Definitive Tests on the Heating and Cooling of Hub Motors

What is this jackshaft doing, you are talking about?

A jackshaft is used to change the RPM and torque of a motor before lead to the rear wheel.
You've seen the Davinci drive that Matt makes for the Astro motors? His astro motor spins at 10.000 RPM. With just a sprocket on the motor axle and a chain to the rear wheel the wheel would have to high gear ratio - meaning max RPM on full voltage would be way too high to be useful, along with too little torque. When a jackshaft is used (think of it as an extra stage of gear reduction) that changes the RPM output and the torque to a more useful combination.

Often with hi RPM motors jackshaft is mandatory for e-bike use, because without a jackshaft the rear sprocket would have to be "larger then the rear wheel" to reduce RPM enough and to bring enough torque to the wheel.

Ah ok I think I have understood. Thanks a lot

Hardergamer said:

John do you mean a side centrifugal cover that exists between the spokes, so it sort of surrounds the hub?

Click to expand...

i think if you covered the motor on both sides with a plate spaced a bit out axially from the motor, and the plates had a hole in the center, and blades going radially from the hole and keep going instead of spokes, I think that would work well .

or with the red painted motor posted a couple posts back maybe adding fins to the inside on the rotor support. there's a lot of space on the rotor plates that could have aluminum fins. maybe a hole at the base of the radial fin and a hole at the end.

Hummina Shadeeba said:

Hardergamer said:

John do you mean a side centrifugal cover that exists between the spokes, so it sort of surrounds the hub?

Click to expand...

i think if you covered the motor on both sides with a plate spaced a bit out axially from the motor, and the plates had a hole in the center, and blades going radially from the hole and keep going instead of spokes, I think that would work well .

or with the red painted motor posted a couple posts back maybe adding fins to the inside on the rotor support. there's a lot of space on the rotor plates that could have aluminum fins. maybe a hole at the base of the radial fin and a hole at the end.

Click to expand...

I already had the same idea. But you will have to take care about the fins that they don´t accidently touch something which isnt supposed to get touched, like the motor cable or the chain.

i was thinking the space where the red motor has fans attached it could have spinning blades attached on the the side of the rotor's support. there's a lot of room in there and they would be away from the windings or the chain

I am not quite sure where the difference is between some fans and spinning blades..

But I have also another idea: Some sort of shovels on the holes of the side cover. This way, the sidecover would function as a big fan as it "shovels" the air inside through itself like a big fan
If this isn´t a good idea then I just dont know what to say anymore.. When I think about it twise I just realize this idea is perfect as the radius is just as big as possible which supports a bigger air flow! Means: The bigger the wholes = shovel of every single hole + the further away every single hole is from the axle, the bigger the air flow! Why did no one else come to this idea before????

GENIOUS!

ps: I think I will have to make a patent to this idea haha :lol:

PSS: Cowardlyduck could you please try this out? I would love to see if this idea works man! This HAS to work for sure!!!

the scooping action, which wants to pull air in, often fights the cetrifugal throwing of air to the outside. i think if the motor part is not moving a scoop is good and if it's moving to go with the centrifugal throwing. as with the race car above.

or maybe a scoop at the entrance hole for air as close as possible to the center of the motor's spinning plate, then the blades internally to throw that air out radially,and then holes furthest from the center of the wheel where the air exits.. a scoop as close as possible to the center pulling in and then it gets spun out.

But why as close as possible to the center? I mean the speed of those "shovels" would be way higher, the further they are away from the axle.. Isn´t it?

the spinning of the wheel will always throw the air out centrifugally even if it were simply a bunch of plates stacked with a small air gap between them, in fact that works well. if it's spun the air comes to the perimeter just as any matter being spun will. If you can feed those blades from the center then they can do their thing, but if you have scoops at the circumference of the motor it will not have a path to flow to and where would it come out. Even the many fans in the red motor above..where is the air coming from and going i wonder? there's programs that can simulate, that i've never used.

i thought same idea as you. scoops as far to the outside as possible and the spinning scoop, when going between 3 and 9 oclock on the wheel, will be going much faster than the wheel and suck a lot in, but none-the-less there was the centrifugal force. I made a version with scoops and it was not as effective as the cetrifugal throwing blades

i cant find a good image in my searching for centrifugal pump. theres some examples of solely tightly spaced disks being used that shows how you dont even need blades and it's very effective. maybe that method would be simpler to make than blades.

