Fram Endurance Flashlight Test in canister

[Ronn]

So I just tested two XG2 both manufactured Plant A June 25th 2024 Shift 1.

Using a funnel, I poured M1 0w40 down the center hole.

Within minutes, both drained through and therefore FAILED.

I have another one arriving in a week. Let's see if it is a different date code. If so, I will test it. In the meantime, I will contact Fram.

Seriously though, shouldn't Fram be doing such a test as part of QA?

Shouldn't anyone doing ISO testing be doing such a test both before and after a filter is tested? It is such a simple check on filter integrity.

Thanks
Question still remains
How does this seepage affect overall performance, especially with a superior media?

 

[ZeeOSix]

He’s essentially suggesting the reverse of what OP did pouring oil from the other end.
If there is leakage. there should be drop in oil level from tube side via “gravity” as OP suggested.

Only thing is if the bypass valve also is a leaker (ie, a metal-to-metal style bypass valve), then you won't be able to tell which one is leaking, or if both leak you won't know the leak split between them.

 

[ZeeOSix]

Question still remains
How does this seepage affect overall performance, especially with a superior media?

Controlled ISO efficiency testing that measures the effect of leaks on efficiency. If a filter is leaking 10%+ of the flow in use, I wouldn't say that's just oil "seepage" like seen from a 1/4" deep layer of oil on a surface driven by gravity - it can't be very deep or it will go into the center tube holes and into the media. When there is actual significant dP across a leak path, it will flow more than that test shows.

 

[CharlieBauer]

Only thing is if the bypass valve also is a leaker (ie, a metal-to-metal style bypass valve), then you won't be able to tell which one is leaking, or if both leak you won't know the leak split between them.

The shape of leaf spring / bypass valve puts the bypass valve higher in this test. All the oil drained out so it is for certain that there is a leak at the interface of the leaf spring to the end cap.

In any case, any leak in any filter and perhaps the speed of the leak gives you a data point on whether you're going to use that filter or not.

 

[Ronn]

In any case, any leak in any filter and perhaps the speed of the leak gives you a data point on whether you're going to use that filter or not.

Thanks for doing that leak test.
Mitsuman47 is going to perform a particle count and hopefully a flashlight leak test soon. I would like to see this done on back to back OCs as well. Albeit not an ISO test, it will be revealing nonetheless.
On another note..
Keep in mind, any bypass getting through is previously filtered oil (relatively clean) and will ultimately be get filtered promptly again (15 sec?). It seems many here are making the “worst case”assumption that this small amount of bypass could have critical contaminants that weren’t captured by the filter (1000X greater area than leak?) AND would then settle in the engine avoiding removal in subsequent passes through a very effective filter. Ok… but highly unlikely. It will be interesting to see particle count results from this upcoming OC using the Fram FE.

 

[Glenda W.]

I cut open an Ultra XG2 from Amazon today. Date code A41781, which comes to June 27, 2024 from a calculator and the date code stickie above.
Bob persons may want to know,
1. The can pops open about 1/8 in.
2. The media is not pink but light tan. Thicker and fuzzier than typical paper from just looking at it.
3. Louvers, seem open and not jagged.
4. The bypass valve was not removed, can see minor ruffles on the top flat where it seals.
5. Valve leaks light, on ends the most. Pressing slighty nothing got better on this first look. Not rotated, as found.
6. Through the louvers inside I can see wire backing, like diamond shape, silver color. Not the square black mesh.
7. Poppet black seal didn't show light leak.
8. Strong chemical smell, have to put outside, wash hands.

Have to delve more into it when have more chance. Was worth it to see for $4.50 including tax.

Thank you for the contribution! I find this filter very interesting. Seems to be a hybrid OG/new style. I very much look forward to pics!

 

[Pablo]

So I just tested two XG2 both manufactured Plant A June 25th 2024 Shift 1.

Using a funnel, I poured M1 0w40 down the center hole.

Within minutes, both drained through and therefore FAILED.

I have another one arriving in a week. Let's see if it is a different date code. If so, I will test it. In the meantime, I will contact Fram.

Seriously though, shouldn't Fram be doing such a test as part of QA?

Shouldn't anyone doing ISO testing be doing such a test both before and after a filter is tested? It is such a simple check on filter integrity.

