Definitive Tests on the Heating and Cooling of Hub Motors

Emmett said:

Don't know what you mean about "lots of water getting in".

Click to expand...

Lets say the worst case scenario is that it's a rainy day and your bike is parked outside. That gap where the oil is getting out very well may let water in, this could be a little or a lot of water depending on many things. Thermal expansion creates a positive pressure inside of the motor which likely has a lot to do with the volume of oil you see coming through the bearing race gap, it works the same way when the motor cools down, so the motor will sort of continually 'breathe'. Even if it's just humidity in the air, or drops of water from a rain storm, oil slowly gets pushed out and water might slowly get sucked in. In my mind, the best thing to do is to have the axle as the point for breathing, much less oil will get to there with the right amount of oil, and the slotted axle probably isn't going to be anywhere near as exposed to things like rain. This might be a good time to mention drip loops.

Plenty of the reason why we would use seals instead of shields is to keep splash and rain water/dirt out and lubrication in, having a sizable gap around the races pretty well defeats the seal. The water may emulsify (mix) with the oil and degrade it, some feel it may evaporate, but it's likely dependent on the addatives in the oil, some oils won't mix with water, I don't know what would happen with castor oil, but in any case, it isn't ideal to have the oil mixed with water. You are far better off then leaving your motor bare, of course.

https://www.youtube.com/watch?v=gwlbAJLzI_w This is the video I mentioned, by the way, if anyone is curious.

Rather than making another post, I'll just tack this on here.

liveforphysics said:

It does matter which ATF gets used, but not for conductivity reasons.

It's because the detergents lead to foaming in high speed operation, which leads it to building pressure and forcing out the bearings or wherever it can, along with the foam not doing as good of job as oil for cooling.

Thinest non-detergent ATF if your winner. I know this one doesn't foam even with 7,000rpm rotors spinning in it:

http://www.redlineoil.com/product.aspx?pid=137&pcid=9

Click to expand...

I did some searching in both search engines and keywords on various pages for things ATF related to detergent and foaming but didn't find anything that really told a story on how to tell which have detergents and which do not, how do you tell?

How were you able to determine that there was no foaming? My guess was either you performed a simple experiment in a bucket with a mixer, or you looked inside a motor after running it, my speculation is worth nothing to your experience.

Thanks

Yes, this is common sense if you have had the motor open while upgrading things (wire size, etc...). A skim coat if high temp silicone sealant has worked well for me on the bearings. Same for the hub covers. Believe it or not, the oil can/will still weep between the phase wire, and the phase wire coating (yes inside the wire) to outside the hub. Since I haven't found a solution to this, I simply allow the drop of oil to exit harmlessly away from the disc brake. It doesn't cause me a problem other than knowing it is there. This has been discussed in the past. I've been waiting 2 years for someone to come up with a solution (no such luck). My best guess is that some sort of automatically opening pressure valve needs to be installed in the hub cover (near as possible toward the axle), so that when:
-hub heats up
-air in hub gets hot and expands
-pressure builds forcing hot air out the valve (and not oil since centrifugal forces put the oil on the perimeter of hub).

I researched valves for a while but couldn't seem to find one small enough to fit my application. So for now I put up with a harmless tiny drop of oil.

The leak between my axle and inner bearing race is very slow. No water is getting in there now. I often ride through puddles or mud, but never ride or leave the bike in the rain. I'll seal that gap soon to keep oil in.

Plus I'll change the oil as often as necessary. I have two plugs on the right side on opposite sides of the axle near the magnets, so super easy to change the hub oil.

I have a 1.5mm breather hole on the sprocket side. Oil weeping doesn't even make it to the spokes. It only needs a 3 sec rag wipe before and after a ride. So no pressure differential between my hub core and the outside. For riders doing commuting, I can see the need to improve the breather solution. A 1mm hole at an angle should keep most people happy. Of course drill it on the sprocket side.

