Filtering vs. Flow and Testing?

Tempest · Feb 6, 2007

This is in regard to This interesting article on filters and flow that I found while searching. The main thing I am focusing on is:

Quote:

The HP-1 and the PH3600 use the same filtering media, the HP-1 just has considerably more media area.

The PH3600 is rated 15 microns SPFE at 3 GPM, the relief valve opens at 6 to 12 psi. At 9 to 12 psi, the relief valve will flow 0.1 GPM, at 15 to 17 psi, the relief valve will flow 3 GPM maximum, due to its orifice size. Fram has not tested the pressure drop-vs-flow beyond 3 GPM, but the pressure drop is
The HP-1 is rated 20 microns SPFE at 10 GPM, the relief valve opens at 9 to 12 psi. The relief valve characteristics are similar to the PH3600. HP-1 pressure drop-vs-flow data:

3.2 psi @ 3.2 GPM (This seems incongrous with the PH 3600 data. The engineer was unable to explain the apparent discrepancy.)
4.1 psi @ 4.8 GPM

4.8 psi @ 6.4 GPM

5.5 psi @ 8.0 GPM

Does this sound typical? Do larger filters (all else being equal) deliver lower pressure?...but more flow?

Quote:

Frankly, I'm a little nervous stuffing 12 GPM thru a filter that's only been tested to 3 GPM.

Is this typical? Are there GPM standards for what filters have to pass?

Any comments on these, or any other part of the article are appreciated and welcomed. I don't know enough to fully formulate proper questions....

Gary Allan · Feb 6, 2007

The numbers are surely correct ..but misleading. They're only good in a comparative way.

I'm unsure how automotive filters are tested for pressure at flow ..but you can throw most of the numbers out the window when it's in an engine. There are very few situations with a fully functional engine where you should see any flow restrictions. The restrictions have no bearing on flow until your oil pump is in relief. 5gpm is 5gpm. This is not easy for many to grip on.

Those numbers have to be into something like a zero pressure environment. When an engine is downstream of it ..it gets harder to develop pressure. It fractionalizes any resistance to flow.

12gpm would be a tremendous amount of oil flow for a gasoline engine. I think that my HV oil pump is rated @ 9 gpm @ 3500 rpm (I don't recall exactly). It's about double what a normal pump is for my application.

One thing here that I can't quite flush in my head ..

Quote:

4.1 psi @ 4.8 GPM

4.8 psi @ 6.4 GPM

5.5 psi @ 8.0 GPM

( if you plot these points, its very linear )

The only thing that I see is that the two readings/indications have a common median point of reference. There's no proportion to the PSI/flow rate. That is ..they form two individual straight lines ..but I don't see them tracking each other on the same slope. From the median +/- .7 PSI results in +/- 1.6gpm. But reduce it .7 and 1.6 ..extend that downward and you have more pressure bottom then you do flow to divide.

Tempest · Feb 6, 2007

Quote:

5gpm is 5gpm. This is not easy for many to grip on.

So if I'm reading you correctly (and trying to recall my physics class), with a given GPM, the pressure would vary directly by the cross section of the tube the oil is flowing through? And that you WILL get 5GPM (for example) so long as the pump is able to put out enough pressure? I would assume though that working at a pressure would put more strain on the pump and shorten it's life.

Quote:

The restrictions have no bearing on flow until your oil pump is in relief.

I assume there is an "over pressure" valve on the pump that will open and reduce pressure output by the pump?

From the article:

Quote:

As the flow rate increases, the fluid face velocity at the filter media increases, making it more difficult to trap fine particles.

I suppose this makes sense. In this case, would a larger filter (all else equal) for a given application not only potentially last longer, but actually filter better? I'm thinking that a larger can/media surface for a given flow will drop the pressure therefore slowing oil flow through the media, improving filtration. Am I off on this?

Gary Allan · Feb 7, 2007

Quote:

So if I'm reading you correctly (and trying to recall my physics class), with a given GPM, the pressure would vary directly by the cross section of the tube the oil is flowing through?

Yes ..however (I really have tried to shorten this ..but I failed miserably ).....

