ML20215N581

Part 21 Rept Re Cracking of Engine Cylinder Heads.After Each Shutdown,All Indicator Cocks Should Be Opened & Engine Slowly Rotated to Look for Signs of Cooling Water Cracks. Part 21 Rept Will Not Be Submitted

ML20215N581
Person / Time
Site:	Cooper
Issue date:	10/29/1986
From:	Lambert W COOPER BESSEMER CORP.
To:	Haass W NRC
References
REF-PT21-86 QCG-3277, NUDOCS 8611070027
Download: ML20215N581 (7)

Text

i 50-298 COOPER BESSEMER RECIPROCATING w.w October 29, 1986 Our Ref: QCG-3277 U. S. Nuclear Regulatory Commission Mailstop_EWW332 Washington, D.C. 20555 Attn: Mr. W. P. Haass

SUBJECT:

Nebraska Public Power District Cooper Nuclear Station Standby Diesel / Generators We are sending you herewith a copy of our repcrt covering the cracking of engine cylinder heads. Please note from the report that because this problem is not generic, we do not propose to submit a 10CFR Part 21 Report.

W. If. A. Lambert Manager, Quality Control File: S0-8585 Encl.

kll 8611070027 DR 861029 ADOCK 05000298 Uncoln Avenue PDR k\

Grove City, eennsylvania 16127 .

(412)458-8000 Telex: 86-6613

.monmso,se cournessons . ooron.on,us cournessons . eome eso,~es 4F l

3 %\ -

October 23, 1986 QCG-3268 REPORT .

NEBRASKA PUBLIC POWER DISTRICT (NPPD)

Cooper Nuclear Station, Brownville, Nebraska KSV-16-T Cylinder Head Cracking-Copy Distribution D. T. Blizzard-Director of Engineering, G.C.

D. E. Brazeal-Aftermarket, Tulsa P. R. Danyluk-Manager, Engine Engineering,'G.C.

T. O. Garrison-Aftermarket, Ddessa V'?iP;TNaass-Nuclear Regulatory Commission - Washington, D.C.

G. Horn-Divison Manager, Nuclear 0perations, Nebraska Public Power District J. M. Horne-Manager, Analytical and Compressor Engineering, G.C.

T. W. Kearns-Manager, Application Engineering, G.C.

G. L. Miller-Aftermarket, Mt. Vernon C. Rousselle-Aftermarket, New Orleans M. A. Schleigh-Supervisor, Design Engineering, G.C.

B. R. Sedelmyer-Director of Operations, G.C.

R. L. Spetka-Manager, Aftenmarket, G.C.

K. R. Young-Aftermarket, Mt. _ Vernon 4

6 9

e 6

9

October 23, 1986 QCG-3268 REPORT ,

~

NEBRASKA PUBLIC POWER DISTRICT (NPPD)

Cooper Nuclear Station, Brownville, Nebraska

! KSV-16-T, Cylinder Head Cracking -

i 1.0 Introduction 1.0 Two standard connercial KSV-16-T's were built in 1970 to provide -

standby power at the Cooper Nuclear Station. These engines, rated at 4000 KW (5560 BHP) at 600 RPM, went into service in 1973 and have performed with minimal difficulty. Both engines were built before current NRC regulations became effective, and therefore, manufacturing traceability was not required.

1.2 1e engines are identified as follows:

C-B Serial No. NPPD Engine No.

7102 2 7103 1 1.3 Hours run and number of starts as of October 18, 1986:

Engine No. 1 Engine No. 2 No. of Starts 785 761 No. of Hours 1434 1215 -

1.4 During the current " refueling outage" at Brownville, NPPD had scheduled preventative maintenance programs for each engine. Engine No. 2 was removed from service early October and a diagnostic was performed. Two heads were removed for the purpose of inspecting the _

valve seats. With the two heads removed from the engine, cracks were observed on the fire dome surfaces. Subsequently,- all 16 heads were removed and delivered to the Cooper-Bessemer facility in Tulsa i for inspection. Twelve had cracks in evidence. A crisis developed because sufficient heads to replace all 12 cracked ones were not available, and quoted deliveries would take 25 weeks.

i e

y .e, m e--- ,

. Page 2 QCG-3268 October 23, 1986 2.0 Discussion 2.1 All cylinder heads are subjected to thermal stresses of varying magnitudes which are conditioned by their internal structure. These stresses are tempered by occasional / transient combustion upsets as.

well as external considerations such as load (overload), valve and fuel timing (high peak pressures), cooling water temperature, air manifold temperature, etc.

2.2 In 1970 when these engines were built the state-of-the-art head was a " gas / diesel" (Part No. G18-11-6Af 2) configuration as opposed to a straight diesel head which is currently available. The original head contained gas passages, not required for a diesel head, and therefore, the internal structure was more complicated. While the " gas / diesel" head has proved itself under optimum conditions, it has also demonstrated an inability to withstand abnormal combustion chamber upsets. It should be noted that the most common combustion

. upset arises from incorrect fuel timing which results in high peak pressures and high temperatures associated with this. (Pressure rise is a consequence of temperature rise).

