There was one (EG or TG) where the leaf spring wouldn't sit all the way down because the center tube was in the way - either bent inward or too long, can't recall. A few bad examples doesn't mean they are all like that. Anyone reading this forum for a while knows every filter brand has some issues now and then that or outliers, not the norm. And some have issues that are the norm, and with enough data it becomes apparent which ones those are.
I haven't compiled all the examples, so it's possible I'm remembering wrong, but it seems to me that there have been many different Fram failures of various types. In the past few months, it seems like virtually every Fram filter has had seal leaks, badly uneven pleats, core misaligned with the media, threads cut wrong, etc. etc.
I generally agree with you that all manufacturers are going to have a few bad products, and straight statistics will confirm your statement that one shouldn't make too firm of a conclusion based on those occasional bad examples.
But taking a more Bayesian approach, the clear context of recent Fram defects means the
prior probability of defective samples from Fram is higher than for other manufacturers, so the
new evidence of a defective product will result in a much higher
posterior probability of a defective Fram filter.
I.e., the context here makes individual observations of Fram defects far more significant that it would be for other manufacturers, so a more Bayesian (rather than straight) statistical approach is more appropriate, and that approach makes Fram products more suspect based on individual defect observations than it would for other manufacturers.
At least, that's how I'm seeing it. And I think that mathematical reality is the basis for many to come to the same conclusion via simple intuition.
I would before using a rippled leaf spring sitting on a metal end cap with no seal between them. Fiber end caps have an advantage that they provide a natural sealing material.
Agree with you on that theory-- and it was speculated that perhaps the fiber endcap models might be immune to ruffle-induced leakage because of it. However, recent examples seem (at least to me) to indicate that the fiber endcaps are leaking as well.
I believe it was you that made the reasonable speculation that the fact that even the fiber ones didn't seal might indicate that in addition to the ruffle problem, the stampings might be so far out of flat that the springs couldn't even seal against fiber.
It's probably more of a regular basis to not have an issue than to have an issue at the leaf spring seal on filters with fiber end caps. When I see way more instances of a problem with metal leaf springs sealing on a fiber end cap then I'll say there's an on-going issue.
That's the correct approach to the science of determining the problem, but as a practical matter I would guess that most will be hesitant to take the risk of using filters that might end up providing those additional instances that would confirm an on-going problem. I know I sure would be!
Most will probably be like me and go the other way-- I'll start using them again once I see few/no instances of leaks.
And as mentioned earlier, the leaf spring needs to be put back into it's original position on the end cap before doing a light leak test on it. It takes a set into the end cap while it's assembled, and the impression isn't totally uniform, so the leaf springs seal can't be accurately evaluated unless it's back in the same position is was in before the filter was cut open.
Agree with that.
