Oil cooling your hub- NOT snake oil!

[John in CR]

I did a little research on ATF oil ...
it seems ATF breaks down above 200 degrees F or 100 C . thermal stability is important.
...
3. high thermal stability (so it doesnt break down into carbon or sludge)

Please provide a link to the source of the researched information indicating ATF turns to sludge at 100degC.

Uh oh, I found this little blurb about Dexron III: here http://www.speedway.com/MSDSList/0130SPE012.pdf

"Carbon monoxide, carbon dioxide, hydrogen sulfide, oxides of sulfur, oxides of nitrogen, oxides of zinc, aldehydes and hydrocarbons., Thermal decomposition may produce hydrogen sulfide and other sulfur-containing gases at temperatures greater than 150 F"

Another interesting tidbit from http://www.mattscaddyshack.com/1/automatic-transmission-fluid.html

Automatic transmission fluid will provide 100,000 miles of service before oxidation occurs under normal operating temperatures or about 175 degree F. Above normal operating temperatures, the oxidation rate doubles with each 20 degree increase in temperature. Car engine overheating will decrease the life of transmission fluid. Example: At 275 degrees F, the fluid has a life of only 3,000 miles!

275°F is 135°C, and there's little doubt windings can quickly be pushed that high in tens of seconds in high load low rpm conditions.

While those aren't definitive, it makes sense that ATF isn't designed for cooling off copper heating elements, which is what many transform their motor windings into by using controller settings too high. The 4000W mentioned a couple of posts up is a good example. Some rigs are fine in stock form at that kind of peak power limit, while others can be melted in minutes at even lower. Each must be tuned for its specific combination of components, loads, and use.

John

 

[salty]

Hi

I did a little research on ATF oil and other oils that would be suitable for hub motors.. based on the data provided on this blog.. the stator temp can get over 200 degrees F, and it seems to be around 130 C with 4kw input power..

it seems ATF breaks down above 200 degrees F or 100 C . thermal stability is important. a good oil in hub motors would have these qualities:

1. high dielectric
2. non corrosive
3. high thermal stability (so it doesnt break down into carbon or sludge)
4 good thermal conductivity

I would think a good candidate would refrigerant oil either mineral based or alkybenzene oil those oil have all those qualities .. grater than 300 degree before break down. does not attack the enamel on the motor windings.
and some oil go as high as 400 degrees..

great work on this site..

newbie (retired aerospace engineer)
salty

salty
100 µW
100 µW

P

 

[teklektik]

Uh oh, I found this little blurb about Dexron III: here http://www.speedway.com/MSDSList/0130SPE012.pdf

"Carbon monoxide, carbon dioxide, hydrogen sulfide, oxides of sulfur, oxides of nitrogen, oxides of zinc, aldehydes and hydrocarbons., Thermal decomposition may produce hydrogen sulfide and other sulfur-containing gases at temperatures greater than 150 F"

Although undoubtedly true, this is a Health/Safety notice. The standard operating temp for ATF is 175degF and different varieties have rated viscosity specifications to 100degC. This health notice is simply saying don't snort it - even below normal operating temperature. Although it's interesting that that ATF outgases toxic chemicals, it probably isn't too noteworthy in this discussion.

Another interesting tidbit from http://www.mattscaddyshack.com/1/automatic-transmission-fluid.html

Automatic transmission fluid will provide 100,000 miles of service before oxidation occurs under normal operating temperatures or about 175 degree F. Above normal operating temperatures, the oxidation rate doubles with each 20 degree increase in temperature. Car engine overheating will decrease the life of transmission fluid. Example: At 275 degrees F, the fluid has a life of only 3,000 miles!

275°F is 135°C, and there's little doubt windings can quickly be pushed that high in tens of seconds in high load low rpm conditions.

While those aren't definitive, it makes sense that ATF isn't designed for cooling off copper heating elements, which is what many transform their motor windings into by using controller settings too high. The 4000W mentioned a couple of posts up is a good example. Some rigs are fine in stock form at that kind of peak power limit, while others can be melted in minutes at even lower. Each must be tuned for its specific combination of components, loads, and use.

Although I'm not swept off my feet by ATF statistics presented by a Cadillac sales department trying to sell ATF changes, the numbers presented there are not unbelievable for non-synthetic ATF. Choosing a less terror-inducing number from the table suggests that at 255degF (123degC), ATF will last 6200miles before oxidation occurs. 'Oxidation occurs' is a key phase here, and it's relevance to cooling seems open to conjecture. In any case, this is a workable distance before swapping out the ATF (for me) and a synthetic would certainly fair better (no claim in the table as to the type of ATF but being Cadillac it's probably Dexron IV).

