Good! Someone finally oil cooled a hubbie!! HS3540

[Burtie]

A breather without leaks ...

Have found an easy way to modify the vent hole, to prevent any loss of cooling oil.

Cut about 1/2 inch length of the plasic tube that is sometimes used as extensions on aerosol can nozzles.
(It has about a 1mm bore and a 2mm OD.)

Use silicone gasket sealent to bond this into the angle vent hole (angle described by andynogo at the top of the previous page), so that it is almost flush with the outside surface of the hub, but protrudes inside the hub by about 5 or 6mm.

In my experience, this is sufficient to prevent any oil which is running down the inside of the cover, from leaking through the vent hole. But still allows the hub to breathe, preventing leakage through the axle seals, and thus avoiding oil contaminating the brake disc.

Burtie

[sn0wchyld]

if your going though all the trouble to liquid nitrogen cool, may as well use a high temperature superconducting coil too

im rooting to discover one that works at room temp. solve all our problems.

mike

hehe... wont that just offload all the issues with heat in the motor to exploding fets in the controller? sorta the same reason why no ones been able to run Hobbykings new giant motor under any sorta load?

but yea... i've been dreaming of the same thing... it'd also make getting nice high amp wires in via teh axle a snap... just use the same guage as your hall wires or less! lol.

edit
all in all great to hear the successes with this... has anyone thought about doing some custom or modified side covers with a enlarged surface area? ie make the covers look more like heatskinks? that should improve things even more...

[PeeHell]

The real world data from Burtie seems great (~70c cooler !). Could some of you who tried this great mod detail this procedure in steps ?

I mean:
What are the specifics area that you sealed ?
What size of bit did you use to drill the vent so it can fit the venting "straw" perfectly ?
Could you use the "venting" hole to fill the hub with oil so you'd only have one hole to drill ?
Is the "Zzzz" sound heard when accelerating attenuated ?

[Burtie]

Seal both covers and the axle slot.
Drill the vent hole opposite side to the disk, close to the axle, just big enough to insert the aerosol tube (perhaps 2 or 2.5mm diameter).

I think it probably best to drill a separate filler hole big enough to sensibly inject the oil with a syringe (maybe M3 or M4), thread it and fit a bolt with an 'o' ring under the head, to seal it.
Position the filler hole to make sure the protruding bolt head will not interfere with the derailleur or the disk brake caliper.

Yes, the resonant audible buzz I experienced, seems to be attenuated significantly by the presence of the oil.

I am still amazed at how well this oil bath mod cools the windings, removes all the 'temperature anxiety' from my hub motored ebike riding

Thanks to bigmoose for bringing this technique to our attention at ES.

Burtie

[dfar]

Just wanted to state that I tried the ATF cooling method and am really impressed with the results.

Before the introduction of the oil my HS3540 would get up to 90 Celsius +++ within 15min of heavy riding. With the ATF I cannot get the temp over 55 Celsius with full throttle riding. Overheating is now not the limiting factor in the performance of the bike.

I did not drill any vent wholes in the hub motor and sealed the sides with silicone. It seems a small amount of ATF leaks out through the bearings and axle and onto the sprocket. This unintended leak effectively "oils" the chain therefore I don't really have a problem with it.

The only thing I am thinking about altering is drilling and tapping a filling valve so one could top up the ATF without taking the side cover off the bike. Maybe make it into a filing/venting valve.

[sn0wchyld]

Just wanted to state that I tried the ATF cooling method and am really impressed with the results.

Before the introduction of the oil my HS3540 would get up to 90 Celsius +++ within 15min of heavy riding. With the ATF I cannot get the temp over 55 Celsius with full throttle riding. Overheating is now not the limiting factor in the performance of the bike.

I did not drill any vent wholes in the hub motor and sealed the sides with silicone. It seems a small amount of ATF leaks out through the bearings and axle and onto the sprocket. This unintended leak effectively "oils" the chain therefore I don't really have a problem with it.

The only thing I am thinking about altering is drilling and tapping a filling valve so one could top up the ATF without taking the side cover off the bike. Maybe make it into a filing/venting valve.

sounds good! what sort of power levels are you running/what size wheel? and when you say heavy riding, do you mean riding WOT lots with no pedalling?

Im getting more and more tempted to do this, since I have a spare set of side covers sitting on my desk. I'm trying to think of a way to increase the surface area of the covers though... probably overkill for sub 5kw levels by the sounds of things but I'd rather go too far than not far enough...

