How much metal is OK in the oil filter?

[stownsen914]

How much metal is acceptable in the oil filter at an oil change? After a "learning experience" with another engine earlier this year, I am now much more rigorous about checking the filter at each oil change.

Picture below. Not a very good one, but it gives an idea of the number of metal flakes I see. It doesn't look like much metal, but there is some visible. This is from a 3.5L race engine with about 10 total hours, 3 hours on this oil and filter. The filter element is from a Canton canister style filter that is placed in the scavenge side of the oil system before the tank and cooler, so theoretically should catch anything coming out of the engine. Is this amount of metal OK? Or should I ideally see NO metal at oil changes after several hours of track use in a race engine?

Thanks.

Scott

[tharbert]

Do you have your oil analyzed?

[Steve@Rennsport]

Canton filters pass more metal than OEM ones do so if you see any metal, its time to open the engine up to find the source.

Its not going to get better.

[stownsen914]

I'm sending out an oil sample today, actually.

I thought Cantons were non-bypass filters, meaning nothing gets through?

So should I see NO metal at all in a filter?

[KTL]

How much gets thru the Canton depends on the media's micron rating. The typical Canton media cartridge is 8 micron. That's pretty good

https://www.cantonracingproducts.com/cgi-bin/commerce.cgi?search=action&category=2500

However it's also important to recognize how many GPM the filter can flow. That's always been my concern with the small Canton that can fit on the filter housing on the engine case (the one that replaces the oil cooler). Is 12 GPM enough? I don't know that answer. https://www.cantonracingproducts.com/product/25-184/25-184---CM-34quot-BILLET-SPIN-ON-FILTER-20MM-THREAD-2-58-O-RING/

The OEM style disposable can filters typically have a rather low bypass valve pressure rating, do they not? For instance the Mann 940-29 filter for a '78 to '94 911 scavenge filter (equivalent to Mahle OC-54 or OEM filter) has a bypass valve cracking pressure of 2.5 bar = 37 psi. I don't know how that translates to actual engine oil pressure. But it would seem that at high rpms the OEM filter enters bypass mode and a good amount of oil is therefore bypassing the filter?

[stownsen914]

For what it's wrth, I'm running the remote Canton filter that is 4.62" tall, and which I understood flows 40 gallons of 180 degree oil per minute.

[chris_seven]

The smallest particle that you would expect to see with the naked eye would be between 40-50 microns if you have relatively good eyesight.

These particles look much larger than this and compared to 8 microns are massive.

There are many ways to rate filters Absolute, Nominal and Mean and they all differ slightly but I would have thought any engine filter should take out particles of this size.

[turbo owner]

Originally Posted by Steve@Rennsport

Canton filters pass more metal than OEM ones do so if you see any metal, its time to open the engine up to find the source.

Its not going to get better.

This + a dozen,

I can show you what your bearings probably look like, I have specks just like yours imbedded in my bearings and an engine that was about to fail.

[stownsen914]

Steve's comment above about the Canton filters has me wondering now if he means that it's not the 8 micron filtering threshold as much as possibly poor design or quality potentially allowing particles larger than 8 micons through ...

[KTL]

They are rather basic in their design. Anti-drainback valve is simply a rubber ring sandwiched between the filter cartridge and the machined inlet/outlet body. Not a whole lot different than a disposable can filter. The cartridge seals to the end cap with an o-ring for the ID of the cartridge and the OD of the end cap.

In my limited experience with these, the ends seal well and the oil has nowhere else to go but thru the media. Mine sealed so well that the cartridge bulged a bit on the ID.

Engine Bearing Failure- Disassemble Heads Too?

The above incident is what prompted me to bring up the question about filter flow capacity. I wasn't very inspired by Canton's belief that engine assembly lube products created additional restriction. The oil I used was straight 30W for initial break in.

[stownsen914]

Kevin - I read the thread you linked above. Interesting that the cage actually deformed! I wonder if flow is somehow reduced by running the Canton filter "backwards" with flow going from outside to inside ... especially if debris starts to build up in the flow path, given the small space between the filter cartridge and the filter body.

[KTL]

I don't know if the flow capacity is different if you run the filter backwards on a canister filter like yours. My filter is a traditional style similar to common disposable filters. The flow goes outside-in like a common filter and so that would be the normal direction of flow.

