New Oil Filter Study

[ltslimjim]

The 5 filters you mentioned are considered top of the line, and pricing reflects this as all can't be had for less than $10(as much as $15+), while a PureONE and Wix/NG are considered competitors and are considered mid-grade options; their prices too reflect this as they can be had from $4-$7.

The 'grades' pretty much match pricing. Ironic, huh?

 

[Sayjac]

Originally Posted By: ltslimjim
The 5 filters you mentioned are considered top of the line, and pricing reflects this as all can't be had for less than $10(as much as $15+), while a PureONE and Wix/NG are considered competitors and are considered mid-grade options; their prices too reflect this as they can be had from $4-$7.

The 'grades' pretty much match pricing. Ironic, huh?

Though I have no doubt that all the filters finishing at the top are fine filters, that is a point. And, as has been previously mentioned here, even filters with the very same media showed very different results. So yeah, one of those enjoy the dissection pics, ignore the grading results, studies.

 

[Alex89gp]

This test does not go far enough. Statistically, no valid conclusion can be made about the filtering capabilities of each brand with just this data. The author would have to test each brand and model multiple times (with a new filter each time), take an average, and compare that average to the other brand averages by using statistical functions. By just testing one filter for each brand, it's not even known whether a poor test result is an outlier.

 

[Seguino]

Curious that the Wix and the supposedly identical NAPA Gold have different amounts of media.

 

[jmsjags]

has anyone tried to get in contact with the author of the study to ask him about some of these discrepancies? we need to get him on this forum so we can put him in the hot seat

 

[Gary Allan]

Originally Posted By: Seguino
Curious that the Wix and the supposedly identical NAPA Gold have different amounts of media.


For all we know they were not produced using the same batch of media. That is, I'm sure that there's more than one way to skin a cat and a given run with a given media may result in different square inches being required. One may have been made yesterday and one may have been made a year ago.

I do not know the process here, but have seen enough in such kind of manufacturing to know that you make filter material and you test it and grade it. "It is what it is". You don't reject anything unless it cannot meet some spec. It would then fall into a grade that "could" be selected for a given run at a given ^2" ...etc..etc.

Again, these are just a few of the "well suppose" scenarios that could explain it.

 

[Pete C.]

Pleats are counted, either electronically or manually, using a +/- X number of pleats. Example, If 55 pleats is the desired, right on the money count, The spec might say 55 +/- 5 pleats. You might notice some elements have ink on the paper where the seam is. This is the marker for where to cut the element.

 

[Seguino]

Originally Posted By: Pete C.
Pleats are counted, either electronically or manually, using a +/- X number of pleats. Example, If 55 pleats is the desired, right on the money count, The spec might say 55 +/- 5 pleats. You might notice some elements have ink on the paper where the seam is. This is the marker for where to cut the element.


Thanks for the explanation, Pete.

 

[dvldoc]

What was up with this cardboard end capped Fleetguard model, I expected it to look like at least a Baldwin not a Fram.

 

[Sayjac]

Yep, that Fg looks like nothing more than a rebadged orange can. Besides the same "engineered media" end caps, the bypass and base plate look the same. Hmmm.

 

[ffxfina3]

Originally Posted By: dvldoc
What was up with this cardboard end capped Fleetguard model, I expected it to look like at least a Baldwin not a Fram.


Agreed, that is very interesting and disappointing to see. I've got some FG LF3614 filters. I won't cut open a new one. But I'll cut open the next used one I have.

-Bryan

 

[morris]

i see lots of information about the filter, one brand didnt have enough flow. filter didnt get tightened enough, the weather was bad outside, my dog bit me after breakfast. BUT NO one says any thing about the amount of oil the stock oil pump pumps. i worked in a experimental shop building trenchers, we had pumps that pumped 10 gpm, 20, 30 40 50, 70,80 gallons perminute. ill guess its about 4-5gpm at 2,000 engine rpm.

 

[river_rat]

Originally Posted By: morris
...NO one says any thing about the amount of oil the stock oil pump pumps...

That would be useful info to have sometimes, but is not published anywhere for most models. You'd have to measure your own output per revolution I guess...which would be difficult... unless you had a replacement pump on the bench you could measure.