Does anyone have any CAD or 3d drawings for the sidecovers by chance? Speaking to the guys at my uni workshop about this, they mentioned that I might be able to get some free CNC time to make some custom sidecovers with enlarged surface areas... my cad skills are pretty non existant atm though, so a template to work off would be a big legup...

[dfar]

sounds good! what sort of power levels are you running/what size wheel? and when you say heavy riding, do you mean riding WOT lots with no pedalling?

heavy riding = lots of starts and stops with relatively heavy acceleration to keep up with traffic (0-60km in 7-10sec)

82 volts hot off the charger. pulling upwards of 5-6k from dead stop to max speed, also using in the city with regen activated for breaking. Wheel is 26 inch, and as previously stated using a HS3540. With stop and go using 5k (not sure of the avg) to accelerate up to a max of 65-70km/h avg 35-40km and using regen for breaking motor has only been as high as 60 Celsius after 25 km of riding. I like to pedal when riding but even when using only the motor I can't seem to get the temps past 60 Celsius. *note ambient temps are avg 8 Celsius (sweater weather) when riding, The real test will come mid summer but so far I'm really impressed with the results.

Can't comment on long term impacts but I will post back if I have any issues.

[kfong]

This is really good news, nice hearing the actual field reports from Burtie and Dfar. I will seriously consider doing this.

[bobale]

It seems a small amount of ATF leaks out through the bearings and axle and onto the sprocket.

Wait a second. Bearings are usually greased with thick grease. How will introducing light oil affect their life? Will ATF wash off thick grease, and leave bearings dry if you decide not to put any more oil into the motor?

[parabellum]

It seems a small amount of ATF leaks out through the bearings and axle and onto the sprocket.

Wait a second. Bearings are usually greased with thick grease. How will introducing light oil affect their life? Will ATF wash off thick grease, and leave bearings dry if you decide not to put any more oil into the motor?

Mxus, GM, Hx have sealed type bearings, so even if thick grease is replaced with ATF (with created pressure), oil is being retained inside the bearing for quite a while.
I have different problem, my phase wires are drooping oil constantly. It may be because of gravity, my wires are tied on axle height and bike is leaned to wire side if parked or capillary action since I do not think that even parked oil level comes to the axle.
Pressure must be discarded, breathing tube/hole is present.

[amberwolf]

My guess is that the ATF is not going thru the actual bearings (unless they haven't got grease on them anymore or are open-shell bearings with no walls to hold the grease in), but rather thru the inner or outer faces of them, where they meet either the axle or the hub cover. On some of the motors I have, bearings are a loose fit either on the axle or into the covers, and there would easily be a path for something as thin as ATF (or water!) to wick around them. On others, they're a good press fit or even interference fit, and sealed well because of it.


The wire leakage is probably from wicking action, as ATF gets splashed onto the wires at the phase coils and then runs down or wicks up them, then gets sucked thru the sheath openings and out the wire bundle. It can be stopped but you'd have to open the motor and clean the ATF from the end of the wire sheath/bundle, and the axle opening inside, and then seal them up with something ATF-resistant. I'm not sure which sealant is best for this, and wouldn't be damaged by the ATF and also still provide waterproofing.

[hjns]

Seal both covers and the axle slot.
Drill the vent hole opposite side to the disk, close to the axle, just big enough to insert the aerosol tube (perhaps 2 or 2.5mm diameter).
I think it probably best to drill a separate filler hole big enough to sensibly inject the oil with a syringe (maybe M3 or M4), thread it and fit a bolt with an 'o' ring under the head, to seal it.
Position the filler hole to make sure the protruding bolt head will not interfere with the derailleur or the disk brake caliper.
Yes, the resonant audible buzz I experienced, seems to be attenuated significantly by the presence of the oil.
I am still amazed at how well this oil bath mod cools the windings, removes all the 'temperature anxiety' from my hub motored ebike riding
Thanks to bigmoose for bringing this technique to our attention at ES.
Burtie

Hi Burtie and others,

Very nice thread!! Thanks for the comprehensive explanation. Please, forgive me my noobness, but I just have a few questions:

1. Seal covers and axle slot with what sealant? Anything specific that can endure exposure to ATF oil?
2. Where to get something like the aerosol tube?
3. Should you protect the windings / hall sensors from exposure to the oil with something? Is the (low) conductivity of the oil not a danger to the electrical function of the motor / hall sensors?
4. Would this also work with cooling a controller?

