federal Regulations DG diesel generator DR/QR design review / quality revalidation EDG emergency diesel generator FaAA Failure Analysis Associates LOCA loss-of-coolant accident LOOP loss of offsite power LP liquid penetrant H/S maintenance / surveillance NRC U.S. Nuclear Regulatory Commission 0G Owners' Group; the TDI Diesel Generator Owners' Group OGPP Owners' Group Program Plan 0/R operability and reliability PNL Pacific Northwest Laboratory QA/QC quality assurance / quality control SCE Southern California Edison Company SONGS-1 San Onofre Nuclear Generating Station Unit 1 SWECo Stone & Webster Engineering Corporation TDI Transamerica Delaval, Inc.

TER Technical Evaluation Report 4

xi

REVIEW AND EVALUATION OF TRANSAMERICA DELAVAL, INC., DIESEL ENGINE RELIABILITY AND OPERABILITY -

SAN ON0FRE NUCLEAR GENERATING STATION UNIT 1

^

1.0 INTRODUCTION

Following an extended outage of its San Onofre Nuclear Generating Station Unit 1 (SONGS-1), Southern California Edison Company (SCE) submitted a return-to-service plan to the U.S. Nuclear Regulatory Commission (NRC). A problem area identified by NRC is the reliability of the Transamerica Delaval, Inc.

(TDI) diesel engines used by SCE as emergency power sources for the SONGS-1.

These engines' operability and reliability (0/R) have been brought into ques-tion by a major crankshaft failure in one TDI diesel (at the Shoreham Nuclear Power Station in August 1983) and other problems reported by TDI diesel owners in nuclear and non-nuclear installations.

The San Onofre Nuclear Generating Station Unit 1 is served by two TDI Model No. DSRV-20-4 engines. Both are nameplate rated by TDI at 6000 kW, and operate at 450 rpm with a peak cylinder firing pressure of 1150 psig. SCE has designated these engine-generators as DG #1 and DG #2. The most recent infonnation provided by SCE specifies the emergency loads as a maximum of 4350 kW for DG #1 and 4443 kW for DG #2 under design basis accident conditions coincident with a simulated loss of offsito power (LOOP).

In response to the NRC concerns about the SONGS-1 TDI engines, SCE provided a comprehensive analysis of engine performance history and actions taken to assure their 0/R. This analysis was described in several documents and related meetings with the NRC staff and NRC's consultant, Pacific Northwest Laboratory (PNL).

The Pacific Northwest Laboratory has been requested by NRC to review and evaluate these documents and SCE's overall effort to assure their engines' operability and reliability. This technical evaluation report (TER) expresses 1.1

)

PNL's conclusions and recommendations regarding the capability of SCE's TDI standby emergency generators to serve their intended function at SONGS-1.

1.1 ORGANIZATION OF REPORT This technical evaluation report is organized as follows:

e Section 2.0 provides relevant background information on the known problems and efforts toward their resolution by both SCE and the TDI Diesel Generator Owners' Group (an ad hoc group of similar TDI engine ;

owners) who have united to resolve these mutual concerns, e Section 3.0 presents a review and evaluation of SCE's rasolution of known problems in 16 engine components identified by the Owners' Group through a review of TDI engine operating history. _

e Section 4.0 presents PNL's review of SCE's proposed maintenance and \

surveillance (M/S) program.

e Section 5.0 documents PNL's evaluation of SCE's engine tests and results, as well as the utility's plans for further testing.

e Section 6.0 presents PNL's overall conclusions and recommendations regarding the suitability of the two diesel engines to perform their intended function as emergency standby power sources for the SONGS-1.

1.2 LIMITED APPLICABILITY OF CONCLUSIONS PNL has reviewed the basic documents supplied by SCE, has participated in various meetings with SCE and NRC, and has observed a few components of the DG #1 and #2 engines as disassembled for SCE's crankshaft inspection. Concur-rently, PNL also has reviewed various relevant Owners' Group documents and participated in their meetings with NRC, and has drafted technical evaluation '

reports on some elements of the Owners' Group Program Plan (0GPP).

This TER on the SONGS-1 DG #1 and DG #2 engines precedes completion of the 0GPP and its appropriate implementation by SCE. This document also precedes the OG planned full plant-specific design review / quality revalidation (DR/QR) analyses of the engines. Therefore, PNL cannot reach final conclusions 1.2

regarding the SONGS-1 engines' operability and reliability to perform indefinitely their expected design function. PNL's conclusions regarding the long-term 0/R of the engines are subject to full completion of all elements of the OGPP and SONGS-1 DR/QR programs and implementation of their findings.

Hence, PNL has evaluated all components in light of expected operating

.' conditions and patterns at SONGS-1 over a period of time corresponding to one reactor operating cycle, which PNL understands to be approximately 18 months from plant restart. By that time, all phases of both the general 0GPP evalua-tion and implementation and the plant-specific SONGS-1 DR/QR program should be complete and ready to implement. Because these actions will represent proposed resolution of the TDI engine issues at SONGS-1, it is expected that NRC will draw its final conclusions regarding the long-term operability / reliability of the SONGS-1 engines at that time.

The considerations and recommendations presented in this TER are sometimes expressed in terms of "until the next reactor refueling outage." However, in using this phrase, PNL does not intend to imply (unless specifically stated otherwise) that tne engines or their components are therefore unreliable or inoperable for their intended use over their normally expected life. l 1.3 REPORT PREPARATION This report is based in part on a review of documents cited in Sec-tion 2.0. The PNL team also visited the SONGS-1 on October 3 and 4, 1984, while the DG #2 engine crankshaft oil holes were being inspected. At that time the PNL team, together with NRC, also met with appropriate SCE staff and man-agement concerned with diesel engine operability / reliability. A meeting among representatives of SCE, NRC, and PNL was held on October 22. and 23,1984, to

-. review findings on the investigation of the cracks in the region of the oil holes in the crankshaft.

i The following PNL staff members and consultants were involved in this review and evaluation, and authored this report:

e D. A. Dingee, PNL project staff e H. M. Hardy, diesel consultant to PNL 1.3

a e P. J. Louzecky, diesel consultant to PNL o T. W. Spaetgens, diesel consultant to PNL.

Others whose contributions were considered in formulating the conclusions include PNL' Assessment of Diesel Engine Reliability / Operability Project team members J. M. Alzheimer, M. Clement, S. D. Dahlgren, R. E. Dodge, R. H. Ferris, W. W. Laity, J. F. Nesbitt, J. C. Spanner, and F. R. Zaloudek; and consultants i.

S. H. Bush, A. J. Henriksen, A. Sarsten, and J. V. Webber (representing Ricardo Consulting Engineers). The report editor was A. J. Currie. ,

9 1.4

.' i l

l

2.0 BACKGROUND

This section presents background information on efforts undertaken by the TDI Diesel Generator Owners' Group and by Southern California Edison Company to resolve the problems identified in the TDI diesel engines.

~f 2.1 OWNERS' GROUP PROGRAM PLAN

. Thirteen nuclear utilities that own diesel generators manufactured by Transamerica Delaval, Inc. (TDI) have established an Owners' Group to address questions raised by the major failure in one TDI diesel engine at the Shoreham Nuclear Power Station in August 1983, and other problems in TDI diesels reported in the nuclear and non-nuclear industry. On March 2, 1984, the Owners' Group submitted a plan to the U.S. Nuclear Regulatory Commission out-lining a comprehensive program to requalify their diesel generator units as standby emergency power sources.

The Owners' Group Program Plan describes a two-phase approach for resolv-ing the known and potential problems in TDI engines:

o Phase 1 addresses the evaluation and resolution of significant known problems in 16 components. These problems were identified by the Owners' Group through a review of the operating histories of TDI engines in nuclear and non-nuclear service.

e Phase 2 entails a comprehensive design review and quality revalidation (DR/QR) to identify critical components of TDI engines in addition to the 16 referred to above, and to assure that these components are also adequate for their intended service.

The Owners' Group Program Plan also describes a program element for special or expanded engine tests and component inspections, as appropriate, to verify the adequacy of the engines and components to perform their intended functions.

The NRC staff has determined that the TDI diesel reliability and QA issues must be adequately addressed before additional licensing action is taken to authorize the operation of nuclear power plants equipped with these engines.

2.1

9 At NRC's request, PNL reviewed the Owners' Group Program Plan. The results of that evaluation were reported to NRC in PNL-5161, Review and Evaluation of TDI Diesel Generator Owners' Group Program Plan (Pacific Northwest Laboratory June 1984).

Section 4 of PNL-5161 deals with considerations for licensing actions for nuclear stations prior to completion of the implementation of the Owners' Group .,

Program Plan. Reconsnendations in that report relevant to SCE's restart of the San Onofre Nuclear Generating Station at this time are: ,

1. The engine should be inspected per Section 2.3.2.1 of PNL-5161 to assure that the components are sound.

2. Preoperational testing should be performed as discussed in Sec-tion 2.3.2 of PNL-5161.

3. The engines should receive enhanced surveillance and maintenance.

4. The engine should have AE pistons; otherwise, " lead-engine" tests as described in Section 2.3.2 of PNL-5161 should be completed.

The first three recommendations are self-evident; namely, that the engine has sound parts, that appropriate preoperational tests have been satisfactorily completed, and that a suitable program of surveillance and maintenance is established to assure future performance. .

The significance of the last recommendation, that the engine be outfitted with AE pistons, is that relevant service experience (107cycles) exists to empirically support the satisfactory analytical res,u,lts for engine loads corresponding to a brake mean effective pressure (BMEP) limit of 185 psig. The SONGS-1 engines do not meet these piston criteria. Accordingly, the adequacy of the modified AF pistons under SONGS-1 service conditions was specially reviewed (see Section 3.2.6). '

2.2 SAN ON0FRE NUCLEAR GENERATING STATION ,

In its efforts to establish the operability and reliability of SONGS-1 TDI diesel engines, SCE has conducted tests and inspections and has provided NRC 2.2

~

with relevant letters and reports. Documents considered by PNL in preparing this technical evaluation report are listed below, e a letter dated January 6,1984, from D. M. Crutchfield (NRC) to K. P. Baskin (SCE), " Request for Additional Information Regarding Transamerica Delaval Emergency Diesel Generators - SONGS-1" - This document posed questions regarding the SONGS-1 TDI engines and requested SCE to respond within 30 days.

. e a letter dated February 7,1984, from R. W. Krieger (SCE) to D. M. Crutchfield (NRC), " Docket No. 50-206, NRC's Request for Additional Information, Transamerica Delaval Emergency Diesel Generators, SONGS-1" - This letter requested additional time to respond to the questions posed in NRC's letter of January 6,1984.

e a letter dated February 29, 1984, from M. O. Medford (SCE) to D. M. Crutchfield (NRC), " Docket No. 50-206, NRC's Request for Additional Information, Transamerica Delaval Emergency Diesel Gen-erators, SONGS-1" - This letter provided NRC with the responses to many of the questions raised in NRC's January 6,1984, letter.

e a letter dated June 29, 1984, from M. O. Medford (SCE) co D. M.

Crutchfield (NRC), " Docket No. 50-206, NRC's Request for Additional Information, Transamerica Delaval Emergency Diesel Generators, SONGS-1" - The 36 attachments to this letter provided information to support the SCE contention that the SONGS-1 TDI engines are a reliable source of onsite power.

e a letter dated July 26, 1984, from D. M. Crutchfield (NRC) to K. P.

Baskin (SCE), "Transamerica Delaval Diesel Inspection Requirements for Restart of San Onofre Unit 1" - This letter provided the results of NRC's review of the information submitted by SCE in their June 29, 1984, letter. It identified items of inspection, maintenance and surveillance, and preoperational testing that SCM must complete prior to the SONGS-1 restart.

e a memorandum dated August 20, 1984, from C. H. Berlinger (NRC) to C. I. Grimes (NRC), " Diesel Generator Inspection Requirements - San 2.3

Onofre 1" - The enclosure to this memorandum provided the bases for conclusions reached by NRC staff regarding measures to be taken prior to the SONGS-1 restart.

e a letter dated August 28, 1984, from M. O. Medford (SCE) to W. A.

Paulson (NRC), " Docket No. 50-206, Return to Service Requirements Regarding TDI Emergency Diesel Generators, SONGS-1" - This letter e provided additional information regarding planned TDI inspection, tests, and maintenance and surveillance in response to NRC's letter dated July 26, 1984.

e a letter dated September 18, 1984, from D. M. Crutchfield (NRC) to K. P. Baskin (SCE), " Request for Additional Information Regarding TDI Diesel Generators - SONGS-1" - This letter requested information concerning cracks found by SCE in the inspection of the oil holes in the crankshaft of DG #1.

e a letter dated September 20, 1984, from C. H. Berlinger (NRC) to C. Ray (0G), " Request for Additional Information Regarding Transarrerica Delaval (TDI) Emergency Diesel Generating Units" - This letter requested the OG to examine their member utilities' records to determine the occurrence of electrical transients and to evaluate the generic and plant-specific effects these transients may have had on the diesel units.

e San Onofre Nuclear Generating Station Maintenance Order No. 84092733000 dated September 25, 1984, describing the torsiograph and pressure test procedure for the DG #1 crankshaft.

e a letter dated October 10, 1984, from M. O. Medford (SCE) to W. A.

Paulson (NRC), " Docket No. 50-206, Diesel Generator Stress Analyses ,

Reports, San Onofre Nuclear Generating Station Unit 1" - The i

enclosure to this letter summarized the results of stress analyses performed by Failure Analysis Associates on five major components of

the SONGS-1 EDGs.

e the minutes of the October 22, 1984, Owners' Group meeting held in Bethesda, Maryland.

2.4 l

e a letter dated October 26, 1984, from M. O. Medford (SCE) to J. A. Zwolinski (NRC), " Docket No. 50-206, Return to Service Requirements Regarding Transamerica Delavo; rmergency Diesel Generator, San Onofre Nuclear Generating Station Unit 1" - This letter and its enclosures provided SCE's response to NRC's request (letter, September 18,1984) for information related to the results of an NRC-requested (letter, July 26,1984) inspection of the SONGS-1 EDGs.

4 9

I 2.5

3.0 COMPONENT PROBLEM IDENTIFICATION AND RESOLUTION This section documents PNL's evaluation of SCE's efforts to identify and resolve site-specific problems noted with the 16 components known to have significant problems. These components were previously identified by the

, Owners' Group through a review of the operating histories of TDI engines in nuclear and non-nuclear service.

The overall SCE inspection process is reviewed first. Then PNL's review, evaluation, and conclusions regarding SCE's component problem resolution efforts are described. It is important to note that PNL's conclusions incorporate, generally without stating, SCE's comitment to the surveillance and maintenance program described in Section 4.0 of this TER, as well as the utility's commitment to appropriately implement the applicable recommendations of the OG concerning these components as soon as practicable.

3.1 SCE INSPECTION Following the Shoreham crankshaft failure in August 1983, a concern was raised regarding the reliability of the TDI engines at other nuclear utilities.

In January 1984, NRC requested SCE to provide a complete report documenting the parts inspected, the procedures used, and the resultant findings.

In fact, the decision to thoroughly disassemble and inspect both engines was made by SCE in 1982 after they heard of the cylinder block deck cracks from the cylinder head studs to the liner bore at the Shoreham Nuclear Power l Station. Therefore, the engine disassembly and inspection was well underway when the NRC and 0G directives for complete inspection reports were issued. At the time of the Phase 1 inspection the SONGS-1 engines had been run for a total

of 1190 " engine hours", of which at least 180 hours0.00208 days 0.05 hours 2.97619e-4 weeks 6.849e-5 months on each engine was at a load above about 4500 kW. This total time included about 400 starts on both DG #1 and DG #2 onsite at San Onofre and about 300 starts on DG #1 at the TDI manufacturing plant.

l l

l i

3.1 l

3.1.1 Inspection Procedures Because of SCE's favorable TDI engine operating experience and previous inspections, only about 25% of the basic engine parts were inspected at or during each refueling. However, when SCE was advised that the Shoreham engines and others developed cracks in their blocks from the liner bore to the cylinder head studs, they ir.stituted an inspection program to remove the cylinder head,

pistons, connecting rods, turbochargers, and other engine components for a critical inspection of these parts, including the block. ,

This inspection left intact the engine base, crankcase, cylinder blocks, crankshaft, and gearing (plus miscellaneous other components attached thereto, some of which were themselves inspected in place). On each engine, detailed inspections were performed on a sampling basis on all 16 components with significant known problems. In addition, some 45 other components were )

inspected (some from only one of the two engines). Visual inspection, dimensional checking, and testing by material comparator, liquid penetrant, '

magnetic particle, ultrasonic, radiography, and eddy-current techniques were utilized, all to criteria established by the OG (where available) or to cri-teria thought by SCE to be conservative where Owners' Group criteria did not exist or were not fully detailed. Records were maintained for each part (or group); these records were then subjected to a process of review, evaluation, and disposition, which was designed to maintain quality control.

The same procedures used to inspect and evaluate the disassembled compo-nents were, in general, applied to all replacement parts (connecting rods, push rods, cylinder heads) . Furthermore, SCE reported that they had previously enhanced onsite QA/QC oversight procedures at TDI's plant so that all parts of significance were adequately surveyed and the documentation checked before the parts left that plant.

