MULE1.2 Axial flux test motor/bicycle specific

[rhitee05]

I like the overlapping coils method as well, but It doesn't seem to be compatible with cores, and it looks like we're going with cores on this one, which is probably a good thing for this first motor.

You could do this with cores, but it'd be a PITA to wind. That's how induction motors and some other types are wound. Basically, you end up with parts of 2 or more coils in each slot.

This would be true only when using core's right? You're saying you want the core OD (coil ID) to match the magnet size. Everything for coreless seems to imply to make the magnets match or better, overlap the coil on the ID and OD.

Actually, the opposite - I was talking about the coreless case. Visualize the flux as a cylinder the same diameter as the magnets passing through the stator. You want the ID of your coil to be the same size or larger than this cylinder. If you make it smaller, than the turns in the center of the coil are only seeing a portion of the total flux and will be less effective.

If you're using a core, you should be able to make them smaller because you've got iron to help direct the flux so it goes through the center of the coil.

 

[Miles]

I like the overlapping coils method as well, but It doesn't seem to be compatible with cores, and it looks like we're going with cores on this one, which is probably a good thing for this first motor.

So, the choices for a cored 2 rotor design are:

Double layer aligned pairs with common core.

Double layer aligned pairs with individual cores.

Double layer overlapped pairs with individual cores.

Single layer.

Or?

 

[rhitee05]

Or?

Just to confuse things more, one other winding design option.

View attachment TORUS Motor.JPG

The literature calls this a TORUS-type motor. The core is doughnut-shaped with the windings wrapped around it toroid-style. Look carefully at the polarity of the magnets and the flux path - they're arranged in opposing N-N, S-S pairs. The flux enters the core, passes circumferentially through the coil(s), then exits and returns to the next pole. Each magnet rotor has a separate magnetic circuit, instead of the flux passing straight through from one to the other. This can be built with the windings in slots, or just wrapped onto the surface of the core.

I don't know if there's anything intrinsically better about this setup than the through-flux design, just a different construction method. If you did the non-slotted version, it would be fairly easy to make the core since it's just a ring of iron (preferably laminated). You could also make it water-cooled pretty easily by running water pipes through or next to the core.

 

[Miles]

This really is a CADoodle...

It's quite tricky to work out a design with the core common to both coil layers..... With separate cores for each coil, it would be dead simple..... I guess you could fill in the structure with resin, once its all working...

 

[Thud]

Thanks for the illistration Miles,
I been trying to catch up again, a short nap & you get passed right by

I was having trouble visualising Erics lapped coils & contemplating the senario with flat windings.
I see now they are slightly off set to attach the centers out of the flux path.
Is that correct?

would this require clocking the rotors the same few degrees for proper alignmet also?

Any reason the cores couldn't be seperat entities that could be joined upon final assembly?

If not just resing against each other may get the flux manipulation we desire?

I know the Launchpoint & cisro motors are heavily lap wound with Jillions of stator teeth. but again we are looking at coreless motors, I speculate the extream low end tourqe is not where those motors are making their efficancy ratings.
I am looking for Copper tape & the EMI tape looks closer to our requirments. The heavyest gage I find on the stained glass sites is 1.5 mils =.037mm. that seems thin to me. Thoughts?

 

[Miles]

I was having trouble visualising Erics lapped coils & contemplating the senario with flat windings.
I see now they are slightly off set to attach the centers out of the flux path.
Is that correct?

It shows aligned layers with a common core. There is a gap between the coils.

Below is Eric's design with the offset layers (but with cores). Do these cores make sense?

 

[Miles]

Any reason the cores couldn't be seperat entities that could be joined upon final assembly?

If not just resing against each other may get the flux manipulation we desire?

Good point. I don't know. It would certainly make a huge difference to manufacturability.

 

[spinningmagnets]

Here's a simple AF from a German site. "looks like" iron plates behind the magnets, aluminum cylindrical body. The stator spins, so maybe its brushed? no reason the coils can't be stationary and the magnets rotate (with optical synchronization). Here's a translation:

The disk motor was from before the time of the rare earth magnets, which are well suited for the use of Alnico magnets, because these entities must have the stomach for a long energy optimization (keyword bar magnets). Besides the very good dynamic properties, the design is short in length. The disk rotor drives available today have rare earth magnets used in the design principle of the permanent magnet dc machine and synchronous machine.

http://www.energie.ch/at/sonder/

 

[rhitee05]

Miles, nice CAD work. You've captured the design I was suggesting quite well.

For a DIY iron-core motor, I think your first drawing with non-overlapping coils would probably be easier to build. You could do the overlapping coils, but it would be more difficult to construct so I don't know if it'd be worth it. If you went ironless, which is what I believe Luke is interested in, I think overlapping is the way to go.