Thanks very much!!

[parabellum]

1. Seal covers and axle slot with what sealant? Anything specific that can endure exposure to ATF oil?
2. Where to get something like the aerosol tube?
3. Should you protect the windings / hall sensors from exposure to the oil with something? Is the (low) conductivity of the oil not a danger to the electrical function of the motor / hall sensors?
4. Would this also work with cooling a controller?

Thanks very much!!

1 and 3) use gasket maker silicon (in every auto parts store) for halls and sealing covers, then special winding varnish/polyurethane spray or epoxy for winding (From hardware store)
2) I bought big syringe in the drugstore and took section of the needle it is SS I think.
4) Try and report back . I think you will have mayor problems to seal the controller.

[Burtie]

I didn't bother coating the halls, or windings, with anything.
I dont think exposure to the ATF should harm either.
If I am wrong, and mine breaks, I will let you know.

One point brought up in the other thread, make sure any loose internal wiring is secured with tie wraps/glue so that the hydraulic force of the oil does not continually flex them and cause them to break somewhere.

The pipe I used for the breather, is the thin tube that sometimes comes with aerosol cans of oil, that is pushed into the nozzle and used as an extension.

[hjns]

Hi Burtie and Parabellum,

Thanks for the responses! When I open up the motor, I can add a temp probe and I will make sure every wire is strung up nicely with wrap ties.
This weekend I will try and find this oil resistant silicon gasket. I did some reading on ATF oil and it is really supposed to be non-conducting. Therefore, my question is why one would cover the windings with the special epoxy?

[parabellum]

Therefore, my question is why one would cover the windings with the special epoxy?

Maybe erosion? Remember, "water cuts trough rocks" give it time. It also holds windings and those funny sticks between tooth slots in place, I personally also like that varnish smell.

[dfar]

After about 1 month of riding the hub motor with ATF fluid I have been forced to open up the hub motor to repair the axle (nothing to do with the ATF fluid except that It maybe allowed me the widow to push this motor hard enough to fatigue the axle enough to break).

anyway when I opened it up there was still a lot ATF oil left inside, less than what I put in, but enough to cool the motor effectively but keep it from excessively dripping from the motor after use. I did not measure how much "optimal" oil was left (I still could I guess it's still in the container) but it looks to be a maximum 100ml. also it was noted that after the motor was dissembled ATF fluid was slowly draining (over the course of 2 days) out of the motor. I would assume this would be oil trapped between the stamped stator plates. if filling the motor up for the first time this would also have to be taken into consideration. This definitely helps the heat transfer.

I know you all want to know what the motor looked like so I'll get on with it. It was exactly as I had first purchased it, no rust, no degradation of the wire coatings in the wire or in the silicone phase/halls, the factory epoxy on the halls were unaffected.

so, so far so good. once I get my new axle I'll continue to test.

[hjns]

Cool, dfar! Thanks for the update. Pics look smooth, or should I say oily...

[parabellum]

anyway when I opened it up there was still a lot ATF oil left inside, less than what I put in

Where is most of your oil coming out? I am riding over 1 months (not everyday) and have a feeling I only lost few droops. Am I underestimating oil loses?
P.S. My MXUS has hole trough axle center not groove like yours. Thanks for your review and hope you fix it soon!

[John in CR]

I just ran across a warning about Neo magnets here http://www.mceproducts.com/knowledge-base/article/article-dtl.asp?id=2

"Do NOT use Neodymium Iron Boron magnets under the following conditions:

* In an acidic, alkaline, or organic solvent (unless you are going to hermetically seal the magnet inside a can)
* In water or oil (unless hermetically sealed, or you are prepared to accept a limited life)
* In an electrically conductive liquid, such as electrolyte containing water
* In a hydrogen-containing atmosphere, especially at elevated temperatures. Hydrogenation, a process where the Hydrogen molecule will react with the NdFeB, will occur, causing the magnet to rapidly deteriorate
* Environments containing corrosive gasses, such as Cl, NH3, Nox, etc.
* In the path of radioactive rays "

On all the motors I've opened, the nickle plating on the magnets is always scraped up or worse, so there's no way I'd consider them hermetically sealed.

[bigmoose]

John, have you considered all the magnets that are part of motorcycle alternators and have been bathed in engine oil, in fact hot engine oil,for their entire lifetime?