3.1.2 Results/SCE Conclusions SCE's Phase 1 inspection results for the SONGS-1 engines are presented in '

a report dated June 29, 1984.

3.2

Based upon the engine inspection results, SCE concludes that the two TDI diesel generators installed in SONGS-1 are adequate to perform their intended function.

3.2 PNL EVALUATION PNL's review of SCE's component requalification efforts is documented in this section. Each generic problem component is discussed individually. The components are presented in a sequence reflective of their location within, on, or about the engine. The sequence generally progresses from bottom to top (that is, structural components, power train components, ancillary and auxiliary systems and components, on-engine and then off-engine).

The components are described in terms of their function, operating history, and status as determined by the Owners' Group and SCE. Then, PNL'S evaluation and conclusion (s) are presented for each component.

It is noted that the TDI engines at SONGS-1 will be operated at a BMEP of about 114 psig. This load is low for the model DSRV-20 engine. In discussing the 16 components, reference may be made to this reduced engine load.

3.2.1 Engine Base and Bearing Caps Part No. 02-305A Owners' Group Report FaAA-84-6-53 3.2.1.1 Component Function The base itself supports the crankshaft and upper structures, and carries the thrust of the cylinder combustion loads to the main bearings. The shaft is bedded in half-circle bearings set within " saddles" in the base. The bearing caps are structural members that hold the upper bearing shells in place over the shaft main journals while also absorbing the upward, reciprocating piston inertial loads. The studs and nuts hold the cap (shaft) in place. A failure

' of base, cap, or bolting would allow shaft gyration or misalignment, poten-tially leading to shaft fracture and seizure, sudden engine stoppage, and possible ignition of crankcase vapors (termed crankcase explosion).

3.3

3.2.1.2 Component Problem History Two basic problems have occurred that warranted 0G evaluation of this component as a generic issue:

e Saddle structures were found to be cracked in one engine base (an inline DSR4-8 engine) at Long Island Lighting Company's Shoreham Nuclear Power Station.

e On a non-nuclear application of a DSRV-16-4 engine a nut pocket failed on the through-bolting from the crankcase. '

3.2.1.3 Owners' Group Status Failure Analysis Associates (FaAA), as discussed in the Owners' Group report, has analyzed the base, bearing saddles, bearing caps, nut pockets, and bolting / nuts. FaAA has concluded that the base assembly components have the strength necessary to operate at full rated load for indefinite periods, provided that all components meet manufacturer's specifications, that they have not been damaged, that mating surfaces are clean, and that proper bolt preloads are maintained. They concluded that the failed nut pocket was due to impuri-ties in the casting material. The pattern of saddle fractures at Shoreham was determined to be related to improper procedures in disassembly.

3.2.1.4 SCE Status There have been no instances of failure or evident deficiency on either DG #1 or DG #2 at SONGS-1. FaAA advised SCE (FaAA-84-6-54, Evaluation of Emergency Diesel Generator Crankshafts) that the stress levels were the highest between cylinders 3 and 4, and 8 and 9. Based upon this information, SCE removed the main bearing caps from journals No. 4 and No. 9,10, and 11 of DG #1 to inspect the crankshaft. At the same time, the caps, upper bearing shell, and engine base were visually inspected. On DG #2, caps, upper bearing, '.

and base inspections were also secondarily inspected during inspection of the crankshaft in the oil hole area. a These inspections did not reveal any indications or cracks in the base or caps on either engine. The engines were reassembled in accordance with l 3.4

. 1 1

1 requirements on cleanliness and bolt torque. SCE contends these components are adequate to serve their intended function as required for nuclear standby service.

3.2.1.5 PNL Evaluation and Conclusions

, There is no indication of widespread, generic failure in bases or caps of TDI DSRV-20-4 engines, either in general or at SONGS-1. Hence, there is no basis for fundamental concern at SONGS-1.

Furthermore, these DGs operate at about half design load. On these bases PNL concludes that the engine base and bearing caps in both DG #1 and DG #2 are acceptable for their intended service at SONGS-1.

3.2.2 Cylinder Block Part No. 02-315A Owners' Group Report FaAA-84-5-4 3.2.2.1 Component Function Cylinder blocks (with their associated cylinder liners) contain the pistons. They are bolted to the vee-engine crankcase. The cylinder heads are bolted to the blocks; hence, the blocks are subject to the power forces from the cylinders. They also support the camshaft and other miscellaneous compo-nents, and serve as the outer containments for the jacket water system.

Depending upon its nature and location, a structural failure can lead to >

inadequate support of the primary combustion and mechanical forces, with concomitant sudden engine shutdown.

3.2.2.2 Component Problem History Several incidents of cylinder block cracking have been reported in non-I nuclear TDI engine applications. In nuclear service, indications have been reported to date on engines at Shoreham and Comanche Peak. None has resulted in emergency shutdown or catastrophic failure. A number of engines have continued to operate many hours with known cracks.

Two basic problems have occurred that warranted OG evaluation of blocks as a generic issue:

3.5 r -- ---m'- - - * - - - - - - - - ' - -

~~'c+--- - ' - ' - - ' - - -

o Supports for the camshaft were found cracked in Shoreham's inline engines, e Cracks have been found in the upper reaches of various blocks. Most have been on the top surface, between the cylinder liner opening and adjacent head stud openings (ligament cracks); some, however, have been between stud openings in adjacent cylinders. At Comanche Peak 4

" vertical" cracks have occurred in the area of the liner landings, but have not extended to the top surface. At least one instance of a circumferentially oriented crack along the liner landing (support ledge) itself has been reported, in the V-16 engine at St. Cloud, Florida.

3.2.2.3 Owners' Group Status -

Failure Analysis Associates performed strain gauge testing combined with two-dimensional analytical modeling of block tops and liners. In their report, FaAA concluded, in part:

e Eventually, depending upon a combination of sufficiently high load and operating hours, and/or engine starts to high load, cracks can be expected to initiate between stud hole and liner openings. However, such cracks are predicted to be benign (i.e., nonpropagating) if the block materials are free of deleterious materials and properly cast, and if engine loads therein remain below the nominal 225 psig BMEP rating. FaAA notes'that some engines (such as those at Duke Power Company's Catawba Nuclear Station) have operated many hours at loads at and exceeding these levels without even initial crack indica-tions. This is, in FaAA's opinion, Indicative of the conservative nature of their evaluation.

e e The initial development of ligament cracks between the stud hole and liner opening will increase the likelihood of more serious cracks developing between stud holes of adjacent cylinders. Therefore, the

Owners' Group has recommended that engine blocks with ligament cracks 3.6

~ _ .

1 be inspected by the eddy-current technique prior to the engine's return to standby service after any period of operation above no load.

e The FaAA report implies that, if block material is equal to or better than typical Class 40 grey iron, is properly cast, and has no initial cracks between stud holes of adjacent cvlinders, the block of any engine can be expected to survive the service requirements of any LOOP /LOCA event (even to 225-psig BMEP).

3.2.2.4 SCE Status To date, no cracks have been reported in the engine cylinder blocks of either DG #1 or DG #2 at SONGS-1. The liners, however, have not been removed from the blocks, so there has not been an inspection for liner support ledges or landing area cracks.

SCE plans to conduct cylinder blocks surveillance inspections whenever a cylinder head or liner is removed. The current plan is to inspect about 25% of i the heads, pistons, and connecting rods at each refueling, at which time the block area will be inspected.

3.2.2.5 PNL Evaluation and Conclusion Cylinder block cracking has been reported in both nuclear and non-nuclear service. However, no cracks in the blocks of the engines at SONGS-1 have been reported. PNL notes that the maximum firing pressure for the SONGS-1 engines is about 1150 psi, compared to the firing pressure of 1670 psi at a BMEP of 225 psig, the pressure corresponding to the engine nameplate rating. This reduced firing pressure is seen by PNL as an important factor in reducing the probability of cracks developing.

i e j The OG/FaAA report has not received full PNL evaluation to this date.

( However, PNL views FaAA's conclusions as reasonable and useful in evaluating i . the SONGS-1 block. PNL has, however, expressed some concern over the effec-tiveness of eddy-current testing if cracks do not penetrate to the surface.

i 3.7

PNL notes that SCE is comitted to a program of inspecting for cracks during each refueling. The tests will include both liquid penetrant and eddy-current inspection of all possible crack areas.

In light of the favorable inspection results, the comitments made to monitor, and the site-specific load expectations, PNL concludes that the engine blocks of DG #1 and DG #2 of the SONGS-1 can be expected to operate reliably -

for their intended duty and can be placed in service at least through the next refueling cycle. The PNL concern regarding the inability of eddy-current ,

inspection to detect subsurface cracks is also mitigated by the SCE commitments and engine load limits.

3.2.3 Crankshaft Part No. 02-310A Owners' Group Report FaAA-84-4 3.2.3.1 Component Function The crankshaft receives the reciprocating power strokes from the cylinders (via the pistons and connecting rods), converts them to rotary motion, and transfers the shaft power to the generator. It also drives the gear train that operates the camshaft, which, in turn, operates the cylinder-head valves, fuel injection pumps, governor, etc. The crankshaft is supported in journal bearing mounted in the engine base. The crankshaft begins as a forged steel billet, which is subsequently formed into the crankshaft configuration by a further process of forging, punching and twisting, after which it is totally machined. In TDI vee engines, opposite pistons are connected to the same crankpin by an articulated master / slave connecting rod arrangement. By means of holes drilled throughout the crankshaft, pressurized oil is picked up from the main journal bearing supply points and transmitted to connecting rod ,

bearings, link pins, wrist pins, undersides of the pistons, and other parts.

The crankshaft is subject to a variety of very complex stress fields. ,

These include direct and torsional shear stresses and bending stresses due to the piston thrusts; inertial effects of rotating and reciprocating masses; torsional, axial and flexural vibration stresses; bending stresses due to overhung flywheel; 5ending stresses due to wear-down in main journal bearings; 3.8

~

.v J

and variation in external support alignments. These nominal stress combina-tions are augmented in local stress fields due to the stress-raising influence

of oil holes and crankweb/ journal transition zones. Residual stresses due to f forging and straightening procedures, operating conditions, and accidents also l

affect the final stress spectrum. The machined surfaces of the crankshaft Journals and crankpins are subject to damage from oil impurities, bearing deterioration, and excessive heat. Crankshaft failure can come, therefore, in various ways. At its worst, it may actually fracture (through fatigue) and separate, leading to immediate stoppage and probable significant conjunctive

! damage to other components. Other types of failure (such as cracking) may

evidence themselves less destructively and more gradually, and can sometimes be

., detected via monitoring, surveillance, and maintenance activity including periodic crankshaft deflection checks which help to prevent crankshaft overstressing in bending by disclosing; e gradual shifts in shaft supports internal to the engine (such as significant bearing wear or deterioration) e gradual changes in external engine support, as in a concrete j foundation, loose foundation bolts, or a shift in the shims between the foundation rails and the engine base.

3.2.3.2 Component Problem Histor_y j Three failures of DSRV-16 crankshafts of current design have been

reported, all in the non-nuclear industry. Two of these failures have been f attributed to excessive torsional vibration stress; no cause has been suggested for the third.

Three failures of DSR-8 crankshafts, having 11-inch diameter crankpins,

, have occurred at Shoreham. Two of these were cracked and one was a fracture; all three were the result of fatigue. These three crankshafts were replar.ed with shafts of newer design which utilize larger 12-inch diameter crankpins.

3.9 l----._.___.____..._.._- __ __ _ _ _._ _.-.__,.,_.._ - _.__ _ _ _ ,

? '.

l- .

1 i

3.2.3.3 Owners' Group Status -

The Owners' Group analyzed the DSRV-20-4 diesel generators at SONGS-1.

j Their analysis was documented in the FaAA crankshaft evaluation report (FaAA-84-6-54), which covers both the Midland (V-12) and the San Onofre (V-20)

{ emergency generating units. -

[- The crankshafts in the two V-20 engines at SONGS-1 have 13-inch diameter j crankptns and 25-inch diameter by 5-1/8-inch thick crankwebs. The FaAA report j provides the analytical, compounded results of the forced-damped torsional i response at 8750 kW and between 427 and 472 rpm, when 24 harmonic orders of j excitation are considered. The maximum nominal stress (t4018 psi) is shown to occur between throws No. 8 and 9; the stress between throws No. 9 and 10 are 5%

lower as per Table 7.4 ibid. FaAA concluded (page 8-1) that "the crankshafts l are adequate for their intended service, provided a torsiograph test is

conducted to confirm the torsional system natural frequencies and measure the j combined response". This report did not address the matter of the resonant 1 vibratory stresses in the lower part of the speed range.

j At this writing, the Owners' Group is finishing the re-evaluation of the j dynamic response of the V-20 crankshaft / generator dynamic response throughout the entire speed range as a result of additional torsiograph and ancillary testing conducted on DG #1 on September 27, 1984.

1 3.2.3.4 SCE Status

! SCE confirmed to NRC and PNL in a meeting at SONGS-1 that they had checked r J

on misalignment possibilities between the engine, flywheel, and generator.

l They found the alignment to be entirely acceptable; that is, they found no shift of any external support bearings or shims, and no weardown of the bear-

[ ings that might suggest alignment problems. This included confirming that hot g l- and cold crankshaft web deflections were within manufacturer's specifications. .

i j

i SCE has monitored closely the matter of quality assurance during ,

procurement, manufacturing, shipping, and receipt of their two TDI diesel generator units. Pre-award survey quality audits followed by conformance surveys conducted by SCE at the TOI factory, including the inspection and witnessing of manufacturing processes and tests, and qualification testing of j l 3.10 ;

k I

i 1

- ~ _ _ . . ___ ___ _ _ _ ... . - . _ - _ _ _

. 1 i

l the finished diesel generator units according to SCE-approved procedures, had satisfied SCE that the two units conformed to specification requirements.

Factory qualification testing has included 300 fast-start and fast-loading

sequences (4000 kW 115%) of DG #1. On three of these starts, the loading was thereafter elevated to 5000 kW for 1/2 hour. At SONGS-1 the number of a additional starts to the time of the recent inspection is 372, bringing the total to 672 for DG #1. For DG #2, for which shop qualification fast-start testing was not applied, 387 total starts are recorded to date. The engine logs show that the operating hours at or above 4500 kW are at least 183 for DG #1 and at least 188 for DG #2, while the engine-hour meter indicates total operating time for both units at 1190 hours0.0138 days 0.331 hours 0.00197 weeks 4.52795e-4 months .

Since April 1977 the two diesel generator units at San Onofre have been subjected to various surveillance tests, including monthly load tests (a 1-hour run at approximately 4500 kW, increased in 1984 to 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months ' duration); oper-I ability manual start tests (a 5- to 10-minute run at no load, also increased in 1984 to 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months ' duration); loss of bus start tests (a 5- to 10-minute run at no load); and refueling interval tests (overspeed-trip test,1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months at 45 kW, I load rejection; total 3 to 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months ).

As a result of the concern generated primarily by the failure of the three I

V-16 crankshafts of current design, NRC recommended that an inspection of the main journal oil holes be made on the SONGS-1 DG #1. This inspection was to identify particularly the No. 9 and No.10 main journals (i.e., between throws 8 and 9, and between throws 9 and 10), in view of FaAA's analytical results I referred to above. Attention was also directed to the journal /crankweb transition fillets of the same throws.

SCE reported on their examination of the oil holes of Nos. 8, 9, 10, and 11 of DG #1 on August 8, 1984. Using fluorescent dye penetrant, the No. 9 main journal was found to have two cracks at opposite ends of the oil hole, running through the radius and into the bore, at 45' to the crankshaft axis. The longest crack was approximately 3 inches with a depth of just under 3/16 inch occurring at the junction of radius and bore. With the No. 9 throw at 9 o' clock, the opposite end of the oil hole displayed two more cracks (again at opposite sides of the hole) but running in a direction of some 75* to the 1

3.11 w w w m gr -ww-- +w- ---

r- - - -

3--ey-e-+ erg - g---w-tw-gw w gegy,,y-vw---wrweisw--w me --- w-v--eu--v------wwTep w,----- w~-we,,--ewgr pp-e-- *i-y---wm -,-nr,ye- vc-e-e-

crankshaft axis. The maximum length was approximately 2 inches and the depth 1/16 inch at the radius / bore junction. These two pairs of cracks at opposite ends of the hole were at 65 to 70' to each other i.e., they were close to exhibiting the 90* 'X' formation typical for cases where torsionally-induced alternating tensile stresses have caused fatigue cracking at points of high stress concentration and in areas where the mean torsional stress is moderate.

The No.10 main journal hole also had a pair of opposite cracks, the =

longest being 1-5/8 inches with a depth of 1/8 inch and running at 40* to the crankshaft axis. Here, also, the other end of the oil hole had two shorter cracks running at 45' to the crankshaft axis and at 90' to those at the other end of the hole. The No.11 main journal had one small crack 5/16 inch x 1/100 inch running at 45' to the crankshaft axis.