In a simplified example (since I've never taken a formal physics class ..it's all I can give you

) look at a 1' pipe @ 5gpm ..look at a 1" pipe @ 5gpm. The only difference is velocity and pressure ..more accurately "back pressure" generated by the downstream restictions (pipe friction/cross section/length/etc). Both have 5gpm. As long as the 1" pipe doesn't generate enough "back pressure" to reach the pressure limits of the pump ..the oil/fluid has NO CHOICE but to accelerate when it encounters restrictions, orifices, decreases in conduit diameter ..etc..etc.

The problem is that the rest of our lives revolve around pressure dependent flow ..not flow generating pressure. When you turn on your faucet the flow depends on the pressure applied and the transmission line that it's supplied by. There internal frictions and bends can alter your flow rate. Same with your wall outlet. 120V applied and the current is determined by the resistive load. It doesn't work that way with an oil pump ..at least for the most part ..as long as they're below their relief limit. It's going to generate X gallons of flow at X rpms (minor losses that we'll ignore). A pressure will result. If the pressure is below the limit of the pump .. any and all intermediate restrictions are expressed by pressure elevations ..or "backpressure". This is where the oil had to accelerate at it's mandated flow rate. Just like a rapids in a river ..the flow is the same in gpm ..but the water is traveling at a faster rate over less of a cross section. The engine itself is the biggest restriction that the oil flow will see. As long as the pump isn't in relief ..the filter has to have some proportional subordinate resistance to the engine (any and all downstream restrictions). This, typically, adds up to very little. It doesn't matter how thick the oil is or how dense the media is ..nor the flow rate. As long as that pump isn't in relief ..that filter will barely see any PSID. It can't ..it's more or less an immutable bylaw.

Most look at the filter as a restriction ..which, compared to an open pipe draining on the floor ..it is ..but when you attach that filter to a pipe that has a .1" outlet ..and have a very high pressure limit on your pump ..you can have all kinds of flow and the the filter won't even see it in terms of PSID because it is but one slice in a vastly restrictive downstream environment. Sure the pressure applied may be high ..but the backpressure (just beyond it) is almost as high due to downstream restrictions that are miles higher in magnitude than the filter.

Here's an example that may make sense. You get off the turnpike ..it's bumper to bumper to the toll booths (the filter) 12 lanes wide of toll booths ...20 cars deep each booth. You figure you'll be screaming after you pay the toll ..until you see the traffic jam is all the way to your destination and that is what backed up the traffic at the toll booth. In your engine the roadways are always fully enveloped ..always filled to capacity ..always in a traffic jam and playing kinetic bumper cars. Only the ease of movement (energy expended moving it- expressed in back pressure generated) is varied due to viscosity at a given flow rate.

Quote:

As the flow rate increases, the fluid face velocity at the filter media increases, making it more difficult to trap fine particles.

This is true ..at least to the point that the faster the particle is moving ..the harder it is to trap. There's some industry article(s) that, more or less, state that there's a blow through effect. They also test filters using "pulsing" of flow. This produces a lower capture rate. It's not something that the end user can typically alter outside of using a larger filter of the same known media quality.

Tempest · Feb 7, 2007

Excellent post Gary! Well worth the read

, and thanks for taking the time to type it!

Quote:

It's not something that the end user can typically alter outside of using a larger filter of the same known media quality.

So from your post, this would simply be because a given amount of oil is passing through a larger filter surface. Pressure would have nothing to do with it.

I understand more clearly about what you are saying about the engine being the major back log. I did some research a few years ago into dust collector ducting for both my home shop and my employer. One thing I found to be key was the size (diameter) and length of runs. Smaller ducts are less efficient because there is more drag per unit of air passing through them, but that air moves faster. Larger tubes are more efficient, but move the air slower. (Speed of the air is important so that the dust does not fall out of the stream and clog the pipe.) And the more air one pushes through the pipe, the higher the pressure/speed/friction.

Likewise, the small passages in the engine would have considerably more friction (resistance to flow) due to their small size vs. the rather large diameter of the filter/return port. They are also much longer in length, increasing the friction even more. The filter media is very thin with a large surface area so it = low friction.

Learning a lot here! Love it!

Gary Allan · Feb 8, 2007

Quote:

So from your post, this would simply be because a given amount of oil is passing through a larger filter surface. Pressure would have nothing to do with it.

