ML20080B777

Affidavit of SG Christenson Re Suffolk County Contention 1 on Emergency Diesel Generators.Each Generator at 4,890 Horsepower & 3,500 Kw Is Overrated & Undersized to Meet Operating Requirements of FSAR Section 8.3.1.1.5

ML20080B777
Person / Time
Site:	Shoreham File:Long Island Lighting Company icon.png
Issue date:	02/06/1984
From:	Christenson S SUFFOLK COUNTY, NY
To:
Shared Package
ML20080B778	List: ML20080B777 ML20080B791
References
ISSUANCES-OL, NUDOCS 8402070264
Download: ML20080B777 (9)

Text

. . .. . - -- . _ . .

COCKETED UERC ATTACHMENT 1 UNITED STATES OF AMERICA 'N FB -5 P2 57

! NUCLEAR REGULATORY COMMISSION

~

i Before the Atomic Safety and Licensing Board i

i i )

In the Matter of )

)

LONG ISLAND LIGHTING COMPANY ) Docket No. 50-322-OL

)

(Shoreham Nuclear Power Station, )

Unit 1) )

)

2 AFFIDAVIT OF STANLEY G. CHRISTENSEN Stanley G. Christensen, duly sworn, deposes and says as fol-lows:

1. My name is Stanley G. Chrintensen. I am a professor of marine engineering at the United States Merchant Marine Academy, Kings Point, New York. I am currently one of the expert consul-tants for Suffolk County, New York. My principal area of i

i expertise is diesel engines,mphinawhich I have been involved for i

over forty years. My experience includes serving as chief engineer in charge of diesel engines in marine applications, and teaching a variety of courses on the design, manufacture, opera-tion and repair of diesel engines and their components. I am the l

i author of numerous publications concerning diesel engines, including Lamb's Questiuns & Answers on the Marine Diesel Engine, I a standard reference bouk. My professional qualifications are set i

forth in more detail in Exhibit A to this Affidavit. I have been

! B402070264 840206

! PDR ADOCK 05000322 PDR G

,Qgg

1 g

. .  ; providing services to Suffolk County concerning the emergency diesel-generators ("EDGs") at the Shoreham plant since June 1983.

2. The purpose of this Affidavit is to explain my technical

• 5 judgments regarding Suffolk County's proposed supplemental EDG contention I. In reaching these conclusions I have reviewed and analyzed all of the reports and memoranda issued by Failure

-Analysis Associates concerning EDG parts and components which were made available to Suffolk County, as well as reports by the NRC I

Staff, documents concerning the EDGs furnished to the County by i

LILCO, and documents received by the County from owners and I operators of other diesel engines manufactured by Transamerica Delaval, Inc. ("Delaval"). I have also visually inspected and taken measurements of certain parts and components of disassembled j

L EDGs, and I have visited Delaval's facility in Oakland, Cali-1 fornia, inspected the drawing of the Delaval cylinder head, and i

held discussions with Delaval personnel regarding the cylinder 4

heads. Finally, I have consulted various articles and texts as

appropriate.

j 3. I have concluded that at 4890 horsepower and 3,500 kW, i

i each EDG at Shoreham is over-rated and undersized to meet the l operating requirements of FSAR Section 8.3.1.1.5 for continuous i

operation (8,760 hrs.) at 3,500 kW and overload operation for 2 ,

i l hours in any 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period at 3,900 kW without reducing the 1

maintenance interval established for the continuous rating.

l

4. Each EDG is a Delaval model R-48 diesel engine. The

! Delaval R series diesel engine was developed in the mid-1950's, i

i with an initial rating of approximately 250 horsepower per cyl-l 6

i )

Y

. inder. The horsepower of the R series engines was increased to 611 horsepower per cylinder in the EDGs at Shoreham without any fundamental redesign of the engine. Instead, changes were made in the design of particular components. For example, in the late 1960's Delaval introduced the R-4 model (the same model as the

-Shoreham EDGs) with the current cylinder head design that in -

/?

creased the flow of air through the engine; by this change coupled O

with an increase in operating speed, engine horsepower was boosted without a major overall engine design change.

