I am predicating this conversation on a healthy engine; leaks are exclusionary to the debate.
UOAs may or may not be "a waste"; that depends upon how you fassion and interpret the term. I would agree that on small sump systems, it can often be cheaper to KNOW you flushed out POTENTIAL contamination, than the UOA and KNOW you don't have any. However, in this example, the OP has a sump that is fairly large, and with mutiple filters, the cost of an O/FCI is far less than a UOA. Knowledge is cheaper than guesstimate action in this case. So we move on ...
The add-pack, the filter and the OCI are used in conjunction to take a three-pronged approach to wear reduction.
When it comes to filters, PCs (particle count analysis) is the best way to judge filter performance. PCs can accuratly report on a filters ability to filter. But that is only an INdirect inference as to how they would control wear. People often point to the (in)famous GM filter study, but if you understand the DOE and how they manipulated the inputs, you'd not hang your hat on that study as anything relavant to the real world (and they directly admit as such towards the end of that study!).
Conversely, UOAs can also be used to judge filter performance in an indirect sense. UOAs can infer how well a filter is doing it's job, but as it is only 1/3 of the contributors, UOAs cannot speak directly to filtration.
But ...
UOAs are probably the cheapest way we have to look at wear directly attributed to the "wear reduction package" (filter, OCI, add-pack)
And so, we can use UOAs to look at wear metals and soot/insolubles and make a conclusion about how well the combined tri-effort is doing.
Genearlly, wear rates continue to drop all the way out to 15k miles even with conventional fluids, in most engines. This is proven in thousands upon thousands of UOAs from all manner of engines; gas an diesel, large and small, air and liquid cooled, etc. Also this phenomenon is proven in SAE studies. (See my normalcy article for mounds of concrete proof.)
One of the major contributing factors to wear reduction is the reaction of the add-pack to oxidation. It is actaully DESIRABLE to have some oxidation of the oil; it forms the tribo-chemical barrier that sit between the metal parts in addition to the hydrodynamic film barrier. In fact, the higher the heat, the greater the film barrier (within a reasonable limit that does not scorch the lube). The longer the oil is run, the more barrier that develops, and then wear becomes almost non-existent for a flat-rate period of exposure. (see SAE 2007-01-4133).
So, filtration is only a partial contributor to wear reduction, and in fact, once a minimum threshold is met, "better" filtration does not show any significant correlation to wear reduction. Again - look at the massive amount of data in my UOA research, and see the unique examples (Vulcan 3.0 and Dmax engine) of singular UOAs showing no correlation to wear with variation of filtration.
When it comes to the topic of media blinding off, I have this to say:
There are two separate scenarios here; I'll answer each ...
1) If the filters are opening into bypass with any regularity because they blind off, it must be at a point PAST 15k miles, or the wear rates would see a statistically significant jump up before that point. Hence, a 4k mile FCI is ludicriously short. Wear data proves that blinding of media does not happen early.
2) If the filters are actually blinding off early on as you claim, and opening with regualarity, it apparetnly has ZERO effect on wear. Why? Because wear rates continue to drop towards 15k miles and often beyond. If the media is blinded off, and no longer filtering, then what does that really say of the "need" for fitlers at all? When the data shows wear rates continuing to drop towards 15k miles, and we accept that your position of media blinding off early as true, then does that not indicate that filters have practically no effect on wear? In short, for your position to be presumed right, we'd have to see a statistically significant shift in wear UPWARDS at some point before the 15k miles. BUT WE DO NOT!
So, the reality is that it really does not matter from which position you argue. Whether or not the bypasss blinds off early or later, wear is generally dropping in the VAST majority of engines, all the way out to 15k miles.
Additionally, soot and insoluble counts would take a significant escalation if the bypass was opening often as well. They are contributors to wear, although they are not the only source of wear. And yet the data does not show that happening with any regularity either; soot sees a somewhat positive parabolic slope to occurence rate, depending upon each engine family.
You point of filter media blinding off is moot. It either does not happen often (as I suspect and Jim's experiment is premilinarily showing), or it does happen but has no effect on wear. 4k miles is total joke; it's easily within the reach of any typical normal filter. Most filters can easily go 10k miles, and premium ones can go further than that.
Finally, of all the data I have, there is no correlation that shows a causation of filter degradation when O/FCIs are pushed out past a year or two. Filter left on for up to 4 years show no shift in wear rates as opposed to those in shorter applications. Data speaks volumes here. Simple, and easy to understand. All you have to do is read the data for what it is. I don't rely on supposition and rhetoric. I base my statements in facts gleaned from reams of data directly attributed to UOAs in massive market applications.
Now - I would appreciate it if you would NOT misinterpret my points here. I am NOT stating that filtration is worthless or can be ignored. I am NOT saying that filters last forever. What I am stating is that the HUGE amount of data over tens-of-thousands of UOAs I have, show that once filtration meets a decent min performance level, it can hold that level for a LONG time, and wear continues to drop (for most engines) out to 15k miles. And it is imperative to understand that wear control is achieved by three things; not one.