With 100ml of castor oil the cooling gain is amazing. I ride a Stealth Fighter on dirt trails. Previously my rides could be over in 30 mins and I was slowly but surely killing my hub with heat. Now I can ride for a full battery drain and 2 hrs of fun.

Also the motor sounds smoother and pulls better, compared to when my stator is cooking. The motor works so much better for me that now I see my controller temp going up and down. I can ride a lot faster/harder for longer. Previously it was all about built up hub heat.

Maybe 120ml is the best. I don't see the point in needing more. With 100ml, any oil drag at 50 to 65km/h is not noticeable, and drag from tires and wind is surely 10x more. If there is oil foaming going on then so be it. It's probably a good thing.

If oil leaks out via inside the cable sheathing, then try a zip tie on it.

I don't think you understand. The oil is between the copper wire and the wire insulation (next to the wire strands). Cable tying isn't going to do squat.

itchynackers said:

I don't think you understand. The oil is between the copper wire and the wire insulation (next to the wire strands). Cable tying isn't going to do squat.

Click to expand...

this is a known problem. guys doing submersed pc cooling in transformer oil reported that the oil came dripping out at the other end of the wires, travelling through meters of wire insulation w/o any pressure involved.
happened to my bike wiring as well. the oil found it's way into the controller. i couldn't believe it when i saw it. but i guess this could depend on oil type used?!

itchynackers said:

I don't think you understand. The oil is between the copper wire and the wire insulation (next to the wire strands). Cable tying isn't going to do squat.

Click to expand...

You are right. I thought you meant down the tube bundling the insulated wires. With no pressure build up in the hub, that's real interesting. It does make some sense how oil propagates over metal surfaces. However it sure wont bother me with my bike.

Yes, Mercedes got a lot of problems in some models with oil travelling thru the wires destroying the sensitive electronics meters away. Water does the same, but not as extreme as oil. Se http://en.wikipedia.org/wiki/Capillary_action



One solution that helps is to cut the wire, and use the oiltight jack close to the oil source. The other solution to this problem is changing the wires to the anti-capillary wire:


On the positive side, having oil cooled wires do sound even more high-tech than oil cooled motors .

but where do you find some?

Another way to do it is to tin the end of the wire back to the insulation, then clean and apply some epoxy between the insulation and the tinned part. This seals the end of the wire. You could probably use silicone or adhesive lined heat shrink tubing also.
Tinning the end fills in all the spaces between the strands and makes it like a solid wire.

fechter said:

Another way to do it is to tin the end of the wire back to the insulation, then clean and apply some epoxy between the insulation and the tinned part. This seals the end of the wire. You could probably use silicone or adhesive lined heat shrink tubing also.
Tinning the end fills in all the spaces between the strands and makes it like a solid wire.

Click to expand...

That's a great idea! If anyone is doing some testing soon, this option should be thrown in the mix

cal3thousand said:

fechter said:

Another way to do it is to tin the end of the wire back to the insulation, then clean and apply some epoxy between the insulation and the tinned part. This seals the end of the wire. You could probably use silicone or adhesive lined heat shrink tubing also.
Tinning the end fills in all the spaces between the strands and makes it like a solid wire.

Click to expand...

That's a great idea! If anyone is doing some testing soon, this option should be thrown in the mix

Click to expand...

It should be easy enough to test this out, a glass of ATF and some stranded wire, I am just a little curious about the implementation. Might the process of tinning also be capable of melting the housing and aid in the creating that seal? Which epoxy would be ideal for this application? I suppose it can be added to the tests, one wire plain for control, one just soldered, one soldered and epoxied.

The process were discussing is sometimes referred to as wicking. One solution to this, though not perfect, involves the knowledge that wicking ends at a connector much in the same way a well tinned and sealed wire (in theory) would do the same.

Essentially, one of the setups I've been engineering for a few motors I am designing is adding thread to the hallow axle, and screwing in a connector. This will create a tight seal and wicking will stop at this set of connectors. There are down sides to this application, but it's pretty easy to implement.