I found the deformed filter when changing the oil and decided i'd better not risk starving the engine if the filter was indeed a restriction. So I put the oil cooler back on the engine since I already had the engine out of the car on account of needing to fix a leaky cam housing thrust plate paper seal. Next time out I wrecked the engine for certain because I under-filled the oil tank. Usually I put 4 or 6 quarts of oil in the tank and crank it over to draw the tank down before filling it up more. I forgot to fill it up to desired level and I believe I was running on only approx. 1/2 capacity. So it appears my engine fail was a series of problems. It started with a possible oil filter choke and finished with a substantially reduced supply of oil. Expensive lesson learned.

[stownsen914]

Kevin - gotcha. I haven't seen one of their stock replacement style filters, so didn't realize how they worked. Either way, pretty interesting that the cage deformed. Something was definitely going on there ...

[67dubcab]

If the filter is the Canton I have pictured Then I would replace it immediately!
This filter is in constant bypass mode as the end-caps are not sealed to the element housing. I went with this originally as I liked the no-bypass feature and the micron filtering rating, the filters lack of filtering killed my engine in less than 10 miles.

The signs of fail started upon opening the box as when I unbolted the cap the o ring was crushed when assembled- I contacted them and they sent me some extras to replace it but they too would not fit in the machined groove, fixed with swapping out to the smallest one and some careful assembly.
The next sign was the supplied -16 fitting I ordered from Canton almost touches the filter when screwed into the filter body.

[67dubcab]

Continued..
I solved the adapter problem by welding a-16 directly to the housing and porting the -12 stock threaded port.

At this point I think that if this can flow and filter at their claimed levels with that design constriction I must be golden! More like golden flakes of metal after I went to change the oil after my first start-up.
The filter was located just before the dry sump tank and with its no by-pass and particle rating there should be No metal in the dry sump oil, instead it was full of particles large and small. Fail complete.

I took off the top of the oil filter housing and the end cap came off with it, after pulling out the rest of the filter i could see instantly what the issue was, the end caps were crudely stuck to the rest of the filter with clear sealant that was allowing big gaps for oil (and metal) to slide on by.

[stownsen914]

67dubcab - that sounds like a really terrible experience, sorry to hear that. You lost the motor because of this?

In my case I have a second filter (a factory racing filter) located where the engine oil cooler went on the older 911 motors. The Canton is more of inspection filter in my case. But the lack of quality of the filter element that you experienced is rather disturbing. I haven't had that happen to me, but maybe that just means that stuff is getting by and I am not aware?

I have another car that ate a rod bearing after 65 hours. I have seen what's in the pre-tank/cooler filter, also a Canton (lots of copper), and will be curious now if the filter did it's job and kept it out of the tank. I'll know when I drain the oil ...

[67dubcab]

Hey stownsen914,
Yep, engine is dead, sitting in my shop waiting to be opened to assess damage.
I had contacted Canton to discuss my experience with their product and they of course denied any product malfunction.
I explained that the end caps were not sealing and even proposed a simple fix of bedding the ends in a pool of sealant instead of the hit and miss of silicone from a caulking gun.
I did this myself with a spare filter by filling the end caps with liquid metal and then flipping it over and filling the other end after it set.
After a short run of maybe 100 miles (on a new used 3.2 engine) I checked the filter.
End cap seals intact but filter buckled and collapsed! so apparently the filter needs the leaking end caps to maintain flow!

[stownsen914]

Wow, that sounds like a pretty terrible experience ...

Question - were you running the oil inlet through the bottom of the filter housing?

[KTL]

Originally Posted by 67dubcab

After a short run of maybe 100 miles (on a new used 3.2 engine) I checked the filter.
End cap seals intact but filter buckled and collapsed!

That sounds/looks familiar. Here's the filter element from the Canton 25-184 mounted on my 993 filter housing that was installed on my 3.2 short stroke



It's this filter https://www.cantonracingproducts.com/product/25-184/25-184---CM-34quot-BILLET-SPIN-ON-FILTER-20MM-THREAD-2-58-O-RING/

The above picture was prior to having the massive bearing failure I linked above in post #10. Also note that the oil used in this filter was 30W oil. Point of mentioning that is a thinner oil like 30W would place less of a burden on the filter than would a typical 20W50

Seems to me a filter like this that filters down to 8 microns is restricting oil flow too much. Anything less than 30 microns, I believe, is too small anyway to create significant damage? That micron threshold is just something I recently read and isn't based on any sort of technical documentation/report.

[stownsen914]

As a point of comparison regarding filter size threshold, I just checked the System1 web site, and their finest oil filter is 30 microns. Maybe 8 is too fine/restrictive?

Kevin - Interesting pattern on the collapsing cage of your filter. The three collapsed spots look almost timed at 120 degrees apart from one another.