Guy's the centrifugal fan has been tried plenty of times here before. John in CR has implemented some of the best I've seen. There's also been a few examples of it being tried where the blades have come loose and caused carnage inside the motor. One ES member even started 3D printing some intake scopes to sit under the disc brake....they were called the under disk impeller (UDI) from memory.
Do some searches and you should find a few examples.

I won't be doing any centrifugal setups for a number of reasons:
- Cooling is only active when moving, and proportional to speed. I need cooling mainly for steep hill climbs which are slow and I often like to stop at the top.
- There is barely any room on the outside with the chain on one side and the disc brake/calliper on the other side.
- There might be enough room inside for a small setup, but the risk of the blades contacting wires, or coming loose is too great IMO...and I have my fans I don't want to remove now.

All that being said, I think if a manufacturer was to build a hub motor with centrifugal cooling built in, it would be possible to make it work very well and not have any of these limitations apart from the lack of cooling when slow/stopped.

Cheers

Cowardlyduck said:

Guy's the centrifugal fan has been tried plenty of times here before. John in CR has implemented some of the best I've seen. There's also been a few examples of it being tried where the blades have come loose and caused carnage inside the motor. One ES member even started 3D printing some intake scopes to sit under the disc brake....they were called the under disk impeller (UDI) from memory.
Do some searches and you should find a few examples.

I won't be doing any centrifugal setups for a number of reasons:
- Cooling is only active when moving, and proportional to speed. I need cooling mainly for steep hill climbs which are slow and I often like to stop at the top.
- There is barely any room on the outside with the chain on one side and the disc brake/calliper on the other side.
- There might be enough room inside for a small setup, but the risk of the blades contacting wires, or coming loose is too great IMO...and I have my fans I don't want to remove now.

All that being said, I think if a manufacturer was to build a hub motor with centrifugal cooling built in, it would be possible to make it work very well and not have any of these limitations apart from the lack of cooling when slow/stopped.

Cheers

Click to expand...

You are correct, Pendragon aka Ken had the UDI. I received on of his units and ended up ruining it when I tried to modify it. As memory serves me, I contacted ken for another unit, and he was in the middle of either buying a house or building a shop....something like that, and it didn't happen. Anyway it was a one piece unit and when attached to my drill, really moved the air around like Ken had designed it to do.

Not being, what is the word, condescending?, but why the talk about venting air into the hub and centrifugal fans?

I thought that the FF + hubsinks beat all this out and eliminated everyone's overheating? If you can simply pour some FF into the motor and then slap on a hubsinks and that eliminates your overheating why bother with these other methods.+

Being a user of a fan setup I would say if FF does the job don't bother with anything else. Unless of course someone still overheats with FF + hubsinks, but nobody is complaining.

I thought that the FF + hubsinks beat all this out and eliminated everyone's overheating? If you can simply pour some FF into the motor and then slap on a hubsinks and that eliminates your overheating why bother with these other methods.

Click to expand...

Lets start with the knowns, Sketch has real world data shows that FF and his hub sinks cool significantly better than say just using FF or just venting a motor. That whole synergy thing. So I guess comparing apples to oranges, what would need to happen to accurately to compare which cooling method is the best, we would need to use the same motors and running gear, and literally do field comparisons to log which one is more effective. Is forced cooling with CD's internal fans and FF better than hub sinks with FF. Hell if I know, but we know they both cool better than no force air, no hub sinks, and no FF. I think the deciding factor is what way the rider wants to go, does the rider want the hubmotor completely enclosed, or doesn't mind venting. One of the pros of the sealed hub motor is minimal dust and dirt exposure to the internal part of the motor. The con is, vented motors tend to exhaust and vent condensation and limit rust formation compared to non vented motors. This I have personally experienced. That said, does the FF act as an anti rusting agent. Sure it does, at least to the areas its in contact with. And dont' get me started about oil bath cooling and hub sinks, 150cc of oil sloshing around the inside of the motor will keep it from rusting, but its messy as hell when splitting the cases.

Rix, I'm going to have to do my own testing on this FF. I actually have two v3 cromotors and even an unused 17" rim, just need to lace one back up again. I guess I'll have to buy some new spokes as I don't want to delace another motor.

I would probably be one of the best persons to see how well FF performs because I have two of the same motors, and one of the motors is already modified for ducted air cooling with a fan.

I could then compare the two and see right away if the FF works as well or better than air cooling. The modified one I have for air cooling works great and has, for the most part, eliminated overheating on all but the steepest hill climbs. If the FF was able to match that or even come close then I would say don't bother with air cooling, just use the FF for its simplicity.

If however I find the FF has delays in the cooling, or I actually have to stop the bike and wait sometimes for it to cool back down, or even worse it overheates too fast when really pushed then I can say ducted forced air cooling is better and this whole FF vs forced air cooling deal will be closed.

