And “"Therein lies the rub"…
We still don’t have any actual EMPIRICAL DATA that substantiates this claim!
If you make that claim, then you can't keep saying it's not an issue based on BRs testing. Saying two conflicting things, and only accepting one you think it true, regardless of the "no actual empirical data" mantra.
The small “gap“ relative to the OVERALL area of the filter itself is minuscule! Furthermore, there’s absolutely no accounting for the fact that the small amount that could get around the filter (“bypassed”) is promptly RECIRCULATED back into the filter continuously.
It's not about the size of the gap relative to the area of the filter. It's a leak, and if it's leakage percent based on the total flow through the filter media, then that's how it needs to be analyzed. The bigger the leak gap, the more the leakage percent, and the more it lowers the efficiency of the filter. It's not that hard to understand.
Sure anything that leaks past the media gets recirculated for another chance to be caught. That's really not how a filter should work, and filters aren't suppose to have a constant internal leak past the media. But if you like that then go for it ... but others who understand how filters should actually work really aren't buying into that "but it might get caught on a different pass" excuse as a reason to use a leaky filter.
Lastly, I would not simply “dismiss” all the results that WE DO HAVE (BR Tests and used oil particle counts) that support the superiority of this filter “in action”.
So now BR's data is "empirical data" when you said before we have no "empirical data". It's only "valid empirical data" when it fits your narrative, lol. Like said before, based on the PC model and just from pure common sense, a non-leaking filter that ISO 4548-12 tests at 99% @ 20u can not still be that efficient with any internal leakage. And if it tested at 99% @ 20u with an internal leak, then it would have to be a very small leakage ... it would have to be around 1% or less of a leak.
How do you know that the filters BR tested that ranked about the same as the inefficient Boss (99% >46u per M+H) didn't have leaky leaf springs? That's basically the only way they could all rank around the same because there's a big difference between them and the Boss per official ISO 4548-12 efficiency testing.
All you have are “theoretical results” based on assumptions. SHOW me one legitimate instance where this filter FAILS TO PERFORM as advertised…”99% @20microns” which btw, is backed by ISO tests that you claim are “ambiguous”. The burden is on YOU…. not me , arm chair observations… assumptions and calculations notwithstanding.
Get an ISO 4548-12 test done on some leaking vs not leaking filters and you'd see the difference. The PC model basically shows what would happen in an ISO 4548-12 test comparing a non-leaking filter to one with say a 10-15% leak going on. The ISO testing Ascent did now 4 years ago (back in 2021) showed the RP did well, but I'm betting it also didn't have a ruffled leaky leaf spring. And who knows exactly when Fram ISO tested the FE ... most likely before it even hit the streets back in April 2023. So that that were ISO tested probably had a non-leaking leaf springs. Back when they knew how to stamp a smooth flat leaf spring.
But the bottom line is based on some of the major leak gaps seen on Champ Lab made filters (including the FE), those filters could have pretty reduced efficiency due to the leakage, depending on the level of internal leakage. The PC model shows that, and it's a valid analysis of the affect of the leak on the efficiency. You can't prove otherwise, and probably don't really understand the physics, but that's fine as long as you think that leaky filters are still doing what you want them to do. When is the last time you cut open an FE you used to look at the leaf spring and leak gap?