ML20084F381

Provides Comments on Engine Start & Operability from Review of Standby Diesel Generators & Reliability Rept Submitted on 840220

ML20084F381
Person / Time
Site:	Grand Gulf
Issue date:	03/30/1984
From:	Laity W Battelle Memorial Institute, PACIFIC NORTHWEST NATION
To:	Berlinger C Office of Nuclear Reactor Regulation
Shared Package
ML20084F372	List: ML20084F381 ML20084F388
References
NUDOCS 8405040107
Download: ML20084F381 (12)

Text

_ _ _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ _

ENCLOSURE 2

/ 74 OBattelle Pacdic Northwest t,acoratories P.C. ses 999

. Ricniand. Wastungton U.sA. 79352 March 30, 1984 renon. ison 375-2780 re 15.:r4 Mr. Carl Berlinger Division of Licensing Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Comission j Washington, D. C. 20555 l l

Dear Mr. Ber11nger:

i

SUBJECT:

GRAND GULF NUCLEAR STATION STANOBY DIESEL GENERATORS: RELIABILITY REPORT SUBMITTED TO NRC BY MISSISSIPPI POWER AND LIGHT COMPANY LETTER DATED FEBRUARY 20, 1984 In responsa to your request of Frid4y, March 23, PNL reviewed the sub-ject report and discussed it with you by telephone on Thursday, March q f Those who participated in the review are identified in the en-closed sumary of our coments and conclusions.

participated in the telephone conversation: D. A.Four of the reviewers Dingee, A. G. Henriksen (consultant), 8. J. Kirkwood (consultant), and myself.

You asked during the above-mentioned telephone conversation for our coments on the issues of engine start and engine operability. Our coments are as follows:

Engine start - On the basis of tests at Grand Gulf (summarized in ranle 1-2 of the subject report) and at the Shorenam Nuclear Power Station, the Transamerica Delaval, Inc. (TDI) diesel engines start with a high degree of reliability. We have seen no evidence to suggest that there is any problem peculiar to TDI engines N start-ing them within required time constraints.

Engine coerability - The information available for our review is not sufficient to provide a basis for predicting engine acer-ability.

In particular, unresolved issues that pertain to key comoonents (e.g., connecting rods and cylinder heaos) need to be addressed before engine operability can be predicted with reason-able certainty.

Additional Section II of the enclosed coments on this issue are included in summary.

e405040107 840425 PDR ADOCK 05000416 S PDR

Mr. Carl Berlinger March 30, 1984 @MHelle Page 2 Please don't hesistate to call me if you have any questions on the com-ments contained in this letter or in the enclosure.

Sincerely, '

J wni .

Walter W. Laity PNL Project Manager WWL:fo Enclosure cc: M. Plahuta, DOE-RL e

1

I ATTACHMENT Review of Minninnioni Power 1 Licht fuhmittal previdine the NRC with TDI Eneine status Renart Da+=d Februarv 20. 1984 I, Rasis of Review This review incorporates the comments and discussions of the following staff after approximately a one-day reading and a one-day working session: l PNL Core Team ('I Ricardo Engineering (J. V. Webber, et. al .)("I S. H. Bush, consultant ("I

8. J. Kirkwood, consultant A. Henriksen, consultartt This review focussed on an evaluation of the current reliability and operability of the MP&L TDI engines to meet the requirements to serve as backup power at the Grand Gulf nuclear plant operating at full power. This review dealt with the information provided in the MP&L letter to Mr. Harold Denton, dated February 20, 1984.

The review addressed the 16 generic issues identified by the TDI Owners' Group that were addressed in the MP&L submittal. Additional considerations are also noted.

The presentation generally follows the order of issues addressed in the MP&L submittal; the review of items where MP&L has effected repairs or modifications is provided first. A summary position follows as section II. This is followed by a review of the MP&L response to concerns for other issues raised at the January 26, 1984 Owners' Group meeting at Shoreham. Finally a review of the MP&L Testing and Maintenance Program is provided.

