Thermal paste on hub motor windings. Dumb idea or dumb idea?

[CUDAcores89]

I had this idea today about how to cool my hub motor, but it sounds pretty stupid and I see several drawbacks, but I wanted to know what endless sphere thought.

I currently run a 9x6 generic ebay hub motor at 1500w continuously and i'm trying to see if I can run this thing at 3000w. Nearly all my riding is done on very flat road or roads with very low grades, so I've never had a problem with my hub motor overheating at even 1500w.

Stuff that people have tried to dump more power into hub motors include:

Ferrofluid (statorade)
External heatsinks mounted to the hub motor (hubsink)
Air holes drilled into the side of the motor

Something just occurred to me when I was thinking about the purpose of thermal paste on a CPU (I am also into building computers). The goal of thermal paste is to bridge the gap between two metallic surfaces. After looking at how a stator is wound, you can see small spaces in-between the motors windings and the stator core as shown:



The copper windings are the only thing that's actually creating more heat, so our entire goal should be to move the heat away from the copper windings as best we can.

Here was my idea. What if we covered the outside of the copper windings and fill in the gaps between the windings with a thin layer of thermal paste? What this would do is serve as a nearly direct connection to the stator core and close any air gaps, allowing the heat to easily pass through the thermal paste directly to the stator core which could act as it's own heatsink.

We wouldn't have to worry about the thermal paste just flying off the winding's because the copper doesn't move, so that wouldn't be a problem but I can think of other problems (more on that later).

On top of this, you might be able to benefit significantly from adding small 40mm fans to the inside of the hub motor to cool the motor even more. I know this has been tried on the forums before, but I don't know what the success rate was.

The biggest drawback I can think of to doing this is dirt and mud will get stuck to the thermal paste. To solve this problem, you could use some high temperature aluminum tape that sits over the copper windings to effectively "hold" the thermal paste in place and protect it from dirt and grime getting in.

The second problem is doing this might not actually help at all. As seen by the picture above, there are what looks like plastic spacers that sit inbetween the windings and the stator core, and plastic is a poor heat conductor nullifying even trying this.

The third problem with all of this is even after coating the windings in paste, it might not do much. Since the heat has nowhere to radiate but back to the inside of an already hot hub motors internals, there might be little to no effect.

The only way to test this would be for me to take apart my hub motor two times (my hub motor is currently apart, waiting for ferrfluid to come in) and test it each time. I'm not willing to do that because I don't want to risk stripping the threads on the aluminum plates and I am not interested in buying a new hub motor just for this test, so that won't happen.

But does anyone else speculate that doing this would have an effect at all? Or would this be a waste of time. I am a big proponent of the scientific method but in this case I really can't do a good test unless I was willing to open up my hub motor twice. Once for a before test and once for an after test.

Any thoughts?

 

[eCue]

I like your thinking but the paste might be a barrier to air flow , it has me thinking could add internal "fan"" blades to the Motor windings themselves so it spins like a fan inside the wheel blowing heat off the winding directly.
Not sure if it would create airflow over the winding , I think it would

 

[amberwolf]

Theres a thread about definitive testing of heating/cooling hubmotors which contains many ways of doing such, including testing of potting the windings in thermally conductive epoxy. I dont recall any of the specific results, but they should be easy enough to find if you use the thread search tool (at the top of each page of thread).

Testing of fins on the covers to stir the air near the windings has also been done, some in that thread, some in others (such as some old stuff by John in CR)

 

[dogman dan]

Well, the dumb idea is running that type motor on 3000w in the summer.

Bottom line is you need a bigger motor for 3000w, or cooler weather. I used to run 3000w all winter, then when the hub overheated in the spring, I'd swap out the motor with another, and run 1500w all summer. The motor would generally just toast the halls, and I'd fix it by the next spring. 3000w was nice, so I'd push it into spring till it cooked. :lol:

When I got a bigger motor, all that stupid shit ended, although I did have to get used to riding a bike with a heavy wheel.

For now, assuming you are in the north hemisphere, time to stick to 1500w. Those motors can take 3000w, meaning about 2000w continuous when cruising full speed, for 30 min or so. They are pushed way into the inefficient range of use at 3000w, as is clearly showed by the rapid overheating. I don't understand the motor theory, but its something to do with the lamination thickness of these cheap motors, and eddy currents. Even a dumbass like me eventually does understand this, if the motor is heating up that fast, you are pushing it too far beyond its design limits. For daily use, wasteful of your battery, for a race, you'd get beat by the guy not wasting his watts into heat.

