Definitive Tests on the Heating and Cooling of Hub Motors

RageNR said:

I watched that race. Had no idea you guys participated here on the forum. Pretty cool.

Click to expand...

Yeah I've been blowing up motors there for the last few years now. haha
Funny you should mention it, here's a pic of me closely approximating your avatar! (once it was clear I wouldn't catch the guys running hubsinks I decided to clown around )

would the heatsinks not start becoming effective until the motor was saturated with heat?

Click to expand...

Well, you could argue that if you're not getting your motor it's a non-issue either way.
The heat does need to soak across from the windings to the sidecovers and to the spine of the hub though, so yes technically they're not doing alot until you get the motor quite warm. But as long as there's a temperature delta between the surface of the motor and the environment the sinks will enhance it.

RageNR said:

I'd still like to see someone figure out how to make an external cooling solution work, with radiator, coolant, and fan.

Click to expand...

Has been done, should be some threads here on ES about it.

I think Linukas even made a kit for it.

I'd guess that motors that have controllers inside them, like the magic pie, and the stromer / ultramotor, would probably benefit from FF + rotor-based heatsinks, if they are being used at their higher power levels constantly or very often.

I don't yet know how to connect up to the Stromer motor I now have to make it's internal controller operate (waiting on response in the stromer threads) but once I figure that out, it might be interesting to test it on the trike, as that's a lot more workout than it was meant for, and see how it works as-is, then with FF and heatsinks (though I'd have to come up with my own versions).

Once testing is done to satisfy curiosity, then it would go on to run Bill's bike, where it would probably be used very infrequently and at very low levels and speeds, so it's not worth buying stuff for it that will only be used for a test.

Hyena said:

Funny you should mention it, here's a pic of me closely approximating your avatar!

Click to expand...

LOL! Good show sir, good show. *golf clap*

Hyena said:

The heat does need to soak across from the windings to the sidecovers and to the spine of the hub though, so yes technically they're not doing alot until you get the motor quite warm. But as long as there's a temperature delta between the surface of the motor and the environment the sinks will enhance it

Click to expand...

I can believe that. My theory behind it is, it is there to keep the motor from overheating more so than as a function to lower the overall thermal load.
Not sure this is an accurate term to put on it, but it's sort of like a "thermal limiter". Therefore its most important function is to protect the motor from overload.
The additional benefit is that it will, by nature of physics, also help lower the temp. to a degree.
I'm clearly not as smart as you fellah's. Sometimes an explanation from a simpleton is easier on the feeble minded.

amberwolf said:

Has been done, should be some threads here on ES about it.
I think Linukas even made a kit for it.

Click to expand...

Oh? I've searched for something like this several times, but never found anything. Only people talking about the idea.
If I can not find it, I might bother you for a link.

off to investigate...

Maybe I misunderstand what you're looking for, but this is the search I used to find that:

https://endless-sphere.com/forums/search.php?keywords=liquid+cool*&terms=all&author=&sc=1&sf=titleonly&sk=t&sd=d&sr=topics&st=0&ch=300&t=0&submit=Search

That only looks in thread titles, so I'm sure there's more posts about this stuff buried in threads (possibly even in this one!).


Here's one about Linukas' kit being used
https://endless-sphere.com/forums/viewtopic.php?f=2&t=65954&hilit=liquid+cool%2A#p992078
tested and worked, but edventure didn't ever finish posting the data/results/etc.

and posts by linukas
https://endless-sphere.com/forums/viewtopic.php?f=30&t=63142&p=989682&hilit=mxus#p989682

https://endless-sphere.com/forums/search.php?keywords=cool*&terms=all&author=linukas&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search


and doctorbass' project
https://endless-sphere.com/forums/viewtopic.php?f=2&t=39933
but i guess he never even tested it after going thru all that work

Hyena said:

Wait, CD are you running FF with a hub peppered with holes around the rim ? It doesn't leak/fling out ? I tried some in a hub with an unplugged hole and it rapidly leaked out (though to be fair it was a lower temp grade and I did boil it)

Click to expand...

Indeed I am/have been for some time now. The verdict was still out on whether or not it would leak out, but I think my little experiment shows it won't.
Clearly the hold of the magnets is strong enough to keep the FF in place. I have not run my Leaf motor past about 75Kph though, so I imagine at some point the FF might let go and fly out the holes. That being said, it will only do this if the holes are centrifugally further out than the magnets and there is a path for the FF to flow out to the holes.
With some careful placement of the vent holes not too close to the perimeter you could theoretically run a motor at any speed and not have the FF come out. The only catch with this is any dirt that then gets into the motor is gonna have a hard time getting out also and will absorb into the FF.