However, there is a way you can take advantage of this idea without getting so fancy. Simply create a drip loop where your motor connectors are and wrap a little bag around it. The bag will catch the wicked oil, or so I imagine, haha.

One caution:

If the wire already has oil on it, it will make soldering properly very difficult. Either use new wire or carefully clean all the oil off the part getting tinned with some kind of solvent.

What I did on a few setups was to tin the end and make the connection with a piece of heat shrink over the wire but far enough away to not shrink when I soldered the connection. After cooling, I put silicone glue over the joint and slid the heat shrink over that. I heated the shrink tubing and excess silicone squeezes out of the ends. This makes a very robust seal.

I'm probably remembering this from someone else's suggestion, but could a solid wire be brought out of the hub, thereby interrupting the wicking path?

gogo said:

I'm probably remembering this from someone else's suggestion, but could a solid wire be brought out of the hub, thereby interrupting the wicking path?

Click to expand...

Hard to get through the bends, and 3 of anything doesn't fit a circle well. Instead you could just use the solid core for a very short length, and seal the transition point to stranded very well so oil can't get to it and wick.

Making the end solid core using solder sounds like the best way to me, and Fechter's method of sealing by shrinking onto uncured silicone sounds like a real winner.

Why even bother talking about the minutia of using oil without talking about adding exterior surface area? It's absolutely required for a meaningful improvement good for all conditions. It matters little that heat can be transferred more readily from the stator to the shell if the same bottleneck at the outer surface of the shell remains. This limitation is why in the years since oil fill was introduced, that not a single person has used it to extract anything approaching extreme power from a hubmotor.

Don't fall for the anecdotal evidence of climbing Pikes Peak, because while it was a great testament to electric power, running a 9C at about 3kw continuous with pedal assist in temps near freezing is not evidence of a better cooling approach, especially when stator temps got so high that Itchy had to back off the throttle for a while. Justin already proved what a stock 9C can dissipate, and considering low temps on the mountain, the oil filled hubbie didn't really fare much better unless you want to say the 9C was running some embarrassingly low efficiency up the mountain.

The closer you get to continuous operation, the less benefit oil fill has...ie unless you increase the external surface area of the motor. The improvement will be in direct proportion to the increase in surface area, and if you add some kind of blades to the mix to increase the velocity and turbulence of the exterior flow over the shell's surface, you can greatly increase the already increased benefit. These changes are even easier than trying to stop oil from leaking, yet remain totally ignored that I am aware despite being the critical element of real success with the oil fill approach.

I have no interest in getting oil on my tire, brakes, or carport floor, so I'm not leading by example on this one. I set the air cooling mark for hubmotors at 107mph, or 27kw peak input on a daily rider pushing 400lbs, however you want to look at it. Until oil cooling shows that it allows drastic improvements in potential performance it's not the panacea so many think, and that can't happen without the simple process of increasing external surface area.

I'm going outside to drain the coolant out of my radiator. Its the same surface area, so shouldn't make a difference right? Lol.

itchynackers said:

I'm going outside to drain the coolant out of my radiator. Its the same surface area, so shouldn't make a difference right? Lol.

Click to expand...

Go back to scratching your balls in the corner, because when it comes to heat transfer you'll never get it just like you don't get what's invalid about your comparison.

Isn't is easier to add the external oil radiator and a small electric pump (car and motorcycle junkyards are full of it) than adding the alu sinks outside the rotating hub? MXUS3000 (3kW air cooled, 8 kg) is only about the twice cooling area vs small Q100H (good for max 1 kW oil cooled, 2 kg). We need something about 10 times bigger area. Sooner or later we must accept motorcycle cooling soulutions (big radiator, fluid cooling) as the only way of getting into the 10+ kW terrirory.

Pictured: Gearbox oil cooler under the engine antifeeze cooler on a motorcycle.