(*)W. W. La tty, D. A. Dingee, S. D. Dahl gren, M. Cl ement, J. R. Nesbitt; J.

Alzheimer

(**)Ricardo Engineering provided comments by telephone on the basis of a i

review done at their facilities in the U.X. No Ricardo representatives l were available to participate in the meeting at PNL on March 28, or to I review the comments and conclusions documented in the repor*. Likewise, S. H. Bush was not available for review and comment on the conclusions.

i l

l

TIf,Sr-- ev of canclusions and observations The information available for our review is not sufficient to provice a basis for determining whether or not TDI engines at Grand Gulf can meet requirements for emergency service. Major unresolved issues (addressed

, later in this review) include:

o connecting rods o wrist pin bushings o cylinder heads o turbocharger o connecting rod bearings o testing / maintenance plans The reviewers note that action taken during implementation of the Owners' Group Program Plan may reveal issues that have not been addressed in the MP&L report of February 20, 1984. These issues may bear on the operability and reliability of the TDI engines at Grand

, Gulf. Accordingly, the issues addressed in the MP&L report are not necessarily the only issues that will need to be addressed for these engines. ,

An appropriate surveillance and maintenance program migb.t provide a basis for engine operation during the period when the Owners' Group Program Plan is being implemented. We cannot predict at this time i

whether or not the surveillance and maintenance program would be sufficient to ensure that the diesels could be expected to meet al1 of

~

l the ' emergency power requirements described in the Grand Gul f FSAR. This tentative conclusion is subject to the following:

o Identification of the root causes of unresol ved probl ems (e.g.,

rejectable indications in cylinder heads), and appropriate corrective action.

o Verification through inspections currently underway at Grand Gulf that engine components are exhibiting only aormal wear in )

the operating experience accumulated to date.

III. Review Results - Grand Gulf Enaine Renairs and Madificatiens (MPAL Renar* Sactions 7 throuah 9)

A. Pistens

1. Cqnsiderations; o Because the peak pressure in the TDI engine at Kodiak is about 3/4 the peak pressure in the Grand Gulf engine at full power, the operating time at Kodiak is not of as much value as the 2

_. - _ . _ _ - . . . _ _ _ _ , _ . _ _I

l

. ..er operating time in the TDI R5 engine, which operates at a i higher peak pressure. Further, the R-5 tests will only be relevant to the AE piston skirts used in the Grand Gulf engines if it can be. demonstrated that the AE piston skirts used in the former are the same as those used in the latter.

o Our tentative conclusion concerning the suitability of the AE piston skirts is contingent upon finding no rejectable inoica-tions in them following the recently completed 600 hour0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> test at Grand Gul f.

2. canelumien o Subject to the above considerations, the evidence available to the reviewers suggests that the AE piston skirts are suitable for Grand Gulf operation.

8. Cvlinder Heada

1. canaidaratian*

u Crack propagation in a cylinder neac ouring operation may lead to serious damage to the engine and/or turbocharger, possibly resulting in sudden engine shutdown.

o The MP&t. report of 255 of the heads with rejectable indications is very high. The cause -of the rejectable ,fndications has not been identified, o

An analysis of failure rates of cylinder heads of this type operated at comparable loads may be instructive for establishing confidence in the suitability of these heads for engines in nuclear service.

o It would be of interest to know whether the heads operating in the R-5 engine are of the same design and whether they are performing without development of rejectable indications.

2. conclusion o The cause of the rejectable indications has not been identified.

)

Accordingly, there is insufficient evidence to say that the heads will perform reliably.

l . . .

C. Connae+1ne Red Bearines

1. Considerations and canclusion The evidence available to the reviewers is insufficient to conclude whether the problems identified at Shoreham are applicable to Grand Gul f.