Cheers

Cowardlyduck said:

Hyena said:

Wait, CD are you running FF with a hub peppered with holes around the rim ? It doesn't leak/fling out ? I tried some in a hub with an unplugged hole and it rapidly leaked out (though to be fair it was a lower temp grade and I did boil it)

Click to expand...

Indeed I am/have been for some time now. The verdict was still out on whether or not it would leak out, but I think my little experiment shows it won't.
Clearly the hold of the magnets is strong enough to keep the FF in place. I have not run my Leaf motor past about 75Kph though, so I imagine at some point the FF might let go and fly out the holes. That being said, it will only do this if the holes are centrifugally further out than the magnets and there is a path for the FF to flow out to the holes.
With some careful placement of the vent holes not too close to the perimeter you could theoretically run a motor at any speed and not have the FF come out. The only catch with this is any dirt that then gets into the motor is gonna have a hard time getting out also and will absorb into the FF.

Cheers

Click to expand...

If you have FF in the motor, why even bother with air cooling at that point?

Actually, I have seen both of those some time back. The thread for edventures bike wasn't conclusive the last time I looked at it. Prob been over a year now that I read either of those threads.

Not quite what I had in mind on the execution, but the same general idea. I have this bad habit of drifting online when it starts to get late. Like I am doing right now.
You might as well leave your logic outside by the door, cuz ain't no one home upstairs at this hour. LoL
So yeah, I saw both of those. Guess I forgot about it.


I'll tell you what sparked my interest in the idea.
I've worked on lots of different types of commercial equipment over the years. Cooking equip, comm refrigeration, comm A/C, large dish machines, etc.
These large Hobart dishwashers use one really heavy duty motor for the wash pump. Sits outside the the base wash tank near the bottom.
Here is an example of a medium sized unit:

There is a doughnut shaped assembly on the end of the motor that looks akin to a turbo. That's where the impeller is.
What is interesting about these is how they seal the impeller and housing to keep the water from leaking past the shaft and getting into the motor.
They use a press fit ceramic bearing in the backside of the impeller, and another in the housing.
Now I know that is not super special. There are other applications that use similar designs. What had me surprised is the fact that those ceramic surfaces could handle that sort of vibration, jarring, impact, and resistance to chemicals for so long. Sometimes decades. And they ARE A BEAR to change out.

So that had me thinking if a similar setup could be applied to the outer casing of a hubmotor. Obviously, heavily modified.
After seeing how these have held up in these washing machines, I think it might hold up on an ebike.
Those Hobart machines are downright brutal. The dish station workers don't give one F what happens to these $10k+ machines. They let forks and spoons and crap get down in the wash tank, then get sucked into the impeller. And yes there are safe guards to stop that, but they will leave them off because they don't want to bother with cleaning them. *facepalm*
Changed out nearly 20 motors I'd say by now. And at $1k a pop for the motor... well yeah, it ain't cheap.

Anyways, it's an idea I've been kicking around for a while. Just thought I'd throw it out there.


Here you go. Seal looks like this:

Goes in this:

Offroader said:

If you have FF in the motor, why even bother with air cooling at that point?

Click to expand...

If it works, best of both worlds, and besides you can't exactly un-vent and already vented motor. I wanted to try FF together with fan cooling to see if it could work, and I'm pleased to say it does so far.

Cheers

Cowardlyduck said:

Offroader said:

If you have FF in the motor, why even bother with air cooling at that point?

Click to expand...

If it works, best of both worlds, and besides you can't exactly un-vent and already vented motor. I wanted to try FF together with fan cooling to see if it could work, and I'm pleased to say it does so far.

Cheers

Click to expand...

You can unvent that motor no problem, just get some epoxy, block one side of the holes, and fill the hole in. Would take 10 minutes and be a strong bond.

If you just wanna *test* it unvented, just use some clear packing tape over the holes if the surface is smooth.

I have no desire to test it un-vented, or to reverse the venting. As mentioned, so far it works fine even with fan cooling and FF, so I'll keep doing that.

Someone else with a non vented motor should post up pics of the internals after a few thousand kms riding.