The problem is, and I'm not trying to put people down here, is you guys using and testing FF without having used an alternative cooling method are not really good at making any kind of comparison. yes, you can say "FF works great", but what you are testing against is a sealed hub motor with no cooling.

Nothing is worse than a hub motor without any kind of modified cooling on it. It will overheat so quickly and will have to sit for an hour to cool back down once it gets fully saturated with heat. Almost all FF users are comparing it to that and almost anything will be better. My Cromotor was unrideable in the summer without modified cooling after I rode it for about 8 miles and heated it up. After 8 miles it would take only seconds to overheat, and take an hour or more to cool back down. Even after sitting for 30 minutes, I would overheat it back up in less than 2 miles.

Maybe this FF with added heat sinks may work just fine. I would know immediately after just a ride if it offers acceptable cooling performance or not as compared to my forced air cooling which has almost eliminated overheating and I push my motors really hard with constant stop and go acceleration and hill climbs in hot summer heat.

To be totally honest, I would be extremely happy if the FF worked just as well because I don't really want to deal with Fans and the electronics to power the fans. I'd simply rather just use a sealed motor with absolutely no noise.

Just want to mention: I have 8ml ff and Im still able to overheat the motor when riding hard. So ff without fins is still not overheat proof.

Offroader said:

To be totally honest, I would be extremely happy if the FF worked just as well because I don't really want to deal with Fans and the electronics to power the fans. I'd simply rather just use a sealed motor with absolutely no noise.

Click to expand...

You really should try it out now and compare the result by yourself
I believe, as already mentioned, that your edf cooling system will be ahead when climbing slowly around, because it will remove the heat from the windings inside the motor, while FF "only" improves the heat path to the shell.

Rix, I'm going to have to do my own testing on this FF. I actually have two v3 cromotors

Click to expand...

That's why I mentioned it. Having the exact same motors puts you in position to experiment and find out which set up will cool the best, and which one you will like the most. Share your findings.

DasDouble said:

Just want to mention: I have 8ml ff and Im still able to overheat the motor when riding hard. So ff without fins is still not overheat proof.

Click to expand...

Yes of course I would use fins.

Rix said:

Rix, I'm going to have to do my own testing on this FF. I actually have two v3 cromotors

Click to expand...

That's why I mentioned it. Having the exact same motors puts you in position to experiment and find out which set up will cool the best, and which one you will like the most. Share your findings.

Click to expand...

Going to place an order for 10 guage spokes from Buchanan maybe tomorrow. I bought a spare 17x1.6 holmes rim so I will relace my cromotor up to that.

madin88 said:

Offroader said:

To be totally honest, I would be extremely happy if the FF worked just as well because I don't really want to deal with Fans and the electronics to power the fans. I'd simply rather just use a sealed motor with absolutely no noise.

Click to expand...

You really should try it out now and compare the result by yourself
I believe, as already mentioned, that your edf cooling system will be ahead when climbing slowly around, because it will remove the heat from the windings inside the motor, while FF "only" improves the heat path to the shell.

Click to expand...

.

It will be interesting to find out the results for sure. I am going to really stress test the FF + sinks and I will know exactly the limits and what type of riding conditions causes overheating to occur.

The stator acts as a huge heat buffer and if I ride hard and long enough to really heat that up I will notice I will overheat much more quickly when climbing. That is why putting the fan on right at the start of the ride really makes a difference. It is blowing the heat off the windings and not allowing it to soak into the stator.

In the beginning I used to wait until the motor was overheating before putting on the fan. I found that it was better to just put the fan on at a slow speed at the beginning of my riding.

This is sure to cause some controversy ! First my bike : Beach cruiser type frame 24 " x 2.5 " wheels H 3540 with cooling fins , battery 18 x 12 18650 pf cells , controller Lyn 18 FET 100amp . At wot 6000 watt max output . I ride on flat streets in Sacramento . If I take off at wot , overheating after multiple time becomes a problem . With only the fins I can go 12 times in a row 0 to 30 mph before the temperature reaches the C125 degree cut off is reached . Sealed the motor with permatex and added 100ml of soybean oil . Was able to go 48 times 0to 30 at 6000 watts before hitting 125c . So adding the oil increased the runs for 12 to 48 !! Finns and oil were outstanding until the oil started leaking out of the bearings . Flushed out the oil with alcohol and added 5ml statoraid . Redid the same test got 18 runs before c125 . Then it leaked out on the rear brake as well . My test is brutal full 6000 watt runs not cooling between runs . So IF you could keep the oil in the motor , heating problem solved , statoraid , not so much !