Pertinent information for establishing bearing suitability includes the following:

o Radiographic inspecticn of all bearings to acceptance criteria established shell report,by the Owners' Group in a recently issued bearing o

Inspection and documentation of wear patterns of al1 bearings at Grand Gulf to verify absence of abnormal conditions, such as end loading.

D. Push Reds

1. Considmentions o

A push rod failure will ultimately lead to shutdown of a cylinder and will require early shutdown of the engines.

o The MP&L submittal addresses corrective action for the connector push rods but does not address the 25 failure of the main rocs.

o There is no evidence that the new design has been proven to be reliable.

2. Conclusion o

Adequacy of the modifications should be verified througn 100%

inspection of the push rods to establ1sh that no cracks developed during recent testing at Grand Gul f which included 600 hours0.00694 days <br />0.167 hours <br />9.920635e-4 weeks <br />2.283e-4 months <br /> at full power.

E. Crankshaft

1. Considerattens o

Results in the MP&L report of analyses performed indepencently by TDI and Bechtel suggest that the stresses in the crankshaft i

used in the TDI engines at Grand Gulf are acceptable.

o Incipient problems would be indicated by wear patterns on the bearing.

Likewise, hot

• and cold shaft ceflection reacings 4

l l

l -

rcveal alignment problema that could lead to difficulties with shaft, bearings, bearing supports aryd base, j

o The reviewers feel that TDI statistics concerning shaft problems would be pertinent to establish confidence in the Grand Gulf designs.

2. conclusion o

The Grand Gulf crankshaft designs appear to be satisfactory.

This is contingent upon MP&L determination that other crankshafts of this design in sistlar service have not failed due to design deficiencies, an examination of bearing wear patterns, and hot and cold crankshaft deflection readings.

F. L.P. Fuel Line Failure

1. Cannidarations and Ennelumien '

i o

The problem appears to be an isolated one, not generic.

o The problem definition and solution are deemed to be acceptable.

MP&L should determine that no new vibration response problems have been introduced by their solution. .

G. H.P. Fuel Lfna Failure

1. can.1 derat 4 on. ,

o The problem appears to be related to manufacturing rather than design.

o MP&L did not indicate how (or whether) the new lines were inspected to verify absence of the drawseam, o

The line pressure cycles are severe; ranging from near atmospheric to about 5,000 psi and cyc1ing at the cate of 1/2 the engine speed (i.e., 225 rpm).

i

• Hot deflection shutdown readings should be completed within 15 minutes of engine to be valid. i

l

2. conclusion

! o The problem appears to be adequately understood and the solution is acceptable. MP&L shou 1d verify through inspection that the

! new lines are not defective.

H. Cranke== cannerew

1. can iderations and conclusion o

Failure of crankcase door capscrews is relatively common due to difficulty in obtaining even loading at the capscrew panel interface. The reported consequence, namely a piece of the bolt entering the generators, is unusual. The solution (protecting the generator) is accaptable.

IV. Review Probl as;Results - Grand (MP1L Renart Gulf Rannonsen Attach:2:;t 11 to Other Owners' Groun Generie A. cvlinder Liners

1. Tenrideratione and canelunian o

The mei. hod of examination of the damaged i f ner was not stated.

The reviewers agree with the probable cause of the grooving observed in that liner.

o' The correctiive action (replacing the damaged liner) is judged to be acceptable.

B. cylinder Black

1. considerations and conclusion o

The MP&L report does not address whether cylinder bicek cracks i of the type noted at another nuclear installation are present in the Grand Gul f engines. If such cracks are present, the issue needs to be addressed.

on the Grand Gulf cylindor 'Weblocks. have no basis at this time to comment C. Ennine Bane

1. canaiderations and canetunion l

! o

The information presented suggests that the problem stems from a failure in maintenance to apply proper bolt torque. The corrective action (verification of correct preload values in main bearing studs) appears adequate, subject to verification by l

i 6

• MP&L oith TDI of historical data to confirm that the problem deco not involve other than maintenance considerations in installa1ons similar to Grand Gulf.