As you point out, this is a matter of selecting the proper solution for any individual build. For many builds that do not continuously run at elevated temperatures, ATF still appears a viable choice. Thanks for digging up the chart - it reaffirms ATF utility at 100degC and puts out some numbers to gauge effectiveness at higher temps.

 

[John in CR]

I just grabbed the first 2 things I found. If it's off gassing something, then it's chemically changing and isn't the same afterward.

We should get MWKeefer's input, because he starting testing oil in hubbies several years ago, so if anyone has long-term results it would be Mike.

 

[Spicerack]

Well all I can say is that I ran ATF in my 2kw Mac motor for well over a year at 80-100 C at the stator. Last time I pulled it apart it was spotless inside and the only sign off any reaction was possibly some daggy looking heat shrink.

The whole point of using at is that it is super easy to get hold of and does the job (particularly in geared motors which can't be vented). Remember the KISS principle people! :wink:

 

[teklektik]

I just grabbed the first 2 things I found. If it's off gassing something, then it's chemically changing and isn't the same afterward.

Or simply evaporating - a non-chemical change...
IMHO the last place I'd look to evaluate the suitablity for this purpose is a hazmat document that is about health effects not the fluid characteristics proper and that has not a single quantified statement - even about health risk contamination levels never mind issues of lubrication, viscosity, thermal effects, etc. I can see nothing in this document that contributes to evaluating ATF one way or the other for suitability in hub motors.

Well all I can say is that I ran ATF in my 2kw Mac motor for well over a year at 80-100 C at the stator. Last time I pulled it apart it was spotless inside...

The whole point of using at is that it is super easy to get hold of and does the job (particularly in geared motors which can't be vented).

As with Spicerack (and I believe MWKeefer), I'm interested in gear motor use. I find the opportunity quite attractive to capitalize on an engineered fluid that is meant to be a coolant and lubricant, is safe for seals, has foam inhibitors, and is readily available. Personally, for my application I'm much more interested in Justin's results about cooling effectiveness than I am about concerns of breathing the stuff or thermal breakdown...

 

[The Stig]

Hi all, I'm interested in possibly liquid cooling my own MAC in the future at 2-2.5kw. Will look through the thread more tomorrow but could y'all start to put together a review or guide that could be pasted into the first post for liquid cooling noobs like me. Stuff like how this has worked out, any forseable long term problems? How it was done? like what is a "basic closed oil-splash system just like most lawnmower engines"?

 

[neptronix]

A liquid cooled MAC sounds fantastic.
Does it quiet the motor down, by chance?

 

[Sancho's Horse]

Check the wikipedia on ATF for a quick ref. Sulfer has a special relationship with oxygen on the periodic table, and thus is used by our bodies as well as engineers to accept oxidating electrical conditions. Sulfer products are likely additive breakdown. I think that a bit of breakkdown of the various additives as well as the special friction conditions ATF must maintain is largely irrelevant to its ability to conduct heat. It may lose some of its protective qualities for the seals/plastics, etc., but not a huge deal. ATF is a winner in my opinion.

 

[johnrobholmes]

Wires and jackets can be sealed with vacuum and epoxy or silicone. Make a drip loop into the sealing material within the vac chamber, with the jacket opening under the sealant. Pull the vac and get the first bubbles out of the jacket, then clean it all off and let it cure.

 

[The Stig]

Wires and jackets can be sealed with vacuum and epoxy or silicone. Make a drip loop into the sealing material within the vac chamber, with the jacket opening under the sealant. Pull the vac and get the first bubbles out of the jacket, then clean it all off and let it cure.

Wut?

Link? Pics? Something noob friendly!

 

[spinningmagnets]

The benefit of synthetic ATF compared to conventional organic ATF is that the synthetic will remain closer to its designed viscosity whether its hot or cold, there is very little thickening when cold, or over-thinning when hot.

Also, it can run to much higher temperatures before there is any lubricity-breakdown or oxidation. Even when using organic ATF, I am still watching this thread with great interest.

I seem to remember seeing some type of fling-disc near a shaft (on some machine) where it exits the case so no oil would be hanging around the shaft/bearing joint. If the case is vented to prevent heat-derived pressure of the air inside, it seems like even thin ATF would not have an easy exit...

 

[speedmd]

Been following the thread. Came across this paper I thought would be of interest on the topic. Talks about adding tiny amounts of hexagonal boron nitride (h-BN) particles to oil which will greatly increase thermal conductivity. Has the word "nano" (super small) associated with it, so I expect that is somewhat inversely proportional to the cost.
http://www.greencarcongress.com/2012/02/rice-20120202.html

 

[The Stig]

Oil Bath Cooling a MAC Motor
feel free to help me design/write this!


