custom hub motors

hillzofvalp

100 kW
Joined
Dec 25, 2010
Messages
1,887
Location
Somewhere over the rainbow, Canada
I'm dealing with the stupidity in screwing over my 9c with less than 100 miles use. You can see my thread on the mess of a motor. So I'm thinking I might buy a replacement stator, and I probably will. I just thought though how cool it would be to build up a stator that will fit the 9c magnet ring, maybe some custom side covers, and maybe a performance boost... If I get a replacement stator out of another 9c, I'll still have all the parts left from the donor.. as well as the dead stator

I work in a cnc lab at my school and have some downtime to work on my own projects. We have a water jet I could probably use to make the poles, as well.

As of this point, I know very little about motor design, and would like to hear your comments on what could possibly be changed about the 9c stator, etc. I know the 9c has magnets offset a bit.. but I wonder if wider poles could be put in or if maybe longer magnets could be installed.

Maybe there's a performance enhancement really obvious that's missing, maybe something nine continent could not manage to do to do on mold-formed mass-produced motors.. or something that was not economically feasible
 
If you want to turn it into a real performer simply rewind the stator with 3 or 4 turns using thicker magnet wire to get better copper fill. You'd need to use it as mid drive only, though a 4 turn might work in a 16" wheel. Gearing down a much higher Kv motor in that manner will make it capable of much higher power, and the higher rpm will help it dissipate more heat too. The winding goal should be to end up with net copper fill equal to or better than the 8x8 9C, which is their most efficient model.
 
the stock motor is pretty efficient, there's not a whole lot that can be gained for the work it will take, but thinner stator plates could be used. I'd leave the magnets alone, unless you have the skill to build the tools it will take to keep those magnets aligned perfectly as the glue drys. If you want the same speed motor, you keep the number of turns around the poles. But you can vary the strand count. the fewer the better, but you want the highest copper fill, and in some cases that means higher strand count depending on the pole's void space shape. that takes playing with a calculator to discover. I'm not sure if anyone has experimented with multi-gauge/multi-strand motor wrapping, but in theory, you could get higher copper fill and more torque. As for side covers, there's a million good reasons to make your own. I would If I had a lathe and a CNC. I'd also build it to take larger bearings, and Asymmetrical ribs on the in side to kill the harmonic resonances that cause noise.
 
hillzofvalp said:
This is extreme, but would it be effective to machine ($$) fat strips of copper to perfectly fit the poles, and then solder each pole by a thick connecting wire or copper screw terminal?


Yes. But they are called slots where you put copper, and teeth that you wrap copper around. Poles are your magnets.

If you do the solid slot approach, you can make a really efficient motor (you can search google for various folks who have made solid slot motors), however, your controller gets M.C.-Hammered with the pants and the whole bit, so you end up needing to run huge inductors inline with your motor (which makes you end up back at normal efficiency), and your system voltage for a solid slot motor is going to need to be like <12v, so you're going to need to make the wiring all VERY large. However, it can be done, and you can end up with a more efficient system, but it's going to add a few pounds of weight in cabling, inductors, and require a custom controller.
 
What will thinner stator plates do? Why can't they just be on solid unit?

12V.. wow... That would make charging so much easier (a single server supply instead of 4-5 in series).

I think making new covers will be my new project.. cause I F'ed up these ones being a jackass. Couldn't wait until I had time in the lab to do it on mill instead of by hand. bigger bearings.. like that idea. a motor that doesn't require routing wires through axle would be nice.. maybe by using multiple bearings.

Do you think I can take the width of the teeth higher, and then ADD magnets on onto the ends of the ones already in place?

Efficiency gains are great, but modifying this motor to handle significantly more abuse would be nice.
(you have to remember I just spent 5 hours + $150 to build this wheel, and I would hate to relace it to a better motor)


edit : first googled page and I love the solid slot idea! So easy to mill out...
So: Extremely large bearing interface to accept 4 gauge wires... and a controller that can handle 300A bursts.
 
hillzofvalp said:
What will thinner stator plates do? Why can't they just be on solid unit?


Whoa. A solid stator, or even just shorted lams makes it turn like it's packed with peanut butter. You really shouldn't be trying to build a motor, or even thinking about motor building until you can answer that question inside and out, backwards and forwards, and then look at a no-load plot and understand why making them thinner is purely an exercise in busy-work for a motor in the direct drive RPM ranges.

Must read Mile's entire section on motor theory. Then work on understanding what you read. A normal person is going to take a few months to actually let that sift and sink-in to the point they have a clue how they work. Years of motor/flux study after that could lead you to a point you could make some motor designs that are much worse than the worst Chinese hubmotors available. My office-mate has been doing motor design for 10 years now, spent 11 years in college studying it, has hundred-thousand-dollar software to help him model effects, and spends weeks designing a single lamination design, then has them laser cut (because you can't mill lams that have a proper shape), makes proto-types, tests, and goes back to the drawing board.

Make side covers with bearings large enough to pass the wires through. That will very slightly decrease efficiency (larger bearings have a tiny bit more drag), but with large enough wires you should be able to make up that loss and come out even again.
 
THANK YOU! YOU SOLVED MY MOTOR PROBLEM! When the motor scraped the debris... the laminations shorted! ah ha! there's a huge gash in them.

http://endless-sphere.com/forums/viewtopic.php?f=3&t=33403&start=45

How involved do you think going to an aluminum slot design be? What about pros and cons for going to a copper slot design instead? Are there any losses in torque from using aluminum in this case, or is it made up by the increased capacity to conduct current?
 
hillzofvalp said:
In other words, the inner race is so big you can mill out a section to route wires through? or are the balls in the bearing stationary and you put wires between balls?

The former. John in CR has done this, IIRC.
 
Come on AW, I wanted to see him route the wires through the spaces between the balls in the bearings. :lol: My manual labor job suggestion in the college dropout thread was spot on.
 
hillzofvalp said:
link me John.

That was your thread, but my attitude is that you need a healthy dose of callus creating, sweat inducing work over Christmas and summer breaks. You'll appreciate being in school, and while you're working like that your mind can often be elsewhere doing something productive, such as analyzing your motor problem from all angles and not just typing out every question that pops into your head.
 
Back
Top