Let's talk Transmission Maintenance...

[JCL]

Originally Posted by ard

Not sure what you mean.

If a UOA shows low contamination, viscosity in range are you saying "the friction modifiers may still be depleted"????

Or are you saying 'the fluid may be contaminated, and the viscosity low, but since the friction modifiers MIGHT still be OK it should go another 40,000 miles'??

I don't understand why a UOA not picking up friction modifiers (I'll take your word on this) means that UOAs cannot be useful to quantify ATF life.

Sounds like your recommendation is 'since UOAs might not pick up friction modifiers, lets stick with 100k intervals...and once you get to 100k, if nothing is slipping, better not mess with it.'

A

PS You must agree that maintenance strategies are statistically based...so not insurance. Indeed I think it was you that brought up the issue of balancing the risk of failure due to fluid/lubrication issues against the likelihood of sensor/electronic failures.

A fluid analysis is going to show the presence of water or not. Not really relevant, in a sealed system. It is going to show the kinematic viscosity. Automatic transmission fluids have a fairly wide range for viscosity spec (just see the difference between DexIII and DexVI), and viscosity is even less critical given that the transmission automatically compensates for changes in viscosity as clutches wear (using a feedback loop on clutch engagement times, for example). As well, an analysis is going to show dissolved metallic particles. By definition, these are dissolved and are small enough to pass through the filter, and are not abrasive. They can certainly indicate wear trends over multiple samples, but their existence isn't a problem in a sample. The particles that can hurt the transmission are large particles, and you could do a count for those, using a microscope, agreed. But the big difference between various transmission fluid specs is the friction characteristics. Those characteristics are matched to the clutch plates and software during design. It is the only reason there are so many fluid specs, because of the use of various friction modifiers designed to achieve smoother shifts. I just don't know how to test for the condition of those additives using an oil analysis. It isn't typically on the test order form. My suggestion is that the presence and condition of those additives is more important than the things that are easy to test for.

My recommendation is that if there is information that supports longer transmission life through predictive and preventative maintenance, then it could be worthwhile to do that maintenance (if the cost of the maintenance is lower than the cost of shortened component life).

While maintenance strategies are often linked to statistics, that approach leads to fixed maintenance intervals. Modern preventative maintenance (the optimal path) is by definition in advance of failure, so actuarial statistics are usually not available. Decisions need to be made in the absence of full information, as uncomfortable as that is. Otherwise it would be easy. Judgement and analysis of failure modes becomes much more important.