D. Head stude

1. canaidarations and conclumfen o

There is no basis in the MP&L report to comment on this problem.

E. Rnekar Arm canscrews

1. canaidarations and ccaclumfen o

A recent report issued by the Owners' Group on this topic addresses: (1) design, (2) materials, and (3) retorquing.

o MP&L should implement the Owners' Group recommendations.

F. Turhacha reer

1. canniderattene o

The MP&L statements on misalignment as the cause does not provide a convincing argument.

o It is considered unlikely that vibration generated internal to the turbocharger could be the cause. Such imbalance would rapidly lead to destruction of the bearings and rotor.

o It is considered more likely that vibration is caused by engine vibration transmitted inappropriately through turbocharger supports and/or piping.

o MP&L should verify that appropriate consideration has been given to exhaust pipe residual loads on the turbocharger. These loads contribute to the loads on turbocharger mounting bolts, and may contribute to excitation of turbocharge mounting vibration.

2. canclusion o

' There is insufficient evidence to accept the MP&L problem resolution.

3. Obmarvation o

With regard to the turbocharger thrust bearing failure that has been experienced at Shoreham, we concur that the Grand Gul f 7

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

~

diesel engines appear to be adequately protected with an electrically-operated prelube system for normal startup. In the event of a "bl ack start" (i.e., no el ectrical power), however, there appears to be no protection.

G. cannectina Reds

1. Cannidarations o Consequences of connecting rod failure include immediate shutdown of the engine, possibly catastrophic damage, and a potentially severs hazard to operating personnel in the vicinity of the engine, o The reference to marine experience is not necessarily applicable because of differences in engine loading.

o Evidence presented suggests that a reduction in frequency of failures may have been achieved but not necessarily a solution to the problem.

o The 104hour figure is given as the average hours of operation between occurrences. It is not accompanied by a time distribution of failures which may be an important consideration.

2. conclusion -

o The evidence presented does not provide a sufficient basis for conclusions regarding the adequacy of the connecting rods for the intended service. Because of the potentially serious consequences of connecting rod failure, a conservative approach to estabi tshing connecting rod adequacy is called for. This approach should take into consideration such factors as the root cause of connecting rod cracking, appropriate tests to verify corrective action, probable minimum time between failure under worst-case conditions that may be imposed on Grand Gul f engines, and appropriate ongoing surveillance to ensure that the connecting rods remain sound in service.

H. Jackat Water Pomns

1. canniderations and canclusion o The problem appears to be adequately understood and the solution is acceptable.

a

r t

~ '

2.' obnarvation l

l o It is common practice in non-nuclear installations to have an j

electrically driven standby jacket water pump.

I. Air start valve Cannerews I

! 1. Can=1darations and Canclusion o The problem appears to be adequately understood and the solution is acceptable.

V. Wrint Pin Buehf na fa new orchlem area not addrented bv the TDI owners' Groun at thin timm1

1. Cannideratione and Canclusion o

Failure of wrist pin bushings may have serious consequences, comparable or worse than failure of the connecting rod bearings.

o Unit loadings on wrist pin bushings are larger than on connecting rod bearings, o

All eight wrist pin bushings removed from the 101 engine at Shoreham during the week of March 19 were dye checked and found to be cracked. No pattern of cracking was evident.

It was also reported, but not verified, that new bushings received at Shoreham from TDI but not installed, are also cracked. This suggests that the cracking is a manufacturing problem, and if so, it may be present in the wrist pin bushings in the TDI engines at Grand Gulf. Accordingly, we bel ieve that all wrist pin bushings should be dye checked and those found to have cracks should be replaced with bushings that are not cracked, o

This problem needs to be addressed immediately because of the seriousness of the consequences.

