ES Motor Project?

[Miles]

If we could decide on a specification that would satisfy enough people on ES, we could design our own motor. It might be possible to have the laminations made in sufficient quantity to justify the cost of tooling for stamping.

[Edit: Split from: viewtopic.php?f=30&t=40074 ]

[Jeremy Harris]

If we could decide on a specification that would satisfy enough people on ES, we could design our own motor. It might be possible to have the laminations made in sufficient quantity to justify the cost of tooling for stamping.

Are they best stamped, do you think, or would laser cutting be an option?

Given the mix of skill on ES it may well be we could put together a motor that meets the needs of a lot of forum members, perhaps farming out the CNC work and lamination cutting and putting together a kit of parts for a wind-your-own motor. The first challenge would be deciding on a size and design. I'd like to see an out runner that's about 100 to 120 mm stator diameter, with a stator length of around 100 mm or more. A 12 slot, 14 magnet set up would be fine for me, and if we could choose dimensions that suit an off-the-shelf neo magnet so much the better.

Has anyone tried contacting the power croco/power ditto people to see if they have parts available?

[Miles]

Are they best stamped, do you think, or would laser cutting be an option?

I was only thinking of the price. I think stamping is cheaper beyond a certain quantity. In some ways, laser is better - no tooling costs, no need for post annealing, possibility to tweak the design, etc. I guess one could start with laser cutting and then move to stamping if/when the demand justified it.

[Thud]

For your pleasure,
Here is a dxf of some stator configs utilizing standard NEO's & DOM steel for the flux cans.

OUTRUNNER MASTER.dxf

3" 4" 4.5" 5" & 6" motor lay-outs.

(249.2 KiB) Downloaded 49 times

I guess i'll put this here at the risk of duplication when I actually make time to finish the motor prototyping project.

I bought a stock of .008"t shim stock to make laminations & planned on insulateing them with a thin nomex sheet between each lam. The steel is (1080 low carbon) no where near real lamination stock but the only thing in my price & availability range.

After considering my options, I decided photo etching was the only reasonable way a guy in his basment could pull off making a batch of perfect laminations. I Have the screenprint that nests a 3" & a 5" motor in a single pc of stock.
but my shim stock was rolled...I cut it to size on a paper cutter but it has a permanent "cup" in the stock that needs to be relaxed before I can start the printing proccess. (i'll clamp the whole stack between 2 plates & take it to red heat & let it cool slowly to unstress the steel)


Around the new year I purchased the latest Pulsed EDM board from Ben Flemming. Once I have EDM capability it will be a walk in the park to cut proper punch dies for MASS PRODUCTION

Any one have any idea on what it would take to make a nice Laser for a bench top cnc ?(Co2 60+ watts?) I searched a little one night, but A clear example W/enough documentaion for me to plagerize didn't apeare....havn't got back to that one yet.
let the conversation continue.

[Miles]

Thud's layouts (above DXF):

3" OD: 2.54"∅ stator 12t 7p [DLRK]

4" OD: 24t 14p [DLRK]

4.5" OD: 4.46"∅ stator 18t 12p [ABC]

5" OD: 24t 14p [DLRK]

6" OD: 24t 16p [ABC]

[mdd0127]

You are awesome Thud!

If you start making motor kits and the guys in Russia get their full sine controller and computer/meter off the ground, I'll probably get excited about the future of e-bikes again.

A couple of quick suggestions for ultimate motors would be liquid coolant channels in the stator, and a 5mgt driver pulley integrated into the endcap. Also, I like the idea of developing a larger diameter outrunner like the 5 or 6 inch size. It could be really thin in cross section for people with lower power needs and thicker for freaks like me. There are plenty of options already out there for smaller diameters. I think that the increased torque capability and narrower cross section would fit most builds better.

I wish there was some way I could be of more help to this project!

[Miles]

I'd like to see an out runner that's about 100 to 120 mm stator diameter, with a stator length of around 100 mm or more. A 12 slot, 14 magnet set up would be fine for me, and if we could choose dimensions that suit an off-the-shelf neo magnet so much the better.

How about making the stator diameter 112mm?

I think an outrunner is still best at that size. At 150mm and above, it seems less clear cut.

Design for 2 stack lengths? 50mm and 100mm?

I'm keen on 18t 16p. It has a high winding factor and (like 9t 8p) the BEMF is very close to sinusoidal.

[Miles]

Ref: Winding Factors and Joule Losses of Permanent Magnet Machines with Concentrated Windings
Freddy Magnussen and Chandur Sadarangani, Royal Institute of Technology, Department of Electrical Engineering,
Stockholm, Sweden.

[Miles]

What is the range of nominal speeds we need to cover?

[liveforphysics]

I'm keen on 18t 16p. It has a high winding factor and (like 9t 8p) the BEMF is very close to sinusoidal.

