Originally Posted by Cujet
I dispute the "wear metals per mile" hypothesis as either a measure of engine wear rate or an oil's performance. There are many reasons for this and there is not time enough to describe all of them. However, everything from oil related deposits to individual component wear (such as a single, fuel pump cam lobe, such as the Audi failures) can wildly skew the numbers. Leading to meaningless data in either direction.
In the aviation world, it's not only commonplace but 100% required (on some engines) to perform oil analysis. We regularly see worn out engines that are physically coming apart, providing superb wear metal numbers. This aircraft engine camshaft is example number 1, with superb UOA results and a significant wear rate:
With that in mind, I'm much more interested in the particle counts. With my point of view, circulating particulates are a major cause of timing chain wear. Elimination of said particulates, along with sufficient viscosity is and has always been, the answer to long chain life. Furthermore, it seems sufficient viscosity is also a factor in ring/cylinder life, independent of UOA results.
The aviation world sees far different operating conditions and extremes, including lower oxygen concentrations resulting in A/F corrections and considerations, severe temps swings, leaded fuel, redundancy in design (for obvious reasons), different metals used in some applications, different fuel delivery methods, different oil pump systems, etc.
It is true that there are times UOAs will not pick up all wear, especially when the wear particles are larger than what the UOA will detect. But there are LOTS of examples where wear rates were accurately tracked, and used to decipher/discover a wear problem. And there are lots of UOAs that show good wear, and were confirmed in tear down.
Knowing the engine design unique characteristics is key, as it the operational conditions. Engines that don't have a real Achilles heel can fair very well with UOAs to track wear, despite your objections.
You assert that PCs are going to help understand the conditions? OK- to what end? Show me the correlation between PCs and timing chain failure, or more importantly, how we can use PCs to predict timing chain wear, and when it would be prudent to change the timing chain before failure. And what size particle is the delineation of where timing chain wear becomes affected? And is this true for all manner of timing chains, or is it different for link-plate versus roller? Single or double row? Show me any data that helps us use PCs as a predictive tool to understand wear rates and when to change oil and/or a lubricated component.
I don't disagree that reducing a particle counts is a good thing; tighter filtration leads to less foreign material in the bloodstream so to speak. But how does one use the PC tool to a meaningful useful end? Show me the data that indicates it's actually helpful as a PREDICTIVE TOOL and not a reactive tool.
Further, where UOAs can tell us composition of the elements, PCs cannot. PCs tell us size and quantity, nothing more. PCs cannot distinguish between Fe (and the steel components thereof), Cu, Al, Pb, etc. If you saw an uptick in the PC quantity, what does that tell you besides "Hey, there's more stuff in here!" The PC has no ability to direct you to a potential source of the wear metal. In fact, PCs won't even tell you if it's metal, or soot, or some other insoluble. PCs cannot distinguish the make-up of the particles.
You state that UOAs are not accurate, but there are LOTS of SAE studies that show good correlation between UOA wear data and other physical measurements such as component mass weight assessment, electron bombardment, etc. There is even data that shows there is good correlation between PCs and UOA wear data in an SAE study.
I'll entertain your dissension, but you need to bring facts to the table for the discussion please. Make a claim? Back it up please. Show me that PCs are more accurate and useful than UOAs, if that is your claim. How would we use a PC analysis to accurately predict wear trends, and determine potential contributor of the wear condition?