SCE undertook to remove these cracks by increasing the bore of the No. 9 hole from 15/16 inch to 1-1/2 inches and by drilling the No.10 journal to 1-5/16 inches final size. The oil holes were ground and polished subsequent to drilling. This corrective work was done by a seasoned TDI craftsman. TDI and the OG have analyzed the repaired holes and have concluded that the change in stresses is not significant. Subsequent to this particular inspection, SCE had the remaining main journal oil holes and journal web fillets inspected for cracks (No.1 journal could not be accessed without major disassembly). Only very minor cracks were found in main journal No. 8 oil hole; these were polished out.

During the time of the NRC/SCE/PNL meeting on October 3 and 4,1984, the main journals No. 8, 9, and 10 of DG #2 were inspected, using both an eddy-current probe and fluorescent dye penetrant. Journal No. 9 oil ule showed minute cracking on one side of the shaft. The other side was for

to be clean. Both sides of No. 10 journal and both sides of No. 8 jour 31 were inspected and found to be clean. The cracks in No. 9 were considt.ed to be g sufficiently short and shallow to permit removal by polishing.

In collaboration with the OG, SCE undertook an extensive investigation of ,

the causes of the observed cracks and the potential for recurrence. The investigation included taking torsiographs to confirm the torsional vibration characteristics of the engine and to explore the torsional vibrating response during startup, load acceptance, load rejection, and coast-down.

3.12

Although the analysis of the torsiographs is not yet published, pre-liminary results were supplied to NRC and PNL in an OG meeting on October 22, 1984. Briefly, these results are:

e The torsional vibration characteristics are as expected; that is, TDI early analysis is confirmed.

e e A resonant vibration (critical) occurring around 240 rpm engine speed produces torsional vibration free-end amplitudes of up to 3.4* peak-to-peak under transient conditions of slow start and coast-down, 2*

under some fast starts, and approximately 4.5* under one fast start condition. This last angular deflection was determined to be large enough to cause the initiation of cracks in the oil holes. The other 7

transient deflections would be greater than the 10 -cycle endurance limit but were determined to be not large enough to initiate cracks, given the few numbers of cycles encountered to date in the life of the engine.

e The cause of the variability among the fast starts is conjectured to be a result of certain piston firing orders at the time the engine passes through the 240 rpm critical.

e Cracks in the oil holes that are initiated by the large angular deflections can propagate with the lower stresses of both transient and steady-state operations, j In response to the findings at SONGS-1, the OG arranged for inspections of both sides of the No. 9 journal oil holes in crankshafts at two other TDI DSRV-20 engines in municipal power company operations at Homestead, Florida, and Princeton, Illinois. The inspections were performed by a level two NDE p examiner from the Catawba Nuclear Station using fluorescent dye penetrant examination. No evidence of cracking was found. The Princeton, Illinois, engine, rated at 8,800 kW, is reported to have logged over 15,000 hours0 days 0 hours 0 weeks 0 months opera-tion, and the number of starts is known to be much larger than at SONGS-1. The Homestead, Florida, engine, also rated at 8,800 kW, has been operated over 17,000 hours0 days 0 hours 0 weeks 0 months and is estimated to have undergone 4,000 slow starts. Because s

I 3.13

neither of these engines has encountered a significant number of fast starts, this data is offered in support of the theory that fast starts have adversely affected the SONGS-1 DGs.

The SCE and OG interim conclusion and course of action is as follows.

Some of the fast start transients have caused cracks to initiate. These have propagated as a result of both transient and steady-state operations. All

affected oil holes have been repaired and reinspected. Recurrence will be avoided by avoiding fast test starts (NRC is in the process of modifying SONGS-1 technical specifications to remove requirements for fast test starts during monthly surveillance tests). Depending on the outcome of further analysis, SCE will institute other operational procedures such as barring-over the engine to " preferred" orientations prior to all starts. There is no commitment to this at present.

3.2.3.5 PNL Evaluation At this writing, the FaAA/TDI torsiograph tests of September 25 and 27 have not been fully evaluated and reported by the OG. However, some torsio-graph strip charts were received and reviewed by PNL consultants.- Evaluation and conclusions by PNL regarding the reliability of the V-20 crankshaft /

generator systems for interim licensing (to the next refueling outage) are dependent and contingent upon a confirmatory report by the OG concerning the preliminary information presented at the October 22 meetings at NRC in Bethesda. This report presumably will provide complete torsiograph test details and analysis.

To provide a rationale for the PNL conclusions and to provide a better definition of current PNL observations and concerns, the following comments are given.

s e PNL has reviewed the torstograph traces for several startup and coast-down transients and has observed the reported angular deflection responses. The explanation given by the OG (FaAA) appears

reasonable, although not yet formally quantified.

e PNL acknowledges the positive results noted in connection with the No. 9 oil hole NDE inspection done at the two DSRV-20 engines in 3.14

municipal power service in Homestead, Florida, and Princeton, Illinois. Because these engines have operated without crankshaft failure well in excess of 107 cycles at engine loads above those required for the SONGS DGs, the basic steady-state performance of the DSRV-20 crankshaft is confirmed. Without detailed knowledge j concerning items such as 1) crankshaft material strength and method of manufacture, 2) torsional vibration characteristics, 3) damping,

4) starting and coast-down characteristics (viz., faster deceleration with a lighter alternator would lead to fewer high-stress cycles),

etc., PNL cannot finally conclude that these results are entirely

, relevant to predicting fest-start performance. However, the engines have similar power output to that the generator characteristics should be similar and the location of the material, although shifted, ,

should be of comparable magnitude. Thus, in the opinion of the PNL consultants, the absence of No. 9 oil hole cracks in these engines lends support to the OG theory that fast starts initiated cracks in the SONGS-1 DGs.

e PNL believes that other factors could be contributing to the oil hole cracking and should be investigated as a backup effort to the study of fast-start transient effects. For example, it is known that the engine can reach an equivalent of 63 or even 64 cycles per second (480 rpm), which is at or near the 2.5 order critical (476 to 478 rpm). Operation at this critical in combination with a cylinder misfiring can produce large torsional stresses. Another example is that misfiring might have contributed to the variation in angular deflection on the startup transients. Misfiring could be a partial explanation of the lower amplitudes noted on two of the transients.

This is quite opposite to the effect noted above on the 2.5 order critical, where misfiring contributed to a considerable augmentation

. of the vibratory stresses.

j e PNL has concerns regarding the omission of the front-end gear train

! driven system from the forced-damped torsional analysis. PNL l considers it important to establish any augmentation of vibratory I

l 3.15

{

I

torque at the crankshaft gear / idler gear mesh as a result of the gear train characteristics, particularly during, but not limited to, the transient criticals. The vibratory torque at this gear mesh can be correlated with the mean power torque. PNL believes that gear tooth separation is probable (gear mesh vibratory force exceeds mean power force and teeth separate), somewhere in the 210- to 280-rpm range, depending upon camshaft, pump, etc., torque requirements, and the resulting impacting of the gear teeth should be given consideration, along with the relevant gear and bearing design margins, in .

evaluating component reliability. Further, close inspection of all gear teeth is considered to be essential to establishing engine reliability. This inspection should include tooth structural as well as tooth surface conditiors (already inspected).

3.2.3.6 PNL Conclusions PNL concludes, on the basis of information provided on October 22 and discussed above and the NRC actions to remove fast start requirements for monthly surveillance testing, that the V-20 crankshafts, as repaired, will be reliable for operation at SCE required loads (4400 kW) during the interim period to the next fuel outage in early 1986, at which time it is recommended that the oil holes be reinspected.

Although PNL is concerned regarding the front-end gear train, the engines are considered reliable because they will no longer be subjected to monthly surveillance fast start tests. PNL recommends, however, that NRC require that SCE inspect the gears, their teeth, and the idler bushing at the next extended reactor outage, and certainly at the next refueling outage.

PNL's conclusions are contingent upon the supposition that confirmatory analyses of the torstograph data, especially the load application data, do not ,

uncover unidentified problems.

3.2.4 Connecting Rods ,

Part No. 02-340A Owners' Group Report FaAA-84-3-14 3.16

3.2.4.1 Component Function Connecting rods transmit the power thrusts between the pistons and the crankshaft. They are of forged steel, either round or H-shaped cross section, in TDI vee engines there are two types of rods. One is a master rod serving one bank, which is directly connected to the crankthrow (or crankpin)

/ of the crankshaft. . The other is the articulated link or slave rod, and is connected to the master rod via a link pin. The master rod lower end is split diagonally across the crankpin annulus. The mating surfaces, generally known as serrated joints, are machined as racks (i.e., with gear-like teeth) and bolted together.

Rods can fail in various ways. Of greatest concern -in these engines is

, the possibility of breakage in the " rod box" in the vicinity of the link pin and/or failure of the bolts across the mating faces. Major failure will lead to instant damage to major parts of the engine. Lesser failures, as in the bearing support areas or link pin assembly, may lead to damage to bearings and journals, link or wrist pins, or even to piston seizure or a loose connecting rod (with possible major damage to the engine).

3.2.4.2 Component Problem History Various connecting rod failures have been reported from the non-nuclear field. One failure mode was in the link-rod blade-to-pin bolting, due to loss of bolt preload. The other principal mode of failure was fatigue cracking of connecting rod bolts and/or the link rod box in the area of the mating threads. No connecting rod failures have occurred in nuclear service.

3.2.4.3 Owners' Group Status In light of the problems in a few specific installations (non-nuclear),

e the OG undertook an in-depth evaluation of all known potentialities. The OG determined that the loss of preload on bolting between the link rod and link pin stemmed from an incorrectly sized locating dowel. The dowel's excessive length prevented proper bolting contact between pin and rod. Under firing-load conditions the dowel would yield sufficiently in compression that the bolt pre-load would relax, with resultant fatigue problems. Replacement with dowels of

proper length, followed by proper bolt preload, corrects the incipient problem.

3.17

The second failure mechanism was fatigue cracking of the cross-joint connecting rod bolts and/or the link rod box at the mating threads. TDI attributed these rod cracks to " thread fretting," which they concluded resulted from distortion of the rod bolt under operating loads in the area of the mating threads. The distortion cocid occur even if the bolts had been installed with the originally specified bolt preloads. The Owners' Group addressed this ,

concern for the two versions of the connecting rod, namely the original design equipped with 17/8-inch bolts and a later design in which the rod boxes were equipped with 1 1/2-inch bolts. Stress analysis, including finite element studies, was done by FaAA. They concluded that both designs are adequate for the service intended, provided connecting rod bolt preload is regularly checked within specified time limits that are related to engine load requirement. How-ever, the rod with the 1 1/2-inch bolts was found to have an 8% to 9% greater margin of safety than the rod with 1 7/8-inch bolts because the related rod-box structure is more massive with the smaller bolt configuration. 'The Owners' Group recommends eddy-current inspection of the rod box threaded hole.

, Implementation of this recommendation has so far proven to be impractical.

Another area of concern was that of possible sideways bowing of the connecting rod, sometimes coming as a consequence of the forging process. FaAA computed the consequences and established a functional tolerance limit against which connecting rods should be checked.

The last area of possible failure was in the wrist pin (or piston pin) bushings, considered by TOI as a component of the connecting rod assembly.

Several original and replacement bushings, particularly at Shoreham, were found to have indications on both inner and outer surfaces. FaAA evaluated these as interdendritic anomalies (casting defects), having little functional significance but best replaced where encountered.

s 3.2.4.4 SCE Status SCE performed comprehensive inspections and examinations of the connecting ,

rods and related parts on both DG #1 and DG #2 engines during their complete

! disassembly in 1982, with no adverse findings.

t I

! 3.18 l

1

l

~

Visual inspection of 30% of the connecting rods was done in 1984. This inspection revealed no cracks, no fretted serrations, nor any galled bolts.

Liquid penetrant inspections of certain surface areas on the link rod box, j inner diameter of box bushings, and rack surface areas were deemed l

satisfactory. No indications were found on any connecting rod-box external

- surfaces.

In addition:

e Magnetic particle testing was done on all bolts and studs, with satisfactory results reported. .

e Material comparator and hardness tests were performed on th'e rods, rod-box areas, pins, and bushings, including spares. PNL was advised verbally that the results were satisfactory, but as yet had not been reported.

o Dimensional checks of link pin locating dowels were all found to be satisfactory.

e Bolts were torqued to standards, using a calibrated hydraulic wrench, and bolt stretch measurements were recorded by ultrasound.

e Liquid penetrant inspection (on bushings only) revealed no cracking or scoring.

e Magnetic particle examinations of the serrated or rack areas and rod-box external surfaces revealed no indications.

e Dimensional checks of link pin locating dowels all were satisfactory.

SCE did not check for connecting rod bowing because its inspection preceded the applicable OG report. Instead, SCE relied on inspection for 8

bearing and bushing wear patterns as evidence of rod distortion. The results were deemed acceptable. SCE inspected the bolts and mating threads at the rod-box joint on both DG #1 and DG #2. In 1984 only 30% of the rods were inspected; however, in 1982 all the rods and bolts were examined, and no indications of fretting or scuffed surfaces were found. SCE reports that no unfavorable indications were found on relevant areas of the wrist-pin bushings on either engine as examined by liquid penetrant (LP) methods.

3.19

As a consequence of their inspections and analyses, SCE believes that connecting rods of both engines now are fully adequate for their intended standby nuclear service.

3.2.4.5 PNL Evaluation and Conclusions TDI and the Owners' Group each have conducted extensive investigations and analyses of the connecting rod failures. PNL has not been able to reach final '

closure on the sufficiency of their results, but generally concurs with their conclusions as to the failure mechanisms, subsequent corrective actions, and ,

overall operability and reliability of the components if given sufficient surveillance and maintenance.

SCE has addressed the generic issue of potential connecting rod problems through a sampling disassembly and inspection and appropriate analysis, replacement and refurbishment. SCE has taken action to assure proper bolt preloads. However, PNL notes the following:

e No eddy-current tests of rod-box threads were conducted per 0G recommendations because SCE did not consider the results reliable.

PNL concurs with this action based on data available on this technique published to date.

e SCE did not lap the mating surfaces or check contact by " blueing" techniques in the reassembly of the' engines. (" Blueing" is a process confirming mating surface contact area by using a thin surface coating of a chemical that, when pressed or rubbed against a mating surface, will indicate where contact is achieved.) However, SCE said the surfaces looked good and no blueing was necessary.

e PNL does not agree with SCE's contention that bearing and bushing wear patterns will, of themselves, establish clear proof of rod ,

straightness. If indeed such bowing were to occur in forging, the process of boring for wrist, link, and crank pins should provide end-connection alignment. Then only rod flexure, over a prolonged period, would show up in noticeable bearing wear patterns. Mean-while, potential fatigue stresses would accumulate. However, this 3.20

area remains one of concern, subject to findings of the OG strain-gauge testing on connecting rods.

PNL concludes from all available and confirmatory information that the connecting rods on both DG #1 and DG #2 at SONGS-1 can safely and reliably perform their intended function through the next refueling cycle. This is stated with the following rationale and provisos:

e The SONGS-1 engine connecting rod boxes have 1 1/2-inch bolts (rather than 1 7/8-inch bolts); this provides a rod-box structure with a higher factor of safety in the area of concern.

e The SONGS-1 engines have very few service hours logged relative to service of engines when rod-box failures occurred (many thousands of hours).

e Engine operations will be limited to loads of about a BMEP of 114 psig. This will limit firing load effects on rod-rack surfaces and bolting, and on any rod bowing.

e SCE has agreed to conduct an adequate reinspection of the rod-box bolt tension at acceptable time intervals. SCE has agreed to also examine the condition of bolt-hole entrances and teeth, and rod straightness, at the next refueling outage on 30% of the rods not inspected previously.

3.2.5 Connecting Rod Bearing Shells l

Part No. 02-340B

Owners' Group Report FaAA-84-31 3.2.5.1 Component Function The connecting rod bearings interface the connecting rods with the crankshaft. They are of cast aluminum alloy with a thin babbitt overlay, and

, are furnished in two identical halves. They are lubricated under pressure, and l

a substantial flow of oil proceeds through machined channels in' the shells from the drilled crankshaft oil holes to the passageways within the connecting rods

! and on to the pistons and intervening bearing surfaces. The upper bearing half l is subject to the piston firing loads and therefore is more critical.

3.21

Failure can occur through inadequate oil flow or pressure, excessive or unplanned loadings, structural anomalies (from design or manufacture), or fatigue and erosion of the babbitt layer in crucial areas. Bearings are also subject to particle or chemical contamination from the oil, including water, or even the wrong oil selection for the duty, any of which can lead to failure.

The failure mechanism generally is gradual, and its onset can be detected by '

prudent surveillance of oil and filter conditions. However, a substantial structural problem, excessive cylinder loads, or heavy water contamination can ,

lead to rapid failure. This can affect the shaft, sometimes with irreparable results.

In light of the several conditions affecting bearings, the need for replacement is not uncommon. However, in customary service, bearing life generally is measured in multiples of 104hours, given reasonable service conditions, i

3.2.5.2 Component Problem History No significant failures of the TDI OSRV-type diesel engine connecting rod bearing shells have been reported in nuclear applications. However, some have been replaced because of deterioration due to inservice conditions or because they were found to be in nonconformance with Owners' Group recommendations regarding voids in the base material.