 

[ZeeOSix]

Originally Posted By: morris
i see lots of information about the filter, one brand didnt have enough flow. filter didnt get tightened enough, the weather was bad outside, my dog bit me after breakfast. BUT NO one says any thing about the amount of oil the stock oil pump pumps. i worked in a experimental shop building trenchers, we had pumps that pumped 10 gpm, 20, 30 40 50, 70,80 gallons perminute. ill guess its about 4-5gpm at 2,000 engine rpm.


Guess I'm missing your concern. The test seemed to only be a measure of comparable filtering performance, not the PSID vs. Flow characteristics of each filter.

 

[morris]

to:ZeeOSix answer: if a filter will only pass 10 gpm, and the engine pump will only pump 5 gpm,at 2,000 rpm. then at 4,000 rpm it "might" pump 15 gpm, over running the filter. in the early 60s chrysler found that the 413,426 race engines, oil pump couldnt keep up at 6,000 rpm. they found the suction passage in the block and the pipe to be to small. so they made a larger passage on some engines, and a duel external suction piping on others. dont forget ANY pump will push better than it sucks

 

[ZeeOSix]

Originally Posted By: morris
to:ZeeOSix answer: if a filter will only pass 10 gpm, and the engine pump will only pump 5 gpm,at 2,000 rpm. then at 4,000 rpm it "might" pump 15 gpm, over running the filter.


The pump might put out 15 gpm through the filter/engine, but only if the pump is putting out 15 gpm before it goes into pressure relief. Not many engines in stock street cars can pump more through a filter than the filter can flow. Most higher tier oil filters can flow 10~12 GPM of hot oil before the filter will go into bypass.

PureOne filter, 5 PSID at 12 GPM (5w-30, 200 deg F oil).

Originally Posted By: morris
in the early 60s chrysler found that the 413,426 race engines, oil pump couldnt keep up at 6,000 rpm. they found the suction passage in the block and the pipe to be to small. so they made a larger passage on some engines, and a duel external suction piping on others. dont forget ANY pump will push better than it sucks


The oil pump can only put out what it can take in.

 

[john1944]

Originally Posted By: morris
to:ZeeOSix answer: if a filter will only pass 10 gpm, and the engine pump will only pump 5 gpm,at 2,000 rpm. then at 4,000 rpm it "might" pump 15 gpm, over running the filter. in the early 60s chrysler found that the 413,426 race engines, oil pump couldnt keep up at 6,000 rpm. they found the suction passage in the block and the pipe to be to small. so they made a larger passage on some engines, and a duel external suction piping on others. dont forget ANY pump will push better than it sucks


That must have been some fight with the external suction piping having a duel with each other. I could understand it if there were two or dual pipes.

 

[river_rat]

I forgot which P1 number this was for?

Originally Posted By: ZeeOSix

 

[xnighter]

So, if people do not like this visual method and some people think that the pore blockage test could skew the results, should optical/laser particle count test be used instead? Should it be used in conjunction with another test as confirmation or optical particle count by itself is sufficient? What test and procedure do you think will be conclusive, short of testing hundreds of filters to get statistically valid results? Thanks.

 

[ZeeOSix]

Originally Posted By: xnighter
So, if people do not like this visual method and some people think that the pore blockage test could skew the results, should optical/laser particle count test be used instead? Should it be used in conjunction with another test as confirmation or optical particle count by itself is sufficient? What test and procedure do you think will be conclusive, short of testing hundreds of filters to get statistically valid results? Thanks.


This is the best method ... but something a "garage lab" can't do.

ISO 4548-12:2000
----------------
Methods of test for full-flow lubricating oil filters for internal combustion engines -- Part 12: Filtration efficiency using particle counting, and contaminant retention capacity.

This part of ISO 4548 specifies a multi-pass filtration test with continuous contaminant injection and using the online particle counting method for evaluating the performance of full-flow lubricating oil filters for internal combustion engines.

The test procedure determines the contaminant capacity of a filter, its particulate removal characteristics and differential pressure.

This test is intended for application to filter elements having a rated flow between 4 l/min and 600 l/min and with an efficiency of less than 99 % at a particle size greater than 10 microns.