How much oil did you pour in?

 

[CharlieBauer]

How much oil did you pour in?

Filter 1 and Filter 2 respectively. Filter 1 pic was taken a minute or so after filling as I had to fetch my phone. Filter 2 was taken almost immediately after filling.

 

[CharlieBauer]

The math says that with 10% bypass (no way to know if this is a conservative or non-conservative bypass amount without a very involved calculation) in a 99% efficient filter, the particulate count going to the engine is 12.2% of the particulate count that is generated by the engine on each oil pass. That assumes all particulate is >20 microns. With no bypass it is 1.1% of the particulate generated each pass. There are two ways to look at this. One is to say that the filter with the bypass has 12 x as much particulate going to the engine. Another take is that the filter with the bypass has 12% of the particulate that the filter with no defect has leaving the engine (yet still in the engine). A lot depends on where the particulate is generated and where the oil flows from there. A lot of speculation could be made about the impact of this.

I've done some number crunching and extrapolating based on the Ascent and Brand Ranks data. I calculate that the Royal Purple that Brand Ranks said came back with a particle count of 57.8 per million for 21-38 microns is a 91.3% efficient filter for the micron range of 20-40.

I calculate that the Royal Purple that Ascent tested would be 99.3% over the same range whereas the AC Delco that Ascent tested which is rated as a 98% filter for 25 to 30 microns would be 97.7% for the range of 20-40 microns. The Purolator Boss would be (my calculation) 77% efficient for particles 20 to 40 microns and the WIX XP 71.4% efficient over the same micron range.

So I would say a leaky Fram Endurance / Royal Purple / Amsoil can take quite the hit in efficiency. Not the worst but the leak does mean larger micron particles are not being filtered at all whereas a Boss with less overall efficiency would catch all of those larger particles.

Brand Ranks also tested the Amsoil which for them does not appear to be a leaker and pretty close to the Royal Purple tested by Ascent. My calculations show it as 99% for the 20-40 micron range.

The Ultra that BR tested I calculate at 94.9% and the Fram Endurance 97.3%. So possibly leaky but less so than the Royal Purple they tested.

 

[Ronn]

So I would say a leaky Fram Endurance / Royal Purple / Amsoil can take quite the hit in efficiency. Not the worst but the leak does mean larger micron particles are not being filtered at all whereas a Boss with less overall efficiency would catch all of those larger particles.

Correct me if I’m wrong, but those former filters you mentioned have a very high efficiency rating. “Should” a larger particle make it through the bypass gap, it’s almost certain it would be caught in the next pass 15 or so seconds later. Odds of going through tiny gap a second time would appear to be astronomical? This is why, at least in my mind, superior filter efficiency would greatly mitigate against small breaches in the bypass. This isn’t to say those gaps are “acceptable”, but it appears the manufacturers may have thought so

 

[Glenda W.]

Two .75”x.020” are not tiny gaps when we’re filtering micron sized particles. Best way to mitigate is to have a properly constructed filter with no way to bypass the media during non bypass events. Plus it’s the smaller sized particles that cause the most wear, especially over extended intervals. Single pass efficiency is also extremely important to keep as few particles from circulating and accumulating over time as possible. Other high efficiency filters are available without the potential for this defect.

 

[CharlieBauer]

Correct me if I’m wrong, but those former filters you mentioned have a very high efficiency rating. “Should” a larger particle make it through the bypass gap, it’s almost certain it would be caught in the next pass 15 or so seconds later. Odds of going through tiny gap a second time would appear to be astronomical? This is why, at least in my mind, superior filter efficiency would greatly mitigate against small breaches in the bypass. This isn’t to say those gaps are “acceptable”, but it appears the manufacturers may have thought so

The particle test numbers that you yourself have posted up ARE after the particles have gone around and those are the ones that did not get captured by the highly efficient media precisely because there is a gap to get through.

These particles are expressed as ppm. There are lots of them not just 57.8!

The highly efficient filters should show almost zero particles in the 21 to 38 micron range after 4.4 grams of test dust goes through them. There should certainly be zero above 38 and there is not.

 

[Glenda W.]

I have another test I’ll perform this afternoon. It may or may not be a game changer. I’m sure it will cause more debate….