Various problems were encountered in the in-line TDI DSR engines at Shoreham. Edge cracking occurred, allegedly due to inappropriate bearing shell overhang beyond the support structure.

3.2.5.3 Owners' Group Status i The OG has investigated both the Shoreham (inline) and Grand Gulf (vee)

bearing shells. On their behalf, Failure Analysis Associates conducted various analyses. They concluded that the bearings are suitable for the intended service, provided 1) they conform to the manufacturer's specification and .

2) they meet the criterion for subsurface voids developed by FaAA for the Owners' Group. Indeed, on the basis of their analyses of the Shoreham bearing conditions (reflecting their brief service operating hours and loads, etc.),

3.22

FaAA concluded that the different, but generally similar, bearings in the vee engines can expect a 38,000-hour life at full load, if void criteria are met.

3.2.5.4 SCE Status SCE reported that in 1981 all the connecting rod bearings were replaced on both engines. This change was made in connection with the piston modifica-tion. Both DG #1 and DG #2 connecting rod bearings were inspected on a 30%

sampling basis in 1984 by the following means:

e visual inspection for scoring, galling, cracks, or excessive wear e liquid penetrant inspection for linear indications e radiographic inspection for internal defects.

The inspection results for DG #1 and DG #2 were:

e A number of bearings, as reported, showed some visible markings; they were deemed reusable, except as noted below. (Wear was not referenced.)

e One bearing showed a linear indication via LP test and was replaced.

Except for the above, the inspection showed no defective bearing shells. All were deemed reusable. SCE concludes that, with the evaluations and replacements made, the connecting rod bearing shells in the reassembled engines are " adequate for service through the next cycle of operation."

3.2.5.5 PNL Evaluation and Conclusions PNL representatives did not see the shells during the inspection of either engine, but the subject was reviewed with SCE staff at the San Onofre meeting on October 3,1984 PNL has reviewed the OG bearing reports.

+ Based on 1) a review of applicable OG reports, 2) the SCE inspection results, 3) the engine loading limitations, and 4) SCE's commitments to monitor oil pressure and condition, PNL concludes that the connecting rod bearings for both DG #1 and DG #2 will be adequate to operate reliably until the next refueling outage.

3.23

3.2.6 Piston Skirts Part No. 03-360AF Owners' Group Reports FaAA-84-2-14 and FaAA-84-5-18 3.2.6.1 Component Function The piston (as an assembly of piston crown and piston skirt, along with

rings, piston pin, et al.) receives the thrust of inertia and combustion and transfers it to the connecting rod. The cast steel crown carries the direct .

combustion pressure and thermal conditions; the skirt, of ductile iron, actually transfers the load to the piston pin / connecting rod and guides the reciprocating motion of the piston within the cylinder. Such a two-piece piston structure is relatively common to large, modern, high-output engines.

( In general, failure is most apt to reflect excessive pressure and thermal stresses, of both high-cycle and low-cycle character. Durability is affected by material selection and fabrication quality, as well as design characteris-tics. A crown separation will require immediate shutdown; it is likely to lead quickly to serious cylinder, head, and rod damage, and to piston seizure, with adverse impact on the crankshaft and possible crankcase explosion. Hence, adequate attachment of crown to skirt is a concern.

3.2.6.2 Component Problem History TDI has utilized several skirt designs, designated models AF, AH, AN, and AE, in their R-4 engines. Most early nuclear service engines were furnished with AF and AH skirts, although one plant received AN skirts.

Cracks have been found in a number of AF skirts, including earlier con'-

figurations of the TDI engines at Grand Gulf and Shoreham. The area of sensi-tivity was at a " boss" where the bolts join the crown and skirt together. Some ,

skirts also have the problems at the interface of an internal, circumferential rib and the piston-pin boss.

3.2.6.3 Owners' Group Status The Owners' Group consultant, Failure Analysis Associates (FaAA), analyzed the AF piston skirt design and concluded that the AF skirts will crack at the TDI nameplate rating (i.e., 248 BMEP), but that cracks will not propagate to 3.24 l

l

the point of actual functional failure. Modified AF skirts were considered to be adequate for service at full loads, provided they were 100% inspected for cracks in the stud boss area prior to use, and provided they are inspected periodically. The Owners' Group also concluded that, at reduced engine loads, the modified AF pistons would be suitable for extended service, provided no relevant indications were found during inspection of the stud boss area. FaAA calculated that the critical piston skirts stresses at SONGS-1 if the engines operated at 6000 kW would be 24.9% lower than those occurring at Shoreham.

3.2.6.4 SCE Status An inspection was initiated in about 1982 to look for cracks in the cylinder block. At this time the pistons, cylinder heads, rocker boxes, and connecting rods were removed from both engines. The pistons were inspected for cracks by liquid penetrant; one was found with a crack. This piston was replaced.

The cracked piston skirt problem was reviewed with TDI. They advised upgrading all the pistons by replacing the solid skirt-to-crown bolt arrangement with the double set of Belleville washers for holding the skirt to the crown. This change reduces the skirt boss stresses (see FaAA report 84-5-18). SCE followed TDI recommendations and, at the time the above change was made, all the pistons were also heat treated to improve their crack resistance.

A 25% sample of the modified AF piston skirts was examined in 1984 using OG draft criteria for inspection and acceptance. It was reported that the piston skirts were in good condition, showing no cracks or signs of scuffing.

The wrist pins and wrist pin bushings were also examined at that time.

SCE reported that both the wrist pins and wrist pin bushings were in good

-' condition and showed no sign of distress.

3.2.6.5 PNL Evaluation and Conclusions The following factors have been considered by PNL in reading conclusions regarding the adequacy of the modified AF piston skirts for service in DG #1 and DG #2 to the next refueling outage at SONGS-1:

3.25

i e The diesels at SONGS-1 employ AF piston skirts that have been modified and inspected per 10 CFR 21 notices. i e A 25% inspection was performed on both engines in 1984 following over l 180 hours0.00208 days 0.05 hours 2.97619e-4 weeks 6.849e-5 months of operation at or above 4400 kW on each engine, and no l cracks of any kind were found.

e According to the TDI Owners' Group, AF piston skirts have operated in the DSRV-20 engine at Homestead, Florida, over 10 7 cycles (740 hours0.00856 days 0.206 hours 0.00122 weeks 2.8157e-4 months ) at loads of 4000 to 5000 kW. Subsequent inspection (visual) revealed no cracks in any of these skirts. This result is consistent with FaAA analyses concluding that AF skirts are satisfactory for long-term service at conditions of operation below full-rated load.

Although this conclusion by the OG has not been finally evaluated by PNL, it is considered reasonable, especially in view of the very low engine loads at SONGS-1.

e FaAA calculated that the stresses at SONGS-1 DGs operating at 6000 kW would be about 25% below the stresses at the Shoreham plant. Opera-tions at 4400 kW at SONGS-1 will reduce the stresses even further. t To amplify this point, the rating of the engines at SONGS-1 corre-sponds to a BMEP of about 114 psi. This engine rating produces a firing pressure, as reported by SCE, of 1150 psi. This pressure is considerably below the 1670-psi reported pressure of the Shoreham engines.

On the basis of the above considerations, PNL concludes that the modified

{ AF piston skirts can be expected to operate reliably through the next refueling outage.

3.2.7 Cylinder Liners l Part No. 02-315C l

l Owners' Group Report FaAA-84-5-4 ,

3.2.7.1 Component Function Engines of this size and character are designed with individual, removable cylinder liners, which fit inside the cylinder block. The liners contain the 3.26 l

i

l pistons and are capped at the upper end by the cylinder head. Thus, they act as containment for the firing forces, subject to the stress and heat thereof, and the reciprocating travel of the pistons. The outer surfaces are cooled by jacket water circulating within the block. The lower end is sealed against an opening in the block floor with 0-rings. The upper end has an external, a circumferential ledge, which seats on the block's " liner landing." The head is gasketed and bolted in compression against the upper liner annulus, to seal in the high-pressure combustion gases. The liner is of nodular iron, selected for

~

5 its strength, castability, and durability against the scraping action of the pistons and rings.

' Liners generally do not fail, but they can be adversely affected by inadequate or inappropriate lubrication, the forces and heat of the combustion processes, the character of the pistons and rings, and the quality of fuels.

Failure most often is in the form of scoring by broken rings or carbon deposits, or " scuffing" by the action of the piston on the cylinder walls, due to one or more of the factors mentioned. If such conditions are severe enough, a piston will seize and cause significant damage to liner, head, and connecting rod, and even to the crankshaft. A crankcase explosion can result.

3.2.7.2 Component Problem History Only one incident of cylinder liner " failure" in nuclear service is known. This failure occurred in 1982 at Grand Gulf when a piston crown separated from the skirt during testing of the Division II engine and marred the liner.

3.2.7.3 Owners' Group Status The OG included considerations of liners in their study of cylinder

, blocks. Two concerns were uncovered:

e The TDI design calls for the liner to protrude slightly above the top

( ,

deck of the block, to ensure a tight, compressive fit against the head and gasket. However, this produces bending moments in the head and substantial shear stresses on the cast iron liner landing of the l

3.27 l . .. . . _ - - . . _ . __

block. Both aspects ara suspect in some of the real or incipient failures in those components. TDI has approved remachining to reduce the protrusion.

e The design also calls for a tight fit between the outer ring of the liner ledge and the matching counterbore of the block. There is some concern by the Owners' Group that this could increase hoop stresses

in the block, which might lead to block cracks. TDI has approved reducing this fit in the cylinder block.

3.2.7.4 SCE Status As reported verbally to PNL on October 3,1984, SCE verified that all cylinder bore dimensions were satisfactory in its 1982 inspection. All liners also were inspected for signs of interior wear, scoring, or scuffing. No scuffing or scoring was found; however, no liners were removed from the cylinder blocks. At this time all liners were honed to ensure that the new skirts and rings would seat properly.

In 1984 SCE visually inspected 30% of the liners in place in DG #2 and found no unusual wear.

SCE concludes that the liners in these two EDGs at SONGS-1 are adequate for nuclear standby service.

3.2.7.5 PNL Evaluation and Conclusions PNL representatives viewed the bottom ends of a few liners from the side cover opening during the crankshaft inspection. The liner surface looked good. In PNL's view, SCE has adequately inspected and honed the liners surface. Regarding the possibility of hoop stress induced cracks, PNL notes that there were no relevant findings for liners at SONGS-1 and that SCE has agreed to inspect the blocks for cracks when the liners are removed. SCE plans 1

! to implement any Owners' Group recommendations in this regard. Therefore, PNL,

! concludes the liners are suitable for nuclear service for the next refueling .

outage.

3.28

L 3.2.8 Cylinder Heads Part No. 02-360A Owners' Group Report FaAA-84-15-12 3.2.8.1 Component Function

.- The cylinder head caps the cylinder, providing (along with the liners) the 4

enclosure needed to direct the combustion forces against the piston. Its lowest surface, facing the cylinder, is known as the firedeck. In the TDI design there are two inlet valves and two exhaust valves in the flat surface of the firedeck, plus the fuel injector air starting valve and test cock. All

these openings and their associated passageways have to be cast into the structure of the head, which, in itself, must also contain substantial internal jacket water passages for cooling. In addition to the firedeck are a top deck or enclosure and an intermediate deck providing structural rigidity and control over jacket water flow.

The head is bolted to the cylinder block via a number of studs extending through the head from the block. On top of the cylinder head are two more components: the subcover or rocker box, which supports the valve actuating mechanisms, and a light top cover.

The TDI R-4 heads are cast of steel alloy. Casting a head of this complexity is difficult, particularly in steel. The internal passages are achieved via casting cores, which are challenging to hold in place during casting. Consequently, such heads have had a tendency to have uneven and/or incomplete sections. These can lead to a variety of flaws or indications, some of which can be repaired in the manufacturing and machining process.

Failures have tended to be mostly rather superficial linear indications i with no consequential results. However, some deficiencies lead to warpages or cracks. The latter, if through to the jacket water passages, will result in

. leaks of water into the cylinder when the engine is down, and of combustion gases into the jacket water during operation. The former can result in water in the ;ylinder, which could severely damage head, piston, rod bearings, and shaft on startup.

3.29

3.2.8.2 Component Problem History Numerous reports on TDI cast steel cylinder head failures are available from both the nuclear and non-nuclear industry. For identification purposes, TDI cylinder heads have been classified by FaAA as Group I, II and III, all under the same part number. Group I heads include those cast prior to October 1978; Group II heads are those cast between October 1978 and September 1980; and Group III comprises heads cast after September 1980. The distinctions involve both design changes to facilitate casting control and general quality control improvements. Most instances of cracked heads have involved Group I.

Only five instances of water leaks in Group II and III heads have been reported, all in marine applications. Most of the reported cracks initiated at the stellite faced valve seats.

The most recent known head failure was reported by Mississippi Power &

Light relevant to their Division I TDI diesel engine at Grand Gulf (Letter AECM 84/0401 to NRC dated July 30,1984). It reported a 2-inch through-wall crack in the right exhaust port casting surface between the valve seat area and the exhaust valve guide. This allowed jacket water to leak from the head cooling passages into the cylinder cavity, and was detected by barring-over the engine with cylinder cocks open. The specific head group classification of this head was not reported. However, the affected head was supplied with the engine and hay undergone 1500 hours0.0174 days 0.417 hours 0.00248 weeks 5.7075e-4 months of operation. Of this total, approximately 335 operating hours were at 100% load (7000 kW, 225 psig BMEP) and 31 hours3.587963e-4 days 0.00861 hours 5.125661e-5 weeks 1.17955e-5 months were at 110% load. This failure is still undergoing investigation; however, because MPAL knows of no occurrence of other similar failure, it concludes this was a unique, isolated event.

3.2.8.3 Owners' Group Status Failure Analysis Associates' mechanical and thermal stress calculations, which did not include finite element calculations, concluded that Group I, II, ant III heads, as designed, are adequate for the service intended. The report -

rec;nmends that Group I and II heads be 100% inspected by liquid penetrant, mag 4atic particle, and ultrasonic testing to determine firedeck thickness. For Grot 4 III heads, sample inspection as described above is recommended. For all 3.30

three groups of heads, FaAA recommended rolling the engine over before manual startup, with-cylinder cocks open, to assure no water has leaked into the cylinders. l 3.2.8.4 SCE Status In response to concern over heads of Group I, as originally furnished on the SONGS-1 engines, SCE advised that in 1981 all the heads were inspected visually and with liquid penetrant. In 1984, 30% of the heads were inspected and seat cracks were found; however, they were shallow. The heads were considered acceptable for use. No other cracks were found.

SCE reported that, during each refueling, about 25 to 30% of the cylinder heads are removed from the engine for inspection. These heads are sent to TDI for upgrading. In this way SCE will upgrade all the heads for DG #1 and DG #2. )

SCE stated that the new heads will be inspected before their use and that the firedeck thickness will be checked on old heads when removed from the engines for inspection.

3.2.8.5 PNL Evaluation and Conclusions .

In general, PNL concurs that, with the following provisos, the Group I TDI R-4 heads should serve satisfactorily on both DG #1 and DG #2 engines through the next refueling cycle:

e Engines should be air-rolled over with cylinder cocks open 4 to

' 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months , and again at 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months , after any operation, and thereafter prior to any planned start, to detect any water leakage into the cylinders.

e The engines should be limited to their emergency rating of about a

. BMEP of 114 psi, as advised by SCE.

e SCE should document its commitment to rework the heads on a 25 to 30%

basis at each refueling outage.

8 l

3.31 1

1

3 3.2.9 Cylinder Head Studs Part No. 02-315E Owners' Group Report Emergency Diesel Generator Cylinder Head Stud Stress Analysis (SWECo March 1984).

3.2.9.1 Component Function

~

Eight studs per cylinder are used to bolt the heads to the cylinder block. Together they transmit the power load from the head to the block.

Head bolts are not normally found to stretch or break; however, these occurrences are possible, due to faulty design, materials or fabrication, or excessive firing pressure. Fatigue failure is a greater concern, given reasonable operating conditions. This will occur if preload is insufficient and the bolts go through many cycles of loading. Once a bolt yields or breaks, its neighbors must carry increased burden, and the head is unevenly stressed.

This generally results in escaping combustion gases, with the attending hazards of heat and fire, as well as physical and metallurgical damage to head and block.

3.2.9.2 Component Problem History i

TDI has employed two basic stud designs recently. One is of straight j shank diameter. There has been concern that its tight fit within the block i

stud opening, coupled with inadequate preload, could put side thrusts on the block and contribute to block fractures. A second design uses a necked-down shank. This design not only avoids any possible stud-to-bore contact, but reduces the preload needed to maintain positive stresses during the firing l cycle.

l To date, no failure of cylinder head studs has been reported in the

! nuclear industry. However, some isolated failures have been reported in the t non-nuclear field. The cause has not been established.