Or a more efficient motor

Air cooling is theoretically great in terms of KISS, but with the insides of current motors having a predilection for corrosion it's not suitable for all applications.

John in CR said:

Why even bother talking about the minutia of using oil without talking about adding exterior surface area? It's absolutely required for a meaningful improvement good for all conditions. It matters little that heat can be transferred more readily from the stator to the shell if the same bottleneck at the outer surface of the shell remains. This limitation is why in the years since oil fill was introduced, that not a single person has used it to extract anything approaching extreme power from a hubmotor.

Click to expand...

You're making some assumptions about the definition of the motor overheating problem. If say 60% of e-bike owners are like you and care about cooling capacity with continuous operation, sure that's a majority, but it would still leave 40% of us who are on/off the power and might benefit hugely with well sealed oil cooling. That's a very good reason to "bother".

Of course more external surface area and air turbulence could help remove exterior heat. But that aint my limiting factor any more. I have a super slow left bearing oil leak, and finally I can run my battery empty.

The words "drastic increase" are not enough to do justice to the performance gains I got with oil in the hub.

All my riding is on/off power. That is 0 to 3700W through a H4065. Full power is usually for 3 to 10 secs bursts, but up some hills I am at 1500 to 3700W for a few minutes.

Before I added 100ml of castor oil in my hub, I could ride normally for 20 to 35 minutes, and then my stator temp sensor reached a programmed redline, and my motor power was limited to about 20%, which I realise is for good reason. But the ride session is effectively over. Touch the hub for 5 secs and it would burn my finger tips, and my skin is not thin. The motor had to sit for 2+ hours to cool naturally or I had go back to my garage and blast compressed air through the hub breather holes, and even then the problem would be back 15 mins into the re-attempted ride. On the trails I could not explore certain areas because of the heat limitation of my hub motor.

I would try to take it easy and keep the stator temp lower for longer. Giving the hub exterior more time to dissipate heat. The hub felt extremely hot to touch, and coasting often and even long downhills seemed to achieve nothing to reduce the redlined stator temp.

As I mentioned earlier, I'm not interested in venting my hub to let dirt, water and mud in there. Where I ride, it's all sandy clay dirt, dust, rocks, creeks and bog holes. Yes I could add more surface area, but that seems like a lot of effort, plus I still think there is something else going on which you are not considering.

So I put the oil in. The difference is more than just remarkable. With the same riding habits, the stator temp is typically in the 20 to 40% range, and it holds there! The motor feels like it pulls harder and smoother. What's interesting is that with a cooler stator I can now get my controller FETs up to a much higher temp. So I'm feeding the motor more amps, while the motor stator is staying much cooler.

The surprising thing with oil, is that late in a ride, and going as fast as I am capable on the trails, the hub exterior doesn't feel any hotter than before. In my subjective bare fingers touch test, the exterior peak temp feels cooler than the earlier times when my stator was cooked and surrounded by air. I understand the heat must be dissipated through the exterior surfaces, and the external heat dissipation is now greater earlier in my ride. But the net increase in heat transfer seems too good to explain it. With oil and a cooler stator I speculate that my motor is now producing less heat for the same or even higher work load. Yes on/off power riding. It just works.

I wonder if you've increased the average exterior temperature of your motor but not the peak. That would go some way to explaining where the extra thermal energy might be going.

Also, as you indicate, cooler motor windings have lower resistance and so produce less heat. Resistive heat losses are a vicious a circle...

John in CR said:

Why even bother talking about the minutia of using oil without talking about adding exterior surface area?

Click to expand...