I sure hope you will stick with air cooling. This will give the rest of us time to surpass the performance of your monster motors!

John, I went and read your link in detail. The issue was with BONDED Iron Neo magnets that are held together with an organic binder I believe. Note how the headings and sub headings are indexed in the article you cited. This is not a problem with sintered Iron Neo magnets.

As we have all seen condensation will rust sintered Iron Neo magnets in a fortnight. Oil makes them look new for years. Join the revolution John...

[dfar]

anyway when I opened it up there was still a lot ATF oil left inside, less than what I put in

Where is most of your oil coming out? I am riding over 1 months (not everyday) and have a feeling I only lost few droops. Am I underestimating oil loses?
P.S. My MXUS has hole trough axle center not groove like yours. Thanks for your review and hope you fix it soon!

Well I lost more oil when I first filled it up (possibly With too much oil added) and by the end of a couple weeks of riding the amount that leaked was significantly less. The oil would leak out between the axle and bearing and then drip onto the rear freewheel which would then drip onto the tire which would then be thrown up towards my seat/back. When enough oil had been "naturally" drained out of the motor only enough oil to keep the chain/freewheel nicely lubricated would be evident, not even enough to cause drips and the max temps the motor could reach were still high 60's degree Celsius.

[John in CR]

I sure hope you will stick with air cooling. This will give the rest of us time to surpass the performance of your monster motors!

Bigmoose, I was just following a link in the technical section. I have no experience with motorcycle alternators, do they use neo mags? Also, no the indexing doesn't indicate the warning is just for bonded magnets, since it's the final part of the article and has a typeface equal to both the scintered and bonded sections. If you tell me it applies only to bonded, then I'll take your word for it, and send a note to MCE to correct their article.

I'm still trying to find a good reason to give oil a go rather than simply follow the crowd, since if I did that I'd have a mountain of burned up motors. I do need rain, mud, and beach bikes, so oil is attractive for that over a sealed motor.

The heat still needs to get transferred to air, and if it's done at the covers you're probably looking at well under 1kw in max heat rejection according to my calculated estimations. If you want to pass what my motors are doing using oil, then you'll need a radiator and a pump. The main motors I've used are barely more than the H35 in terms of stator size. It's the turn count and wheel size that makes them outperform what the vast majority use.

BTW, I only have one monster motor, and I just got it back on the road today with a controller that I hope will feed it, but Hubmonster has never been warm much less hot. That was when it was sealed, and now that it's ventilated it probably never will, at least not with a controller I can afford.

The new high efficiency motors I'm testing can't be oil filled without modification to give up their series/parallel shifting capacity, but their smaller outside shell makes them a poor candidate anyway. Those little motors definitely aren't monsters, and though they outclass typical ebike hubbies by a pretty wide margin in terms of build quality, design, and performance, they're too different to directly compare. Once I get around to ventilating one, I'll call it the MiniMonster in your honor, and it will most assuredly laugh at the oil drippers.

If you want to talk about design changes to include proper oil cooling that will result in a sub 20lb hubmotor that is economical and more powerful than can be accomplished with air cooling then I'm all ears. Until then let's see an oil filled hubbie running continuous at greater than 200wh/mile performance levels, because we're already hitting that and still haven't found the limit with an air cooled motor that could be called an HS40 if Xlyte made it. Don't forget Zombiess bike ran at 88wh/mile to win for the ebikes last week.

John

[Farfle]

I need to go thrash it around and get a screenshot, but the air cooled doublepie with my inexperienced 200 pound self has exceeded 200wh/mile average on the socal racetrack, it did overheat but not to the point of destruction, The temp sensor did pass 190C (thank you 220C rated wire ). BUT about 1KW of the heat its making is iron losses due to a construction error, so when I get in two more pies ill have to do another test.

[John in CR]

I need to go thrash it around and get a screenshot, but the air cooled doublepie with my inexperienced 200 pound self has exceeded 200wh/mile average on the socal racetrack, it did overheat but not to the point of destruction, The temp sensor did pass 190C (thank you 220C rated wire ). BUT about 1KW of the heat its making is iron losses due to a construction error, so when I get in two more pies ill have to do another test.

There you go BigMoose, another air cooled to try and surpass. To be fair I was talking about our 200mm stator 40mm wide motor.

Farfle, don't you have an extra pair of covers? I've got some ideas that should improve the air cooling fairly significantly over the current version. The fun you guys are having up there almost makes me want to move back stateside.

John