3.2.9.3 Owners' Group Status .

Stone & Webster Engineering Corporation (SWECo) has analyzed both the old design studs and new necked-down studs developed by TDI to minimize potential 3.32

cylinder block cracking, and has concluded that both stud designs are adequate for the service intended, provided proper stud preload is applied.

3.2.9.4 SCE Status The SONGS-1 engines were furnished with the straight-shank stud design, and to date no failures have occurred. The studs have not been replaced, nor have the top two threads been machined off these studs to help reduce the possible cylinder block cracking from the studs to cylinder liner bore.

In 1981 all cylinder head studs were retorqued. In 1984 30% were visually inspected and retorqued following engine disassembly and inspection. The engines operate at a BMEP of 114 psi and the firing pressure is only about 70%

that of the Shoreham engines.

3.2.9.5 PNL Evaluation and Conclusions PNL concludes that the studs now installed on DG #1 and DG #2 will be reliable for nuclear standby service, at least through the next refueling outage. This conclusion is based on the following considerations:

e No failures of cylinder head studs have occurred at SONGS-1.

e SCE's action of inspecting and torquing the studs is deemed acceptable.

t e SCE has confirmed to NRC that these bolts were installed with proper preloading.

3.2.10 Push Rods l

Part No. 02-390C & D l Owners' Group Report FaAA-84-3-17 e 3.2.10.1 Component Function Push rods transmit the cam action from the camshaft on the engine side to

, the intake and exhaust valves in the head. One main rod extends from the camshaft to the subcover where it acts directly on the intake valve rocker arms. The second main rod transfers cam action to an intermediate rocker in the subcover and'on through an intermediate push rod to the exhaust valve 3.33

rocker arms. They are subject to high-acceleration compressive forces as they respond to the cams. Fundamentally, these are steel tubes with rounded ends, to fit the various mating sockets.

A failure would, at the least, reduce valve action and, thus, cylinder performance. Total inoperability of a cylinder could result, but would not necessarily lead to immediate engine shutdown. Because these components are -

always in compression, failure modes are limited, assuming reasonably good design. -

3.2.10.2 Component Problem History TDI push rods originally had tubular steel bodies fitted with forged and hardened steel end pieces, attached by plug welds. An estimated 2% reportedly developed cracks in or around the plug welds. A " ball-end" push rod design introduced later consisted of a tubular steel body with a high-carbon steel ball fillet-welded to each end. This design proved to be prone to cracking at the weld. A third design, consisting of a tubular steel body friction-welded on each end to a forged plug having a machined, hemispherical shape, was then introduced. This third configuration is referred to as the friction-welded design.

3.2.10.3 Owners' Group Status Because industry (both nuclear and non-nuclear) had expressed concern about the continued integrity of TDI push rods, the TDI Owners' Group included the component in the known generic problem category for specific study and i resolution. Failure Analysis Associates has performed stress analyses as well as stress tests to 10 7cycles on a sample of the friction-welded push rods, at j

conditions simulating full engine nameplate loading. No sign of abnormal wear or deterioration of the welded joints or ends was observed. Other nuclear ,

owners have run these versions in actual service, with no adverse results, 7

beyond 10 cycles.

FaAA, in their analyses, concludes this design is serviceable as required, but does provide stipulations for inspection and action, including destructive examination of a random sample from each plant.

3.34

l 3.2.10.4 SCE Status During an early inspection of DG #1 at SONGS-1, a few push rods with the ball ends were found cracked. Even though they did not cause engine shutdown, they were all replaced with the forged end friction-welded push rods.

Furthermore, during the visit to San Onofre, PNL was advised that all the push

.* rods in DG #2 had also been replaced with the forged end rods. PNL was also advised that a magnetic particle inspection had been made of all the push rod

, ends.

SCE concludes that the new rods, as installed, are reliably serviceable for their standby nuclear service.

3.2.10.5 PNL Evaluation and Conclusions After reviewing the FaAA report, the SCE actions and reports, and examining push rods in extended service elsewhere, PNL concludes that such rods -

of the friction-welded design are satisfactory for their intended purpose in both DG #1 and DG #2.

3.2.11 Rocker Arm Capscrews i

Part No. 02-390G Owners' Group Reports Emergency Diesel Generator Rocker Arm Capscrew Stress Analysis (SWECo March 1984, July 1984).

3.2.11.1 Component Function 2

The rocker arm capscrews bolt in place the rocker arm shaft in the subcover assemblies. They are fairly standard bolting materials. A failure would weaken or cancel the restraints on a rocker shaft and cause malfunction of intake or exhaust valves. Reduced engine output would result.

4 3.2.11.2 Component Problem History Rocker arm capscrew failures at Shoreham have been reported. There have been no reports of similar failures elsewhere.

1 3.2.11.3 Owners' Group Status Stone & Webster Engineering Corporation, a consultant to the Owners' Group, has performed stress analyses of both the original capscrew design with 3.35

a straight shank (the type that failed at Shoreham) and a newer design incorpo-rating a necked-down shank. SWECo has concluded that both designs are adequate for the service intended. They have attributed the failure at Shoreham to insufficient preload.

3.2.11.4 SCE Status SCE examined the capscrews from the disassembled heads with favorable results. No defects were found in the capscrews (or rocker boxes). However, no material verification of the capscrews was performed, as had been

recommended by the OG. SCE has committed to this verification at the next refueling outage.

The capscrews were reinstalled at recommended torques. SCE concludes--

subject to favorable material verification--that the capscrews are properly serviceable for their intended function on the SONGS-1 engines.

During the rocker arm capscrew inspection, the rocker boxes were also inspected for cracks in the capscrew hole area. Cracks in this area have been found at other installations, but none was found on either of the San Onofre engines.

3.2.11.5 PNL Evaluation and Conclusions PNL concludes, from the OG analyses and the favorable inspection results at SONGS-1, and from observation of high-cycle operating results on identical components elsewhere, that SCE's conclusion is acceptable. PNL notes the SCE connitment to a material verification at the next refueling outage.

3.2.12 Turbochargers .

Part No. MP 017 (Model 65G)

Owners' Group Report FaAA-84-6-56 g 3.2.12.1 Component Function The turbochargers on the SONGS-1 TDI DSRV-20-4 engines are Model 65G units manufactured by the Elliott Company. These turbochargers (four per engine) provide pressurized air to the cylinders for combustion of more fuel than would be possible with a "normally aspirated" engine. The turbochargers consist 3.36

1 principally of a turbine, driven by engine exhaust gases, directly driving an air compressor wheel. The associated housing ducts the air and exhaust to and from the rotors; the exhaust inlet guide vanes direct the exhaust gases toward the turbine wheel blades. Turbine speed changes with engine load (i.e., gas volume, pressure and temperature), with maximum speed depending on specific turbine selection and design parameters.

Because close tolerances and high rotating speeds are necessary for effi-ciency, and because temperature levels can approach 1200*F at the exhaust inlet, all components are sensitive to temperature, pressure, structural loads (vibrations), and contaminants or particles in the gas and air streams. The radial and thrust bearings require particular care and lubrication.

Vanes and blades are sometimes lost due to heat and vibration, or frac-tured by impact of particles, such as bolt heads, fractured vanes, or valves.

Undue stresses from connected exhaust piping or inappropriate supports can cause rotor wear at stator interface. Inadequate bearing lubrication (ar.d the cooling the oil provides) leads to bearing failure. Depending on the severity of the situation, shutdown can come quickly, but usually is not immediate.

3.2.12.2 Component Problem History i Various problems have occurred in the turbochargers on TDI DSR-4 engines i

in nuclear service. The principal one has been the rapid deterioration of the combination turbine thrust / radial bearings. There also have been concerns over missing exhaust inlet vanes, missing or broken bolts joining the exhaust ,

manifold to the turbocharger at the inlet, and broken bolts and welds in i support mounts. To date, thrust bearing problems have evidenced themselves at the Comanche Peak, Shoreham, Catawba, and San Onofre nuclear plants.

i Because nuclear EDGs have, to date, had unusual quick-start requirements--

and are tested extensively to assure reliability for such duty--the owners and l

l TDI investigated the failure parameters early in the history of such service.

It was recognized that the bearing and bearing lubrication systems inherent in l

the 6SG design did not provide adequate lubrication on the bearing thrust pads and rotor thrust collars under fast startup conditions to high loads. TDI instituted two steps of modifications in an attempt to address this problem; 3.37

- ----. 9ge--~

one instituted and modified the oil drip system and the second provided for manual prelubrication prior to planned starts.

3.2.12.3 Owners' Group Status On behalf of the Owners' Group, FaAA undertook an extensive study of causes of reported failures in nuclear service. The net result was an affirmation of inadequate startup lubrication. Briefly, the resulting -

recommendations were:

e Retain and use a " drip system" that directs a small flow of oil toward the bearings at all times in standby, but increases the flow of oil,to 0.35 gph. (Higher flows are apt to flood past the bearing into the exhaust manifolds and create fire risk.)

e Provide and use an auxiliary prelubrication pump to direct substantial flow to the bearings immediitely prior to planned startups.

e Maintain oil filtration at 10 microns or better and utilize spectrochemical and ferrographic oil analysis regularly, o Enhance bearing inspection programs. At least one bearing should be inspected at a station following every 100 starts of any nature.

Inspection should also be done following 40 starts without manual prelube.

i An 0G supplementary report dealing with turbocharger vanes and inlet capscrews has yet to be released.

3.2.12.4 SCE Status In 1980 and 1981 the SONGS-1 thrust bearings were found to be unacceptable (bearings were wiped) and were replaced. At that time SCE instituted the drip ,

j and the full flow auxiliary prelube system on DG #1 and DG #2. When the engines were inspected again early in 1984, the turbochargers revealed no significant wear. Between 1980/81 and 1984, DG #1 and DG #2 had undergone over 100 additional starts and 100 hours0.00116 days 0.0278 hours 1.653439e-4 weeks 3.805e-5 months of additional operation.

In addition to these maintenance efforts, SCE has committed to the basics of the OG plan for turbocharger modifications, operations, and maintenance, 3.38

l including inspection of the turbocharger thrust bearings if any engine experiences 40 fast starts (starts without manual prelubrication of the bearings). Also, consideration is currently being given to coast-down lubrication of the turbocharger bearings.

After making the cited changes, and in light of the OG/FaAA analyses that

. claim satisfactory O/R if their recommendations are followed (to which SCE has agreed), SCE concludes these turbochargers now installed will adequately perform their intended function .through the next operating cycle.

3.2.12.5 PNL Evaluation and Conclusions PNL has reviewed the FaAA report referenced above, the results of the June 22,1984, meeting among representatives of FaAA, the Owners' Group, NRC, and PNL, and the inspection data presented by SCE. PNL also has examined the prelube system at other, similar plants. On these bases, PNL concludes that a similar new prelube system now installed on the diesels at SCE will provide sufficient additional lubrication to augment the protection of the turbocharger bearings during planned fast starts. Further, in PNL's view, the few unplanned fast starts that may occur without prelube will not lead to bearing failure prior to the next refueling outage. According to Failure Analysis Associates, as confirmed in a telephone conversation between PNL (W. Laity) and FaAA (T.

Thomas) on July 20, 1984, the shortest known time-to-failure of a turbocharger thrust bearing subjected to " dry" starts (for which no bearing prelubrication was provided) occurred at the Shoreham Nuclear Power Station. That bearing experienced at least 62 " dry" starts before failure. The operating history of the SONGS-1 OGs suggests that each engine is likely to experience few, if any,

" dry" starts in a given operating cycle.

PNL also notes that SCE has established a planned program of relevant surveillance and maintenance and, at the next refueling outage, has agreed to implement the OG recommendations for inspections. SCE has also committed to comply with OG recommendations regarding capscrews, vanes, and mounting and supports that may result from the Phase 2 DR/QR.

PNL notes that the engines at SONGS-1 are run at a BMEP of about 114 psi. This engine rating is only slightly above the naturally aspirated 3.39 I

rating of diesel engines. The air intake manifold for both banks is joined together, so if one turbocharger should fail for any reason during emergency operation, the other three turbochargers would supply sufficient air to sustain operation until a repair could be effected.

On the bases of the above considerations, PNL concludes that the turbo-chargers et SONGS-1 on both DG #1 and DG #2 are adequately operable and .

reliable unti? the next refueling outage.

3.2.13 Jacket Water Pump Part No.02-425 Owners' Group Report Supplement to Emergency Diesel Generator Engine Driven Jacket Water Pump Design Review (SWECo July 1984).

3.2.13.1 Component Function The engine driven jacket water pump furnishes water to the engine jackets (i.e., the cylinder block surrounding the liners) and to the heads. Water is also sent to the turbocharger jackets. They are customary centrifugal pumps, driven by a power takeoff from the front-end gear case.

Without the pumps (or an emergency backup), the engine will quickly shut down due to excessive temperatures. Such pumps generally are trouble-free, but occasionally develop problems of shaft seals, bearings, and drige mechanisms.

3.2.13.2 Component Problem History A TDI engine at Shoreham has experienced a jacket water pump shaft failure. There is no history of failures on jacket water pumps designed for the DSRV-20-4 engine.

3.2.13.3 Owner s' Group Status l Stone & Webster has investigated the jacket water pumps as installed on the TDI inline and vee engines. They reviewed these jacket water pumps from the standpoints of mechanical design, material suitability, and hydraulic .

performance. Stone & Webster found the pumps such as those installed on the l

l 1

l 3.40 i _ - _ _ _ _ __.__ -

San Onofre DG #1 and DG #2 engines to be acceptable, with a recommendation that a limiting torque be established for the pump shaft nut holding the " external spline" in the shaft taper.

3.2.13.4 SCE Status During the engine overhaul and inspections after approximately 200

operating hours at engine loads at or exceeding 4400 kW, the water pumps for DG #1 and DG #2 were examined. The water pump drive gear of DG #1 showed a

. heavy wear pattern. The drive gear was replaced. During another inspection a similar condition occurred. At the third inspection it was concluded that th'e water pump drive was not aligned properly, so it was realigned in accordance with the manufacturer's recommendations, in 1984. At the time of the inspec-tion, an oil leak had been noted. However, it was repaired by replacing a gasket.

The water pump on DG #1 was disassembled for the 1984 inspection and will be assembled by TDI. TDI will also torque the impeller onto the shaft properly. The water pump for DG #2 did not exhibit wear or apparent alignment problems.

The predicted satisfactory operation of water pump drive was based upon theoretical torsional vibration calculations made by FaAA and reported in FaAA 84-6-54 (June 1984). Recently engine DG #1 was torsiographed. The preliminary results of these torstographs confirmed the earlier calculations.

SCE concludes that, with these steps taken and the spline nuts properly torqued, the pumps are suitable for their intended service.

3.2.13.5 PNL Evaluation and Conclusions On the basis of the analysis, inspections, and repairs cited above, PNL p concludes that these pumps are serviceable for their intended use in the SONGS-1 DG #1 and DG #2.

3.41 i

3.2.14 High-Pressure Fuel Oil Tubing Part No. 02-365C Owners' Group Report Emergency Diesel Generator Fuel Oil Injection Tubing (SWECo April 1984).

3.2.14.1 Component Function ,

The high-pressure fuel oil tubing carries the fuel oil from the cam-driven injection pumps on the engine sides to the injector riozzles (spray nozzles) in the heads. This oil is under pulsating and quite high pressure (-500 psi to 15,000 psi once each cycle); hence, any flaws in the steel tubing or fittings used, or any breaks caused by vibration or other factors, will release an oil spray in high-pressure bursts, with consequential personnel and fire risks.

3.2.14.2 Component Problem History High-pressure fuel tubing leaks have developed during preoperational engine testing on Shoreham and Grand Gulf engines. No other failures in nuclear applications have been reported.

3.2.14.3 Owners' Group Status '

Stone & Webster has analyzed the failed high-pressure fuel tubing and has concluded that the failures originated in inner surface flaws that were intro-duced during fabrication. If, through eddy-current inspection, the inner surface condition of ned tubing is found to be within the manufacturer's specification, Stone & Webster has concluded the high-pressure tubing is

suitable for the service intended. It was also recommended, however, that all future replacement lines be of a superior material and be " shrouded" with a i, sock to protect against open oil sprays in the event of future leakages.

(

3.2.14.4 SCE Status g j In 1984 a boroscope examination was conducted on the fuel lines on both i engines; no defects were found. On the basis of this favorable inspection ,

6 result and the operation of the engine to over 2 x 10 cycles without a fuel l line incident, SCE has decided to continue engine operation with the existing high-pressure fuel oil tubing.

3.42

[

l i

SCE has agreed to inspect the high-pressure fuel tubing at the next refueling outage by the eddy-current technique as recommended by the Owners' Group. SCE stated that the inspection is contingent on the availability of operable eddy-current test equipment. They plan to do this inspection on a sampling basis.

3.2.14.5 PNL Evaluation and Conclusions On the basis of 1) the favorable boroscope inspection, 2) the successful

- operation to at least 2 x 106 cycles, and 3) the light loading on the SONGS-1 diesel generators, PNL concludes that the fuel lines for the SONGS-1 diesel engines can reliably serve their intended function until the next refueling outage. At that time PNL recommends that SCE conduct a 100% inspection (rather than a sampling inspection) to Owners' Group specifications.