You loose the ability to use Sevcons and Kellys when you increase the pole count. Unfortunately, low pole count motors have additional iron weight and torque ripple.

[Thud]

For design considerations, I am letting "on the shelf" magnets & available tube diameters dictate the motor design.

A 112mm (ish) stator fits lovely into a 5" shell with 3mm thick magnets.

At 16 pole pr's we would have to use 10mm wide magnets. (but an 18t 8p fits with gaps) we can increase copper & concentraight buy stepping to 21t 14p here is a screen shot side by side....along with my 24t 16p


Take your pick.....

Now for some question of my own:
given the same direction of the turns in an ABC wound motor.....will inductance be greater when compaired to another configuration? the 18t motors are typicly AaABbBCcC wound & it looks like an inductance crippiling effect we have observed on the colossis motors.
fizzin to get my learn on....

[Farfle]

Hmm, if we make a monster low kv outrunner could it do 20 kw on a single reduction?? If so, I would gladly donate my two lathes and mill. If you need anything made; shafts housings, bells etc...
Just send me cad files.

[Miles]

Hmm, if we make a monster low kv outrunner could it do 20 kw on a single reduction??

The size we're considering at the moment wouldn't be able to achieve that.

[mdd0127]

10kw is enough to scare most people, even experienced riders on a sub 100lb machine so I think that the sizes you're considering are just perfect.

[Jeremy Harris]

I agree, start with something that's around that power level max, with a known to be good pole count/slot arrangement and that uses stock materials. The 5" tube, 3mm magnets, 112 mm stator sounds a good starting point to me. Motor length could be adjusted for power easily enough - longer would be better in terms of electrical efficiency, from proportionally less end turn loss, but at the cost of poorer cooling. Still, the power freaks could always look at liquid cooling the bearing housing themselves, by making up a custom carrier.

That's one advantage of going down the "sort some parts as a collective" route, people will be free to use the hard to source parts (stator lams, maybe the machined cans unless they happen to have a biggish lathe) as-is but could make up their own shafts, bearing carriers etc to suit their own project. Even a small Chinese bench lathe could turn out a bearing carrier without too much trouble.

[John in CR]

I'm keen on 18t 16p. It has a high winding factor and (like 9t 8p) the BEMF is very close to sinusoidal.

I like the 18t, but I wonder about the 16p. The 93% peak efficiency hubbies I have are a 150mm X 50mm long stator with 18t, and the windings are beautiful with very little end copper. They have 20 magnets though and the magnets are curved. How is the 16p and advantage over 20? Does the better shaped BEMF make it easier to drive?

John

[Miles]

How is the 16p and advantage over 20?

Reduced iron losses. At the speed of a hub motor, it's not that important.

That was the reason for this question:

What is the range of nominal speeds we need to cover?

[nieles]

what about your axial flux motor miles, did you abandon that project?

[Miles]

what about your axial flux motor miles, did you abandon that project?

No, I'm still working on it. It would be too expensive/difficult for a project like this, though, and it's unproven.

[Farfle]

Doh, I punched in 20kw without stating it as a burst rating, 10kw cont would be awesome on a single reduction...

[John in CR]

How is the 16p and advantage over 20?

Reduced iron losses. At the speed of a hub motor, it's not that important.

That was the reason for this question:

What is the range of nominal speeds we need to cover?

With the approx diameter 5" motor you wanted I'd want to go with the 100mm stator length. Could handle a 10-12kw input in the 2000-2500rpm range?

BTW, did you ride someone's high powered rig and now you need one too or was I wrong in thinking you were after the lower power end of the spectrum?

John

[Miles]

Could handle a 10-12kw input in the 2000-2500rpm range?

That's around 45Nm of torque. I'd guess not.

BTW, did you ride someone's high powered rig and now you need one too or was I wrong in thinking you were after the lower power end of the spectrum?

It would be good to have something I could run at around 1200 rpm. I'd only want a 50mm stator length.

[Miles]

At 16 pole pr's we would have to use 10mm wide magnets. (but an 18t 8p fits with gaps) we can increase copper & concentraight buy stepping to 21t 14p here is a screen shot side by side....along with my 24t 16p


Take your pick.....

21t 14p would have greater cogging. If we want to run this at relatively slow speeds...

24t 14p has quite a low winding factor (0.76).

My vote is with 18t 16p.

BTW Thud, you wrote 16 pole pairs. I just assumed you doubled up on the magnets for curvature?

[Thud]

I did not, but tha is an option i supose....I been using nomenclature to fit my moods lately... so I have it drawn a little wrong.

I'll re-sketch it for the 8 pole pr count & adjust the tooth size acordingly for a better "looks about right" design.

back in a bit

[Miles]

I've done a parametric core model. So, we can tweak the parameters to however we want them...