Miles' DIY motor thread

[HAL9000v2.0]

Hi Hal.

It would mean substantially more machining. This way, you just need to trim the edges with a cut-off saw. If you want anything other than a parallel inner core, I think you might as well machine the whole thing from a laminated blank.

I don't get this. your core plates are all different so why it would ask more machinning? Just make this inner part of a "H" on every next part a little wider. That leads us to next question: Whay would you like parallel inner core? When you add area thet you get extra inside isn't this better then the parralell? Must draw this...to many parameters... I think I can't explain it with words...

 

[Miles]

Hi Hal.

It would mean substantially more machining. This way, you just need to trim the edges with a cut-off saw. If you want anything other than a parallel inner core, I think you might as well machine the whole thing from a laminated blank.

I don't get this. your core plates are all different

But they're not. You assemble identical plates and then trim the top and bottom to width.

There will be around 50 plates in each core.....

 

[HAL9000v2.0]

Aha... I see... and if you can cut different parts will you make it trapezoid?

 

[Miles]

Probably. I'm not sure, though.

I think it would make more sense to machine directly from a laminated block than to assemble multiple sizes of plate and then have to do post-machining, as well.

 

[liveforphysics]

Whay would you like parallel inner core? When you add area thet you get extra inside isn't this better then the parralell? Must draw this...to many parameters... I think I can't explain it with words...

What you're trying to say is that you would want the empty spaces to be parallel, which means the stator core centers must be tapered. This would allow a large improvement in available area for copper.

Hey Miles, I will make you an offer.

I know lams are almost always just stamped to make them dirt cheap to manufacture in quanity.

However, if you want to write up all the code, and specify the material and the bit and what ever else, so you could do a few hundred lams on each sheet in my CNC.

You could do rows however many poles you're doing wide, then make each one shrink a bit as it goes back along the sheet. For such a small motor, I bet just a few thin sheets of 0.5mm silicone steel would be all you need.

It's easy enough for me to put something in the machine and hit go, I would be glad to help a friend if you're interested.

 

[HAL9000v2.0]

I just wanted to offer him a laser cut but you beat me... :lol:

I will go with Mu metal for core. What do you think?

 

[liveforphysics]

If I had a 4th axis setup all ready...

We could use that excellent spiral wound thin ribbon idea to make a core. Just a smear of epoxy from a sponge or something as it's rolled would keep things insulated. You could get something like that setup yourself, make a mounting plate for the back that will bolt to my 4th axis, then I could plunge cut it from the outside to make the parallel areas for the copper, and then the slits at the top to make winding easy and to magneticly seperate it.

If I get my 4th axis setup and working before you get lam's made, this might give the best possible end product.

 

[Miles]

What you're trying to say is that you would want the empty spaces to be parallel, which means the stator core centers must be tapered. This would allow a large improvement in available area for copper.

This is something that isn't clear to me. Does the cross-sectional aspect ratio of the iron affect its saturation level?

 

[Miles]

then I could plunge cut it from the outside to make the parallel areas for the copper.....

I can't see a way to do that, on this scale?

With an even number of slots, perhaps EDM or laser?

 

[HAL9000v2.0]

That I also want to know...

Will the hole in the middle of the core help? Of course shaped like offset path of coil let say -5mm.

 

[Miles]

This is an approximation of the machined one-piece core.

 

[HAL9000v2.0]

That is the design I ment.
I this what Luke idea was? To wound and glue metal ribon and machine it?

 

[Miles]

Yes, I think so.

It was also what we were discussing, here:

I was wondering wouldnt you be able to just wrap your core material then machine it the way you want.

I did consider whether or not it would be possible with post machining but, it would be too difficult to create teeth with such a small diameter core, I think.

 

[HAL9000v2.0]

... it would be too difficult to create teeth with such a small diameter core, I think.

on even number of poles I will do it with wire cut.

 

[Miles]

I would be interested to try that.

 

[HAL9000v2.0]

Can you do the wound/epoxy stuff?

 

[Miles]

Can you do the wound/epoxy stuff?

Probably.

How much heat is generated with EDM?

 

[liveforphysics]

With the layers being insulated, and the part essentially being a giant many layer inductor, I wonder how well it would like being wire cut. It seems like it would have a good ability to absorb the current spike when the wire arcs (perhaps not cutting? cutting more slowly?), but maybe I'm just over-thinking it.

 

[liveforphysics]

Can you do the wound/epoxy stuff?

Probably.

How much heat is generated with EDM?

The HAZ is very small with flood or submersion EDM. I can't help but wonder if the wire's path wouldn't be stopped by trying to mush epoxy out of the way, or if it would block needed coolant flow to remove the kerf material particles.

 

[HAL9000v2.0]

I just spoked with guy with wire cut. He never cutted souch laminated part. another thing it is that he needs to mount third axis but this is not problem. Part can be submersed completly in fluid so cooling is not problem. this is machine

 

[Miles]

It would be great to try out this method.

Need to use high-temperature epoxy, I think.......

 

[HAL9000v2.0]

Problem is also how to hold the part. magnetic holder is out...

 

[Miles]

Maybe we should we try an experiment with a flat laminated block, first? It would be easy to create the blank and wouldn't need 3rd axis set-up etc.

 

[Miles]

However, if you want to write up all the code, and specify the material and the bit and what ever else, so you could do a few hundred lams on each sheet in my CNC.

You could do rows however many poles you're doing wide, then make each one shrink a bit as it goes back along the sheet. For such a small motor, I bet just a few thin sheets of 0.5mm silicone steel would be all you need.

It's easy enough for me to put something in the machine and hit go, I would be glad to help a friend if you're interested.

Luke,

You're very kind. I was thinking to use 0.25mm thickness material, though - that's 50 different shapes..... There would still be a limit set by the tool radius.

 

[liveforphysics]

5 axis machine would be soooo nice.

[youtube]qLJxMUw51N8[/youtube]

I'm looking at buying one of these little guys for my mill:

http://www.doughtydrive.com/products.html


Anything on a mill will always be limited by the needed diameter of the end mill, but I think a 1.5-2mm radius on the corners wouldn't be such a bad thing.