Noob's Guide to liquid cooling a Hub Motor

Parts needed:
-High temp RTV silicone OR High temp Epoxy.
-Mass airflow sensor cleaner. (Auto parts)
-drill (Optional depending on steps taken below.)
-Mobile 1 Synthetic ATF(Automatic Transmission Fluid),
-Big syringe(from pharmacy/medical supply store but it will be junk a month after it being used. I would recommend a syringe from a Motorcycle or Auto parts store like this one: http://www.sickbikeparts.com/catalog/product_info.php?products_id=77 )
-Recommended: Oil absorbent pads. like pig mats. Also rubber gloves and a ventilated area for working with sensor cleaner.


Gettin 'er done:

1. Open motor cover.

2. Vent 1: Drill (and maybe tap) hole for vent on the non-disc-brake side... Could be just a small hole, could be a tapped hole with a bolt screwed in loose, this could let air in and out but limit the amount of oil getting out. A small bolt with a hole drilled through the middle could work.
OR
Vent 2: a really thin breather tube that goes through the axle and then connects to an empty syringe(or small bag/bladder thing) on the outside. That way the syringe(or bag) would open and close as air expanded in the motor from heat, but the system would still be closed and no oil could escape.
An alternative, if you have a lathe could be to drill a long thin hole in the end of the axle that doesn't have a wire coming out of it, and then a hole in the motor somewhere to link it to the inside space of the hub, then follow above instructions for vent 2.
OR
No Vent, I've gone with no vent and I know others have too. the motor will either breath through the sealant in the Axle's Wire Hole, or through the bearings(possibly lubricating them).

3. Seal the wire hole in the axle.
-First use Sensor Cleaner to clean everything the sealant will touch, let it dry for a while.
-Next Try to push sealant into ase wires with high temp silicone or epoxy. (I used liquid electrical tape, will report back on how it does. not recommended for high power as its only recommended for up to 200 degrees F)

4. Seal cover with small bead of silicone (1/4 - 1/8 bead?). Put thread locker on screws. I just tightened the screws on my Ezee a lot and it held fine.

5. Fill motor 1/4 to 1/3 full with Mobile 1 synthetic ATF transmission fluid. I filled my mine through the disc-brake bolt hole, you may need to open a second bolt hole, higher up than the first, to allow air to escape as you add oil.

Done.


Your suggestions are necessary! and welcome

Also does anyone know a source of really really thin high temp tubing? Maybe some kind of wire with the copper pulled out?
How about the part we would use for the syringe/bag/bladder needed for vent 2 - ideas?

 

[h0tr0d]

Been following the thread. Came across this paper I thought would be of interest on the topic. Talks about adding tiny amounts of hexagonal boron nitride (h-BN) particles to oil which will greatly increase thermal conductivity.

I knew that boron nitride was superb for increased lubricity, which is great for the internal gears of a GD hub. The thermal advantage just makes it perfect!

Great find!

 

[teklektik]

I knew that boron nitride was superb for increased lubricity, which is great for the internal gears of a GD hub. The thermal advantage just makes it perfect!

Actually, this is a bad feature for gear motors.

Manufacturers of sprag clutches recommend against using extreme pressure (moly) lubricants or lubricants with low friction modifiers. Quite simply, the freewheel (sprag clutch) will not engage.

 

[speedmd]

I wonder if we can get a bit sent to Justin to add to test oils on the oil side of the definitive cooling tests. Any one out there with connections to the paper's author or know of possible sample sources

 

[The Stig]

... Quite simply, the freewheel (sprag clutch) will not engage.

This is one thing I've been afraid of... I'm guessing synthetic ATF tranny fluid doesn't fall into this category of extreme moly friction lubes...

I was actually surprised the original poster(Spicerack) didn't have trouble with his clutch. unless i missed something. Is synthetic ATF not ideal for avoiding clutch slypage? what would be?

 

[teklektik]

... Quite simply, the freewheel (sprag clutch) will not engage.

This is one thing I've been afraid of... I'm guessing synthetic ATF tranny fluid doesn't fall into this category of extreme moly friction lubes...

No, it doesn't. ATF is specifically intended to support operation of the torque converter, allow friction-based mechanisms like sprag clutches (as in gear motors), bands, wet plates, etc to operate, and is watery-thin so it can be pumped to operate internal hydraulic servos. It is good at shuttling heat up to the tranny oil cooler. It seems appealing for potential use in hub motors because of these characteristics.

Synthetic oils simply have a base stock that is synthesized from chemicals instead of being directly refined from crude. The primary advantages offered by synthetic base stocks are chemical and mechanical characteristics that increase service life - not markedly improved lubricity per se.

Molybdenum disulfide is a solid material in the graphite family that is used in high load greases and in other specialty lubricating products. This stuff is in a whole different amazingly slippery world of lubrication from ATF.