5. A diesel engine is a complex piece of machinery whose parts are in dynamic inter-relationship. Thus, a change in the design of a major component, such as a cylinder head, a crank-j shaft, or a piston, is likely to effect other engine components and, indeed, the performance of the entire engine. In my judg-ment, the piecemeal changes made in the design of particular

} components of the Delaval engine and other steps taken to more i

i than double the horsepower rating of the R series engine have i resulted in an EDG which is over-rated and undersized. For that reason, numerous parts of the Shoreham EDGs were found on dis-

! assembly to be damaged, and it is likely that additional damage will occur to parts and components with EDG operation at 3,500 kW or higher power. My conclusions are supported by the factors dis-cussed below.

6. I have concluded that the original crankshafts with the

. 11" pin cracked principally because they were undersized and 4

poorly designed, although my reasons for this conclusion are somewhat different from those of Failure Analysis Associates 1

. ("FaAA"). The standards for crankshaft design in large diesel engines (like the EDGs) used throughout the world by international

! insurance companies are those of the American Bureau of Shipping, Det Norske Veritas, and Lloyd's Register of Shipping. These standards are more realistic than the voluntary standards of the Diesel Engine Manufacturers Association, an organization composed of six domestic engine manufacturers (including Delaval), whose

! standards are not approved by the American National Standards Institute or any recognized international standards organization.

FaAA failed to judge the adequacy of the original or of the re-4 placed crankshafts (with 12" pins) against the internationally recognized standards for crankshaft design.

7. From measurements of the replacement crankshafts and other data concerning their specifications, I have performed the  ;

necessary analysis to determine the adequacy of the replacement i

crankshafts under the applicable standards of Lloyd's Register of Shipping. My analysis shows that the replacement crankshafts are inadequate for operation at overload power (10% above 3,500 kW) for periods more than 15 minutes. Beyond that period the crank-shafts can be expected to develop one or more nucleation sites.

I

! While not necessarily visibly evident in the earlier stages of

, overload, such sites would later cause crankshaft deterioration and possibly failure.

I

8. My analysis also establishes that even at normal loads up to 3,500 kW, the replacement crankshafts are only of marginal j .

l design, with little capacity to safely handle additional potential l stresses caused by such factors as unevenness in cylinder firing 9

r-__ .. -.,.-__,._,,--,-,,v.,,,, 7, ._,_,,,w,4_ - - - . . . 4.,,e , rec--,,,,....,y.,3.,-....w-,,---.. .r.,,.,.,- ,, e . , . - , - . , -

pressures, or failures of other components such as bearings, cyl-inder heads, or jacket water pumps. Accordingly, one cannot have confidence in the reliability of the replacement crankshafts even when not operating at overload power.

9. The greater weight of the replacement c::ankshafts and the consequent increase in the whirl created by internal unbal-anced centrifugal forces acting conjointly on the center main bearing will cause excessive wear on the main bearing between the number 4 and number 5 cylinder. In addition, the effect will loosen the main bearing and cause ntisalignment of the crankshaf t, which could lead to crankshaft failure. While the effect of in-creased crankshaft whirl may be somewhat ameliorated by the. fact that Delaval has reduced the weight of the replacement connecting rods, that design change might weaken 'ihe rods and cause problems in other bearing locations.

10. The cracking of 23 of the 24 pistons in the three EDGs evidences that the pistons were of inadequate design to withstand firing pressures to which they were subjected in the over-rated EDGs, even after a relatively low number of operating hours. This is another indication of the lack of design integration as the series 4 engines were uprated in horsepower. Although I have not yet been able to examine a drawing of the replacement "AE" model Delaval pistons, I have analyzed information regarding their di-mensions and design. My analysis leads to the conclusion that the model "AE" pistons are also of inadequate design to withstand operating conditions. The ratio of the height of the piston skirt relative to the diameter of the cylinder bore (referred to as the

sliding length bore ratio) is too low . This indicates that the bearing pressure of the sliding part of the piston skirt will be

' relatively high, causing overheating which can lead to increased wear on piston rings and to piston skirt cracking. The fastening studs for holding the piston crown to the skirt appear to be in-adequate. The tin plate surface of the skirt, while providing some benefit by preventing scuffing of the skirt, allows dirt to collact and thereby reduces piston clearance, which leads to over-heating and scoring the cylinder walls. The position of the piston pin relative to the piston skirt is undesirably high, pre-sumably to accommodate a longer connecting rod. This increases 1 the bearing surface pressure on the piston skirt, which leads to overheating that, in turn , may cause the skirt to crack. Finally, the "AE" model piston is a new design which is unproven. Its operation, according to Delaval, is limited to testing in an R-5 test engine and field operation in a single engine. This does not provide sufficient empirical data to reach positive conclusions about the "AE" piston. Piston failure can have adverse conse-

! quences on the crankshaft and other components, and can degrade

{ EDG performance and even cause engine failure.