I've seen you go on to basically attack the idea of oil cooling several times and then compare it to your opinion of extreme superiority in venting a motor. While at the same time, you often admit that this is merely a discussion about comparing surface area and that your opinion and attacks aren't worth while. You are absolutely right in that expanding surface area is essential, but this is only true if your goal is to reach for massive increases in continuous power output. Thanks for the engineering advice anyway. I am pretty certain most who have the intelligence to grasp ideas like thermal and aerodynamics can also come to the same conclusions you have, most just won't be a total ass about it and consistently spam a forum concerning it at every given opportunity. We can have a discussion about oil cooling a hub without you telling everyone what a dumbass they are for not talking about what you think is important. Sealing a wire is a whole lot easier compared to expanding surface area but sealed wires are an important piece of the puzzle either way and a super worthwhile discussion, immensely useful. It is incredibly disappointing for you to try and derail this thread and discussion the way you have.

Adding oil to a geared hub motor, for example, makes the motor produce less noise, improves drive train efficiency, improves gear and bearing life, and inhibits corrosion all while giving a modest increase of continuous power output capacity of around 50%, all for an extremely small amount of cost. Oil cooling is worth it, don't try and say otherwise because you seem to have some kind of ventism religion(likely fueled by commercial bias). Yours is a pointless argument that derails threads. The reality is, many of us don't want or need multi-kilowatt motor capacity, and for some that do, most aren't likely to be naive enough to think that adding oil to a 500w motor is suddenly going to turn it into a 5kw motor just like that, not even venting a motor gives that kind of gains.

To summarize, I could put oil into a motor, and then pump it to a radiator that would make any vented motor seem like it was limited just through having more surface area by using a larger radiator seeing vast improvements in thermal capacity. But I am not going to do this because I don't want that kind of capacity, if you do, that is great, but the value of adding a dollars worth of oil to a motor to give it tremendous improvements is totally worth while without adding the motor's weight in heatsinks to it.

John_in_CR said:

It matters little that heat can be transferred more readily from the stator to the shell if the same bottleneck at the outer surface of the shell remains. This limitation is why in the years since oil fill was introduced, that not a single person has used it to extract anything approaching extreme power from a hubmotor.

Click to expand...

This statement is so filled with confirmation bias and error it makes snowflakes do a tap dance on the surface of the sun. The very reason you drill holes in the side covers of a hub motor is to eliminate the limitation of using air to transfer heat from the stator to the shell, the bottle neck wasn't the shell, it was the rate of heat transfer from the stator to the shell, air has poor thermal conductivity, oil does not, how you can come here and make the statements you do without a shred of shame will remain a mystery. Oil cooling things isn't an idea that is new or recently introduced. For example, oil cooling drive trains for cars and adding a radiator to that loop to increase surface area is an old idea. What oil cooling a drive train achieve is massive gains in the amount of power that the drive trains could handle, it also wildly improved the longevity of various moving parts, you'd be a fool to run it dry or use grease instead of oil because even if you reduce some of the friction via the grease, the heat will still stick around and do bad things.

Punx0r said:

I wonder if you've increased the average exterior temperature of your motor but not the peak. That would go some way to explaining where the extra thermal energy might be going....

Click to expand...

the temperature that the motor can get rid off is the same as before. what has dramatically changed is the way the heat takes. before there was air between the stator and the magnets/covers, now there is oil which takes the heat from the windings and transfers it very quickly to the case. and the case is in fresh cooler air - contrary to the windings. that way the case heats up earlier and faster. and the hotter it gets the more heat it can shed (delta temp is higher). but the maximum heat it can get rid of is the very same as before.
without oil the windings just overheated before they could transmit all their heat to the case. with oil the windings stay cooler all the time. and peak temp is reduced dramatically.
that's it.

Can someone compare temperature inside the oil cooled Q100H motor (or other smaller hub motor) vs the outer shell temperature? I have no temperature sensor inside, so I can't do this myself.

fellow said:

Can someone compare temperature inside the oil cooled Q100H motor (or other smaller hub motor) vs the outer shell temperature? I have no temperature sensor inside, so I can't do this myself.

Click to expand...

the temperature inside is almost the same as outside. that's the whole point of it. or am i missing something?