Manual Transaxle oil cooler: FILTERING

[Sodo]

I have started several threads related to different aspects of transaxle oil cooling systems. This thread is focused on FILTERING. (I hope)

The idea being, that an over-engined transaxle absolutely MUST generate more steel trash in its oil. And swimming in these metal particles, wear out your 'parts' quicker, generating MORE steel trash (metal fines) in your oil. And it "compounds" like interest in the bank. More trash creates more trash.

Common wisdom is that you need the BEST oil for an over-engined transaxle. This makes sense, but if you think that since you have the best oil that it's OK to run it "dirty",,,,, that CANNOT work. ANY metal fines in the oil are hundreds, if not thousands of times bigger than the oil film thickness, therefore, if you want an over-engined trans to last, you MUST keep the oil clean. IMHO it's more important to have clean oil than to have specialty oil.

Why care about specialty EP additives (Extreme Pressure additives) when you can SEE sparkles in the oil which is visible metal-to-metal contact? If specialty oil made more than a FEW % difference in lubricity, this would reduce running temperature, and could be PROVEN by oil sellers who would trumpet this at the TOP OF THEIR LUNGS,,,,, right? Not hearing it. But more importantly and irrefutably, a little better lubricity will never overshadow the effects of oil contamination by metal fines.

So anyway, specialty oil probably makes a few percent difference but CLEAN oil is a no brainer, its not probable, its guaranteed. A tiny 5 micron steel chunk might be like an apartment building tumbling down the river, in terms of 'oil film" thicknesses. Roller bearings are not like the journal bearings in an engine. Journal bearings have larger clearances and can tolerate fine trash. Ball bearings have almost a point-contact and require clean oil. There is a reason that your engine does not have roller bearings, and if it does, they are low-load applications.

I spoke with an engineer from SWEPCO (whose focus is mainly machinery) and asked him if 12 microns was sufficient for a transmission and he said "can you filter to 5 microns?,,,the cleaner the better." So I took that as a "NO," that 12 microns is not sufficient. But it's certainly better than 140 microns (the Weddle filter). Reading machinery journals I see filtering to 3 microns. But in any case we KNOW a ball bearing is a point-contact, and that point is much less than 3 microns. I don't know how many microns are the normal surface undulations of a ball bearing and its race.

There is a VERY important ball bearing on the Vanagon transaxle, the mainshaft ball bearing. It's a KNOWN problem with big engines, it can't take the power, and overheats. I think its this bearing that has generated the vanagon transaxle wisdom of "70mph max (sustained) speed". Certainly squirting cooled oil at it will help that number though.

So anyway, for those who have oil cooling pumps on their over-engined transaxles.....ONWARD to discussion of filtering.

On my trans I used a Parker Hydraulic filter that I got on eBay for $30. It has a 12 micron element and here's a closeup of the matting where you can SEE small metallic particles that were stopped by the filter. It has a flow rating of 20gpm at 3,000 psi. I am pumping 2 gpm of 145*F 90wt thru it and it goes thru at about 16 psi. There are others available. Another pretty good one (I think) is this 7GPM unit. I have not heard of a test on this unit yet. It has a refief pressure of about 18 psi which would allow it to bypass until the oil warmed up.

I am also fooling around with a magnetic filter where you replace the 140 micron screen (Weddle filter) with a slug of neobdymium balls. Here's a pic. Pictured is $33 of spherical magnets (648 count). It looks like the Weddle filter can hold about $100 worth. There are lots of smaller aluminum FUEL filter cannisters available, I'd prefer a smaller one. This slug pictured has 648 balls, with surface area of 50 square inches. That oughtta hold a decent amount of steel particles. I'm not sure how you clean the element. Not even sure if it works. Would be a real bummer if I dropped it on the floor of my shop!

Heres a $33 magnetic filter by magnefine. I've seen them cut open (google images: "magnefine") they are not very impressive, just a paper filter and a big donut magnet (prob 2 sq. inches of surface) but a decent price. This one, Magnom Mini ($149) looks better, I think PCForno is running that unit.

Well that's enough typing, for a hot sweaty afternoon. Hope other members find this project interesting.

======== REFERENCE =======
Some of my other transaxle / cooling / oil threads. Others are working on this project too, but I can barely keep up with my own. Feel free to link other threads that members will find useful.

- Transaxle Failures every 10-20k miles ...what the heck started all this?
- the initial query into "oil cooler ports"
- Drilling for 4th gear and mainshaft ball brg oiling with some good electrical controls discussion on p5
- Drilling for the Pinion oiling
- Homebrew fittings for transaxle cooling add-on (no drilling)
- Electrical controls for Cooling system
- Periodically purifying expensive specialty gear oil (by gravity)
- Drilling transaxle port with tranny in-car (squirt 4th gear even if you have oiling plates!)

