Definitive Tests on the Heating and Cooling of Hub Motors

[macribs]

Have you opened up the motor yet? Or maybe looked of exploded view drawings of dd hub motors?

Here you can see the magnets attached to the outer shell, the shell get hot when motor is put to work. The harder the motor work the hotter it gets.
The heat does not have many ways out of the hub, so heat is radiating from motor to the outer shell of the motor, including side covers, maybe some thru the axle and wires also. But there is bottle neck. Air. Air is a poor coductor of heat.

Inside the outer shell is the rotor, windings and so on. Between is a thin layer of air. That thin layer of air makes if hard for heat to radiate to the outer shell. If the air gap could be bridged somehow that would allow for more heat to travel faster to the outside of the hub. Hence FF.

You got a motor with large outer shell. And your motor can easily shed a lot of heat. But the air gap still is restricting the heat flow out of the motor. When you fill FF in the motor the magnetic FF will get drawn to the magnets of the motor and make a bridge where there used to be air gap. Now heat can travel faster to the outside of the motor where the speed wind will help cool motor down.

 

[Offroader]

Just for thought, but FF is only about 5 times better conductor than air. However, 10 ML of ferro fluid will not fill the entire gap, so it will still mostly be air between the stator and magnets.

While FF is 5 times better than air, aluminum is 2000 times better conductor than FF, or 8000 times better than air. So FF is still a very poor conductor to bridge that gap. This is why some people lap their computer heatsinks because reducing that gap even by .001 mm or whatever makes a large difference in temperature because less thermal paste is needed.

What I am thinking is maybe work should be done to reduce the gap between magnets and stator, and fill the gap with more fluid. Not sure if this is possible, but more work needs to be done to increase the heat transfer to the case then just FF.

 

[liveforphysics]

Just for thought, but FF is only about 5 times better conductor than air. However, 10 ML of ferro fluid will not fill the entire gap, so it will still mostly be air between the stator and magnets.

While FF is 5 times better than air, aluminum is 2000 times better conductor than FF, or 8000 times better than air. So FF is still a very poor conductor to bridge that gap. This is why some people lap their computer heatsinks because reducing that gap even by .001 mm or whatever makes a large difference in temperature because less thermal paste is needed.

What I am thinking is maybe work should be done to reduce the gap between magnets and stator, and fill the gap with more fluid. Not sure if this is possible, but more work needs to be done to increase the heat transfer to the case then just FF.

Something to keep in mind with all the thermal conduction comparisons with aluminum or other non-liquids, is that they don't transfer heat through mechanically flowing from a hot place to a cold place.

Fast moving air can easily beat even solid silver metal at conducting heat and/or cooling. Flowing liquids can beat most any solid, even diamond (best thermal conductor of non-nano materials).

A working fluid is not about how well heat moves from one side of the static working fluid to the other side of the static fluid (it's datasheet thermal resistance value). A working fluid is about how much you can get it to wet onto hot places and then wet onto cooler places to transfer energy. If it somehow had no heat conductivity at all through the fluid itself, it would make little to no difference provided it could still become heated by wetting into hot things and then getting slung/flung or magnetically pumped to cooler surfaces to transfer heat.

 

[cal3thousand]

Just for thought, but FF is only about 5 times better conductor than air. However, 10 ML of ferro fluid will not fill the entire gap, so it will still mostly be air between the stator and magnets.

While FF is 5 times better than air, aluminum is 2000 times better conductor than FF, or 8000 times better than air. So FF is still a very poor conductor to bridge that gap. This is why some people lap their computer heatsinks because reducing that gap even by .001 mm or whatever makes a large difference in temperature because less thermal paste is needed.

What I am thinking is maybe work should be done to reduce the gap between magnets and stator, and fill the gap with more fluid. Not sure if this is possible, but more work needs to be done to increase the heat transfer to the case then just FF.

And in this comparison, what's the constant? volume or mass?

