Motors! How do we mod them?!

[Kodin]

So I've been following Justin's definitive cooling thread for a while and it's been giving me more and more questions as time goes on.

First off, I was surprised to learn that stronger magnets may increase a motor's torque, but may decrease it's efficiency due to losses elsewhere. Questions I've had recently include the following:

For an outrunner hub-motor, does the ring the magnets are mounted on contain iron or is it simply aluminum? What is the purpose of the iron? To focus the magnetic field inward? Magnetic shielding? Something else?

Are there better options than iron for this purpose? Maybe copper? I'm thinking along the lines of thermal transfer properties, though magnetic efficiency would also be good.

Are there better options for stator core than thin-lamination iron stacks? So far I see great improvements in cogging going from a 0.50mm stator slice size to 0.35mm; would going to 0.20mm be advantageous or is it better to go to some sort of insulated composite like a nanoparticle-saturated epoxy formulation?

I'm mostly curious about what options are available to improve various aspects of a "standard" size hubmotor. (LeafBike MXUS, QS, etc...) Most use common parts like magnet ring diameter, side covers, etc, so identifying items that could be replaced for better performance or at least different tuning would be advantageous across the board. If you have input, I'd love to hear it.

 

[amberwolf]

Have you read any of Miles' threads and posts about motors and motor design? If not, you might want to look thru his posts for those, as there is a lot of good info in there that will answer some of your questions (and/or others you haven't asked yet). I don't have a link to specific threads, but if you look up his profile and search his posts or just list them and glance thru, it might be easy enough to find.

 

[spinningmagnets]

On outrunners from Hobby King, yes the outer shell is steel The magnets are glued to the inside). On outrunners that are designed to work on RC boats, the baseplate can easily be machined to be very thick aluminum, and have a water-channel running through it.

Due to the typically high RPM's of outrunners, they have responded well to air-cooling by an impellor.

The Joby motor used 0.20mm laminations, so it is possible, but...more expensive.

 

[cycborg]

For an outrunner hub-motor, does the ring the magnets are mounted on contain iron or is it simply aluminum? What is the purpose of the iron? To focus the magnetic field inward? Magnetic shielding? Something else?

Are there better options than iron for this purpose? Maybe copper? I'm thinking along the lines of thermal transfer properties, though magnetic efficiency would also be good.

The iron is there to constrain the magnetic field. You want the field to be concentrated in the gap between the magnets and the stator pole faces. Without a ferrous backing material, the magnetic field on the outside of the magnets would mostly be lost to the environment; the backing creates a magnetic path so that the field is directed tangentially from the N pole of one magnet to the S pole of neighboring magnets and vice versa. Copper, of course, does not suit this purpose at all, being non-ferrous.

Are there better options for stator core than thin-lamination iron stacks? So far I see great improvements in cogging going from a 0.50mm stator slice size to 0.35mm; would going to 0.20mm be advantageous or is it better to go to some sort of insulated composite like a nanoparticle-saturated epoxy formulation?

Thinner laminations are better for minimizing eddy currents, but the problem is that as laminations get thinner, the proportion of ferrous material in the stator decreases, and the proportion of non-ferrous coating between laminations increases. The result is that for a given stator volume, you reach magnetic saturation at a lower magnetic field strength, and consequently the motor's maximum torque is decreased. That's aside from the manufacturability issues of thinner laminations.

A composite is an interesting idea, although again you'd want to keep the ferrous fraction as high as possible, and in addition you have to consider strength and brittleness since torque is transmitted through the stator teeth. So instead of nanoparticles, maybe a steel fiber mesh?

 

[speedmd]

It would be interesting if a significant portion of the back iron could be moved away from a thinner magnet ring at higher RPM's on a out runner. Centrifugally jacked?

 

[macribs]

How do the motor designers calculate the needed back iron? There have been a few here that has shaved off the flanges between the spokes and also shaved the back iron to loose weight.

As it seems those mods work great to reduce weight performance wise we don't know really because AFAIK no one that has done the mod has had motor on dyno before and after shaving.

Is it safe to assume the back iron rings are over engineered on the most common powerful DD hub motors?

There was some talk about D shaped magnets for the mxus I think. What is the benefits of D shape magnets over regular flat ones?

Are there really any other viable ways to mod our runners beside cooling, over volting and weight s(h)avings?[strike]I don't think re wind of motor is doable for many of us, nor is required as we can easily order the correct turn motor for our usage .. .[/strike]

 

[John in CR]

Rewinding a motor doesn't accomplish anything that a change in voltage and current can't do...unless you're talking about fixing the loose windings of hair thin copper commonly used on cheap Chinese RC outrunners.

 

[bobc]

Somebody should rewind their motor with pure silver to reduce I2R losses.
'fraid it won't be me..
Generally a good rewind can improve copper density in the stator slots and thus reduce resistance. Keeping the windings cool also helps because of the temperature coefficient of resistance (copper is about 400ppm/degree - 100degrees is +40%).
The multistrand thin wire the outrunners generally come with mean that a lot of the windings cross section is varnish; you can make a significant improvement to series resistance by rewinding.
This actually makes you a more powerful motor. If you think motor excess temperature is created by the I2R dissipation in the windings, the same temperature limit would be generated by more I if R were reduced. Torque (and therefore power) is proportional to I.

 

[cycborg]

liveforphysics did a lot of trial and error rewinding his legendary flaming dicks motor. Bottom line was he ended up with only a couple percent better copper fill than stock.