ML17304A971

From kanterella
Jump to navigation Jump to search
Part 21 Rept Re Ksv Standby Diesel Generator Rocker Arm Failure.On 890104,exhaust Valve Rocker Arm Installed in KSV-20-T Engine Broke at Plant.Investigation Revealed That Castings Cracked During Mfg.Review Underway
ML17304A971
Person / Time
Site: Palo Verde  Arizona Public Service icon.png
Issue date: 01/18/1989
From: Stolba F
COOPER BESSEMER CORP.
To:
NRC
References
REF-PT21-89, REF-PT21-89-006-000 PT21-89-006-000, PT21-89-6, QCG-5310, NUDOCS 8901240490
Download: ML17304A971 (14)


Text

~

AC CElZMTED DI IBUTION DEMONS ON SYSTEM REGULATORY INFORMATION DISTRIBUTXON SYSTEM (RIDS)

ACCESSION NBR:8901240490 DOC.DATE: 89/01/18 NOTARIZED: NO DOCKET FACIL:STN-50-528 Palo Verde Nuclear Station, Unit 1, Arizona Publi 05000528 STN-50-529 Palo Verde Nuclear Station, Unit 2, Arizona Publi 05000529 STN-50-530 Palo Verde Nuclear Station, Unit 3, Arizona Publi 05000530 AUTH. NAME AUTHOR AFFILIATION STOLBA,F.B. Cooper Bessemer Corp.

RECIP.NAME RECIPIENT AFFILIATION NRC No Detailed Affiliation Given

SUBJECT:

Part 21 rept re KSV standby diesel generator rocker arm failure.

DISTRIBUTION CODE: IE19D COPXES RECEIVED:LTR ENCL SIZE:

TITLE: Part 21 Rept (50 DKT)

NOTES:Standardized plant.", 05000528 S Standardized plant.t'~". 05000529 Standardized plant. 05000530 g

RECIPXENT COPIES RECIPIENT COPIES A ID CODE/NAME LTTR ENCL ID CODE/NAME LTTR ENCL PD5 LA 1 0 PD5 PD 1 1 CHAN,T 1 1 DAVIS,M 1 1 DAVIS,M.J. 1 1 INTERNAL: AEOD/DOA 1 1 AEOD/DSP/TPAB 1 1 ARM TECH ADV 1 1 NRR CRUTCHFIELD 1 1 NRR VARGA,S 1 1 NRR/DEST/ADE 8H 1 1 NRR/DEST/ADS 7E 1 1 NRR/DLPQ/QAB 10 1 1 NRR/DOEA/GCB 11 1 1 NRR DRIS/VIB 9D 1 1 NUDOCS-ABSTRACT 1 1 01 1 1 RES/DSIR/EIB 1 1 RGN1 1 1 RGN2 1 1 RGN3 1 1 RGN4 1 1 RGN5 1 1 EXTERNAL 'INPO RECORD PDR CTR>>'RC PPDR, NSIC SILVER,E"

, 1 1

1 1

NOTES: 1 1 NVZE 'ZO ALL "RIDS" RZCIPIEFIS

PLEBE HELP US K) RKRJCE WASTE.'2GACZ IHE DOCUMENT CKKQKL DESK, ROOM Pl-37 (EZZ. 20079) KO EIaIKQQZZ YOUR NAME FROH DISTEKBUTION

'LISTS K)R DOCUMENTS YOU DGNiT NEEDi TOTAL NUMBER OF COPIES REQUIRED: LTTR 28 ENCL 27

~ ~

< I

~ COOPER INOVSTRIC5 e

COOPER. BESSEMER RECIPROCATING l F. Bruce Stotba Vice President and General Manager January 18, 1989 Our Ref: QCG-5310 Office of Inspection and Enforcement U. S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Director of Inspection and Enforcement

Subject:

KSV Standby Diesel Generator Rocker Arm Failure Arizona Nuclear Power Project, Palo Verde, Arizona On January 4, .1989, an exhaust valve rocker arm installed in a KSV-20-T engine (identified at Unit III A), broke at Palo Verde. Although the inlet and exhaust valve rocker arms are not designated as safety related components, we believe the intent of 10 CFR Part 21 should be met by reporting the incident to you.

Attached is a copy of our report reference QCG-5309 which details the occurrence, our analysis, and conclusions. Also enclosed is a copy of a letter addressed to the Louisiana Power and Light Company (Waterford III) advising the Utility of the incident and its impact upon their operation. Similar letters are currently being sent to the'following plants:

Pennsylvania Power and Light - Susquehanna, Berwick, PA Coliiionwealth Edison Company - Byron, Braidwood, Zion, IL Nebraska Public Power - Brownvi lie, Nebraska Niagara Mohawk - Nine Mile Point 2, New York Houston Lighting and Power Co.- - South Texas Project, TX Please note that we are reviewing the safety classification of rocker arms with a view to upgrading their status.

890i240490 S90iia PDR ADOCK 0500052S S PDC Uncoln Avenue Grove City, Pennsylvania 16127 (41 2) 458-8000 Telex: 86-6613 INTEGRAL ENGINE COMPRESSORS ~ MOTOR. DRIVEN COMPRESSORS ~ POWER ENGINES

0

'.S. Nuclear Regulatory Comission January 18, 1989 Page Sincerely, F. Bruce Stolba FBS/gs Attachment cc: D. T. Blizzard - GC J. R. Rasor - NV D. E. Brazeal - Tulsa C. Rousselle - NO T. 0. Garrison - Odessa C. A. Robbins - MV W. R. Haass - NRC B. R. Sedelmyer - GC T. W. Kearns - GC K. R. Young - HOU W. H. A. Lambert - GC W. P. Purcell - ANPP W. Simko - ANPP File: K-5f OCOOPER wSMSTSS S COOPER BESSEMER RECIPROCATlNG

I OCG-5309 REPORT Arizona Huclear Power Project Exhaust Rocker Arm Failure 1.0 Introduction At 02:50 hours on Mednesday, January 4, 1989, engine III A (SH-7188) was started for a four hour operability run. By 03:24 the engine was running at full load and all operating parameters were normal. At 06:02 the engine shutdown on "overspeed" and subsequent investigation revealed that the number 8 left exhaust rocker arm had broken.