 

[Ohbse]

Hey guys I am new to this so I wanted to ask some questions to get some know-how about ff. I have a QS V3 motor and also have ordered now 10ml ferrofluid 1110 from a german guy from this forum. He said it will reduce the heat about 40%. I have talked about this with some experienced guys in this and they have told me that it is unrealistic that 10 gram of liquit can cover 40% produced heat of a 45.000 gram heavy bike + 70.000 gram heavy rider.
Could you please explain me how the ff works / why it works?

FF effectively bridges the gap between the heat producing elements (windings) and the heat removing elements (Hub shell/covers). This dramatically increases ability to remove heat. 40% reduction in temperature is absolutely possible. What's more amazing as you could achieve that difference with only using half of your 10ml tube. This has been proven empirically and via experience of many, many people now running FF. The silent majority in this case are those that no longer have heat issues.

Offroader - you have invested more time thinking and writing about this topic than it deserves - buy some FF, install it and see for yourself the difference. It is *very* easy to remove, certainly compared to conventional oil cooling and takes mere moments to do compared with many of the other elaborate methods you have spent lots of time contemplating. The science is already done, see the thread title - definitive testing has been completed and it proved, definitively, that FF is an excellent solution.

Independently of the FF advancement, Sketch has come up with an excellent solution for the other side of the equation and has also definitively proven his hubsink solution dramatically improves cooling ability. If anybody running good quality FF and hubsinks mounted correctly still has heat issues then their controller is incorrectly configured or they are running a motor very inadequate for their requirements. a QS 50mm v3 with FF and hubsinks is capable of serious performance for extended periods of time, far, far in excess of what I could personally use despite being young fit and not afraid of breaking rules. At this end of the spectrum I do not believe there's really a problem left to solve, unless perhaps you're morbidly obese.

 

[DasDouble]

It is *very* easy to remove,

Oh really? How do you manage that? I did now know, that it is easy to do that ..

What do you guys do that the FF doesnt get whiped out at the sides of the motor when running fast? Merlin told me I should use some silikon. Do you guys agree? And maybe use some special sort of silicon or just some random one?

Cheers

 

[Emmett]

A working fluid is not about how well heat moves from one side of the static working fluid to the other side of the static fluid (it's datasheet thermal resistance value). A working fluid is about how much you can get it to wet onto hot places and then wet onto cooler places to transfer energy.

Very interesting. Thanks.

I do not believe there's really a problem left to solve ...

I agree with that. Using a liquid in the hub and hubsinks, it is a done deal, in my world. A non-issue. Forget it and ride. Previously my bike was effectively useless.

What do you guys do that the FF doesnt get whiped out at the sides of the motor when running fast? Merlin told me I should use some silikon. Do you guys agree? And maybe use some special sort of silicon or just some random one?

See my post from 8 days ago. The steps for sealing the side covers. https://endless-sphere.com/forums/viewtopic.php?f=2&t=48753&p=1260015#p1257759

If you want to seal the axle slot for your wires, then see this post https://endless-sphere.com/forums/viewtopic.php?f=2&t=48753&p=1260015#p1257955

 

[Ohbse]

Oh really? How do you manage that? I did now know, that it is easy to do that ..

Can of Brake clean or something similar does the trick. Worth trying on your wire enamel to ensure no reaction before you hose down the stator, but unlikely to have a blend that will cause problems these days.

You may have some insignificant residue remaining in the nooks and crannies, but probably much less than general road scum and bearing debris that seems to get everywhere anyway.

 

[markz]

I was going to paint my stator with High Heat Rust-Oleum, but reading FF dissolves paint I wont be doing that now. Maybe look into alum fin heat sink.

 

[Cowardlyduck]

I was going to paint my stator with High Heat Rust-Oleum, but reading FF dissolves paint I wont be doing that now. Maybe look into alum fin heat sink.