In the vacuum of my perfect world scenario ..with no qualifications, imo, no it wouldn't. It's surely beneficial in comparing one filter to another in how it will handle the times when you're in relief or are subjected to conditions outside the confines of "normal" (for example: an engine with 200k on it and worn pump gears that has lingering HLA noise that it relieved with changing brands in filters- there are always exceptions). One would imagine that if one filter had an increased flow at a given pressure ..and had a poorer Beta number, that the media had a larger average pore size. OTOH, if it had a better Beta number to another filter and like flow at pressure, you can assume a closer distribution ..more uniform - pore size that is consistent throughout the media (I think I expressed that correctly). There should be a triangle of elements that you can work in there in manipulation. I'm not sure if the article's PSID/FLOW numbers were done on the filter ..or the filter media. The media tests, if Grease's Filter Study (performed by our fellow member, Schultz of PALL filtration) are how those figures were determined, they use swatches of media to get the readings (was that too much strung together?).

Your observations in dust particle suspension (and in stuff like spray drying - shop vacs - etc.) bear out here too ..or so I reason. Aren't all the pores in a filter like various sized ducts ..where varied relative velocity would either promote or inhibit snagging particles?

427Z06 had an observation that I'll share here. It's not assured as to its validity - but it strongly suggests that all three of these filters use the same media and were tested under the same flow rates. Again - this isn't confirmed - ..but this is what one would expect if the same media was used and assuming that the square inches of the media was somewhat proportional to the size of the can (note qualifying disclaimers).

UPC Number: 765809513488
Principal Application: Various Chrysler/Jeep (82-07), Various GM, Saturn (85-07), Lexus (90-07), Saab (67-07), Suzuki (86-02), Toyota (88-07), Yugo (86-89), Harley-Davidson, Various Lawn&Garden, Farm Equip.
All Applications
Style: Spin-On Lube Filter
Service: Lube
Type: Full Flow
Media: Paper
Height: 3.404
Outer Diameter Top: 2.921
Outer Diameter Bottom: Closed
Thread Size: 3/4-16
By-Pass Valve Setting-PSI: 8-11
Anti-Drain Back Valve: Yes
Beta Ratio: 2/20=21/37
Burst Pressure-PSI: 275
Max Flow Rate: 7-9 GPM
Nominal Micro Rating: 19

Gasket Diameters
Number O.D. I.D. Thk.
Attached 2.734 2.430 0.226

Part Number: 51516
UPC Number: 765809515161
Principal Application: Ford/Mercury (81-07), Chrysler/Jeep (02-07), Mazda Trucks (94-07)
All Applications
Style: Spin-On Lube Filter
Service: Lube
Type: Full Flow
Media: Paper
Height: 4.828
Outer Diameter Top: 2.921
Outer Diameter Bottom: Closed
Thread Size: 3/4-16
By-Pass Valve Setting-PSI: 8-11
Anti-Drain Back Valve: Yes
Beta Ratio: 2/20=14/31
Burst Pressure-PSI: 275
Max Flow Rate: 7-9 GPM
Nominal Micro Rating: 20

Gasket Diameters
Number O.D. I.D. Thk.
Attached 2.734 2.430 0.226

Part Number: 51515
UPC Number: 765809515154
Principal Application: Chrysler Family of Cars/Trucks (57-70), Ford Family of Cars/Trucks (57-00), Mazda Trucks (94-00), Toyota Landcruiser (71-96) - (Two Quart version is 51773)
All Applications
Style: Spin-On Lube Filter
Service: Lube
Type: Full Flow
Media: Paper
Height: 5.178
Outer Diameter: 3.660
Thread Size: 3/4-16
By-Pass Valve Setting-PSI: 8-11
Anti-Drain Back Valve: Yes
Beta Ratio: 2/20=13/23
Burst Pressure-PSI: 270
Max Flow Rate: 7-9 GPM
Nominal Micro Rating: 19

Gasket Diameters
Number O.D. I.D. Thk.
Attached 2.834 2.462 0.200

Quote:

The filter media is very thin with a large surface area so it = low friction.

It would appear so in a comparative way with the rest of the engine.