I think you could make the cores separately, wind them, and assemble the whole lot together at the end. The suggestion I gave Thud in a different thread is to use ferrite powder with epoxy or some sort of resin in a very dry mixture and mold that into cores. That would give the non-conductive properties you'd like to have (low eddy losses) but be much easier than laminations. For the simple version, you could just make a core that fits into the center of the winding. For somewhat better performance, you could mold teeth onto the front of the cores as well to give the usual slotted structure. But, if you mold them one core at a time you could still slide the coil on whole without the hassle of winding between teeth. I found a picture that illustrates this:



It would take some work to figure out how to make all that and make it structurally sound, but should be much easier than any kind of laminated core.

If I have a little time later, I'll see if I can work up a few FEM models to see how those various configurations would fare.

 

[spinningmagnets]

Another AF to add to the discussion: "The Apex motor is an axial-flux permanent-magnet motor, thus similar in design to Wavecrest motors, Ultra Motors and those designed by Cedric Lynch (Agni)"

 

[MitchJi]

Hi Todd,

2) Size- the outrunner I selected was size specific to cram into the rear triangle of a small bicycle frame. The
larger motors that have enough power create other issues, width moves most out of the crank circle.
I think thinner is a winner if we are not going to settle on a hub motor.

One question. D reported that an Astro 3220 (which has more than enough power for most builds) will fit (based on measurements from Matt) between the cranks on the Stinky.

A lower kv would be a huge advantage but I don't see a big advantage to thinner for between the cranks. OTOH if it's very thin for mounting near the rear wheel something like Mud's build where a much thinner motor (one inch thick or less would be great) would be really nice:

Here is a Board room table I installed in Maryland last spring.

Beautiful job on the table 8)

 

[Miles]

http://www.apexdrivelabs.com/pdf/Apex_Motors_and_Generators_Oct_2008.pdf Good to see torque density figures.....

 

[John in CR]

http://www.anythinginstainedglass.com/metals/foil.html

This is just the first placed I found, but it seems they have thickness and width that would be appropriate for our needs. I've got a mega stash of kapton tape in various widths all ready.

The problem I see always comes back to what to do with that pesky center tap... If it just crosses over all the coils on it's way out, it's going to be inductively cooked, as well as lower efficiency. I know there has to be some clever solution to this, I just haven't quite thought of it yet... Maybe if I were to start to try an explain a non-solution in words, a real solution would jump into Jonathan's head, and then we would have something.

Ok, bedtime for me.

Questions related to the inductive cooking of the center tap leading out:
1. Would a thick round wire be subject to less heating than the flat ribbon copper?
2. If you could drill a hole through the side of the coil without creating a short, and pass the center tap through that hole dead center on the ribbon, would it be subject to less heating? If that works, then instead of the hole, what about stacking two 1/2 width ribbon rolls into a parallel sandwich with the center taps running out through the middle of the sandwich?
3. Do turns of a coil have to complete each turn? OR could you loop wire around the edges of the leaves on a 4 leaf clover to create a 4 turn coil without the wire ever physically crossing itself?
4. Do all of the high efficiency production motors that we're trying to improve upon have overlapping windings in their stators? How many turns in each coil are we looking at anyway?

Feel free to flame any stupid questions from me. I'll be unfazed and it might be a good stress outlet for you guys if you need one.

John

 

[Thud]

Well, again a lot of ground to cover,

If I have the facts correct, as described, there is an issue with the offseting of the coils per Erics suggestion, As drawn we are going to lose the flux density of the coils if we can not "pinch" (as luke described) the coils generated flux as tightly as possible with the Magnetic rotors. We have come nearly full circle from the original bobbin coil concept. Again we have 2 coils w/cores open on an end.(maybe not as open but I think you see the parity)
The original sketch looked better in that regard as the offseting is only a degree or two to make the center tap connection.
I await Lukes Take on this concept.

rhitee:
Study the 2d skecth on page 5- you see the .5"dimensioninthe center, this is my vision of a final molded stator thickness. all the molded cores(should that be the final direction) could be sub assemblies. & even modular depending on how complex we were to take the concept. We are thinking alike in that regard. I need to get by my mothers house & see if all the powderd iron cores are still in the shed(dads old lab) I can crush them in a ball mill & recast them into any shape desired. it seems this would be the ideal stuff for our cores.

I never expectded this thread to hit 8 pages before we made 1 shaft turn.....must be a guy thing

 

[RWP]

I am delighted and charmed by all this creativity and brain power. You guys really rock

I have done a good bit of leaded glass in my day and this included a bit of work with copper foil.
I have no comment on the science and nature of the discussion but I have questions about this specific material as something suitable for the proposed purposes.

The foil, as made for the leaded glass copper foil method, is not a precision material and I question, wonder about, is dimensional stability - any dimension - in a specific roll or specific batch. Or differences between manufacturers rolls and or batches.