I wouldn't go so far as to state it dissolves paint.
It happened to react with the Red insulating motor varnish that I used. That does not mean anything for other types of paint, of which there are a huge number of variations and different base types.
I would make a guess that FF would be far less likely to react with oil based paints, but I'm far from an expert in paint tech.

I'll be opening up my Leaf motor in the next month or so. It's had the same treatment as my HS4080; Red insulating varnish, cooling fans, vented, and FF added. I expect a similar result to the HS4080, but will post up the pics also.

Cheers

 

[Baron]

I was going to paint my stator with High Heat Rust-Oleum, but reading FF dissolves paint I wont be doing that now. Maybe look into alum fin heat sink.

I wouldn't go so far as to state it dissolves paint.
It happened to react with the Red insulating motor varnish that I used. That does not mean anything for other types of paint, of which there are a huge number of variations and different base types.
I would make a guess that FF would be far less likely to react with oil based paints, but I'm far from an expert in paint tech.

I'll be opening up my Leaf motor in the next month or so. It's had the same treatment as my HS4080; Red insulating varnish, cooling fans, vented, and FF added. I expect a similar result to the HS4080, but will post up the pics also.

Cheers

How exactly did the ferrofluid react to the red varnish? I searched the thread and I think you wrote it stripped away the varnish only on the steel surface parts? Do you mean just the lams or the magnets too?

Asking because I have a vented motor with the same red varnish and was considering adding ferrofluid. What are the other options in painting/varnishing a stator for venting+ferrofluid? Wouldn't black be a better color paint/varnish anyway?

 

[Cowardlyduck]

How exactly did the ferrofluid react to the red varnish? I searched the thread and I think you wrote it stripped away the varnish only on the steel surface parts? Do you mean just the lams or the magnets too?

Not sure what the reaction was exactly...all I know is what you can see also...the FF was mostly gone, replaced by a gunky residue. It wiped right off both the lams and the magnets...so all the steel surfaces, but not the copper. That could also be because not much of the FF actually touches the copper windings, or it could be due to the windings already being covered in a varnish meaning they held onto the extra layer of varnish better than the steel could.

And after wiping:

All the details are here:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=56965&start=400#p1198704

Asking because I have a vented motor with the same red varnish and was considering adding ferrofluid. What are the other options in painting/varnishing a stator for venting+ferrofluid? Wouldn't black be a better color paint/varnish anyway?

Sure black would be better, but good luck finding motor winding insulating varnish in black. Sure you could just use regular black paint...and I've done this;

But it's not going to protect the motor as well as proper insulating varnish would. I guess it depends on how worried you are about dirt/dust ingress. I ride a lot of off-road in very dirty/dusty conditions, so I need semi decent protection.
I may try re-adding some FF to my HS4080. I didn't bother re-spraying it with varnish since I wiped it off, so theoretically another batch of FF would actually be effective without being consumed now.

Cheers

 

[Offroader]

I decided I'm not going to try out the FF and just modify my new motor for air cooling.

I have a few reasons. I'll just put numbers on them to make the post more readable.

1) I believe alternative cooling methods need to be explored, if I used FF then there wouldn't be more research into the potential of forced air cooling.

2) My last forced air cooling worked great but was not optimized, very inefficient. While I did have the air ducted to have cold air in and hot air out, it was very inefficient.
Inside the motor I did not direct the air through the windings efficiently, so lots of air flowed above the windings (In my cromotor I could fit my finger between widings and side covers, lots of air would blow right above the widings), up to 50% or more of air flow.
Outside the motor I had the intake and exhaust too close and used no ducting so a lot of hot air was being pulled back into the motor.

My fan was smaller at 30MM using thin 24 awg wires, no where near as powerful as my new 35mm fan using larger wires.

My new forced air cooling setup I plan to build into my new motor should have about 3-5x more air flow through the windings. The outside air intake / exhaust will also have a separator/duct outside the motor so that much less hot exhaust air will be sucked back into the motor through the intake holes. This should reduce intake air temps dramatically.