[dobryan]

Lead on McDuff. I do love your willingness to try a lot of out of the box stuff....
_________________
Dave O
'87 Westy w/ 2002 Subaru EJ25 and Peloquin TBD

"To travel hopefully is a better thing than to arrive." Robert Louis Stevenson

MD>Canada>AK>WA>OR>CA>AZ>UT>WY>SD
https://www.thesamba.com/vw/forum/viewtopic.php?t=620646

Building a bus for travel in Europe (euroBus)
https://www.thesamba.com/vw/forum/viewtopic.php?t=695371

The Western Syncro build
https://www.thesamba.com/vw/forum/viewtopic.php?t=746794

[Sodo]

Thanks Dave, glad you're interested. Heres some metal-contaminated gear oil that I (miraculously!) have on hand, imagine that? It's been filtered with a 12 micron filter, but this metal dust is smaller than 12 microns, and has passed thru the filter.

First pic is where I flattened the slug of 648 magnets and dropped it into the oil, let it sit (pic2) and picked it up (pic3). It sure LOOKS like is holding onto some metal fines. I have no way to know if the magnet can hold these tiny particles against oil viscosity and movement.

Here's a pic of a paper towel, where I wiped some black oil off the ball magnets. The (other) magnet is able to lift the paper towel, indicating steel in the black oil.

Well there ya have it folks. It kinda seems like this 648 balls (= 50 sq inches magnet surface area) is a good plan. It's certainly better than the little donut magnet in the magnefine filter (prob 2 sq. inches area). But I DON'T know if it can hold onto metal fines. I do KNOW that the donut magnet in the Magnefine filter can't hold much fines. Although the donut has probably 20x the area of a VW magnetic drainplug magnet.

The plan now is to clean the 648 balls thoroughly (somehow?), and plumb my ACTUAL trans oil thru it, then LOOK & see if the balls hold onto black stuff. Then it's doing SOMETHING good, but this could take awhile....
_________________


'90 Westy EJ25, 2Peloquins, 3knobs, pressure-oiled GT mainshaft, filtered, cooled gearbox
'87 Tintop w 47k 53k, '12 SmallCar EJ25, cooled filtered gearbox
....KTMs, GasGas, SPOT mtb

[pablum]

Best post since i joined. Though you had some fearsome opponents.

[ALIKA T3]

Hi Sodo!

I like the magnetic approach, that was on the back of my head to combine that with a filter, but having somehow a plastic can with baffles or something with magnet outside for ease of cleaning.

I'm way to busy to keep working to finish up on this transmission project for now for various reasons but I love reading about your tenacity!

You didn't link the oil cooler ports thread it seems like.

Aloha!
_________________
Silicone Steering Boots and 930 Cv boots for sale in the classifieds.
Syncro transmission upgrade parts in the Classifieds.
Subaru EJ22+UN1 5 speed transmission
http://www.thesamba.com/vw/forum/viewtopic.php?t=416343
Syncro http://www.thesamba.com/vw/forum/viewtopic.php?t=4...num+gadget

[Syncro Jael]

Sodo,

I too have been toying with magnetic filtering. I like the ball magnets. My approach was to get a magnet inside that Weddle housing also. Currently I have the Neobdymiun rare earth magnets on the outside of the housing. When I have opened up the housing to clean the filter there is a layer of pasty looking goo where the magnets are holding very fine particles of metal. So I know this way of thinking will work.

I was going to put the magnets inside but was concerned about how they would hold up inside the oil flow. Most have a maximum working temperature of around 176°f. Neodymium Magnets are very susceptible to losing their effective magnetic field at elevated temperatures. The Neo magnet grade should be carefully selected to account for the operational temperature of the application and the magnet’s geometry. All magnet alloy will gain or lose effective magnetic field as the temperature fluctuates. Neo magnets will lose effective field as the temperature increases and are said to have a Negative Temperature Coefficient. When compared to other magnet alloy options, neo magnets lose magnetic field at a faster rate. Neo magnets effectively have a higher Temperature Coefficient (TC) when compared to other commercial alloys.

This high rate of change relative to temperature exposure results in Neo magnets being very susceptible to demagnetizing from elevated operating environments. The magnet may be exposed to a temperature which does not fully demagnetize, but reduces its’ magnetic performance to a state that it no longer is able to provide sufficient magnet field to support the application.

The loss that has been discussed thus far is considered reversible. This means that the magnet will recover the loss as it cools back down. What is typically published by suppliers of magnets is the Reversible Temperature Coefficient (RTC) for each particular grade of Neo for both the Residual Induction (Br) and the Intrinsic coercive Force (Hci). The RTC as well as the recommended Maximum Operating Temperature must be considered when selecting a Neo magnet grade for an application above ambient room temperature.