I know Justin already did test the optimal amount of FF to use, but I do wonder if people aren't using enough.

When pushing a lot of power (i.e. heat) would it not be better to have more FF to have more contact/surface area for the heat to pass through?

Surely those running that much power would not really care about a few watts last to drag from the extra FF. I wanna see someone add 20ml of FF and see if it makes any difference over the regular 5-10ml used.

Cheers

Some Stealth users use about 100ml of castor oil inside. With this amount you will not only fill the air gap between magnets and lamms but also fill the windings with oil. This amount(oil) is probably even better for heat dissipation. I use 10ml ferro in my 35mm Leaf motor.

No wonder you noticed a better result with oil cooling as it fills the airgap entirely (which isn't the case with FF).
I think togehter with a pressure valve there should be no leakage through the bearings or wire outlet. However this only will work well if the membran doesn't come in tough with the oil. Sealing up the sidecovers is no big deal.

burningwings said:

This is sure to cause some controversy ! First my bike : Beach cruiser type frame 24 " x 2.5 " wheels H 3540 with cooling fins , battery 18 x 12 18650 pf cells , controller Lyn 18 FET 100amp . At wot 6000 watt max output . I ride on flat streets in Sacramento . If I take off at wot , overheating after multiple time becomes a problem . With only the fins I can go 12 times in a row 0 to 30 mph before the temperature reaches the C125 degree cut off is reached . Sealed the motor with permatex and added 100ml of soybean oil . Was able to go 48 times 0to 30 at 6000 watts before hitting 125c . So adding the oil increased the runs for 12 to 48 !! Finns and oil were outstanding until the oil started leaking out of the bearings . Flushed out the oil with alcohol and added 5ml statoraid . Redid the same test got 18 runs before c125 . Then it leaked out on the rear brake as well . My test is brutal full 6000 watt runs not cooling between runs . So IF you could keep the oil in the motor , heating problem solved , statoraid , not so much !

Click to expand...

Oh boy, I know a few (hopefully not many) in here are cringing knowing I'm going to comment about this

I thought the statoraid doesn't really leak if you seal the motor covers? Is this common with other people to have it leak?

Your results kind of are in line with how I predicted FF + Fins to act. While they will cool the motor much better over time, stressing it or heating the motor like you have, will now allow the FF to cool the motor quickly enough. Basically, at peak heating, it is going to have its limits on what it can do.

What may be happening is that once the stator heats up a little, the windings can't dump that heat quickly enough into the stator. Since the windings are not thermally connected all that great to the stator, they are surrounded by plastic and air, the windings will overheat quickly under high stress. But this is really debatable because the windings will not overheat if the stator is cold (basically if you go WOT on a hill on a cold motor the windings won't overheat for a long time), so the thermal transfer from windings to stator may be very good.

100ML of oil allows the windings to sit in the oil, allowing the windings to much more quickly cool because of their direct contact with the oil.

Filling the motor up with oil is a great way to cool it but I personally didn't want to deal with leaks. Everyone who has filled their motor with oil has had leak issues and many decided it wasn't worth the trouble.

The question still remains, under normal riding nobody accelerates at wide throttle from 0-30 constantly like that. Does the FF + Sinks allow you to not overheat under your normal riding conditions?

Great way to test the motor. I have found stopping, accelerating, and doing it again quickly is a great way to heat the motor up. Many times it is way too hard to find hills steep and long enough to heat it up like that. What I like about your test is you are testing the limits of cooling based on really stress testing the motor to its limits. Since I'm always pushing my motor to its limits, this is where I need my cooling method to be the most effective. Most peole are not riding crazy like this and this is less of an issue.

You are also comparing cooling methods, not just comparing it to a sealed hub motor. I believe nothing beats an oil filled motor not even my fan setup. Although my cooling setup may just have the advantage at the point the 100ml of oil gets really hot and can't transfer that heat quickly enough through the motor housing.

I think that first of all we should try using the right winding in the motor based on the riding style.
If you constantly accelerating from 10km/h to 30-50km/h and your riding style is mostly like that while using a 4t mxus then you know that your motor only gives you 20-30% efficiency (roughly speaking)

Based on that, going to say a 6T motor will probably have a lot more efficiency at those speeds while at the same time you don't have to use such high phase amps to get the same amount of torque - in result a lot less heat.
So why use a motor that has 90% efficiensy at 80km/h when you almost don't ride around at that speed and instead getting only 30-40% ?
I do believe that you can have a lot less issues going this path.