This should make my forced air cooling much much better than before, and what I had before worked well.

3) I use my bike under the toughest ride conditions. My favorite type of riding is on steep dirt hills which I ride up and down and the limits are usually my winding temps. I can also turn up my fans to max speed during these type of riding. Forced air allows control of cooling speeds while FF + hubsinks does not.

4) My preliminary findings of running my fan inside a sealed motor showed that there was a bottleneck in heat transfer to the motor shell.

5) I'm honestly not convinced FF + hubsinks can come close to the performance of sucking cold air in, and blow hot air out. Until I see evidence I'm going to stick with what I know works. Most people don't push their bike to the limits like I do and FF + hubsinks will be the better option for most. But if you are pushing your bike to the heat limits often, then you need something more.

To me it is all common sense, hold your finger right in front of your mouth and blow on it. This is how the windings will be shedding heat in my new motor. The air will be blowing the heat right off the windings and out of the motor at a fast speed.

How can FF+hubsinks work as well when the heat has so many steps to get out of the motor. There is no way once those windings start to peak in temperature they will shed their heat quicker than air blowing the heat right out of them. I'm not trying to say you FF + hubsinks users have crap, just saying I need something better.

Below is a picture of how my setup works if anyone is wondering.

 

[Ohbse]

I decided I'm not going to try out the FF and just modify my new motor for air cooling.

My new forced air cooling setup I plan to build into my new motor should have about 3-5x more air flow through the windings. The outside air intake / exhaust will also have a separator/duct outside the motor so that much less hot exhaust air will be sucked back into the motor through the intake holes. This should reduce intake air temps dramatically.

My preliminary findings of running my fan inside a sealed motor showed that there was a bottleneck in heat transfer to the motor shell.

I'm honestly not convinced FF + hubsinks can come close to the performance of sucking cold air in, and blow hot air out. Until I see evidence I'm going to stick with what I know works. Most people don't push their bike to the limits like I do and FF + hubsinks will be the better option for most. But if you are pushing your bike to the heat limits often, then you need something more.

This thread is full of actual proof, see justins extensive dyno testing with vented covers vs sealed with and without FF. See the many, many people that had extensive heat issues, added FF and now do not. I'm all for exploring alternative avenues, but this one has been definitively proven to be a dead end, at least in comparison to the advantages of a sealed, silent, much more capable passive solution of increased thermal transmission and increased thermal dissipation.

Air cooling is a non-linear curve. More airflow = more cooling, but at a rapidly diminishing rate. Tripling the airflow forced through your motor (and likely quadrupling the noise output!) will only result in incremental gains. The bottleneck in transferring heat to the motor shell is precisely what FF effectively combats! EDF's are loud as hell, even more so with small high RPM ones with a massive pressure drop! For me that's completely the opposite of desirable, silence is golden, especially offroad IMO - you'll end up just as noisy as the stink bike munters. Even if the fans are as effective (which I don't think they will be), why would you choose the more complex, more risky (blowing all sorts of junk through your windings? Doesn't sound like a recipe for a long motor life) and more power hungry option?

If you're absolutely set on adding fans, a more effective solution would be to combine all of the options. Add FF, add hubsinks, add an external EDF blowing ducted air across the hub. I do not believe (with sane input phase amps) you would be able to overheat this setup before your battery is dead, your controller is molten and your ass is sore.

 

[sketchism]

I decided I'm not going to try out the FF and just modify my new motor for air cooling.