So why keep running a heater that squanders your watt hours? When you could run so much farther at that same speed with a bigger, efficient at 3000w, motor? You just need more magnet to get er done efficiently, simple as that.

Money of course, but soon as you get some, get a big motor.

 

[Ernie]

Thermal paste is meant for filling in micro surface imperfections between two flat surfaces, it's meant to be used very thinly.
Just coating copper windings with it will probably do more harm than good. There are fero fluids to improve hub motor cooling, just use one of those.

 

[rumme]

You say you are riding on flat land and never overheat at 1500 watts continuous ? Then I think you will not have much problem running 3000 watt bursts on flat roads. You will have decent speed on flat roads to help keep air flowing past the hub and aid in cooling. Now, if you were trying to pump 3kw in it, going up steep hills, I think you will see more over heating issues because you lose so much speed going uphill.

I think if you add some heat sinks or a air scoop to divert all the air directly to the sidecovers of the hub motor, you will be fine.

 

[CUDAcores89]

In total, here is what i'm gonna do to this hub motor to try to eke 3kw out of it for my riding style:

1. Ferrofluid. Should be a given.

2. I cut up some 20x20x40mm (LxWxH) heatsinks and I plan to thermally glue these to the hub motor.

3 (this is the big one). I'm going to strap a 12v server fan pointing at the hub motor that will be enabled when I run the bike at full power at 100% throttle. since I am a computer builder, I have learned that even a small amount of active external airflow over something that's getting really hot can help with temperatures significantly. A 12v server fan that spins stupid fast and stupid loud can help cool the hub motor when it's running and at stop signs.

The whole reason i'm doing this is to smoke a dude in my area with a moped in acceleration. For daily riding, I will be running at 72v 45A but be using the 50% speed limit on my new greentime controller coming in the mail.

This should hopefully limit me to roughly 1500w instead for daily riding. The high power is mostly meant as a show-off sort of thing right now so I can beat him in a race that is a few miles long.

I think the reason I was able to run this motor at 1500w is because not only do I spend my time on mostly flat roads, but I rarely stop. I ride is about 16 miles long and I stop a total of 5 times throughout the whole ride.

Anyway am currently looking for a new motor. I made another thread about that somewhere.

The big thing i'm trying to go after though is if I have to spend more than $300 on a laced hub motor, i'll probably just call it quits and run my current configuration. These hub motors are very cheap to replace and I am a 19 year old broke ass college student. If it costs too much money to go 40mph, then i'll give up and live with 30mph (which has proven to be easily doable) and live with that.

 

[Ernie]

Sounds like a classic waste of time.
Move to another neighborhood where there are less hoons, the promotion of eBikes doesn't need more young men driving recklessly, it just results in more restrictions and policing of all eBike users.

 

[eCue]

I 2nd the vote to leave the race alone its too much effort for such a small ego gain
Then again im a efficiency nut repelled by wasteful riding.

 

[Aerialpixels]

I know the thread is old but for the sake of anyone else, I give you kudos for thinking outside the box. Only mistake was asking the opinion of people who don't know the answer but are more than happy to knock you down. I did exactly what you're talking about. I put mx4 thermal paste between the stator and the case. Then I added heatsink to the outside. It was a BBSHD mid drive rated for 1000 watts.

Result? The motor runs cooler on 90v than it did on 48v and easily hits 60mph. Night and day difference.
In fact, the only motor I've cooked was the original on 48v. The nodded one has lasted years.

 

[MadRhino]

I know the thread is old but for the sake of anyone else, I give you kudos for thinking outside the box. Only mistake was asking the opinion of people who don't know the answer but are more than happy to knock you down. I did exactly what you're talking about. I put mx4 thermal paste between the stator and the case. Then I added heatsink to the outside. It was a BBSHD mid drive rated for 1000 watts.

Result? The motor runs cooler on 90v than it did on 48v and easily hits 60mph. Night and day difference.
In fact, the only motor I've cooked was the original on 48v. The nodded one has lasted years.

I had not answered to this thread then, but I feel that I have to clarify something for you: The OP was talking about a hub motor. There is no way to put thermal paste between the stator and the casing, because the casing is the rotor, thus spinning. The only gap that could be filled in a hub motor is between the winding and the magnets, and that is not a place to put paste. Ferrofluid was the proper option for him.

Your solution can only apply to inrunners that have an outside stator inside a casing, and it’s been done successfully before. Some such inrunners are even built with thermal paste out of the factory.