Cheers

Just was watching the F1, and looking at all there brake cooling ducting, and it has me thinking about forced cooling of a hub motor? i no some on here are using it

This has some good info on the ducting ect here http://www.singularmotorsports.com/tech-blog-brake-cooling/

And here too https://www.motorsport.com/f1/news/...ver-solutions-that-divide-the-f1-grid-880644/

Well time to keep reading this thread!

Ducting has been discussed before; has even been used a few times.

(not all these results are relevant)
by post
https://endless-sphere.com/forums/search.php?keywords=duct*+cool*+hub*&terms=all&author=&sc=1&sf=all&sk=t&sd=d&sr=posts&st=0&ch=300&t=0&submit=Search

by first post / topic
https://endless-sphere.com/forums/search.php?keywords=duct*+cool*+hub*&terms=all&author=&sc=1&sf=firstpost&sk=t&sd=d&sr=topics&st=0&ch=300&t=0&submit=Search

Also keep in mind that the speeds those pictured ducts are used at are MUCH higher than what we are doing, even those with "high speed" ebikes.

Air behaves a little differently at those speeds vs ours (not a lot) so the ducts might have to be shaped differently to do the same thing for us.

I think the method of cetrifugally throwing the air out as on the outside of the race car wheel above is universally applicable but can be optimized for the speed the wheel is going by changing the shape, angle, and amount of blades

https://www.youtube.com/watch?v=DIXVtP9GAFE&t=5s
on the other side of this spinning motor is just an impeller with a flat plate very similar to race car above. all those weird shaped ducts somehow are pulling the air into the inner side of the wheel at the same time. double action.

Hummina Shadeeba said:

I think the method of cetrifugally throwing the air out as on the outside of the race car wheel above is universally applicable but can be optimized for the speed the wheel is going by changing the shape, angle, and amount of blades....

Click to expand...

Don't forget that the diameter at the blade tips is also important.

I comparing to the race car wheels an advantage we have is the open spokes, so the action of the centrifugal blades can also pull air from the other side of the motor across the outer shell which aids in cooling.

Of course the best way to prevent a hot motor remains creating less heat through greater efficiency. With my latest build the motor hasn't reached 60°C yet, and I haven't even installed my blades or ducting to send more air toward the motor yet, and that's running a 15% larger diameter wheel than before. Of course my 35kg lighter load helps some, but the biggest difference is going down on the battery current limit from 245A to 200A. I'm still slaughtering motorcycles from the stoplights, so I'm not really missing the extra acceleration of higher current limits. Plus I don't have to be so darn careful with the throttle on take-offs. 8)

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

amberwolf said:

RageNR said:

I'd still like to see someone figure out how to make an external cooling solution work, with radiator, coolant, and fan.

Click to expand...

Has been done, should be some threads here on ES about it.

I think Linukas even made a kit for it.

Click to expand...

Commander Keyne


http://endless-sphere.com/forums/viewtopic.php?f=6&t=26029&start=175#p530781

Cowardlyduck, FERROFLUID DID NOT WIPED OFF THE RED INSULATING VARNISH !!

There is something you missed off... Your first motor where the red insulating varnish wiped off IS NOT because of the ferrofliod.. but because of the HOLES in teh side covers that have cought dust and sand that made teh ferrofluid more abrasive and THIS wiped off the varnish... not the ferrofluid itself. as a proof, your second motor wioth less damage have smaller holes to let sand and dust particules to stick to teh magnet and ferrodluid.. so the paint is LESS abrasive this way. THAT'S the answer you need! :wink:

Cowardlyduck said:

Alright, well in an attempt to distract everyone from this pointless argument I'll throw in an update to my FF + Fan cooling endeavours.

If you'll recall, my last motor (HS4080) ended up consuming most of the FF I added when it reacted with the red insulating varnish.
https://endless-sphere.com/forums/viewtopic.php?f=30&t=56965&start=400#p1198704

The red varnish then mostly wiped off when I removed the FF:

Well it seems this time around it's a different story. I opened up my Leaf '1500W' motor which is basically the same setup as my HS4080 with fan cooling, red varnish, FF and heat sinks, however this time around the results seem much more promising:

As you can see, it's not perfect, and there is some grit that made it's way into the FF, but not so much that I am concerned about it after more than 2000km on this motor since adding the FF. The other nice surprise (compared to the HS4080) is the FF is mostly still there. It does not appear to have reacted with the varnish this time around.
I 'think' I used statoraide type A in this motor, where as I used type B in the HS4080. Does anyone know which was which? That would indicate that one of them reacts with the red varnish while the other does not. Another factor was I did use more FF in this motor than the HS4080. I think I used about 4-5ml in the 4080, where as I used about 6ml in this motor.