 

[The Stig]

No, it doesn't. ATF is specifically intended to support operation of the torque converter, allow friction-based mechanisms like sprag clutches (as in gear motors), bands, wet plates, etc to operate, and is watery-thin so it can be pumped to operate internal hydraulic servos. It is good at shuttling heat up to the tranny oil cooler. It seems appealing for potential use in hub motors because of these characteristics. ...

Oh thats a relief!

So would ATF provide the gears with any cushioning from shock loading?
I imagine it would work pretty well as lube for the gears so maybe that would transfer more of the energy in shock loads to moving the rotor.

 

[Marc S.]

From my limited experiance with ATF cooled Bafang geared hub motors I think better cooling isn't the only benefit you can archive. The better lubrication of the freewheel clutch prooved to be equally beneficial.

I use two Bafangs in my ICE Sprint trike. A little '250W" Bafang SWXH that has about 7000km run time on its clock and a '500W' Bafang BPM II thad did about 4000km.
With both I experianced a stuck freewheel clutch after some use in its stock configuration. While the SWXH developed infrequent freewheel problems after about 1000km at its rated power level (10s LiMn battery and 17A controller) the BPM II cluch got stuck after 3000km but was most the time used with a 15s RC LiPo pack and 35A controller.

Both motors got an ATF filling that 'cured' the stuck freewheel clutch. While the cluch of the BPM was more damaged and started slipping while accelerating a couple of hundred kilometers later the SWXH runs fine since the oil filling. When I opened the BPM I found quite some metal shavings from the clutch stuck to the magnets and some corroded ball bearings (ran the poor thing trough the whole winter and had to park it outside for two month) but the Nylon planetary gears look like new.

The little SWXH actually holds up very well in its current 'torture test' where it is hooked up to a 35A controller and 14s LiPo and does my daily 35-45km commute for more than 900km now. Relacing it from a 20" to a 26" rim didn't make things easier on it as well. While it gets fairly warm to the touch when flogged for 15min uphill at 40km/h it didn't burn up yet. My guess is the motor temperature stays in the 60-70°C ballpark. On level ground it drives my trike up to 48km/h and stays noticably cooler (maybe 45-50°C).

So far I'm quite impressed how that little '250W' oil cooled hub motor runs at about 650W continous and up to 1300W peak and despite my best efforts outright refuses to break down. I will never run a geared hub motor without ATF cooling again.

 

[mwkeefer]

Oil cooling hub motors isn't new, my first oil filled and solid steel geared hub motor was custom built in china back in 2009... worked fantastic and improved the thermal path of my geared hubs but also allowed them to mechanically survive quite well under =>6000w burst and 4000w continuous power levels.

It's nice to see everyone trying with so many other types of lubricant and variations of hubs!

-Mike

 

[The Stig]

Both motors got an ATF filling ...

So what else did you do to the motors when you filled them with how much oil? Did you ever have leaks?

solid steel geared hub motor was custom built in china back in 2009.

Where can I get one of these! Did you use ATF as well?

 

[mwkeefer]

Both motors got an ATF filling ...

So what else did you do to the motors when you filled them with how much oil? Did you ever have leaks?

solid steel geared hub motor was custom built in china back in 2009.

Where can I get one of these! Did you use ATF as well?

I had a few of these custom made, I was going to do a bulk order and bring them to market but the "engineers" at the Chinese company I was using continuously changed my designs and sent me proofs which were wrong... if memory serves these were proof #3 (maybe 4) and even these weren't right as the "engineers" decided to add a "Fill Window", totally unneeded and simply adds point of leakage potential. Needless to say I never had them do a bulk order for me since they couldn't follow simple instructions and designs in CAD.

With that said, once I modded the covers to remove the View Window... these hubs worked like champs (a bit wider than usual at 138/139 mm old).

I used synthetic motor oil, diesel if memory serves.. can't remember the weight, sorry.

-Mike

PS: I'd have to check but I believe I have one of these hubs in stock at the office, it would be available. pm

 

[Marc S.]

Both motors got an ATF filling ...

So what else did you do to the motors when you filled them with how much oil? Did you ever have leaks?

The Bafang SWXH is easy to fill through the holes of the disk brake screws with a big syringe. In case of the Bafang BPM II I needed to drill two of the disk brake screws holes through with a 3.2mm drill.

At first I attached a silicon 'breather tube' in one of the screw holes but now I just let them 'breath' through the motor cable. Works fine with minimal spill.
Leakage is minimal IF you don't fill in too much oil.

50-60ml is a good starting point with the SWXH, 80mm will produce quite a mess for a week till the motor got rid of the exessive oil through the ball bearings and motor cable...
70-80ml is a good starting point with the BPM II.

I wrote a little post about it in my blog btw: http://etrike.wordpress.com/2013/04/21/oil-cooling-a-bafang-swxh-geared-hub-motor/