2.3 Ab; ormal thermal cycling leads to high compressive stresses which, upon shutdown (when the heat is turned off) and due to the hysterisis phenomina of cast iron, results in a tensile stress which, if high enough, will cause a crack. Initially such a crack is likely to be hairline in nature, and following many adverse cycles could slowly propogate to the cooling water jacket. From the foregoing, it can be concluded that if a crack were to progress into the water jacket, an internal water leak would be readily detectable shortly after shutdown by rolling the engine with the indicator cocks open. It can further be deduced that by knowing the extent of cracks, cracked heads can be safely run with the limitation being crack propogation to the extent of a detectable water leak.

3.0 Observations 3.1 All 16 heads from Engine No. 2 were magnaflux tested and were hydrostatically tested for one hour at 90 psi with an initial temperature of approximately 200 0 F These tests were witnessed'and recorded by NPPD personnel with copies sent to interested parties.

Appendix "A" shows.a head layout with typical cracking. The results are summarized as follows:

e

,--.,e, - - , - , - , +,-. - -, _ . . ~ - --

e, , ,,,- . - . .- - , , -- - - . - - - - - , . - w-

Pag 3 3 QCG-3268

. October 23, 1986 Left-Ban k -Right-Bank 1.-No cracks, no leaks -Ok. 1. No cracks, no leaks - Ok.

2. Cracks, no leaks. Exten- 2. Cracks, no leaks - Ok.

sive pitting- Rejected.

3. Minor cracks, no leaks-0k. 3. Cracks, no leaks - Ok.

4. Cracks, leaks - Rejected. 4. Craclis, leaks - Rejected'. -

5. No cracks, no leaks -Ok. 5. Cracks, no leaks - Ok.

6. Cracks, no leaks - Ok. 6. Cracks, no leaks. Extensive

' pitting - Rejected .

7. Cracks, no leaks - Ok. 7. Cracks,'no leaks - Ok.

, 8. No cracks, no leaks- Ok. 8. Cracks, no leak - Ok.

4 SUM 4ARY t-4 Heads - No cracks, no leaks - Ok.

8 Heads - Cracks, no leaks- Ok.

i 2 Heads - Cracks, leaks - Rejected.

2 Heads - Cracks .no leaks, pitting - Rejected.

l f 4.0 Recomendations ,

4.1 The eight cracked heads referred to in Paragraph 3.1 as "0K" may be returned to service. Due to the nature of the cracks, they will not propogate while the engine is running, only during cooldown after shutdown. Therefore, after_each shutdown open all indicator cocks, slowly rotate the engine, and look for signs of cooling water leaks.

This procedure may be followed any time after the first hour following a shutdown. This recomendation should be.followed any time

" gas / diesel" heads are in service.

l 4.2 Install new " diesel" heads (part no. KSV-11-6A#1) instead of

" gas / diesel" heads when possible.

4.3 When installing new diesel heads convert to 30 0valve ' seats from 450 in order to minimize valve seat recession and wear.

4.4 As soon as Engine No. I becomes available, subject all heads to a i magnaflux and hydrostatic test using the.same procedures and acceptance criteria as for Engine No. 2's heads (Paragraph 3.1).

, 4.5 Repair of cracked h'eads by metal-stitching or any other process is not recommended.

,m .,e------.---~ .---- , +-- .-4 ,,..~..-.,4-.o.---%----- . . . ,-- ,- - - . -- - - . - - - , - , - - - - --- -..

Page 4  !

I

.QCG-3268 October 23, 1986 ,

l l

4.6 Following initial start-up after the rebuilding of the engines, and '

with operating parameters stabilized, peak firing pressures should be measured and recorded. This will confirm that combustion is proper or a need to " tune" the fuel system. In'the process a base-line is established for future reference. Subsequently peak firing pressures should be checked at least twice between refuelling outages. It should be noted that examination of peak firing pressures can reveal problems developing with fuel systems as well as mechanical problems (low pressure may indicate a bent rod caused by the admission of cooling water into a cylinder).

4.7 For the long term, consideration should be given to controlling the aftercooler water system niore closely to maintain an air inlet manifold temperature of 90 - 120 F.

4.8 The NPPD engines are the only ones currently in nuclear. service with the " gas / diesel" cylinder head. The problem of-head cracking is unique to NPPD and, therefore, is not generic. Consequently, Cooper-Bessemer will advise the circumstances to the NRC by copy of this report and does not propose to submit a 10 CFR Part 21 Report.

W. H. E ambert File: S0-8585 K5f Misc.

11-1 7-. - -

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

?,

wr

, QCG-3268 APPENDIX "A" October.23, 1986 I e 2 < t g' 2 3 4 E h Q 3 *

< ;;2  :

g -

Ea be

.@g E c ',

e 0 y h .

I

! l lE l 8 r f ,

I I HpWI& l -_ y

1. Spring Itatsiner 3. Spring Keeper 6. Inlet Valve

2. Spring Keeper Sesi

4. ' valve Guide- .

7. Exhaust Valve

5. Compression Srring 8. Valve Seat Composite of typical '

cracks. Ilot every head is as extensively cracked as shown here.

f.  % ,

2

( 1 e

Air Q @' Inlet Side

,D 4'

/

2

\

--Air Start Yalve View on Fire Deck (Unnecessary Detail Omitted) .

l P

_ _