Yeah, straight hubs have a severe problem as is - the heat rejection surface the statorade is improving (the outer circumference) is covered in urethane, which is an awful thermal conductor.
DD motors (with the cans not covered by urethane) will perform better.
I think the performance of hubs can be improved by also adding radial fins to the faces of the motor, either inner or outer, so that the heat transferred to the can can be radiated away.
Jerry was doing this quite recently before carvon finally shat itself and died. Results were in dispute.
yeh I am talking about DD. They do run cooler than hub motors, so with Statorade will be even better.
That’s not correct. Radiation happens at all temperatures.
While technically correct, the amount of heat radiated is directly related to the temperature differential between the radiator and the environment. Hotter radiator = more heat being radiated.
Of course yes but the air inside needs to heat the can up right before it starts radiating?
No, there’s radiation and there is convection, for convection the air needs to heat up and then it heats the can, for radiation the transfer is directly from the stator to the magnets and can, doesn’t depend from the air
Makes sense, there’s no physical connection though? Without the goop we’re discussing now anyway
That’s exactly it’s purpose, to make a physical connection
As I said trying to push the heat through the stator core, to the hanger and using it to cool is almost useless, once you calculate and add up all thermal resistances you see it’s too high, that’s why the “watercooled” motor of the Raptor 2.1 still cooks
You have the stator itself that is iron that’s is a bad heat conductor, them goes to the hanger which is part aluminum, that is a good conductor, and part steel, that’s is a bad conductor, but both of them have a minuscule cross section area
That goes to the heat pipe, that’s is an amazing heat conductor which ends in a heat sink that can’t even be called that, at peak current you have a few hundred watts of loss and system that could only handle way less than that
With the ferrofluid you have the can that is spinning and moving forward through the air really fast, that makes the heat sink way more effective than the hanger one, even if you could make both have the same surface area
I did some simulations a while back trying to see if I could use a 5055 motor as direct drive, and one of the studies was to replace the air in the airgap with oil, the difference was considerable, in trying to find it but I don’t think I saved the graphs
I’ve been mulling over the idea of at some point building a large BLDC motor using copper tubing instead of wire as the conductor and pumping coolant through it. Direct cooling of the coils is ideal, but adds all kinds of other problems - plumbing the phases together while keeping them isolated, having a water pump with enough pressure to drive coolant through the tiny tubes with sufficient volume, and getting tubing of an appropriate diameter to allow a realistic number of windings.
Another idea I’ve had is to build an enclosure around the entire stator (inside the rotor - tricky engineering to keep the air gap small) and filling it entirely with circulating coolant. That would eliminate the pumping/plumbing issues of the tube idea, and the leaking/drag issues of direct filling.
The main issue with this approach is that you loose a lot of cooper to give space for the cooling, less cooper = more loses = more heat
Going inrunner solves most of the challenges with cooling, and bring a lot of others
Yeah, but when you have the heat-rejection totally removed from the heat-generation, you can afford to waste a little extra power because your overall cooling will be better, and then you can do other things to improve your power and efficiency, like running N52 magnets that would normally be destroyed by the heat. The copper tubing I mention has comparatively thick walls and a small ID - the piece I have laying around is about 1/8" OD, and maybe 3/64" ID - plenty of copper still there. Problem is, 1/8" is about 8AWG - winding a motor with solid 8AWG wire is difficult if not impossible depending on the size of the motor.
Just going crazy here, you could use mercury as cooling fluid, so it also contributes as a conductor, you would need three completely isolated cooling circuits and a way for the reservoir doesn’t short all together
If you’re going that crazy, here’s an even crazier idea: Use Nak (sodium-potassium alloy) instead of mercury! It’s lighter and cheaper and not as toxic! (Problem is, it’s highly reactive and tends to catch fire on contact with air, and explode on contact with water.) You can even make a purely electromagnetic pump that acts directly on the fluid using magnetohydrodynamic forces!
I just looked on Grainger and they have some capillary tubing that’s 0.072" OD and 0.026" ID - that’s a capacity reduction of 24% compared to solid 0.72" (13AWG) wire. I think the cooling benefit can outweigh that amount of loss.
@Skatardude10 We can upgrade the actual rear truck to be thicker with more heatsink and surface area which could probably help with cooling off. Worth a shot. I have yet to try it. I could get something done and you can test it. 
The stator/axle area will get the hottest and eventually it will seep over to the motor can/rotor.
Another option is changing the motor configuration for less of a magnetic field which should be able to keep it cooler but the sacrifice of a bit less torque. I think that’s working backwards though as you could just use less amps and things would run cooler.
The Statorade Ferofluid that sounds like an option which is popular for Ebike Hub Motors and what @Pedrodemio has mentioned. I haven’t had a chance but I definitely want to try this.
You should be able to pop the top bearing on the TB DD but getting it aligned may be a bit tricky.
@Pedrodemio - We actually use an aluminum stator axle so the stator is sitting on an aluminum axle not a steel axle.
That definitely helps, so the steel part of the hanger are two inserts on on each side?
I’ve found the simulations using the oil vs air in the air gap, both have the same loading conditions, the only variable is changing the air gap material, with Statorade the results should be a little worse since it is does not fill the slot completely. The first one is with air, windings temperature peak at approximately 104 ºC. The second with oil, we get a peak of just 65°C, meaning we could for example as others said, use N52 magnets for better torque and efficiency or keep the high temp ones and push much more current into the motor
AIR:
OIL
These are the tests from Justin with the Inboard motors, the main takeaway I get from him is that a sealed motor with Statorade has better heat conduction than a vented one, and that is with a hub that is thermally isolated in most of its surface, with our direct drives the diference should be even greater
https://endless-sphere.com/forums/viewtopic.php?t=48753&start=1975#p1363433
https://endless-sphere.com/forums/viewtopic.php?t=48753&start=2000#p1367551
Yeah, of course, but I was commenting about the hot as balls comment, which is incorrect.
As for the temperature differential, isn’t it a given that a temperature differential will exist(because of inefficiency rejected as heat).
Also if you can conduct the heat out of the stator core to rotor, there is a lot of additional surface area for both conduction and radiation cooling to happen.
For sure let me know, i’d love to try a beefier hangar on these, and it’d be cool to know if a larger heat-sink hangar can make a quantifiable difference.
I’m thinking a good way to do this would be to do a few of the same runs one after the other logging ambient temperature, IR thermometer of the cans and hangars at a few points throughout, repeated a few times letting the motors cool between runs while recharging.
I’d also like to know if the statorade works, so maybe incorporating that into the last few runs to see the difference between standard, beefier hangars, beefier hangars + statorade and then back to standard hangars + statorade.
HAHAHahahaaa I was there for this 
what you think of casting indium around the stator and with tunnels out of the motor to the aluminum hanger? that’s my plan.
its a soft low-temp melt metal with good thermal conductivity. that seems the best passive cooing technique.
a bit expensive but not much will be needed. its pretty much solder really.