Demagnetization from Elevated Temperatures
Magnets exposed to heat can lose strength which cannot be recovered. Magnets will lose strength as they heat up, but so long as they are not taken above a certain operating point (maximum operating temperature) the strength will be recovered when they cool back down. If the magnet is taken above the operating point, a percentage of the magnet’s volume is demagnetized and the magnetic loss cannot be recovered without re-magnetizing the magnet.

The operating point is geometry specific. Most magnet providers specify a maximum operating temperature for various grades of magnet alloy, but this must be qualified. The noted upper operating temperature assumes the magnet has an appropriate geometry to tolerate the heat level of the selected alloy grade. The advertised maximum operating temperature for a particular magnet alloy grade is not always sufficient to ensure elevated temperature performance. For optimum operational heat resistance the Magnetic Length of the magnet must be sufficiently “long” relative to the area of the pole.

For instance a 0.250” OD x 0.250” Long Neo magnet, oriented and magnetized through the length, has a magnetic length to pole diameter of L/D = 1. This is a very good L/D ratio and the advertised upper operating temperature may be able to be used as the application’s maximum operating temperature. However, a magnet that is 0.250” OD x 0.125” long has an L/D = 0.5 and this would indicate that the advertised maximum temperature of the magnet must be adjusted downward or a higher heat grade selected. Typically L/D ratios over 0.7 are appropriate for the recommended operating temperatures advertised for Neo, but it is best to consult the vendor and execute thermal testing.


. They do make some high temp magnets that are rated to 300°f.

This looks like a great approach.

EDIT: Looking further into your pic, could you place the magnets in rows where the pleats are on the outside of the filter? Then slide the whole thing into the canister? If any of the magnets came loose the filter would still protect the system. Just a thought.

By the way, where are you finding those magnets so cheap? Everywhere I have looked they are about $0.30-$0.75 each, for N42 magnets.

Keep on keeping on...

_________________
1987 Syncro Westfalia Hightop - NAHT
Subaru EJ25 Forged Frankenmotor, Triple Knob.
Jael = (Mountain Goat)

[Sodo]

[Syncro Jael]

[Sodo]

[Sodo]

I took the slug of 648 magnets out of the oil and hung it overnight. Here's pics.
Some of you may appreciate the cross-pollination between Vanagon Science and other forms.

Looks like the magnets only hold onto the metal as expected.

_________________


'90 Westy EJ25, 2Peloquins, 3knobs, pressure-oiled GT mainshaft, filtered, cooled gearbox
'87 Tintop w 47k 53k, '12 SmallCar EJ25, cooled filtered gearbox
....KTMs, GasGas, SPOT mtb

[Syncro Jael]

[Sodo]

[Sodo]

[Pcforno]

Here's another datapoint. I have a recently rebuilt Syncro tranny now with 3000 miles on it, nearly all at 75 mph on the highway. I have a weddle pump, radiator, 6 micron filter followed by a "magnom" magnetic filter. The pump spray is aimed directly below 4th gear at least the MS bearing on the meshing side of the 4th gear pairing. Took a look today and found nearly zero swarf, contaminants, or otherwise. Tranny temps have averaged 150-160 over the rebuilds first 3000 miles.

[/img]

[thasty07]

http://www.billetconnection.com/cvf_page.html

Now that would be awesome to use. If money was no object

[Sodo]

Awesome. Great to see some other eyeballs & synapses focused on this subject. Lifetime fluid is the way to go for MFRs but for the antique vehicles crowd eternal life is more attractive.

On another note the spherical filter media project is losing its luster. Held some talks with the think tank at Mogfest and intend to do a little more testing to clarify soon. Just wanted to get this posted before anyone orders $33 of magnets. Pretty sure theres a better way.

Basically the magnets in contact with each other diminishes their ability to hold onto metal fines. Which is kinda good for "cleaning" but not so great for "filtering".
_________________


'90 Westy EJ25, 2Peloquins, 3knobs, pressure-oiled GT mainshaft, filtered, cooled gearbox
'87 Tintop w 47k 53k, '12 SmallCar EJ25, cooled filtered gearbox
....KTMs, GasGas, SPOT mtb

[thasty07]

Here is another thing to consider or look more into, I called Derale who make alot of transmission coolers and filters and asked what their lowest micron level filter was, hoping to get into the 10-12 range. Their lowest was 21 microns, and the tech said they wont go any lower because after that you run the risk of actually stripping the oil of its properties if it gets over filtered.

Im gunna call a couple other places to see what they say but it sounds like there could be a bottom end to filtering.

[ejimmi]

[Sodo]

[bluebus86]