Miles' DIY motor thread

[Miles]

I did some rough CADoodling to get an idea of the performance effect it would have.

Thanks Luke,

I guess the ideal case is something like this:

 

[j3tch1u]

I'm not sure that it would matter that the rings weren't concentric? The lam material would only be 0.25mm thick.... Interesting idea, though.....

I was thinking to make a winding/mounting inner boss to match the spiral form.

that's what i was thinking. i'm no expert but maybe just compress them together really hard.

 

[Miles]

Need to get some sample material to experiment with, I think.............

 

[liveforphysics]

You can see a pair of them here doing some frilly stuff.

cool! 5-axis pineapple. Jesus (the carpenter) would be proud

ya mon, there are many ways to peel an apple. btw, you wrote "fluke" instead of "flute". still thinking about your $10,000 thermal imager?

Yep, still thinking about that thermal imager, which is sitting next to me right now. lol I have a $105,000 budget for tools this year at work, and I can't wait to get new toys. Uber fancy vibration analysis stuff, maybe more thermal toys... Mmmm!

I did some rough CADoodling to get an idea of the performance effect it would have.

Thanks Luke,

I guess the ideal case is something like this:

That does look pretty ideal Miles. I'm trying to order one of the 4th axis tables this week. If you can't find somebody better qualified and equiped, send me whatever code you want me to run, and mail me some wrapped up blanks, and I will do my best for you. At the bare minimum we can get a collection of broken endmills.

 

[Arlo1]

That scanner is brilliant! Attention all fabricators on endless-sphere: watch that video. LFP will buy one! $3000 for the scanner and then the 3D printer is magic. One each of those and a casting expert and we're in business. Far out man!

Yeh I like when it spits out the working plastic cresent wrench all as one peice!

 

[Miles]

Just worked out that I need a strip 8 meters long, to make the one-piece core.....

 

[j3tch1u]

miles, while you are at it, maybe you could also make a rhomboidal coil stator to see if it makes any difference in efficiency. would just need to angle your slot cuts in a X-like pattern.

 

[liveforphysics]

AFAIK, rhomboid stator pole shapes are used at the cost of efficiency to try and minimize torque pulses/cogging.

 

[Miles]

miles, while you are at it, maybe you could also make a rhomboidal coil stator to see if it makes any difference in efficiency. would just need to angle your slot cuts in a X-like pattern.

You would have to reduce the surface area in the gap, too, though.............thinks.............It might be a compromise worth making to get closer to sinusoidal bEMF?

 

[HAL9000v2.0]

Just an idea...if you make spiral slots ( like on ICE turbine) then overlaping can be made proportional with surface and distance so there will be no cogging...ofcourse there is problem how to wound konvex/konkave coils....

 

[Miles]

It wouldn't be any more trouble to skew the slots:

 

[HAL9000v2.0]

Yes, something like that. Now the question is; is there benefit? It will definitly reduce resistance but also it will reduce max rpm. Is that right?

 

[Miles]

Yes, something like that. Now the question is; is there benefit?

I don't think there's a simple answer...

What do you mean by "reduce resistance", Hal?
Why should the rpm go down?

This is one of the papers that I posted in the Resources thread:
http://laboratuar.kocaeli.edu.tr/tahriklab/diger/tahriklab19.10.2009_10.59.15diger.pdf

 

[Malcolm]

The copper strip on my Etek is skewed rather like that.

Just looking at that stator, I'm curious, do you gain much of an advantage from narrowing the slots at the top of the core? I guess it's to close the flux path?

 

[Miles]

Just looking at that stator, I'm curious, do you gain much of an advantage from narrowing the slots at the top of the core? I guess it's to close the flux path?

Yes, it makes a big difference. It's to increase the surface area fronting the gap. Doubling the area halves the reluctance....

 

[liveforphysics]

Doesn't the slanted slots always cost efficiency either directly through overlapping poles, or indirectly by spacing the magnet/rotor setup to be optimized with the poles skewed?

Are one of the requirements of your motor to have minimized cogging?

 

[Miles]

Are one of the requirements of your motor to have minimized cogging?

It's not something I'm that concerned about, per se. It would be more important for a low speed motor. This one will have a no load speed of circa 8000 rpm.

The most important thing seems to be to get the bEMF waveform to match the waveform that you are using to drive the motor with - then you get ripple free continuous torque........ Theoretically you could achieve this with block commutation, if it changed instantaneously....

 

[RoughRider]

as long as i know cogging cost efficiency, too.

right?

 

[Miles]

I'm not sure that there is a direct connection between cogging and efficiency.......?

 

[Thud]

Not sure this answers any questions Miles,
courtesy of "The Other Dave" on rc groups:

Actually the cogging effect (technically known as detent)
is due to pole construction. The windings have no effect.

I used to design motor drives for large (300+ ft-lb torque)
DC brushless motors.

The cog or detent is the point at which the center of the magnets
perfectly line up with the ideal magnetic path through
the poles.

This detent can be sharp or soft depending on pole constrution.

A straght pole (no twist) along it's length will have
a sharp detent.

A twisted pole will have less. This is because for different
points along the rotor the magnetic "perfect" point is at a different
point in the rotation of the rotor.

Pole shape also has a factor in the detent amount as does air gap between the rotor and poles.

->The drawback is loss of torque, high detent motors have higher
peak torque and better efficency than low detent motors.<-

The positives are smoother operation.

hope that gives some insight to any one following along....Don't know if there is any impeirical data available for compairison.

 

[bigmoose]

Perhaps we should try to recruit the "Other Dave" from the rc groups? With all the motor projects going on, we could use a good motor design consultant!

 

[Miles]

Thanks Todd,

There are so many different factors, it's difficult to link everything up....

I've always thought that cogging was a bit like gaining and losing height when going up and down a hill - the net energy being nil. But I guess that doesn't take everything into account.....

->The drawback is loss of torque, high detent motors have higher
peak torque and better efficency than low detent motors.<-

I'm not sure why this should be the case, though?

 

[Miles]

More on this here:
http://www.trilogysystems.com/pdf/Anti-Cog_Application_Note.pdf

http://www.esmats.eu/esmatspapers/pastpapers/pdfs/2003/marks.pdf

 

[Thud]

Speculating as allways,
I percieve "cogging" as a gage of potential.
Tight gaps & strong magnets are keys to providing tourque. I also would think straight teeth, aligned/true paralelling with the magnets would provide optimum pulling power.(angularly speaking)

I was into "stock" motor 12th scale RC carpet racing. The quick way to find a good motor was to check the cog effect. (more the better).

 

[drewjet]

What about taking an existing stator that is near the specs you are looking for and remaching it.

The square one os from my dishwasher and the round is from a Currie gear motor