3.2.15 Air Starting Valve Capscrews Part No.02-359 Owners' Group Report Emergency Diesel Generator Air Start Valve Capscrew Dimension and Stress Analysis (SWECo April 1984).

3.2.15.1 Component Function These capscrews bolt ia place on the head of the air start valves, which admit starting air to the cylinder. A failure, or an inappropriately long capscrew, will not keep the starting valve assembly in correct contact with its l seat, with consequential risk of damage as high-pressure combustion gases escape.

3.2.15.2 Component Problem History No actual failures of these capscrews have been reported. However, on i May 13, 1982, TDI reported a potential defect due to the possibility of the

~

3/4-10 x 3-inch capscrews " bottoming out" in the holes in the cylinder heads, resulting in insufficient clamping of the air start valves.

3.2.15.3 Owners' Group Status Stone & Webster and TDI both have recommended that the 3-inch capscrews be either shortened by 1/4 inch or replaced with 2-3/4-inch capscrews.

3.43

--. , , . _ - ..-. - _ - _ _ = _ . -

3.2.15.4 SCE Status Upon receiving a 10 CFR 21 report from TDI in 1982, SCE checked all capscrews, found them to be the correct ' ength, and confirmed that they were not " bottoming out." Torque checks showed the capscrews did not loosen. Also, the gaskets did not lose their crush.

In the supplemental report, the OG recommended that a sampling of -

capscrews be checked for material selection. SCE will comply with this recomendation. c SCE maintains that, on the basis of their 100% inspection with no irregularities having been observed, the capscrews and their reinstallation meet TDI and OG requirements and are adequate for standby nuclear service.

3.2.15.5 PNL Evaluation and Conclusions The inspections and actions taken by SCE to eliminate potential problems are judged to be adequate to prevent any subsequent failures. PNL notes that SCE has committed to conduct a material verification test per the OG recommendations. PNL therefore concludes that, with the continued use of SCE's installation procedures to control torque of bolts, studs, and screws to specified ranges, these components will not present future problems on the SONGS-1 engines. Thus, PNL concludes these components on DG #1 and DG #2 are reliable for their intended service.

3.2.16 Engine-Mounted Electrical Cable Part No. 02-6888 Owners' Group Report SWECo No. DR4-210-013 3.2.16.1 Component Function These cables serve the Woodward governor / actuator and the Air-Pax magnetic I pick-up, both mounted on the engines. Inappropriate cable materials, not able to withstand the temperature or service environment, could lead to short ,

circuits, with adverse impact on the component functions and possible risk to personnel.

3.44

3.2.16.2 Component Problem History Two defective cables were recorded by TDI in a 10 CFR 21 report. Also, a TDI Service Information Memo warned of potentially defective engine-mounted cables.

3.2.16.3 Owners' Group Status Analyses of the subject wiring, and of the recommended replacements, were conducted by Stone & Webster Engineering Corporation, both generically and specifically for San Onofre. The replacement cable and terminations were deemed serviceable for this duty.

3.2.16.4 SCE Status In response to the original service information, a complete review of all engine-mounted cable on the 50 HGS-1 engines was conduct'ed by OG engineering personnel. No defects were noted. New cable was installed in accordance with OG recommendations. Based on this action, SCE did not reinspect the cable during the recent inspection Jrocess.

SCE concludes that the eigine-mounted electrical cable at SONGS-1 is sicicable for its intended ndcitar standby service.

3.2.16.5 PNL Evaluation _and Conclusions On the basis of the acti(ns taken by SCE, PNL concludes that the subject cables on DG #1 and DG #2 are serviceable for their intended use at SONGS-1.

! 3.2.17 Components Possibly Affected by Crankshaft Torsional Vibration Part Number Component 02-340A Connecting rod bearing shells

.' 02-350A Crankshafts i 02-350C Crankshaft gears

, 02-355A Crankshaft to pump gears 02-355B Idler gears 02-4108 Governor accessory drive gear 02-410C Overspeed trip couplings 3.45

-- ----,.--,---.<.----1-+-----e---r----- -

er.---- y-- , - - - - r - - - - * - - , - - ----*-+wr- m---T +

'-----------+-c-"1 r+ -

Part Number Component 02-411A Governor drive 02-411B Governor drive coupling, pin and keys02-426 Engine driven jacket water pump 02-420 Engine driven lube oil pump 02-445 Engine driven fuel oil pump 3.2.17.1 Component Function

~

The above identified components serve various functions, generally clear from their designation. All are deemed by the Owners' Group to be subject to loads resulting from torsional vibrations of the crankshaft.

3.2.17.2 Owners' Group Status The Owners' Group has addressed connecting rod bearing shells and the jacket water pump in its Phase 1 analysis of known problems (see Sections 3.2.5 and 3.2.13 above). The other 10 items identified here are under review by the Owners' Group as part of the Phase 2 component design review / quality revalida-tion activity. PNL has not yet received this Phase 2 analysis for the SONGS-1 DGs.

3.2.17.3 SCE Status Following a request by NRC at the October 22, 1984, Owners' Group meeting, SCE reviewed the potential for problems developing with 12 components identi-fied above. Because the component having the most direct interaction with the crankshaft (gears) evidenced no unusual distress due to the torsional vibration of the crankshaft, it was concluded that these components would continue to perform satisfactorily through the next refueling outage, considering the relatively few anticipated starts.

t 3.2.17.4 Phl Evaluation and Conclusions PNL believes that the reported favorable inspection of the gears is not ,

conclusive regarding the condition of the other affected components. The gears are sized to resist shocks from gear torque reversals, whereas other components in the front end may be more readily damaged.

3.46

O The first concern was that the crankshaft might have resonances at the 19.7 Hz natural frequency of the crankshaft. SCE has informed PNL/NRC that this is not the case; the TDI calculated fundamental frequency of the camshaft system is about 89 Hz. PNL has also reviewed the response to crankshaft torsional vibrations of each of the 12 identified components. In the judgment of the PNL consultants, none require inspection at this time.

6 Based on the information provided by SCE, recognition that the SONGS-1 DGs will not undergo monthly surveillance fast starts, and judgment of the PNL consultants, it is concluded that the 12 identified affected components are suitable for operation until the next refueling outage. PNL recommends that SCE review the response of these components prior to the next refueling outage and conduct relevant inspections at the next refueling outage.

.k

, 3.47 i

i

4.0 PROPOSED MAINTENANCE AND SURVEILLANCE PROGRAM In its evaluation of the Owners' Group Program Plan (PNL-5161, June 1984),

PNL stated its view that a comprehensive maintenance and surveillance (M/S) program is a key aspect of the overall effort to assure future TDI diesel

'4 engine reliability and operability. In docunents dated July 26 and August 20, 1984, NRC recommended using an augmented M/S program developed for the Grand Gulf Nuclear Station as a baseline for a program suited to the San Onofre Nuclear Generating Station Unit 1.

Southern California Edison's response regarding the maintenance and surveillance plan for SONGS-1 was documented in a letter dated August 28, 1984, to NRC. Included with that letter were several enclosures. Table A presented the existing SONGS-1 M/S program. A footnote to Table A listed deviations from TDI recommendations. Table B presented additions to the existing M/S program required to fulfill NRC recommendations.

This section documents PNL's review and evaluation of SCE's response to NRC's request and recommendations. In reviewing the SCE M/S plan, PNL organized the information into four groups:

o major maintenance items e proposed additional maintenance items e operational surveillance e standby surveillance.

Each is described and evaluated in the subsections that follow.

It should be noted that the SCE maintenance / surveillance plan for SONGS-1 has not been reviewed by the Owners' Group. Hence, some modifications to SCE's plan may be needed when the OG completes its review.

k 4.1 MAJOR MAINTENANCE ITEMS

, Table A of SCE's August 28, 1984, submittal lists many more maintenance tasks than are discussed below. Those that are not itemized here are judged to be beyond the scope of this effort, which is focused on key components.

However, PNL has added a few items where there is a perceived need.

4.1

Components classified by PNL as major maintenance items include engine structural and moving parts, as listed in Table 4.1. Parts with previously identified problems are also listed. The NRC, SCE, and PNL positions regarding inspections are displayed in columns to the right of each component / item.

Items are arranged in the same sequence used in Section 3 of this TER (i.e.,

structural components, power train components, ancillary and auxiliary p_

components and systems; generally from the bottom of the engine to the top).

4.1.1 PNL Evaluation and Recommendations .

The SCE M/S proposals do provide coverage of a number of items and systems considered key to maintaining engine reliability and operability. They should be deemed applicable to both engines. In reviewing SCE's proposals, PNL noted several important components and systems that were not incorporated in the list, as well as areas where SCE's proposal should be revised. The items listed in Table 4.1 are deemed by PNL to deserve periodic observation, evalua-tion, and maintenance, as appropriate. PNL's recommendations presented in Table 4.1 related to maintenance actions beyond the first refueling cycle (i.e., PNL concurrence with SCE's long-range maintenance plans) are necessarily tentative because the OG position on these items has not yet been identified and evaluated. PNL feels that NRC should require that the items listed in Table 4.1 be incorporated into SCE's surveillance and maintenance program.

These are:

e crankshaft e valve gear o foundation e studs, bolts e cylinder blocks e valve lifters e connecting rods e gear train e main bearing shells e turbochargers e connecting rod bearing shells e starting air strainers e pistons e lube oil e cylinder liners e overspeed shutdowns e cylinder heads e air start valves The following subsections present discussions to support PNL's recommendations for those cases where PNL recommendations differ significantly from SCE plans.

4.2

~ ,s~ ,,

TABLE 4.1. Major Maintenance Items for San Onofre Nuclear Generating Station Item NRC Guidance SCE Proposal PNL Recommendation Crankshaft Web deflection, After 270 hours0.00313 days 0.075 hours 4.464286e-4 weeks 1.02735e-4 months opera- At refueling Concur with SCE. Measure-hot and cold tion at or refueling ments to be done once each refueling cycle. Hot to start in 15 minutes and complete by 35 minutes.

Inspect oil holes none provided Oil hole in DG #1 to Concur with SCE.

for cracks. be inspected at next refueling outage.

Foundation Inspect and check none provided At refueling. Concur with SCE.

bolt torque

,4 Cylinder Blocks Top surface magnetic none provided none provided 30% heads-off inspection at particle examination next refueling. If negative, for cracks no other.

Connecting Rods Visually inspect and After 200 hours0.00231 days 0.0556 hours 3.306878e-4 weeks 7.61e-5 months opera- At refueling outage. Concur with NRC, except check check bolt preload tion or 9 months and every 285 hours0.0033 days 0.0792 hours 4.712302e-4 weeks 1.084425e-4 months .

before power level exceeds 5%.

Main Bearing Shells none provided Annually, use vibra- Concur with SCE but inspect tion analysis, web two locations at each time deflection, bearing of crankshaft crack inspec-temp alarm history, tions or at each second and lube oil analysis refueling outage.

to determine need for

's disassembly.

i s

TABLE 4.1. (contd) !

l Item NRC Guidance SCE Proposal PNL Recommendation Connecting Rod none provided At refueling, bump Concur with SCE. Also at Bearing Shells test for bearing wear. refueling, inspect four half shells, and radiograph.

Pistons none provided not listed Inspect two sets of pistons at first refueling outage.

Cylinder Liners none provided not listed Visual inspection every refueling outage. Measure /

record dimensions at each disassembly.

Cylinder Heads Air-roll before Ultrasonic for fire- Concur with SCE.

planned start, and deck thickness, 25% at 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months and 24. hours each planned outage.

after each shutdown. Also air-roll per NRC i guidance.

Valve Gear Visually check cams, Visually inspect Concur with SCE.

tappets, and pushrods camshafts, tappets, after 270 hours0.00313 days 0.075 hours 4.464286e-4 weeks 1.02735e-4 months , or at rollers, rocker arms, each refueling, pushrods, and valve springs at each refueling.

Head Studs, Check preload on 25% Check preload on 25% of Concur with SCE.

Air Start Valve - of studs, 100% of cap- studs, 50% of capscrews, Capscrews. Rocker screws after 270 hours0.00313 days 0.075 hours 4.464286e-4 weeks 1.02735e-4 months 25% of rocker arm bolts Arm Bolts or at each refueling, at each refueling.

4

. w .- 's , **

I i

j TABLE 4.1. (contd)

Item .NRC Guidance SCE Proposal PM. Recommendation

Valve Lifters none provided Adjust hydraulic valve Concur with SCE lifters at each refueling.

Gear Train none provided Inspect cam and idler Visual, each refueling.

! gears at refueling. Backlash, thrust, idler j Inspect lash, idler bushing and jets at first bushing, and oil jets refueling.

every 10 years.

Turbochargers Check one turbocharger for Same as NRC but check Concur with SCE. Also, l rotor float, and inspect at refueling only. Dis- disassemble and inspect stationary nozzle ring assemble every 10 years. one turbocharger at bolts after 270 hours0.00313 days 0.075 hours 4.464286e-4 weeks 1.02735e-4 months or next refueling outage.

at refueling.

"Y" Strainers in Clean and inspect Clean and inspect Concur with NRC Starting Air quarterly. annually (Table A)

- m System quarterly (Table B) i Lube Oil Sampling 1) At lube oil filter 1) and 2) same but only Concur with NRC j and Analysis inlet with engine running quarterly 4 monthly, or every 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months i running

) 2) Bottom of sump (check l

for water) monthly.

l l Intake Air none provided Monthly Concur with SCE.

, Strangulation Valves: Test l Air Start Valves none provided none provided Disassemble, check seat contact at each refueling l outage. .

i l

l ll 3

)

4.1.1.1 Crankshaft Deflection Checks SCE proposes hot and cold crankshaft deflection checks at each refueling outage, but does not commit to a time after engine shutdown to initiate and complete these checks.

Two purposes are accomplished in crankshaft deflection checks:

b e detection of gradual shifts in shaft support internal to the engine (most likely being significant bearing deterioration) e detection of changes in external engine support, as in the concrete foundation, or a shift of shims between the foundation rails and the engine base plate. (The foundation will change shape with prolonged engine operation, tending to hump toward the middle due to thermal growth, which must be corrected by appropriately shimming the engine. It may also undergo long-term change as chemical processes continue within the concrete.)

PNL Recommendations PNL recommends that SCE take hot and cold deflection readings at every refueling outage as proposed. The hot deflection checks should be taken immediately after the 24-hour preoperational testing, so as to reflect repre-sentative operational foundation temperatures. The hot checks should be initiated within 15 to 20 minutes after shutdown, and completed as rapidly as possible, preferably within 1/2 hour, starting with the last throw of the engine (generator end). Such a schedule, although strenuous, is deemed achievable.

4.1.1.2 Cylinder Block Top Surface SCE has not addressed cylinder block top surface inspection. PNL recommends an inspection with head removal at next refueling.

In 1984, DG #2 underwent a 30% magnetic particle inspection of the stud areas, with negative results. In a report to the TDI Owners' Group,(a) Failure (a) Failure Analysis Associates. June 1984. Design Review of TDI R-4 and RV-4 Series Emergency Diesel Generator Cylinder Blocks and Liners. FaAA-84-5-4, Palo Alto, California.

4.6

Analysis Associates stated " Engines that operate at lower maximum pressure and temperature than the SNPS [Shoreham Nuclear Power Station] engines (e.g., San Onofre) may have increased margins against block cracking that could allow relaxation of block top inspection requirements." The rated 6000-kW BMEP is 154.3 psig, and the higher emergency load of 4443 kW has a BMEP of 114.2 psig.

, However, without block material quality determination, some doubt remains.

PNL Recommendations

. PNL believes these blocks have a low danger of stud area cracking, but they should have a heads-off inspection between the No. 4 and No. 7 heads at the next refueling. If no crack is found, then no further inspection is indicated.

4.1.1.3 Connecting Rods SCE proposes to check bolt preloading at refueling outages. PNL recom-mends visual inspection of connecting rod boxes and checks of bolt preload every 285 hours0.0033 days 0.0792 hours 4.712302e-4 weeks 1.084425e-4 months of operation or at 9 months, whichever is first.

In light of the history in the TDI engine population (however limited) of connecting rod link-rod box cracking, bolting problems (viz., some galling, some preload relaxation, some failures), and fretting along contact areas of the serrated teeth, some regular visual inspection and bolt retorquing (or equivalent checking) is deemed warranted. The relevant Owners' Group report (FaAA-84-3-14) recomends that the interval on bolt retorquing not exceed 200 hours0.00231 days 0.0556 hours 3.306878e-4 weeks 7.61e-5 months of operation at full load, 248 hours0.00287 days 0.0689 hours 4.100529e-4 weeks 9.4364e-5 months at 85% load, or 285 hours0.0033 days 0.0792 hours 4.712302e-4 weeks 1.084425e-4 months at 75%

load. In making that recommendation, FaAA provided no differentiation 'between connecting rods having 1-1/2-inch bolts and those with 1-7/8-inch bolts.

Although the his' tory of 1-1/2-inch bolting is reportedly better, this configuration is not devoid of problems. Thus, even by the OG analysis, the establishment of an enhanced surveillance plan is deemed prudent. PNL has reviewed the OG analysis and concurs.