If I understand correctly when one winds with copper wire each strand of the wire must be insulated. Copper foil has an adhesive that keeps the coil together and allows the foil to adhere to glass. Is this adhesive up to being an insulator and is it up to the heat it will be subjected to in a motor? Perhaps I misunderstand here and another insulating material will be wound along with the foil.

I hope this adds value to the conversation.

Roy

 

[liveforphysics]

Is this adhesive up to being an insulator and is it up to the heat it will be subjected to in a motor? Perhaps I misunderstand here and another insulating material will be wound along with the foil.

Roy

A layer of Kapton film/tape will be layed upon the strip before winding. It can take absurd temperatures, and insulates to the tune of 7,700v/mil. It's like God's gift to insulating wires if your budget it big enough.

 

[Miles]

If we are going to have the two layers aligned, then could the coil pairs be connected in series? If they were joined together with a bridge between what was the centre taps, both taps would then be on the outside of the coils. It would mean that they had to be wound in-situ, though.

 

[Thud]

Well this is certainly been a learning experiance for me. I want to thank every one for the input thus far.
I woke up early this morning to model the current version as I intend to build it.
here are some screen shots:

left to right part call out
brown-steel flux ring machined with a relief for magnet positioning/bonding & retention
green- posts are stand-offs for clam shelling the 2nd flux ring/magnet assembly about the stator
light blue- hub is a mount for the rotor to shaft
tourquose-main hub holds bearings & is the mounting point for the Stator(not drawn) & the motor stand
white-The stand is for bench running-you can imagine the mounting varients of this system.
The width can be reduced sbstantialy by mounting the motor via the stator & using 2 traditional rotors & dropping the clam shell assembly.

here are some perspective view of the assembly(stator not shown)

stator is in place on the left assembly.

I would like to stay focused on this topoligy for the time being.To elliminate confusing "what if" senarios.
Unless there is a glaring flaw with the design, this is the mechanicle layout for the this phase of the project.
any great ideas & additional thoughts would be better referanced in the "Axial flux motor discusion" thread.

The magnets will be discs for 2 reasons. #1they are available #2 I am financing this project (its all I can afford)
The doors are wide open regarding how we will "plumb" the stator.
the stator will ultimatly be a re-enforced resin casting from custom molds. I already have plans for the first unit But am really looking foward to see the gains we can achieve with everyones input.
If anyone would like the current DXF of the motor I will e-mail it. or can i zip it & post it on the forum?

lets see what the day brings. Thanks Again guys. Todd
(edit: just fixing the spelling goofs for the 50th time)
for referanc the stator on this motor is 4.625" dia. 117mm
over all dia of rotors 142mm

 

[liveforphysics]

If we are going to have the two layers aligned, then could the coil pairs be connected in series? If they were joined together with a bridge between what was the centre taps, both taps would then be on the outside of the coils. It would mean that they had to be wound in-situ, though.

I was thinking of something like that. Spot weld the copper ribbon onto a piece of solid copper wire rod at the full thickness of the stator coil pair. Then coil one coil around it one direction, and smear some epoxy over that coil to give it structure, then spot weld on the ribon for the other side of the copper dowel, and wind it to completion.

I supose it would be just as good (perhaps better) to simply make a double bobbin, and start from the center of the strip of copper, fold the strip as needed to gain alignment for each coil, and wind the coils all from a single strip.

This whole double coil thing is an extra complexity that I wish we didn't have to deal with, but ehh... I guess it's better than the alternative.


-Luke

 

[liveforphysics]

Thud- Me likey everything I see there! Mucho likey!


I will buy the magnets and have them sent to you. I will buy those wicked strong pie-shaped magnets from the wind-generator place. I think the set to make an 8" ring is the smallest they have, they are 1/2" thickness, and n40 strength if I remember correctly.

Anyone wana chip in on these magnets with me? It's fine if not, Im buying them anyways, but if anyone feels like helping out:

Liveforphysics (at) yahoo (dot) com

is my paypal address

 

[Miles]

This is what it could look like:

Yes, it could be a continuous strip, folded at the bridge, as Luke suggested (more difficult to model that.. ).

Edit: NB See corrected version of this illustration here: http://www.endless-sphere.com/forums/viewtopic.php?p=214408#p214408

 

[Thud]

Luke,
Cool, I will draw it up & retitle that motor specificly.
The stator discusion will now be primary.

 

[Thud]

Miles,
I don't think that will work in that configuration. By changing direction at the 1/2 way point, we are reversing the polarity of the EM feild thus canceling each other. the right hand rule is in effect if i understand it correctly.
I wish it were that easy though.

 

[Miles]

If the spirals were wound in opposite directions?

I need a coffee

 

[Thud]

hehe that would do it.....! we could just slit a full width pc of tape & insulate it. too simple!
good work.