After wiping down the magnets and laminations, nowhere near as much of the varnish was removed compared to the HS4080. The small amount that was could be put down to rubbing from when I removed the stator, and/or rubbing from grit in the FF, but I don't think the FF itself reacted with the varnish this time around.

Click to expand...

Doc

Doctorbass said:

Cowardlyduck, FERROFLUID DID NOT WIPED OFF THE RED INSULATING VARNISH !!

There is something you missed off... Your first motor where the red insulating varnish wiped off IS NOT because of the ferrofliod.. but because of the HOLES in teh side covers that have cought dust and sand that made teh ferrofluid more abrasive and THIS wiped off the varnish... not the ferrofluid itself. as a proof, your second motor wioth less damage have smaller holes to let sand and dust particules to stick to teh magnet and ferrodluid.. so the paint is LESS abrasive this way. THAT'S the answer you need! :wink:

Click to expand...

Thanks Doc. That is one possibility for sure, and it's certainly a likely scenario.
That being said, in both motors the air flow goes through the holes, through the centre of the stator where the fans are, then out the holes on the other side. The air is not directed over the magnet gap, which is why I tried using the FF in the first place, since any contaminants are flowing with the air so will not contact the FF.

Anyway, I plan to use this second motor with the smaller holes for a lot longer this time before opening, maybe 5000+km. So if it was contaminants in the FF that caused the varnish to wipe off, next time I open this motor should show me if that's the case.

Cheers

Some students have developed a clever solution to cool their 12kg heavy, 60kW ( ) motor. They say they are able to go up to 200km/h.



http://www.focus.de/auto/elektroauto/gauss-projekt-e-motorrad-elektro-schocker-mit-200-km-h-das-heisseste-geraet-der-iaa_id_4937541.html

Some pictures after having ff in my QS 205 V3 motor filled in for a couple of hundreds of kilometers:

Das double, if you sold this ebike, and were building up a new one, would you still use ferrofluid? Would you still use the QS 205/50H? or would you do something different?

I think I would use Ferro Fluid again as the 10ml of Merlin weren´t that expensive. Except of the time I had to spend to open the motor I have not lost anything in any way. I also think the motor got warmer faster then without but not as much as I expected. In general when Im pushing 10kW and don´t do a constant stop and go, it doesnt get warmer then 80-90 degrees.

I once rode a stock Vector bike, as I took care of one for a customer from Artur and have been allowed to ride it. And as far as I remember, it overheated faster then mine now. But the stock one used its full potentional while I only use 10kW, instead of 14 like the Vector stock one does. So all in all my feeling says that it cools down faster in general, then without ff.

Would I use the same motor? Yes, Im really happy with it (except of that my axle broke, that makes me annoyed really much). But I don´t have any experiences with other motors, as I only own the QS until now. But many other people like doc. bass etc switch from their mxus to the QS v3 because of it´s power and it´s excellent accelleration.
One more negative thing: The cables of it do get easily cut by the disc brake. That´s an absolut negative right here. But it "only" has cut the white protection thing of the cables coming out of the axle. Not the phase cables it self yet (I will upload a picture for that).

What else would I do different? I think I would have driven better if I would have buyed a complete batterypack from Vector. I have built myself a 300 cells one with Samsung 30Q cells and thought I would safe some money here but in the end I had to spend about 6 houres per 60 cells pack => 30 working houres. Well.. I have learned freaking much stuff about electricity, but the welding everything and looking that everything becomes perfect took me a really hard time.

Not: I would go with a motorcycle tire on the front, as the 24" bicycle wheel gets used up way too fast, compared with a motorcycle tire

What would I do again?
-I would buy again a Vector frame, because its quite cheap and I had no problems with it untill now. +Has much space inside +Has a motorcycle seat
-Take an 18" rear motorcycle wheel.
-Buy a Max-E controller (10kW are already amazing, but 14kW will be an everyday satisfaction)

-Cheers

Right side driven motor with jackshaft placed high and out of the way?
Space saving solution while still keeping the motor low and central.