PNL Recommendation The San Onofre DGs are operated well below 75% of full loading. Thus, it is acceptable to use 285 hours0.0033 days 0.0792 hours 4.712302e-4 weeks 1.084425e-4 months as the bolt preload checking interval. There-fore, PNL recomniends visual inspection of all rod box external surface areas 4.7

and bolt preload check each 285 hours0.0033 days 0.0792 hours 4.712302e-4 weeks 1.084425e-4 months of operation after post-inspection reassembly or 9 months, whichever occurs first.

4.1.1.4 Connecting Rod Bearing Shells SCE proposes to measure bearing clearance without disassembly at every refueling outage. PNL recommends a sampling inspection of bearings themselves, as well as bearing clearance, at each refueling outage. s.

The Owners' Group design review report (FaAA-84-3-1) concluded that the bearings were adequate at site loads for the lifetime expected usage. SCE, in -

turn, appears to have based its inspection criteria on these findings. PNL is not in complete agreement with this philosophy because of the duty cycle of the engines and the high number of starts they will experience.

Each engine start effectively increases the rate of wear between 10 to 100 times the normal rate of wear on the bearings. In addition, putting the engines on high loads soon after starting also increases bearing wear rate more than does a more relaxed load application. Thus, the bearing wear may exceed the predicted rate. SCE's approach, therefore, requires modification to allow for visual inspection of bearing sets that may be suffering from galling, wiping, cavitation, or load-induced damage. This can be a sensitive area with aluminum bearings.

PNL Recommendations PNL recommends inspecting these bearings (two sets of pistons) by visual and radiography methods at each refueling outage; obtaining product oil contamination analyses; and monitoring all bearing clearances at every refueling outage.

4.1.1.5 Main Bearing Shells SCE proposes no direct inspections, relying on other indicators. PNL '

reconnends direct sampling inspection at the time of crankshaft inspection in addition to SCE diagnostic evaluation. .

SCE proposes to use vibration analysis, crankweb deflection, bearing temperature alarm history, and lube oil analysis results to determine annually the need for disassembly. No criteria were stated, and the effectiveness of 4.8

)

l l

these means in detecting a wiped or cavitation-eroded bearing is questionable.

. On the other hand, the relatively light loading and generally good bearing performance in other engines suggests a small likelihood of distress so large as to threaten the crankshaft.

In view of the cracks found in these crankshafts, it is certain that

-' future inspections will be required, at least in the No. 9 and No.10 main journals. From the crank arrangement, it appears that No. 6 is the most highly loaded shell, with Nos. 2, 5, 7, and 10 being next. It is suggested that the crankshaft and bearing shell inspections be coordinated.

PNL Recommendations The SCE annual diagnostic evaluation should be used. It should be augmented by visual and thickness checks, coordinated with shaft crack inspections, looking at two highly loaded bearings each time on a sampling basis, at each second refueling and whenever other clear opportunities are provided.

4.1.1.6 Pistons SCE provides no maintenance or inspection plan for the piston skirts. Pit.

recommends a sampling inspection at the first refueling outage.

The original model AF pistons were all inspected and modified by replacing the spherical washers with stacked Belleville washers in 1982. Twenty-five percent of the pistons were inspected again in 1984 with no cracks found.

An analysis of AE and AF pistons done for the OG for a peak pressure of 1670 psig concluded that cracks would start and propagate no deeper than 0.5 inches. San Onofre's rated load (6000 kW) has about 1150 psig and the larger service load (4400 kW) has about 1000 to 1070 psig. The OG has not made

$ final recommendations on inspection intervals for the AF pistons. They will be prepared following analysis of thermal loads and plant-specific operating

. conditions.

4.9

PNL Reconenendation PNL recommends that at least two sets of pistons (four pist(ns) be disassembled at the first reported outage and inspected for cracks per procedures recommended by the OG.

4.1.1.7 Cylinder Liners SCE provides no maintenance / inspection plan for the cylinder liners. PNL

recommends a visual inspection at each refueling outage and a recorded dimensional check for wear at every disassembly.

Cylinder liners now installed in SONGS-1 were machined and hcned prior to installation of the type AF piston skirts. However, SCE did not indicate any measurement of wear on the liners. Because liner wear provides an important indication of engine reliability and operability, it should be monitored whenever possible.

PNL Recommendations All liners should be visually inspected at each refueling outage, to check for any scuffing or metal deposition. In addition, a sample of the liners should be measured for wear at every disassembly, and the dimensions recorded for trend analysis.

4.1.1.8 Gear Train SCE proposes to inspect can and idler gears at each refueling; and gear lash, idler bushing, and oil jets every 10 years. PML agrees with the SCE-proposed frequency for, can and idler gear inspection, but recommends the other gear train parts be inspected at the first refueling outage.

The forward end gear train experiences the maximum torsional motion in the shaft system, producing torque reversals in the gears. The 20-cylinder engine p

will tend to have more amplitude at the gears than do shorter engines. With the discovery of torque-cycle cracks in the shafts, and the torsiograph information indicating large angular displacements, the durability of the gear

train is called into question.

4.10

l PNL Recommendations PNL recommends a visual inspection of cam and idler gears at each refueling. In addition SCE should check backlash, thrust, idler bushing, and oil jets at the first refueling. Results will determine future inspection intervals and procedures.

.. 4.1.1.9 Turbochargers SCE proposes to check one turbocharger for rotor float and inspect sta-tionary nozzle ring bolts at each refueling. SCE also proposes to disassemble, inspect, and repair turbochargers every 10 years. PNL recommends disassembly and inspection of one turbocharger at the next refueling outage, with subse-quent maintenance to be dictated by the findings.

Turbocharger thrust bearings were found to be failed in 1980 and 1981, and were replaced. The engines were then fitted with the drip system and manual prelube system corresponding to the Owners' Group recommendations. Tests done in connection with the 1984 engine inspection (thrust clearance tests) yielded satisfactory results.

The turbocharger on the SONGS-1 DGs sees an operating load of only 69% of the 225-BMEP at which this engine is nominally rated. This suggests the turbo-chargers are adequate for operations. However, starting has proved to be the major concern for these turbochargers, and the SONGS DGs are therefore stressed I

comparably to other DGs in nuclear service. In view of the failure history of these turbochargers and until the Owners' Group recommendations are approved by l NRC, PNL considers it prudent to inspect them, on a sampling basis, at

) refueling outages.

l PNL Recommendations

( PNL concurs with NRC recommendations to check one turbocharger for rotor float and inspect stationary nozzle ring bolts at 270 hours0.00313 days 0.075 hours 4.464286e-4 weeks 1.02735e-4 months operation (or at each refueling). PNL also recommends that one turbocharger be disassembled and inspected at the next refueling outage.

4.11

4.1.1.10 Lube Oil Sample and Analyses The SCE proposal and PitL recommendations differ only with respect to the time interval.

NRC advises that lube oil samples be taken from the lube oil filter inlet monthly or after 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months of engine operation, and at the bottom of the sump monthly; SCE proposes to do both quarterly. ,,

There have been cylinder head cracks leading to water leaking into cylinders. Some cracks but no water leaks have been found at SONGS-1. The air-rolling recommended will not detect minor leakage.

Lube oil checks serve two main functions. They reveal any water in the oil, indications of cracks in water-bound components, or leakage past lower liner seals. Such water can lead to lubrication failures, with potential major damage. Second, lube oil checks reveal abnormal wear of bearings and related engine parts.

It is important to collect and analyze samples with sufficient frequency that adverse conditions are detected early enough to avoid engine damage or outage.

PNL Recommendations

1. Check for water contamination after preoperational testing, and then monthly or after 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months of operation, whichever comes first; collect the sample from the bottom of the sump tank, preferably about 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months after engine shutdown, at the time of the engine roll-over.

2. Check for chemical and particulate contamination and imbalance near -

the close of preoperational testing, and then monthly or af ter

! 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months of operation, whichever comes first; collect the sample j (before the filter) while the engine is running, immediately prior to l l shutdown.

3. After the cylinder head inspection plan (including firedeck thick- '

ness) is complete, the interval should be reconsidered in view of the low load.

4.12 l

l l

l i

4.1.1.11 Air Starting Valves SCE provides no maintenance or inspection plan for the air start valves.

Pit. recommends disassembly and inspection at each refueling outage.

The capscrews mounting these valves have all been checked for inter-ference, and have been properly torqued. However, the large number of starts required of the SONGS-1 engines suggests an inspection plan is needed to ensure their operability.

PNL Recommendations PNL recommends that, at each refueling outage, the air start valves be disassembled and checked visually, and that valve seat contact be verified.

4.2. ADDITIONAL MAINTENANCE ITEMS 4.2.1 Rationale Maintenance on the following items of significance has been proposed by SCE as shown in Table 4.2. These items were not included in Table 4.1 because there is no direct relationship to known failures.

e fuel injection pump e lube oil keepwarm filter e fuel injection nozzle e lube oil filter e governor oil change e heat exchangers and intercoolers e air intake filter e jacket water system flush e fuel oil drip tank e engine balance e fuel oil filter and duplex strainer e governor drive caupling e lube oil in sump 4.2.2 PNL Evaluation PNL believes these items are important and that the PNL recommendations

- are consistent with good practice. They should therefore be given considera-tion in finalizing a M/S plan for San Onofre.

4.13

TABLE 4.2. Proposed Additional Maintenance Items for San Onofre Nuclear Generating Station Item SCE Program PNL Recommendation Note Fuel injection Send to shop for rebuild Verify calibration operation (1) pumps and calibration at every every third refueling.

Other refueling.

Fuel nozzles Pop tests at refueling. Pop test and spray pattern (2) at each refueling.

Governor oil Annual At refueling outage. (3) change Air intake Inspect and clean Concur with SCE.

filter annually.

Fuel oil drip Not listed. Check monthly. Clean as tank required.

Fuel oil filter Inspect, clean quarterly. Concur with SCE; also at (4) and duplex (Filter not mentioned.) pressure drop of 20 psi, strainers Filters to be included.

Lube oil in Change lube oil every Concur with SCE.

sump other refueling or as dictated by lube oil analysis.

Lube oil Change with oil change. Change with oil change or (4) 1 keepwarm filter 20-psi pressure drop.

Lube oil Drain water and sludge Also change with 20-psi (4) filter quarterly. Change with pressure drop.

oil change.

Lube oil and Disassemble, inspect and Concur with SCE.

Jacket water repair, as seen by trend-heat exchangers, ing of temperature and intercoolers pressure, or 10 years.

I Jacket water Alternate refuelings Concur with SCE.

system flush Engine balance Record firing and exhaust Concur, but add firing (5) temperatures. Adjust per peak pressure.

manual at refueling.

4.14 4

I

4 .

TABLE 4.2. (contd)

Item SCE Program PNL Recommendation Note Governor drive Inspect for wear at Concur with SCE coupling refueling Notes

s (1) Fuel injection pumps on the SCE engines have not been a source of problems.

SCE proposes to completely disassemble all pumps at every second refueling outage. Due to the precision and close-tolerance nature of the fuel injec-tion pumps, they can be damaged easily during a disassembly, thus requiring replacement of parts when otherwise unnecessary. Fuel injection pumps can be checked for proper operation and calibration at any reliable diesel ser-vice center; faulty or questionable pumps can then be put aside for dis-assembly. It is important to note that the same test should be performed on all pumps after reassembly, should they be disassembled.

PNL does not otherwise object to pump inspection every second refueling cycle, but suggests SCE verify calibration and operation of all fuel injec-tion pumps at every third refueling outage. Should other tests or operat-ing surveillance (i.e., cylinder firing pressure or exhaust temperature) indicate a potential fuel pump problem, verification of the suspect pump should be performed at that indication.

(2) Fuel injection nozzles are similar to injection pumps, in that very close tolerances are encountered; thus, they are also susceptible to damage during maintenance inspection. Proper testing of the nozzles for leakage,

" popping" pressure, and spray pattern would give a complete indication of the status of each nozzle. Then, only nozzles giving questionable results would need to be disassembled. The same tests should still be performed on all nozzles after reassembly, should they be disassembled.

PNL recommends checking " popping" pressure and spray pattern of all fuel injection nozzles at every refueling outage. Should operating surveillance (i.e., cylinder exhaust temperature) indicate a potential fuel injection nozzle problem, the suspect nozzle should be tested and, as necessary, disassembled.

(3) Along with every oil change there is a possibility of dirt entering the governor. Certainly the few hours of operation will not cause any l significant deterioration of the oil between refuelings.

PNL does not believe that changing the oil annually is wrong, but that the

, somewhat longer refueling cycle is both adequate and safer.

(4) The pressure gages are there to be used, and their use will limit cleanings to those that are needed. The fuel filter should be monitored for drop, because water could greatly restrict it.

(5) PNL assumes " Record firing" means record firing pressures.

4.15

4.3. OPERATIONAL SURVEILLANCE PLAN 4.3.1 Rationale Operational surveillance is necessary to ensure safe and efficient operation of the diesel engine. By monitoring and recording various engine parameters, trends in degradation may be noted, thus allowing preventive maintenance. In addition, trend monitoring permits engine shutdown prior to *.

major engine failure. A listing of recommended parameters and frequency of surveillance is presented in Table 4.3. ;

4.3.2 PNL Evaluation PNL concurs with the SCE operational surveillance plan for those items considered. There are, however, significant items that SCE has not addressed. PNL recommendations and justification for these items are given below.

4.3.2.1 Pre-Turbine Exhaust Temperature Pre-turbine exhaust temperature is valuable because:

e The individual cylinder exhaust pyrometer reports only a time average of highly variable function, e The turbine inlet temperature may be higher than any cylinder exhaust because of more hot puffs per time, and also because of possible exothermal reactions in the exhaust manifold.

e Blades and nozzle rings may be damaged by temperatures above the manufacturer's limit, which PNL believes to be 1200 F.

4.3.2.2 Air Manifold Temperature Air manifold temperature indicates the effectiveness of the turbocharger aftercooler, and its efficiency is dependent on water flow and temperature and ,

on fouling. The effects of elevated air manifold temperatures are reduced maximum load and less efficient combustion. Although such danger is less at 5 the low levels, it is considered prudent to monitor and trend the air manifold temperature.

4.16

TABLE 4.3. Operational Surveillance Plan for San Onofre Nuclear Generating Station itse imC Guidance SCE Proposal PNL Recomumendation Lube oil inlet pressure Hourly unless more Record once Log hourly frequent recording per hour is recossended by manufacturer

. Lube oil to turbocharger pressure Fuel oil to engine pressure Fuel oil filter differ.

ential pressure Left bank air manifold pressure Right bank air manifold pressure Lube oil filter differ-ential pressure Jacket water pressure (inlet and outlet)

Crankcase vacuum All cylinders exhaust temperature Exhaust temperature at turbine inlet Lube oil temperature (inlet and outlet)

Jacket water temperature (inlet and outlet)

Tachometer Hour meter Generator load i t ip Left bank air manifold None liste1 None listed temperature Right bank air manifold I temperature p Fuel oil transfer pump Log hourly unless strather differential strainer is auto /

, pressure duplexed and alarmed Starting air pressure Check hourly Fuel oil day-tank level 1r 3r Check hourly Visual inspection for Monthly or after Monthly or Concur with SCE leaks 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months operation after 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months operation 4.17 r e -, w,- -

4.3.2.3 Fuel Oil Transfer Pump Strainer Differential Pressure This pressure should be continuously monitored and recorded hourly unless the pump is automatically valved duplex with alarm to protect fuel feed.

4.3.2.4 Starting Air Pressure This pressure must be monitored to ensure sufficient pressure is available for restart at all times. -

4.3.2.5 Fuel Oil Day-Tank Level This level must be monitored to assure fuel availability.

i 4.3.3 PNL Recommendation PNL concurs with all operational surveillance addressed by SCE. In addi-tion, PNL recommends that all the items in "PNL Evaluation" be continuously monitored and recorded hourly.

4.4 STAN0BY SURVEILLANCE PLAN 4.4.1 Elements and Rationale Standby surveillance is important to ensure the reliability of the diesel engines. The parameters monitored on a " secured" engine show that it is pre-pared for rapid startup and load acceptance. The two factors that contribute most to this are engine temperature and lubrication. Thus, by keeping the engine warm and all oil passages pressurized, the time lag associated with load acceptance is minimized. In addition, a ready supply of quality compressed air is required for starting the engine. Recommended items of standby surveillance are shown on Table 4.4.

The SCE M/S proposal did not have a " standby" category. Entries in the SCE Proposal column of Table 4.4 are from their overall M/S program, except for g two items discussed verbally with SCE.

4.4.2 PNL Evaluation g The SCE proposal covers a number of surveillance items considered important to monitoring engine condition while on standby status. However, some of the items listed were not considered by SCE. For other items, PNL 4.18

consultants' recommendations differ from those of SCE. In general, j'asti-fication for these recommendations is based on engineering judgment.

Two points regarding the keepwarm lube oil filter are important:

1. Entrained water or bacteria (in the absence of bactericide use) will tend to plug some filter media (or weaken others), and so would gradually change pressure drops.