Certainly eager to see your progress, maybe do a thread for it so we can follow easily

Personally i have found the FF + Sinks to be super effective (obviously i am biased but that is from months of hands-on experimenting during the development process as well)

The main thing I would disagree with is what you mentioned about having Many Steps to remove the heat, especially compared to installing a complex system of fans and vents and powering them,
installing the FF gives the heat a direct path to the air outside, effectively making the stator and magnet ring/shell a single unit as far as heat transfer is concerned, and the sinks then increase the surface area (and increase the turbulence of the air for more effective transfer) used to dissipate that heat markedly

(to borrow from your image)

(Justins diagram)

Regardless of my opinion though, i am keen to see how you go with it,mostly because i know what its like when you get an idea you want to follow through until completion, and if worse comes to worse, your plan doesn't preclude the use of sinks and FF combined with the fans and you can install both for the coolest motor around haha, Let me know when you get close and ill send you a set of sinks to test out

 

[macribs]

How can FF+hubsinks work as well when the heat has so many steps to get out of the motor. There is no way once those windings start to peak in temperature they will shed their heat quicker than air blowing the heat right out of them. I'm not trying to say you FF + hubsinks users have crap, just saying I need something better.

I tend to applaud those who stick to their guns and keep refining, thinkering and improving systems while other tell them not to. History has shown us some really does find the holy grail of their root problem. I agree with sketch, pls make a thread of your experiences.

However in this case I can't help thinking you are wasting time, money and effort while being the only player in the band that is "on time". Regarding FF+heat sinks we have data, we have user reports and we even have users that ride their FF+heat sink hubbies in 40+ degrees celsius weather. Repeatedly accelerate, using regen, climbing steep or even ride on sandy tracks. Then again I've been wrong plenty of times. If your way is better we all learned by follow your progress.

If you are sure your usage will put too much heat in the motor to use FF+heat sink may I suggest to look into even larger heatsink, paired up with heat pipes as well? Then your hub will have closed loop liquid cooling system, air cooling of heat sink, and phase transition cooling taking place in the heat pipes That combo should take care of your heat. You still have a closed system, no debris or sand, dust or anything else will find its way into the motor. And you will benefit from various heat transfer solutions.

Personally I can't think of anything better then that unless you go for liquid cooling with radiator, pump and reservoir. But liquid cooling as well as EDF will add more complexity.

If it was not for the harsh riding conditions, salty air from the sea, as well as all the salt covered roads during winter I would have jumped the FF wagon myself. But due to all the salt I think the only thing that will prevent rust inside the hub is oil, so I am going for oil filled hub.

 

[Cowardlyduck]

I say go for it Offroader. Don't start a new thread though....we already have one....lets revive it.
https://endless-sphere.com/forums/viewtopic.php?f=30&t=56965&start=400#p1202235

I'm a fan (pun intended) of both FF + Hubsinks and Fan cooling methods. I'm only sticking with fan cooling currently as I started out on that path, venting all my motors some time ago to make it work well. If anyone's looking for proof that fan cooling works, it's there...just not a recent as the whole FF + hubsinks development.
https://endless-sphere.com/forums/viewtopic.php?f=30&t=56965&start=25#p946731
And more recent harsher/hotter testing with better fans:
https://endless-sphere.com/forums/viewtopic.php?f=30&t=56965&start=350#p1123565

I'm actually going to be re-doing the Leaf motor from the test above soon, as I accidentally pushed too much voltage through the fans frying them. This time around I will be using 12 fans so performance should increase quite a bit.

Anyway, Offroader, I suggest we quit hijacking this thread or they might kick us out...lets pollute the cooling fans inside hub motors thread and come back here when we've got more results/proof to share.

Cheers

 

[Quokka]

Just installed a set of hubsinks on a QS205 pushing 10kw on my Qulbix Q76R. I am back from my first run with them. The temperature here at the moment is stinking hot. 40 degrees C plus with the desert winds blowing in. Innitial feedback is that these do work. For me though, i was still able to go from 80 to 110 in a matter of a few minutes. I do still need to install my ferrofluid so i know that it will get even better. Will report back when i have liquid gypsy magic installed.
Cool down time is noticeably faster. At a stop it drops 10 degrees in 5 mins, and drops quickly back to 85-90 when just coasting along using 1000w or so.