VI. Cewnmente en the MPIL Oua 11f f eation/Rel iabi k ity 0=anstrat4an hst d na (MP&L Renart hction 11.0) o The test program to demonstrate the adequacy of the TDI engines should be related to the demands that may be placed on the engines under emergency conditions as described in the Grand Gulf FSAR.

o The test program followed after replacement of the piston skirts as described in section 11.3 of the MP&L report appears to meet standard industry practice.

However, the summary of testing presented in Table 11-1 suggests that the AE piston skirts were not q; _

. installed in the Grand Gul f engines during " Tech Spec Testing."

This issue should be addressed in the overall test program for the TDI engines.

o The brief description (provided in the MP&L submittal) of the maintenance program and reliability enhancement testing is not convincing to the reviewers that there will be adequate surveillance of physical conditions and monitoring of operating l parameters to assure continuous availability and operability of the l engines. 1 VII. consultant cancurennen B. J. Kirkwood Adam Henriksen Covenant Engineering m._ , , -

ENCLOSURE 3 t .

OBalfelle a

w : v. r e ..u n n .

e. .

<. ...u. .. .

re : c .' 355-2780 April 16, 1984 Te.. is.a:4 Mr. Carl Berlinger ,

Ofvision of Licensing Office of Nuclear Regulatory Commission U.S. Nuclear Regulatory Commission W2chington, D.C. 20535

Dear Mr. Berlinger:

In response to your request of Monday. April 16. PNL has revfewed the qucctions you raised on the suaject of diesel generators. These questions waro discussed by Dave Dinges and Walt Laity with the following diesel engine consultants who, as subcontractors to PNL, participated in a meeting on this suaject at NRC on April 13: Adam Hendriksen. 8.J. K1rkwood, and Arthur Sarsten. Summarized in the enclosure to this letter are tho assumed operating requirement.1 for the diesel engines, followed by our comments on each of the questions.

Please do not hesitate to call if you have any questions on the enclosure.

Sincerely,

- r anX./) g

/1 / k .

^

Walter W. Laity j PNL Plant Manager WWLari Enclosure i cca M. Plahuta, 00E-RL J

l',

l .

GRAND GULF NUCLEAR POWER STATION STAN08Y 0,IESEL GENERATORS - PNL RESPONSES TO NRC QUESTIONS OF APRIL 15, 1984

1. Mun+ MPtt esndue* an annine _+=me-dawn and inanee+1en? If se, must thin he hath aneinas?

The consultants had a range of opinions about the necessity for complete engine tear-down of both engines. All agreed that at least one enginemust be coepletely torn down. Action.on the second engine would be contingent on findings. If no problem is noted with the first engine, then the second engine can be accepted without tear =down if MP&L can demonstrate through a review of the manufacturers' QA program that these two engines are essentially identical. If the QA program review does not give this assursace the opinion of the consultants varied, depending on the level of assurance. Action throught to be appropriate . ranged from a " sampling" inspection of readily accessible items to a tear-down to inspect the critical components (e.g., vrist pin bushings, conrod bearings, and conrods).

2. Annumine the cn1 angine *maMewn dimel amme defee*m. what must MPRf de ir - % elv. and later? .

The inspection of the % r,-down. engine might reveal information that wculd suggest a meeting between NRC and MP&L. However, aven if the inspection reveals no new information

• all defective parts shoula be replaced.

Possibly the block and engine base could be excepted if cracking is not severe or in critical areas. However, if more recent history and analyses confim the cracks to be serious, these parts also must be replaced. Again, action on the second engine woul d be contingent on findings. If the inspection of the first engine reveals serious defects, these need to be evaluated as a basis for establishing inspection requirements for the second engine.

In the long term, MP&L must be bound to implement an enhanced surveillance and maintenance program (see below) and implement the j

Owners' Group recommendations (currently being formulated) on both j

engines at the first refueling shutdown. '

• Progress by the Owners' Group on generic issues can affect the status of understanding at the time of MP&L licensing.

