I send UOA to blackstone labs for my wife's car as their's virtual no data on how GDI hybrids beat up their oil so I'm familiar with the process.
For this 459cc engine I could fill the engine oil real good, run it a bit, pull an sample before the filter goes on as its only got about 2 hours on the break in oil. Install the filter try not to add a bunch of fresh oil between filling the oil filter and taking the sample then run it for another 2 or 3 hours and send them both in. If there's not enough different to make a difference then I know it will be a waste of time.
But i have noticed unfiltered small engines typically recommend 25 to 50hr oil changes and filtered small engines go 100, 125, to as much as 200hrs. And the filtered engines always last longer if the oil gets occasionally changed. I found that for harsh home use, say mowing several acres to 10 acres, a filtered engine typically always out lasts the mower where an unfiltered small engine the engine going bad is why the mower gets junked.
Honestly, I don't think the venture is worth the effort, because the costs of doing the "experiment" proof (UOAs) would eat up any potential savings. But I suggested it because it's important to "know" (not guess) if the experiment worked.
Another option would be to do the before/after trials and get some PC (particle count) data. That would imply a sense of benefit if the data showed reasonable improvement.
But even the BP system itself wouldn't be cheap. I'm sure you're aware of this, but let's enumerate it for anyone not familiar with the process:
- find a port to tap into the pressure circuit
- get a port-tap device (threaded hose barb or hard-line)
- buy and find a place to install a filter mount; one that would include a restriction orifice to slow the flow rate
- buy a decent quality BP filter (these are not cheap no matter what the brand)
- find or create a port to return the oil flow
- buy hoses and fittings to complete the oil circuit
- then start into the UOA and/or PC process (more expense).
Honestly, shorter OCIs would just be cheaper and easier, than doing all the above. Unless you're just into the joy of tinkering; that's got some intangible merit for many folks.
A better quality air filter (as others suggest) would probably pay off a better ROI and be easier to do.
*********
My general point is that doing the experiment could be fun, but if you want it to have true meaning and credibility on BITOG, you have to be able to
prove it's merit, not just brag about something with some pictures, which validates nothing.
DATA is king; the rest is just bench-racing.