Have 10ml of ff in the hub now...
Went for a few fangs around the block, hotest i could get the bike 80-85 degc. Its not the same hot conditions as the last test but i do think it doesnt get anywhere near as hot. Its night time and the temp is around 25deg c. At standatill it drops 15 deg in 5 mins, so it does seem improved on just the heatsinks. Test was just pushing 40- 80 odd km/hr for around 20 mins. Too many kamakazee bugs to deal with. Those things hurt at 80.

 

[litespeed]

Nice!

Tom

 

[Cowardlyduck]

Great stuff Quokka! What did you use to seal the side covers? Looks like a gasket, is it?

Cheers

 

[Quokka]

Permatex ultracopper. Its a silicon gasket maker- i used to be a gearbox mechanic in a past life and this was our go to gasket maker. Seals anything and is high heat. Motor went back together perfectly torquing up the bolts opposite and then a star patern. Disk brake runs even truer than from the factory. Just have to get a razor blade to clean out the factory silicon ( pretty poor sealing job from factory) and then a wipedown with shellite befor the sealant goes on

 

[Quokka]

Just installed a set of hubsinks on a QS205 pushing 10kw on my Qulbix Q76R. I am back from my first run with them. The temperature here at the moment is stinking hot. 40 degrees C plus with the desert winds blowing in. Innitial feedback is that these do work. For me though, i was still able to go from 80 to 110 in a matter of a few minutes. I do still need to install my ferrofluid so i know that it will get even better. Will report back when i have liquid gypsy magic installed.
Cool down time is noticeably faster. At a stop it drops 10 degrees in 5 mins, and drops quickly back to 85-90 when just coasting along using 1000w or so.

Have 10ml of ff in the hub now...
Went for a few fangs around the block, hotest i could get the bike 80-85 degc. Its not the same hot conditions as the last test but i do think it doesnt get anywhere near as hot. Its night time and the temp is around 25deg c. At standatill it drops 15 deg in 5 mins, so it does seem improved on just the heatsinks. Test was just pushing 40- 80 odd km/hr for around 20 mins. Too many kamakazee bugs to deal with. Those things hurt at 80.

Just gave my bike an absolute hiding. Now running ff and hubsinks. Adaptto on boost with ovs 3 at 120amps. Temp here is mid 30's and i tackled the same route i did in my first test with just hubsinks. Air temp is around 10 degrees cooler. The motor temp stayed cool... really really cool. Was running about 30 degrees cooler on the same hills, but was going faster. Not only does the motor run cooler but the motor seems much more powerful at lower temps?? It feels like an absolute animal now. Does anyone know if torque goes down as motor temp goes up?
I then went in search of something harder in the national park. The absolute hottest i could get was 90 in bush. That was going up a dirt road, up a big hill- maybe 2km long, full throttle. I just couldnt get it up to the same temps as it used to run pre- ff and hubsinks.
Along the single tracks i couldnt go that fast because of all the turns, front wheel spent alot of time in the air, its a real blast to ride, in these parts it was down in the 60's
Final heat test I did was a simple brute speed run. Max power, over 100 for a nice long flat stretch for around 7-8km. Peak temp 95. All up i did around 80km of testing mostly on dirt and nearly had a few big stacks hitting soft powdery sand at high speed really speeds up ones heart rate.
Only way i see that i could hit the temp cut out would be a really steep hill (actually i mean a mountain) at full throttle. There is nothing like that here so i am done
Bottom line, i think it lowered the temps by at least 20deg c. Probably more

 

[litespeed]

Kick ass report!

Looks like a great place to ride.

Tom

 

[Rix]

Kick ass report!

Looks like a great place to ride.

Tom

What Tom said.

 

[DasDouble]

How many ml ff have you guys put into your motor? I have 10ml but did not use the whole stuff because it looked like otherwise it would be too much for 1 single motor..