 

[Ronn]

The highly efficient filters should show almost zero particles in the 21 to 38 micron range after 4.4 grams of test dust goes through them. There should certainly be zero above 38 and there is not.

Got it. What filters available show almost zero in 21-38 range then? Not sure what the “57.8” you were referring to? The BR Fram particle test 21-38 showed “17” for FE and “570” and “34” for Tough Guard and Ultra respectively.
Of course there was confirmation on bypass status of tested filters.

 

[KevinMalone]

Got it. What filters available show almost zero in 21-38 range then?

I compiled the particle count test results from BR since they updated their testing rig. The table below is sorted A-Z by 21-38um

Brand Ranks Rig 2.0			Filtration - Particle Count
			4.4g ISO 12103-1 A3 test dust
Brand/Series	Part #	Efficiency	21-38 µm	38-70 µm	>70 µm
Amsoil	EA15K51	99% >20µm	6.9	0.0	0.1
Fram Endurance	FE3614	99%+ >20µm	17.9	2.0	0.3
Purolator Boss	PBL10241	99% >25µm	31.6	0.3	0.0
Fram Ultra	XG3614	99%+ >20µm	34.0	2.7	0.7
Bosch	3330	99% >20µm	56.4	5.3	1.0
Royal Purple	10-2835	99% >25µm	57.8	0.3	0.0
ACDelco	PF53	98% >25µm	134.2	12.0	2.9
Mann	W7015	?	153.5	0.9	0.1
Mobil 1	M1-102A	99% >30µm	252.9	22.5	1.1
K&N	HP-1001	98% >30µm	264.5	0.6	0.1
STP	S3614XL	99% >20µm	401.1	5.0	0.4
Ford Motorcraft	FL-910S	80% >20µm	450.9	29.8	15.2
Fram TG	TG3614	99% >20µm	570.0	1.6	0.4
Wix	51348	95% >25µm?	617.7	2.5	0.1
Wix XP	51348XP	50% >20µm?	807.1	10.0	1.0
Napa Platinum	41348	?	978.6	2.2	0.1
Napa Gold	1348	?	1103.8	88.8	4.9
Fram EG (leaked?)	PH3614	95% >20µm	1563.3	202.0	101.4
Baldwin	BT223	?	2583.3	92.8	27.6
Toyota	90915-YZZD1	?	2774.0	366.7	158.7

 

[Drewby83]

Is this flashlight test a relatively new process? I used to frequent the forum more often and don't recall these being done several years ago. Could this have always been an issue in the production line where metal on metal construction was involved?

 

[Glenda W.]

So yesterday a thought came to me. Would the gaps seal at operating temperature due to expansion?

I heated the filter for 10 minutes at ~200 degrees.

Thermal expansion test = Still fails.

I had hope and was looking for a good debate to use this filter again.

 

[ZeeOSix]

Mitsuman47 is going to perform a particle count and hopefully a flashlight leak test soon. I would like to see this done on back to back OCs as well. Albeit not an ISO test, it will be revealing nonetheless.

The only way a back-to-back test is going to tell you anything is if it can be verified that one filter was a leaker and the other one wasn't. And both OCIs were under the same conditions and lengths. To see what effect a leaking filters has, all other factors need to be held constant.

 

[ZeeOSix]

Keep in mind, any bypass getting through is previously filtered oil (relatively clean) and will ultimately be get filtered promptly again (15 sec?). It seems many here are making the “worst case”assumption that this small amount of bypass could have critical contaminants that weren’t captured by the filter (1000X greater area than leak?) AND would then settle in the engine avoiding removal in subsequent passes through a very effective filter.

New contamination if getting created all the time before it gets sent through the oil filter ... if it wasn't, there wouldn't be a need for an oil filter. All the debris that is captured by an oil filter was created in the system before it ever went through the filter for the first time.

 

[ZeeOSix]

So I would say a leaky Fram Endurance / Royal Purple / Amsoil can take quite the hit in efficiency. Not the worst but the leak does mean larger micron particles are not being filtered at all whereas a Boss with less overall efficiency would catch all of those larger particles.

All sized particles will be able to go through the leak path gaps. If a mL of oil with a debris distribution of the sump goes through the leak path, all sized particles measured in a PC will be going through that leak path.