1.2 Immediately prior to shutdown, excessive vibration of the air intake piping system was observed. This was causing the overspeed switch on the butterfly valve to trip and reset. Operating records attest to this cycling of the shutdown. Ultimately, an excursion of the shut-down valve occurred without immediate reset following. Thus the shutdown which was not an actual overspeed.

1.3 As the investigation into the failure mechanism progressed it was decided to visually inspect all of the rocker arms on units III A and IIIB for cracks. The exhaust rocker arm on 9 right, Unit III A was found to have a crack similar to the "original" crack on 8 left.

On January 17 the other four engines at Palo Verde were inspected and .

no further rocker arms were found to have cracks.

1,4 The 8 left failure was initially examined by AHPP Engineering and Hetallurgical personnel and a representative piece of the fractured surface sent to Cooper-Bessemer. The results of C-B's evaluation concurred with those of ANPP.

2.0 Failure Review and Analysis

2. 1 It was immediately obvious that the point of ori gin of the fracture was an "old" crack. (See the attached sketch for details). Obvious because there was evidence of a yellow sealer paint on the fracture surface at the thinnest section of the push rod boss. This yellow sealer is applied prior to machining a casting such as the rocker arm, and is "touched-up" if necessary following machining. Ther'efore, the crack occurred during manufacturing.

2.2 It was noted that there was eccentricity of the "push rod end" bo're.

(Exaggerated in the sketch for the sake of illustration). At first the lack of symmetry, caused plainly by a casting core shift, was considered to have influenced the formation of the original crack.

However, when the 9 right exhaust rocker arm was examined. by AHPP, yellow paint was also found in the crack, but the wall thickness was considerably greater at approximately 5/16". The thickness of the failed specimen was approximately 3/16".

l P'age 2 QCG-5309 2.3 The exposed "push rod end" bore for the failed piece showed that the "push rod end", which is installed with an interference fit, had "worked" loose. There was evidence of both fretting and burnishing which suggests that the failure mechanism was that of impact loading with the final fracture following the original crack.

2.4 Mhen the 9 right cracked rocker arm was examined by ANPP it was reported that the "push rod end" was tightly in place and had to be forcibly removed. It is, therefore, postulated that the 9 right did not fail because it had not lost its interference fit. ANPP also reported that the "push rod end" bore exhibited rings consistent with the "push rod end" being inserted in a "cocked" orientation.

2.5 A calculation for "hoop" stress, with no credit for the stiffness induced by the junction of the "boss" to the body of the rocker arm or the flat "top", was made. The purpose was to analytically determine at what wall thickness the casting would fracture with nominal interference fit. This thickness was found to approx-imate .160". It is concluded that for the failed specimen the interference was on or about nominal and that for the cracked (only) piece the interference was greater than nominal.

3.0 Conclusion

3. 1 It is concluded from the .-foregoing (para. 2.0) that the castings were cracked during manufacture. The cause for the cracks being high stresses induced when the "push rod ends" were inserted.

Nanager Quality Control cc: A. Schleigh R. J. Brager D. T. Blizzard R. A. Niklos T. M. Kearns J. N. Horne W. 0. Ferguson B. R. Sedelmyer Fi le: 25-1

I OCG-5309 Crack Origin I (Paint in Crack)

+i J

]I i

// I Thin Wall Crack I

I

/

/ /

I

/ /

/ /

/ /

/ /

Punch Rod End SKETCH SHOWING PORTION OF EXHAUST ROCKER ARM KSV-25-A

COOPER. BESSEMER RECIPROCATING COOPER IIIOVSIIIISS January 18, 1989 Our Ref: QCG-5314 Mr. R. A. Legere Louisiana Power and Light Co.

P. O. Box B Killona, LA 70066-0751

Dear Mr. Legere:

Subject:

KSV Standby Diesel Generators Rocker Arms On January 4, 1989 an exhaust rocker arm broke on a unit at Arizona Nuclear Power Project, Palo Verde. The attached report details the incident, our analysis and conclusions. All inlet and exhaust rocker arms were subsequently inspected at Palo Verde where six (6), twenty (20) cylinder engines are installed. As a result we recommend that you also carry out a similar inspection as follows:

1. Remove the inspection door attached to the cylinder head hood by two bolts. It is not necessary to remove the hood.
2. With a mirror and flashlight check the bosses where push rods/valves engage the rocker arms for cracks.'rom the Palo Verde experience, cracks, if present, will be very evident.
3. Replace any rocker arm found to be cracked and report the replacement to Cooper-Bessemer.

'H. H. A. Lambert Manager, Q.C.

cc: C. Rousselle NO R. A. Miklos - GC T. M. Kearns - GC File: K-5f MHAL/kll Lincoln Avenue Grove City, Pennsytvania 161 27 (412) 45841000 Teiex: 499.7257 A/B CBCORPGRCT INTEGRAL ENGINE COMPRESSORS ~ MOTOR.DRIVEN COMPRESSORS ~ POWER ENGINES

0 'V

~

C C

~

l 4