~ ~ - - - - - - , ,. , - , - - . - - - , . . , , - . , . , , , - , ,,,__,_---,,,.---.,,...,n- - - , - . , - . . . - ,- -. , , + - - .

l . .

l l

3.

AnnemineGrandGoldhome inte eneration. what maintenanen ind insnac*1on reau f r-- s nuet be fanaendt There will be an inspection frequency in the Owners' Group plan calling for a complete tear-down. This should be implemented at Grand Gulf. Subject to that detemination, special attention should be given to selected components as described below.

If defects are noted, the parts should be rep 1 aced. The nature of the defect will detemine if this is all that is required. The owners' Group maintenance program recommendations or experience may be used to establish relaxed (or tightened) inspection frequencies.

Detailed inspection and maintenance requirements should be established in conjunction with the engine manufacturer. This should also include all maintenance / inspection identified by MP&L at the April 13 meeting.

Cvlindes Ham h After engine shutdown the engine should be rolled over with air preneure once each hour for four hours (during cocidown) with the indicator cocks open. Engines not in operation should be rolled over once a day. Any heads found leaking must be replaced.

8L9aine Black and Rama, Inspect once a month or after 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> operation for any cracks.

• No other special maintenance required if any defects found are " noncritical".

Cannec*ine Mader Af*ar each 25 starts a 50 hours5.787037e-4 days <br />0.0139 hours <br />8.267196e-5 weeks <br />1.9025e-5 months <br /> of operation g6-months, all bolts on conrods should be retorquod and theneresults recorded.

4 .

Lube of7 Cheeks, Weekly (or after each 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> operation) for l

water and monthly for particulates and chemical contaminants )

associated with wear of bushings and bearings. Al so col l ect sampl e from bottom of sump and check for water.

checked (no time interval given). The filters shoula also be Canae swa r Monthly spot check (25%) of all capscrews in question.

Other-If per question 1 an engine is not torn down, each 6 months a 255 random check of piston crowns, liner walls, heads, upper block at studs, head bolts, areas around head bolts, and push rods (bothmain and connecting). Also on this engine, a 121/25 check of bearings and wrist pin bushings should be done.

Punh Mode

• After 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> operation, cams, tappets, push rods, etc., shoul d be caecked.

This can be done one at a time with the engine shutcown but without affecting its availacility for service.

~

• 4 Rennt dine the surveillanea creeram. what data accord _ine eheul d be taken in eneine standhv and anarat_ine enndit4nnet To accommodate standby monitoring, the daily, weekly, and monthly actions should be accomplished as identified in the MP&L 0/G Maintenance / Testing Program (received on 4/13/84) except that the oil prassure ff1ter drop should be monf tored daily instead of month 1y. One additional standby monitoring requirement is a shaft deflection measurement every six months.

The engine operating surv'aillance program should include the following.

If alam 1evels are reached on any of these, this indicates the need to switch engines.

o exhaust temperature monitor and alam for each cylinder (continuous) o temperature recording before and after turbo-charger (continuous) o hourly readings on standard temperature and pressures for such items as lube of h Jacket water, intercooler, air pressure, etc.

o accelerometer monitoring (continuous) on all main bearings and the turbo = charger o monitor daily the lube oil filter pressure drop

5. h oraenarmtf enal + .+4ne woul d we 8. =ui-e fel l ow4 na ...- 1v af the 1nen M ad enefnelm)1 -

The manufacturers' standard preoperational testing should be cone. In addition:

o run 10 modified starts (defined as prelube and 3-minute loacing to 40% load)

• o conduct two quick-starts to 70% load and hold for four hours duration o conduct one 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> run at 70% load (to look for excursions in temperature).

l l

, - - - - - , , - - - - . . . - , , . , . - , . - a-. - .---- n, ---a-,,,