11. Cylinder heads installed in the EDGs cracked during l

operation and were replaced with new cylinder heads of the same design but allegedly superior quality. The design of the cylinder heads is inadequate to withstand the high thermal and mechanical i

stresses to which they are subjected during operation. The

thickness of the firedeck is uneven. Delaval's design permits

, thickness variations from as little as .400 inch to as much as

.-r 4-w,. .,m.w--,w ~_-,--.,r,_-.,y., c,,..w,a-r,,--w- ...--y . ..n.,,m. . . . - - . , , . . . , , . . , , , . .~w.... -.cw.-,w-_.- - . , ~ .. - , - , . - .

. .861 inch in a firedeck with a nominal specification of .500 inch.

These variations result in increases in either thermal or mechanical stress due to gas pressure, create a stress riser in transition zones, and lead to cracking. The water passages in the cylinder head do not provide for sufficient cooling, which exacerbates the problem of the uneven firedeck. The studs holding the cylinder head in place are not properly spaced, which creates non-uniformity of stress; this again leads to failures in the higher stressed areas.

12. The replacement cylinder heads are not of acceptable manufacturing quality. Of thirteen replacement cylinder heads for the Shoreham EDGs visually inspected by the NRC Staff, one had a crack 3/8 inch long. Delaval nevertheless found the cracked cyl-inder head acceptable because it had passed a hydrostatic test without leaking, and the cracked head was installed in a Shoreham EDG. See NRC Inspection No. 50-322/83-25. The County's metal-lurical consultant informed me that cracks tend to propagate and grow, and I believe a cylinder head known to be cracked is unac-ceptable for installation in EDGs. Apparently none of the replacement cylinder heads were inspected for cracks by dye penetrant or magnetic particle examination by the NRC or LILCO.

13. Additional cracked heads may be assumed to exist in the EDGs. FaAA reported that (C]orrosion of the cylinder liner and piston crown was observed in cylinders of two engines . . . .

EDG Crankshaft Failure Investigation, October 31, 1983, at 1-5.

FaAA does not indicate whether it determined the source of the

. , I corrosion, or the number of cylinders in which corrosion was noted. Nor does FaAA indicate whether or how it inspected the cylinder heads in the EDGs for cracks after the crankshaft fail-ure. On the basis of existing evidence, jacket water in the cylinder resulting from a cracked cylinder head may have con-tributed to the early failure of the undersized and poorly designed crar.kshaf t in EDG 102.

I

14. The foregoing discussion of certain major components of the EDGs shows their interrelationship and how each is of inade-i quate design to satisfactorily withstand the operating stresses of the EDGs are their rated power. The weaknesses of these major components may be exacerbated by their inter-relationship with each other and with other components during operation. For ex-ample, water in the cylinder caused by a cracked cylinder head could adversely impact upon the pistons and the crankshaft.

Another example is that a minor failure in the subassembly of the l

4 cylinder head could impair the operation of the head valves, which in turn could overstress the crankshaft.

i 15. Additional evidence that the EDGs are undersized for i

their intended functions and over-rated is that exhaust tempera-I tures in the EDGs are very high (approximately 1,100 degrees F.).

Further, the operating experience of the Delaval diesel engines on e

the M.V. Columbia evidences many of the same problems occurred as l

i 5

t j

i

. -9_

4 a

with the Shoreham EDGs; the engines on the Columbia had to be de-

- rated. Finally, the excessive vibration and variations in vibra-tion levels among the three EDGs may indicate over-rating.

/

L .

o g; ;.. % . , 1. --

. , / .a . . ; . .j. . . .

Stanley G. Christensen

,i

! Subscribed and sworn to before me this Ji7 0 day of January, 1984.

i' i

Notary P@lic

/E . ./

• Y" I

T

• IFU?fY t;~~/.'.Y .tPR '!:te of New Yosk l Mo. 00/f78225 C mMW in Nossau county Cc erseen byrcs March 30, .!.9.g 4

i 4

4 4

I t

i i

t >

r I

. - . -.-#,.,.. - . _ ~ . , - -_-._,_,..--,..-.,,w_.,,,,y._..,.. ....--,-..,y ,.m.,..,~,.,- .,m,,,,-~..,_.m.mm_m,.,_____wm,,.-. .,mm,.., s.r..