2. The continuous keepwarm flow through the filters will (purposefully) continually " polish" the oil, with gradual buildup of contaminants in the media; the material scavenged out thereby itself helps filter even finer particles as time continues.

Thus, it remains valid to monitor oil filter pressure drop during standby. The changes are slow enough that a weekly check is deemed sufficient.

4.4.3 PNL Recommendation It appears that the SCE standby M/S program needs to be formalized as a separate entity. PNL recommends that it include the items and time intervals listed in Table 4.4.

i .

l l

4.19 l

TABLE 4.4 Standby Surveillance Plan for San Onofre Nuclear Generating Station Item SCE Proposal PNL Recommendation Starting air pressure None provided Visual checks every 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months Log every 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months .

Lube oil temperature in/out Jacket water temperature ,

in/out q ,

Lube oil sump level Monthly .

Fuel oil day-tank level None provided Room temperature i f

Annunciator test Test every 8 hours9.259259e-5 days 0.00222 hours 1.322751e-5 weeks 3.044e-6 months . Log every 24 hours2.777778e-4 days 0.00667 hours 3.968254e-5 weeks 9.132e-6 months Check alarm clear Daily Check compressor air trap operation 3 p qp Fuel rack and linkage Monthly inspect and lube monthly operation Oilquartgy 011 daily Governor oil level None provided Daily Inspect for leaks Monthly Monthly Check freedom of air butter. Lube quarterly Weekly fly valve and cylinder

Keepwarm oil filter None provided Weekly differential pressure -

Test jacket water for pH, Weekly (a) After makeup or monthly conductivity, corrosion inhibitor Take cylinder compression None provided Each refueling and peak pressures i Air start distributor Quarterly Monthly filter 6

Air start admission valve Quarterly Quarterly strainer (a) This proposed frequency was stated in a telephon'e conversation between PNL and SCE.

4.20

. l 5.0 ENGINE TESTING This section presents PNL's review and evaluation of the engine testing program conducted by TDI and SCE in qualifying the DSRV-20 engines for nuclear service at SONGS-1. The engine tests and results are described first, followed by PNL's evaluation and conclusions. PNL's conclusions relate only to the adequacy of the tests conducted to qualify the engines. The effect on engine reliability and operability of the crankshaft failure that resulted from these tests is evaluated elsewhere in this TER.

5.1 ENGINE TESTS AND RESULTS The SONGS-1 diesel engines (DG #1 and DG #2) have been tested by both TDI at the factory and SCE onsite at San Onofre. These tests consist of 1) shop qualification tests, 2) onsite preoperational tests, 3) routine post-operational tests, and 4) special testing following engine maintenance and inspection.

5.1.1. Shop Qualification Tests Southern California Edison reports that DG #1 underwent 300 start and load tests at the TDI facilities. These tests required the engine to reach synch-ronous speed in no more than 7 seconds and to pick up 3.5 MW load in no more than 8 additional seconds. The engine was also operated at 5 MW for 1/2 hour three times during these shop tests. No comparable shop tests were reported by SCE for DG #2.

Additional shop qualification tests included sequential loading tests to demonstrate the engine capability to handle step loads (0.0 to 4 MW and then 4 to 6 MW), to handle overloads (6.6 MW for two 1-hour periods) and to suc-i

$ cessfully handle load drops (drop 6 MW with only a momentary frequency increase).

8 Tests of the air system bottle capacity were conducted to confirm this system's capacity to store and deliver air when needed. The test consisted of conducting engine starts with air that had been stored for 1 week without compressor operation. Six consecutive starts were successfully accomplished.

5.1

Southern California Edison reports . satisfactory performance of all the tests identified above for DG #1. From this result, they concluded that a satisfactory break-in was achieved and that the engine tested met the requirements of TDI quality standards. By virtue of its comparability to DG #1, DG #2 was assumed to be qualified as well.

5.1.2 Onsite Preoperational Tests '.

Southern California Edison reports that both DG #1 and DG #2 were tested in 1976 and 1977 following installation, to confirm satisfactory setup and -

operation. Engine run-in, vibration, and starting tests were conducted prior to engine operation.

The engine run-in tests included runs at no load, a 4-hour run at 100%

load, and a 2-hour run at 110% load. The air start system was also tested by conducting five starts without recharging.

SCE has recently reviewed e the test procedures and results e startup check list ,

e briefing check list e - test procedure exception and deficiency list for these tests conducted in 1976 and 1977.

No significant problems or unusual abnormal observations were evident from this review. SCE concludes that the diesel engines are functioning per design specifications.

5.1.3 Post-0perational Testing Post-operational testing conducted routinely by SCE consists of:

1. a monthly load test: a nominal 1-hour run of each diesel generator at approximately 4500 kW - (In 1984, the load test was increased to g 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months ' duration.)

4 2. an operability start test: a 5- to 10-minute run at no load, to demonstrate operability with manual start circuit - (In 1984, the operability start test was increased to 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months ' duration.)

5.2

.~ ;

3. a loss of bus (LOB) start test: a 5- to 10-minute run at no load, to demonstrate proper activation of automatic start circuits

4. refueling interval tests: a diesel engine overspeed trip check, a 1-hour load test at approximately 4500 kW, and a load rejection test from above about 2600 kW - These tests involve running the diesels

,. for 3 to 4 hours4.62963e-5 days 0.00111 hours 6.613757e-6 weeks 1.522e-6 months .

A listing of these tests by date was supplied by SCE and is summarized in a, Table 5.1.

The above post-operational tests were conducted with five failures to start on DG #1 and four on DG #2. This produced a start reliability of 98.7%

for DG #1 and 99% for DG #2. SCE reports that the malfunctions noted (e.g.,

inadequate lubrication of the fuel rack, low governor oil level, incorrect light bulb, loose wire in a relay coil) were corrected by improved maintenance procedures. SCE states that no failures occurred once the engines were started. They concluded that the start reliability compares very favorably with the goal value of 0.99. This start reliability, taken together with the engines' demonstrated operating reliability, led SCE to conclude that the engines are qualified for nuclear service.

5.1.4 Special Tests Following Major Maintenance / Inspection SCE has conducted special function tests following major engine modifica-tion and inspection. After the pistons were modified in 1982, both engines were subjected to special tests consisting of a 24-hour, full-load run (22 hours2.546296e-4 days 0.00611 hours 3.637566e-5 weeks 8.371e-6 months at 6000 kW and 2 hours2.314815e-5 days 5.555556e-4 hours 3.306878e-6 weeks 7.61e-7 months at 6600 kW). This was done as part of the

. procedure to seat the new piston rings. In 1984, this test was repeated following the sampling inspection, j Additional testing was requested by NRC prior to SONGS-1 return to service. These tests consisted of:

, e ten modified starts to 40% load (SCE selected a 4500-kW load) - A modified start includes turbocharger prelube and a 3- to 5-minute loading to the specified load (in this case, 4500 kW) with the engine run for a minimum of 1 hour1.157407e-5 days 2.777778e-4 hours 1.653439e-6 weeks 3.805e-7 months .

5.3

'l

s a TABLE 5.1. San Onofre Nuclear Generating Station Unit 1 Post-Operational Tests Operability Refueling Load Start Bus Start Start Tests Year DG #1 DG #2 DG #1 DG #2 DG #1 DG #2 DG #1 DG #2 1977 11 9 0 0 37 47 0 0 1978 19 14 0 0 25 31 3 3 .

1979 19 20 0 0 30 30 0 0 1980 14 17 2 3 28 22 5 3 g 1981 10 13 8 5 32 40 0 0 1982 10 17 7 11 22 17 0 0 1983 18 16 21 16 38 30 0 0 1984 8 9 1 7 4 7 0 0 March Subtotals 109 115 39 42 216 224 8 6 Total starts:

OG #1 - 372 DG #2 - 387 Total engine hours at load above about 4500 kW through March 1984 (including 58 hours6.712963e-4 days 0.0161 hours 9.589947e-5 weeks 2.2069e-5 months following special tests - see Section 5.1.4):

DG #1 - 183(a)

DG #2 - 188(a)

(a) These are considered minimum figures. In a telephone con-versation, SCE reported that many load tests continued well beyond the scheduled 1-hour duration. The engine-hour meters .

have accumulated a total for both engines of 1190 hours0.0138 days 0.331 hours 0.00197 weeks 4.52795e-4 months .

l.

5.4

e?

e two fast starts to 70% of nameplate rating (SCE selected 75%) - A fast start simulates an engineered safety feature (ESF) signal with the engine in ready-standby status.

e one 24-hour run at 70% of nameplate rating (SCE selected 75%).

SCE states that all of the tests identified above were performed satis-

~

factorily. The only unusual circumstance noted concerned an unexpected engine operation on one bank of cylinders for several minutes. This occurred during

, an engine air-roll. Friction retaining clamps on the engine fuel control shaft had become loose and the fuel to the left bank of cylinders remained partially open even though the main linkage indicated that the fuel rack was closed. The air-roll then initiated engine startup. The engine ran on the left bank for several minutes and was subsequently shut down by the overspeed trip circuit that closed the air damper. This situation was reported by SCE to TDI. TDI in turn reported the occurrence to NRC via a 10 CFR 21 submittal in early October 1984. Apparently the SONGS-1 DG #1 did not have a retaining (roll) pin that serves as a backup safety feature to the friction clamp, to assure proper orientation of the control shaft. SCE subsequently machined the control shaft and inserted this safety roll-pin. SCE also confirmed that the roll-pins were in place in DG #2.

SCE discussed the circumstances of the engine start and operation on one bank with TDI. It is TDI's opinion that this off-normal operation would not lead to damage to any engine component.

5.2 PNL EVALUATION As explained in Review and Evaluation of TDI Diesel Generator Owners' Group Program Plan (PNL-5161), engine testing and inspections are key elements

$ of the TDI Owners' Group Program Plan for tying corrective actions together and for verifying adequate results. Engine tests are required to demonstrate that 4 a component or unit will meet load and service requirements without failing.

Satisfactory completion of a series of engine tests is particularly important in plants seeking licensing prior to the full implementation of the Owners' Group Program Plan.

~

5.5

.s The engine tests accomplished by SCE and TDI have subjected the SONGS-1 engines to a large number of starts (compared to service start requirements),

and the engine time at loads above the ESF requirements is reasonable.

However, the total number of operating hours (a) accumulated on either DG #1 or DG #2 is far below the 740-hour (10 7 -cycle) criterion for proving the absence of'high-cycle fatigue failure in engine components. Long-term reliability therefore remains unproven.

The starting reliability is acceptable. Starting failures were reviewed .

and no " generic" concern is evident. Liktwise, the running reliability is acceptable; no cases of premature engine shutdown were reported. SCE has corrected the condition leading to failure of the fuel linkage to properly close. PNL consultants have reviewed the off-normal operation and concluded that no engine damage would likely occur.

5.3 PNL CONCLUSIONS Based on its review of the engine testing performed and the results achieved, PNL concludes that both OG #1 and DG #2 have been tested adequately to confirm their ability to meet the load and service requirements until the next reactor refueling outage. The effect on engine reliability and operability of crankshaft cracks anti indications in certain oil holes as a result of these tests is discussed elsewhere in this TER.

l l

t ,

1

n (a) The engine-hour meter shows a combined operating time of 1190 hours0.0138 days 0.331 hours 0.00197 weeks 4.52795e-4 months .

However, SCE records account for only 370 hours0.00428 days 0.103 hours 6.117725e-4 weeks 1.40785e-4 months on both engines.

5.6

e s ds l 6.0 OVERALL CONCLUSIONS This section presents PNL's overall conclusions regarding the capability of the two diesel engines to perform their intended function as emergency standby power sources for the SONGS-1.

6.1 GENERAL CONCLUSION

, PNL and its consultants conclude that the two TDI DSRV-20 diesel engines 4

at the San Onofre Nuclear Generating Station Unit I have the needed operability and reliability to fulfill their intended emergency power function, at least until the time of the next reactor refueling outage.

6.2 BASIS FOR CONCLUSION This conclusion is based upon the known results of the completed opera-tional and special tests and engine inspections. It also reflects PNL's review and evaluation of actions taken by SCE to resolve generic and plant-specific problems and to effect upgrades of their DGs. PNL's conclusion is based on a current knowledge and evaluation of the ongoing Owners' Group investigation of the 16 engine components with significant known problems. The PNL conclusion pertaining to the operability of the DG #1 and DG #2 engines also reflects SCE's commitment to the timely implementation of all applicable OG recommendations and plant-specific items that may result from the SONGS-1 DR/QR t investigations and to concurrence by SCE to the considerations identified below in Section 6.4 PNL conclusions regarding the adequacy of the crankshaft also reflect the planned NRC action to remove requirements for fast test starts during monthly surveillance testing.

I

, 6.3 LONG-TERM APPLICABILITY in Section 1.2 of this TER, PNL expressed its opinion and rationale that g

it cannot reach final conclusions on the operability and reliability of the SONGS-1 DG #1 and DG #2 standby engines beyond the next refueling outage. This constraint reflects the current incomplete status of the elements of the OGPP

)

6.1 l

(

l' 4

and SCE's implementation of the OGPP. When these are completed, it will be possible to draw long-term conclusions.

In expressing this constraint, however, it is not PNL's intent to imply any inherent unreliability or inoperability of these engines, either specifi-cally at SONGS-1 or in general nuclear standby service.

6.4 SONGS-1 RESTART CONSIDERATIONS The conclusion stated in Section 6.1 reflects PNL's careful evaluation of g the SCE and Owners' Group submittals. Specific considerations have been addressed in Sections 3 through 5 of this TER, and reference should be made thereto for PNL's component-specific conclusions and recommendations. PNL assumes that SCE will agree to modifications or additions to their submittals that appear in these sections. Other considerations are:

e SCE will commit to implement all relevant Owners' Group recommenda-tions in a timely manner.

e The engine testing and emergency service requirements for the SONGS-1 DGs will not exceed the engine load of 4500 kW, and monthly surveil-lance testing will not include fast starts.

e SCE will resubmit to NRC a revised surveillance and maintenance plan incorporating changes and additions such as those identified in Section 5 of this TER.

e SCE will document results of the recent DG #1 torsiograph measurements and results confirming the maximum angular deflection and crankshaf t stresses (including oil hole stresses) resulting from normal, fast, and slow starting, coast-down, and load acceptance and rejection.

o Prior to restart of SONGS-1, SCE commits to:

- conduct cylinder block surveillance for cracks in the flange region I if major disassembly is performed per 0G or SCE recommendations for inspection 6.2

I o

- upgrade cylinder heads on an approximately 25% basis at each outage or extended reactor downtime

- modify its planned inspection of high-pressure fuel lines to provide 100% inspection to OG specifications

- conduct an inspection of front-end gears and gear teeth and their

. support bearings on DG #1 as soon as practical, at least at the next refueling outage

- conduct an analysis of the response of the 12 components affected by the crankshafts torsicnal vibration to determine what, if any, special inspections might be required at the next refueling outage.

The conclusion by PNL regarding the operability and reliability of DG #1 and DG #2 to serve as nuclear standby emergency power sources throughout the next refueling cycle is predicated on an understanding that the completion of items described above will not raise unanticipated problems.

?

I 6.3

l

..t o .=

i 0

PNL-5304 ;

1 DISTRIBUTION No. of No. of Copies Copies 0FFSITE K. Trickett, NE-14 U.S. Department of Energy 17 Division of Licensing Office of Office of Nuclear Energy Nuclear Reactor Regulation Washington, DC 20555 U.S. Nuclear Regulatory Commission ONSITE 5 Washingtnn, DC 20555 ATTN: C. Berlinger (10) DOE Richland Operations Office M. Carrington (2)

D. Corley H.E. Ransom / M. Plahuta D. Eisenhut A. Marinos Pacific Northwest Laboratory F. Miraglia M. Williams 6 Consultants 12 NRC Plant Project Managers H. Hardy Division of Licensing A. Henriksen U.S. Nuclear Regulatory B. Kirkwood Commission P. Louzecky Washington, DC 20555 A. Sarsten ATTN: B. Buckley T. Spaetgens S. Burwell D. Hood 5 Senior Review Panel D. Houston K. Jabbour F. Albaugh T. Kenyon S. Bush E. McKenna C. Hill M. Miller W. Richmond S. Miner L. Williams C. Stahle J. Stefano 22 Project Team l

E. Weinkam

' J. Alzheimer

! 2 U.S. Nuclear Regulatory A. Currie Commission D. Dingee

! Public Document Room R. Dodge

. Division of Technical W. Gintner Information and Document W. Laity (15)

, Control J. Nesbitt Washington, DC 20555 F. Zaloudek Technical Information (5)

Publishing Coordination (2)

Distr-1

.. __ _ _ .. _ , _ _ - - . . - .- - - - - _ _ - -- -- - - - . - , _ . -

ML20214E480

Contents

Text

Title:

1.0 INTRODUCTION

1.0 INTRODUCTION

2.0 BACKGROUND

Navigation menu

ML20214E480

Text

Title:

1.0 INTRODUCTION

1.0 INTRODUCTION

2.0 BACKGROUND

Navigation menu

Search