Cooling Cross Wires in Hub Motor

[chuyskywalker]

So, I burned out some wires in this little DD hub motor (8.5" from a hoverboard).



The obvious solution (which I'll implement on the next one) is a temp sensor with rollback. I may even split the load on the motors further by going from 2 rear wheel drive to all wheel drive

However, it seems that all the real damage was done in these "cross over" connections. I would imagine the explanation is that since these are (mostly) just hanging in space, their lack of a place to dump heat means that they take the brunt of damage when pushing too many amps.



So the question: is there any way to help these crossing keep cool?

I've read up on several of the air cooling threads here and I don't think that's the solution for this application. For starters, this is a single sided motor and most of the test results I saw said that only adding holes on a single side does very little, and it would be quite tough to add holes on both sides. In addition, there is very little space inside this motor, so adding something like internal fans for forced cooling isn't likely. As for adding internal blades to push air around, that's possible, there is a good deal of space in the shell as it's about 40mm wide, but only has 25mm stator/magnets. However, I'm not confident enough to make something that would work.

I might consider ATF for this setup. One of the most cited issues is that, if it leaks, you ruin your brakes. Well, this little powerwheels setup is all regen braking, so there's nothing to ruin. Plus, since it's one sided I only need to worry about sealing one side, which reduces leak chance by 50%! ;D

 

[chuyskywalker]

I could also throw some Ferofluid at this, but I'm not sure how much that will help as, typically, that uses the outer shell to dissipate heat and this thing is wrapped in a nice, comfy blanket of insulating rubber. Which, I suppose, also limits the usefulness of ATF to a certain degree as well, but ATF would a least be able to normalize the heat around the entire space, perhaps?

 

[Hummina Shadeeba]

Could paste the wires down with thermally conductive resin. But maybe those cross-over wires shorted because they’re not pegged down and were rubbing. Another reason to glue them down

You’re going to rewind the motor right?

 

[chuyskywalker]

But maybe those cross-over wires shorted because they’re not pegged down and were rubbing.

Nah; it was definitely the kiddos both sitting on the vehicle while ridding through thick grass. Too many amps for too long.

You’re going to rewind the motor right?

Nah, I can find these hoverboards for, like, $100 used and the motors separately for even less a few places. It would be kinda cool to rewind them into some really low turn count, thicker wire setup so I could up the battery voltage and handle the amps better, but it's simply not worth my time/sanity to rewind a motor.

 

[amberwolf]

I'd say your guesses about what happened and why are correct...except that if you look at the windings that that interconnect leads to, on the stator teeth, they're cooked too, on more than one tooth.

Seems possible that there's a short between that winding and the stator tooth, or there's two windings on the tooth and the short is between the separate phases.

The interconnects you see are probably only one phase's wire in each of them. If that's the case, there's nothing for them to short to, so it doesn't matter if they get hot enough to burn off their entire insulation, as long as the ones on the teeth don't overheat, and the interconnect wires don't actually melt or deform.

If there are multiple phases in each interconnect, *then* it matters for them to stay insulated.

You can test, on a good motor, which of these is the case (but you have to disconnect the main WYE interconnect between the phases inside the motor, which is probably the heatshrunk set of wires in the middle of that close-up pic.


For overheating:

Whether any cooling liquids will be enough, I don't know--if they're not spinning the motors very fast there probably wont' be much circulation except right at the outer circumference.

You might want to limit the phase current by temperature; you can install multiple sensors in there (depending on how much space you have for wires), and place at least one of them on the long interconnect wires from one phase section to another, and another stuffed between windings on a stator tooth if it'll fit.

 

[Hummina Shadeeba]

In ur pic I see the 4 transit/crossover wires all burnt and one looks like it was mechanically pulled apart after already being burnt. Cant see if anything else is burnt except maybe one tooth in background.

It looks to be only one winding on each tooth. terminated wye. 4 wires in parallel. About 10 turns. If you’re interested u could unravel and find how many turns on a tooth those four wires do and then try to wrap one maybe 15awg wire around a tooth as many times as possible and see if can do double turns they did on one tooth, and then from there more turns you put on are lowering kv. It’s an easy assures way to wind w lrk winding every other tooth unobstructed. If u want me to wind it do one tooth w the 15 awg bet could make much better n lower kv is the goal.

Rr I’ll do it for 40$. N bake it with best thermal resin n 230c wire. Just to do it n see and sexy photos and 40$ is that final needed push. If ur interested. Hate shipping n u could leave it at noisebridge in sf n I’m local.

 

[chuyskywalker]

In ur pic I see the 4 transit/crossover wires all burnt and one looks like it was mechanically pulled apart after already being burnt. Cant see if anything else is burnt except maybe one tooth in background.

The interconnects you see are probably only one phase's wire in each of them. If that's the case, there's nothing for them to short to, so it doesn't matter if they get hot enough to burn off their entire insulation, as long as the ones on the teeth don't overheat, and the interconnect wires don't actually melt or deform.

I suspect that maybe these interconnects melted enough that where they lay over other phases they made contact, perhaps. I do agree that these are likely all part of the same phase.

Rr I’ll do it for 40$. N bake it with best thermal resin n 230c wire. Just to do it n see and sexy photos and 40$ is that final needed push. If ur interested. Hate shipping n u could leave it at noisebridge in sf n I’m local. [/B]

That's a very generous offer. I may take you up on it if some of the replacement sources don't pan out. The dilemma with rewinding the motor is that it was part of a pair -- having two motors and different winds would probably have some weird consequences, I'd bet.

 

[chuyskywalker]

Woof; spent a solid 45 minutes trying to get a single arm unwound. They did some kind of hot-glue-like adhesive under the winding and, even then, they somehow also got the 4 pair to wrap under other sections of the wind so you can't just isolate one arm and do a simple unwrap.



Yeeessh

That said, the motor was laser engraved with shengbo35v25h25t19061710017BRUL , which I assume, means it's intended for 36 volts, has 25h magnets (it does), and is a 25 turn wind. I think that might be accurate. No idea what "19061710017BRUL" is, though. Maybe 2019-06-17 as manufacture date with the serial 10017. BRUL is a total mystery to me -- maybe a factory code or something? Anyway, not important.