Originally Posted By: ZeeOSix
Originally Posted By: Wilhelm_D
Originally Posted By: ZeeOSix
That really doesn't answer my question, because there isn't a rational answer to say that only guys on BITOG that happen to cut open oil filters are the guys who get bad Purolator filters with media tears.
You haven't presented a question to answer. You've presented an argument in favor of a hypothesis.
There is not enough objective information for you to reach any conclusions that could be projected onto the universe of Purolator filters. No amount of speculation is going to turn anecdotal evidence into something statistically meaningful.
Nothing prevents you from reaching some personal conclusions as to whether or not you wish to use Purolator filters.
The bottom line is that gut feelings and intuition based on less than "ideal statistical data" is still usually right. I've seen it happen many times in cases just like this.
This is my first post. I had to wait some hours for registration to go through, so I lost some verve, so I'll keep this post relatively simple. I am not trying to rile up old arguments. However, the point is, I've lurked these forums on and off for years, always seeking intelligent discussion on elusive tribology answers.
I was motivated to register last night. Why? Because after days of taking note of the Purolator issue, in my search for an optimal lubrication solution for my built 4AGE, my irritation pushed me to register. I was irritated that almost nobody on this site was calling this Purolator "fiasco" for what it clearly is.
In short, ZeeOSix, you are right. There is a problem here. The statistical analysis angle in this thread is a rather moot point. Your "gut" feeling is actually a quite rationally discernible reality - the failures in this case are completely indicative of a design flaw from Purolator engineering. I will probably do some more reading here and create a more useful, focused discussion topic. But I wanted to support what your "gut" is saying, and point out the problem. Others have touched on it, you included, but most dont see it:
While various factors may contribute to these failures - rigid seam crimp (strain discontinuities), assembly errors, media physical/strength properties, bypass performance, etc - the clear problem is a design flaw. I contend the design flaw is in fact the "V" shaped geometry of the filter media, and its interaction with the bonded ends/plates. Quite simply, the pleats in these zones are unsupported against pressure loads. They experience pressure load with only the tear strength of the media (local to the bonded ends) to resist it, hence tearing ensues. All other pleats are pressed against each other. They mutually support each other so that pressure loading can only deflect the media "so much" thereby limiting strain, and avoiding tearing. The pleats near a seam need to be configured in the same way, but are normally angled away from the seam for purposes of assembly logistics. This was not done with these failing filters. Now, it's possible that these configurations were a random result due to an error in manufacturing, but I am doubting that, as certainly any proper inspection system would have caught the flaw, and that responsibility still falls on engineering.
In short, the "V" configuration is the prerequisite for this failure mode. Other factors may contribute, and failure is not guaranteed. But I assert that you wont find this tearing unless that "V" configuration is present. Note, the vertex of the "V" must be pointing outward, not inward. Again, it is in this arrangement that the pleats will experience high deformation due to lack of mutual support, leading to tearing. My hypothesis is easily disproven, with a single example, if somone can find one.
To briefly touch on and rebut the statistical arguments. Following proper statistical methods for diagnostics is always important. In this instance however, that the fact that BITOG community members are finding these problems, does not pan out to be a predisposing factor. While confirmation bias can skew data, in this case looking closely at the statistical concerns demonstrates no such expectations.
Essentially, the ideal sampling method would be "open 100, or 1000, or whatnot, filters, and report what portion has this flaw." That is certainly not what's happening here. People find flaws, and report, but usually, those that find no flaws, see little reason to report. Again, generally - it's well known that community forums are typically replete with common complaints, but are rarely filled with reports of met expectations. At this point though, it's a different situation. It is clear now that users here are reporting on Purolator dissections regardless the findings, and some are finding no tears. In my casual observation, I'm seeing many more reports of torn media than untorn. Other factors come into play as well, like service duration.
There are a few other things to keep in mind. We must define what "normal" performance is, both for Purolator filters, and filters in general. These failures are absolutely not normal or acceptable by any design engineer's definition, nor are they random failures. They follow a consistent, specific failure mode. This alone indicates a specific cause.
Another thing to keep in mind is that it is safe to assume that this discovery occurred randomly. It could have happened with any filter type which BITOG users may be using, but it happened with this brand. Now, perhaps BITOG users use Purolator more than other brands. Is there a significant enough skew to put us where we are now? I very much doubt it.
What I'm getting at is, if we take the statistical concerns point by point, we can see that there is little or no reason to suspect that we are seeing something "normal" and improperly ascribing them to be abnormal.
If these were otherwise random failure modes, proper statistical sampling methods would be required to correctly ascertain the extent of this problem in the population. In this case, the failure mode is essentially identified as a design flaw. Design flaws effect 100% of the designs containing the flaw.
Finally, I'd say, these flaws may or may not lead to any serious consequences to end users, but it is certainly quite possible that nontrivial engine damage could result. The size of these tears, when deflected under pressure, seem like they may indeed exceed the flow of a bypass valve. I view this as a clear engineering design flaw, that would not have happened with prudent design methods. It's hard to comprehend how a filter maker could make such an obvious error. Possibly a random oversight... Thoughts of industrial espionage/sabotage came to mind... Regardless, my faith in Purolator went to zero. Then again, we cant assume other brands are immune to the same issues. The only solution is to always inspect the actual product being used, both before and after being placed in service.
Again, Id like to see BITOG start something more in line with scientific investigation of these failures. A proper discussion topic, maybe a poll, some manner of assessing the situation. I may start something like this myself, this weekend. I dont expect Purolator to be forthcoming in the interest of the end user. Their first priority right now is understandably to protect their reputation.
Dang, I said I'd keep this short. LOL.