Now you also have to realize that this is, somewhat, thrown out the window when your pump goes into relief. At that point there is no mandated flow. The oil doesn't accelerate to over come restrictions ..and the filter can take a larger chunk of the resistive/restrictive pie. That's what bypass mechanisms are for. A major revelation for me was when a new member, who had his bypass port defeated in his BBC, had a filter collapse and block his oil flow. That, and one comment by Pete C. (of Wix) sorta tied it all together for me ..but not all at once. Pete C. said ..more or less blurted out, actually .."The bypass valve is there to protect the engine!!" ..while up until that catastrophic event of that BBC, I had adhered to the belief that the bypass valve was there to protect the media. Before that guy did me a big favor in some very expensive and unintended R&D it had only occurred to me that the media would be breached ..get ripped and be rendered useless. It never dawned on my that it could result in a total collapse.

Those two events made me alter my view a bit. The bypass valve (or mechanism) is there to protect the engine from oil starvation by limiting the maximum PSID that can be generated across the filter media ...while the bypass valve setting protects the media by limiting the maximum PSID that can be generated across the filter media.

Quote:

Learning a lot here!

Everything you see above came from this site. I had no such grip on it before coming here. It was with a lot of pounding it out and, at times getting spanked, that allowed me to get here. There were a number of resident engineers that provided the proper amount of spanking and input. 427Z06 is one ..and XS650 was another (hasn't been around in a while) ..as well as quite a few others (Ugly3 is surely up for mentioning with his offering of high mileage used filters -if I've left out any who feel that they were instrumental - my apologies). I also had the means to do some of my own testing on stuff. That too was mentored and monitored by those who know how to factor worth out of data.

One final word (at least in this post

). No convention/school of thought lasts forever. That is, sooner or later you'll come across some lubrication situation that can't be resolved with the aforementioned views/principles/etc. So, don't take too much to the bank here. There's always a curve ball in the future that will have you

. It happens to me all the time.

Tempest · Feb 8, 2007

Quote:

Aren't all the pores in a filter like various sized ducts ..where varied relative velocity would either promote or inhibit snagging particles?

This is kind of where I was going with the "filter = less restriction". The pores ("ducts") are very small (increasing pressure/speed), but they are also VERY short, and the media is mostly open, so net PSID on the opposite side is little. I don't however think that the velocity change inside the media would effect filtration as the debris should be trapped on the outside (face)of the media. Maybe I'm wrong?

I can see from the filter specs. you provided that the larger the filter, the more efficient the filtering. (Good catch 427!, and the 51515 is spec. for my truck, so I know it is a good sized filter.) This is what I was trying to say earlier (I don't think I was clear). I would think that a physically larger can would drop the pressure (given a fixed GPM) at/in front of the media, reducing the oil speed, and improving filtration. The larger filter surface would also present less blockage (more holes), AND increase filtration yet again just because there is more of it. This is of course assuming all other variables are equal.

Quote:

had his bypass port defeated in his BBC, had a filter collapse and block his oil flow

Why in the world would someone defeat the bypass?

Once again great posting Gary. What did happen to XS650? He was a voice of reason here. Engineers, of all types, are needed on this board.

P.S. Yet another thing that has me thinking on this. The hydraulic filter from Donaldson that I am looking at (3=Beta200) for my Tranny filter only has as PSID of

Gary Allan · Feb 8, 2007

Quote:

I don't however think that the velocity change inside the media would effect filtration as the debris should be trapped on the outside (face)of the media. Maybe I'm wrong?

You're right except for the "blow through" tendency that filters experience in higher velocity states. I may have misapplied the example (I'm tired-so bear with me

)

Quote:

Why in the world would someone defeat the bypass?

It's a common Chebbie racer thing. I don't know who started it ..but the concept is to keep all the oil filtered and prevent larger wear chunks (that may occur in racing type situations) from doing any damage ..or if a part grenades, you may limit its damage (maybe) ..my reservations were in that if you ruptured the media you could be driving around with NO FILTERING doing more damage without knowing it than would occur without the bypass defeated. This is with high rpm engines that will likely see bypass action as they transition and peak. Now most of the racing filters are like sieves and can withstand more punishment. The guy in question was using an ST Ecore with the nylon cage. It can withstand more pressure without collapse .but fractures instead of just flattening the metal tube. The nylon "cage" allowed larger sheets of media to push through. A very expensive deal for this poor guy.

In any event, I think the practice is unwise under any condition regardless of what filter you have on there. There's no such practice performed on Ford products ..at least none that I'm aware of. This appears to be a Chevy thing.

Filtering vs. Flow and Testing?

Similar threads