Percent Fuel dilution to cause viscosity loss

[tangel]

Good morning everyone. A little background information, I am completing a failure investigation on a Caterpillar D398 diesel generator engine which wiped all the main bearings. The engine components show no damage or indication of loss of lubrication except for the main bearings. Two of the seven main bearings spun, the others have extensive visible damage. I have ruled out loss of oil pressure as a cause and am now looking at the possibility of loss of hydrodynamic film due to fuel-oil dilution. My question is is there a linear relationship based upon the viscosities of diesel fuel and the lubricating oil? My thought process is the 160 gallon sump level would have had to increase by a large (noticeable on dipstick) amount if enough fuel to thin the lubrication oil had entered the sump. The results of the spectrographic analysis of the oil will be a few more weeks, and am trying to find a "smoking gun". Thanks in advance for all comments/suggestions.

 

[Latexman]

Perry's Chemical Engineers' Handbook (7th Edition) reports a "mixing rule developed by Kendall and Monroe[sup]48[/sup] is useful for determining the liquid viscosity of defined hydrocarbon mixtures. Equation (2-119) depends only on the pure component viscosities at the given temperature and pressure and the mixture composition.

?[sub]m[/sub] = ([Σ]x[sub]i[/sub]?[sub]i[/sub][sup]1/3[/sup])[sup]3[/sup] (2-119)

For mixtures of the same chemical family, errors average less than 3 percent, while errors overall average 5–6 percent, with errors of mixed families averaging from 10–15 percent.

48. Kendall, J., and K. P. Monroe, J. Am. Chem. Soc., 39 (1917): 1787."

It's an old, simple correlation that looks fairly accurate. It may be just what you are looking for.

Good luck,
Latexman

 

[Fabrico]

You will likely not find even a general ratio at which damage will occur. Things like engine design, load, oil type, temperature, etc. would significantly affect any threshold.

How many hours were on that engine? When was it last serviced or worked on? Who worked on it? Were the bearings that did not spin damaged by poor lubrication, or contamination from the spun bearings? How did the rod bearings look?

Generally speaking, mains are often the first thing to show signs of wear, trouble, or failure. The condition of the rest of the engine does not automatically point to anything.

 

[BrianPetersen]

The engines that I'm familiar with generally spin rod bearings long before main bearings. How do the rod bearings look?

 

[tangel]

Thanks for that forumla Latexman, thats precisely what I was looking for.

Fabrico, this engine actually has a history of this same exact failure, once in 2003 (approx 8600 hours since complete overhaul), again in 2004 (2653 hours since complete rebuild from '03 casualty) and the engine had 8723 hours on it since the complete rebuild in 2004. (Normal rebuild intervals are 19000 hours) I wasn't trying to find a ratio that would cause failure, but just rule out the possibility of fuel thinning the oil an appreciable amount without greatly increasing the sump level.

The rod bearings all looked fine, which is part of why these failures are confusing. In each case the block was checked for straightness, new main caps installed where the bearings spun and align bored in place... I don't want to get started, the analysis is 11 pages thus far and obviously Im still working on it. Thanks again for the suggestions.

 

[catserveng]

Old CAT SOS (Scheduled Oil Sampling) Guidelines indicate 0-5% fuel dilution "Normal" 10% "Abnormal" 20% "Excessive".

Tests using Setaflash are calibrated OK at 3% and flash above 4%, with fuel dilution above 4% being an indication of fuel dilution.

The D398 is a pretty robust engine, and main failures at the hours you indicate are abnormal, we have a large number of engines going 25,000 to 30,000 hours on bottom ends.

If you are doing a failure analysis, here is some CAT info to provide assistance, contact your local dealer to order,

SEBD0531 Engine Bearings and Crankshafts
SEBF8041 Guidelines for Reusable Parts, Specifications for Crankshafts in Caterpillar Engines
Applied Failure Analysis Guides,
SEBV0544 Engine Bearings
SEBV0548 Crankshafts
SEBV0567 Lubrication

The Noria Corp has a large amount of refence material available regarding PdM and fluid analysis

http://www.noria.com

The peak load area of the connecting rod bearing is the most highly loaded bearing surface, and likely the first area where a lack of lube or viscosity problem shows up. D398 is an old family of engines, and you may have a crank journal profile problem, a bearing bore or line bore problem, or an abnormal loading issue.

Hope that helps.