MULE1.2 Axial flux test motor/bicycle specific

[Miles]

It's a limitation of the topology...

Can we agree on: "it's a compromise worth making"?

 

[Thud]

So their combination seems to be a few magnetic poles & roughly, a bazillion stator teeth & comutation points.
We know their desing is working very well. What about a brushless version Mars I would like to see its guts.

Back to our twin rotor design-

What design parameter is going to yeild greator efficancys:
1) As many stator teeth as we can jamb into it ? (may have to limit turns of copper to make 15teeth)

2) Just enough stator teeth to perform the task & maximise the turns.

I am still of the opinion cores are to our benifit at this scale. The coging losses are at the top end of the rpm range where they wont be in play for us most of the time.

 

[Miles]

The coging losses are at the top end of the rpm range where they wont be in play for us most of the time.

You mean iron losses?

 

[Thud]

yes Miles, still a bit blurry after last night.

 

[Miles]

I am still of the opinion cores are to our benifit at this scale.

I vote for a twin rotor with back-to-back rhomboid/mandorla shaped coiled strip cores & windings.

 

[TylerDurden]

Sure, you can cantilever it from one side. I still don't like like it....

Gotta come up with something better then
Get to CADoodling

Why not flip the stator plates to the basket?

 

[liveforphysics]

Sure, you can cantilever it from one side. I still don't like like it....

Gotta come up with something better then
Get to CADoodling

Why not flip the stator plates to the basket?

It would take me far too long to re-draw in sketchup. That should be reason enough.

Now that you mention it though, it would certianly make connecting the coils on the outside much easier, as well as reduce polar moment of the rotor, and Miles could quit complaining about not having something big and stationary to hug on the running motor.

 

[Miles]

Then we're back to some kind of yoke cores, or a less than optimal magnetic circuit..........

I'll make the twin rotor compromise.....

 

[liveforphysics]

Then we're back to some kind of yoke cores, or a less than optimal magnetic circuit..........

I'll make the twin rotor compromise.....

Not as I'm visualizing it. Same thin flat disks. Just the coils and the magnets swapped places.

 

[Miles]

If you use a halbach array for the twin rotors then, potentially, you get a better magnetic circuit than you can (without difficulty) get with twin stators. No?

 

[Thud]

We seem to be in a loop here.

I vote for a twin rotor with back-to-back rhomboid/mandorla shaped coiled strip cores & windings.

Here is something to look at:

34 pole/21teeth
flux rings on the magents
powered iron cores.
could add stators & rotors,
A yoke to support the main sahft ends dosent seem insurmountable. Size is becoming an issue now.
Anyone care to donate flat copper foil for a build?

 

[Thud]

a rotor assembly built on a cast hub, laser cut iron flux ring,turned to inset magnets & bond them.
open center section will provide opportunity for active airflow cooling.

Input please............

 

[liveforphysics]

If you use a halbach array for the twin rotors then, potentially, you get a better magnetic circuit than you can (without difficulty) get with twin stators. No?

That's right. You can also do a lot to trap the coil flux. The magnets don't have to just barely cover the coils. They can be raised on the outside edges by half the stator thickness each, so they only have a tiny gap for airflow on the outside. On the inside diameter coil side, the magnet can be made to extend a CM past edge of the coil, ans additional thin magnet with radial field could be placed as the base of the axial magnet to push the edge flux to concentrate in the gap at the boundry of the coil. This would make for very secure flux containment.

I've played with 2 different magnetic flux modeling programs now, both sucked (or I sucked at using them). Anyone have experience with a good program, and care to do a couple flux models for us?

 

[liveforphysics]

a rotor assembly built on a cast hub, laser cut iron flux ring,turned to inset magnets & bond them.
open center section will provide opportunity for active airflow cooling.

Input please............

Gotta have provisions for the other ring to piggyback over the outside of it.

That design also leaves very little area for coil size.

Think big pie slices for magnets, or better yet, think radial stacks of 4 magnets each making an HB array. Then some bolt tabs on the outside so the second rotor can clam shell onto the back of the rotor disk to encapsulate it inside, but leave the center of the other disk open for stator mounting and connections.

 

[Miles]

If you have back-to-back coils on a twin rotor design, it's best for the cores for each coil pair to be common (axially contiguous)? Right?

 

[liveforphysics]

It may take rhitee05 to answer that Miles. I'm not seeming to come to a conclusion as to which is optimal.


Thud- that upper picture I totally missed for some reason. Damn phone, lol. The structure design looks good. The spokes in this design would be causing a lot of drag for minimal air movement though. Either shaping them like turbine blades, or running a flat disk would cut on drag a lot, and some very clever passive tricks could be used to generate huge airflow over the coils, which will ultimately be the only thing to limit how much power we can continously pump into this monster.

 

[Thud]

I see where you are going now, with piggy back rotors.
(I am still a couple posts behind-Cad is slowwing me down) that thought is turbine style blades (just tough to draw quickly but simple ducting could also be added in an investment cast proccess. dont get hung up by any of my renderings.
You need to refine your goggle skect skills bro & help me keep up with Miles & your self's quickly darting & turning conversations

Can someone scketch a Halbauch array plan using off the shelf products?
specificly in a circular rotor shape. Also hold the diameter of the array to 6.5" or less.
readily available wedge magnets are only avalible in 22.5 deg. segments. 16 poles
Are we willing to sacfrice pole count (raising Kv)

I am anxoiuse to make something......Lets get a 1st iteration settled on & Built.
we can add & subtract spice & texture till we get happy with our home brew sauce

 

[paultrafalgar]

The very last thing I want to do is dampen anyone's enthusiasm for speculation/experimentation, but I'm thinking: The design of electric motors has gone on for at least a 100 years with a huge financial incentive to making the best design. Is it not likely that the adoption of a current design would be more profitable to us than reinventing the wheel? I suspect that a lot of the best designs are for higher power motors, so thinking in terms of downsizing an existing design might be the way to go. In that vein, can someone explain to me why an Agni motor couldn't be downsized to give less power? There is the problem of patents, of course, and so what about asking Cedric (he's a very "alternative" thinker) to produce a motor downscaled for ebike use? What do you think?

 

[Miles]

The very last thing I want to do is dampen anyone's enthusiasm for speculation/experimentation, but I'm thinking: The design of electric motors has gone on for at least a 100 years with a huge financial incentive to making the best design. Is it not likely that the adoption of a current design would be more profitable to us than reinventing the wheel?

The same argument could have been made in the 1980s, when Cedric Lynch was developing his design

 

[Thud]

paul trafalger wrote:
The very last thing I want to do is dampen anyone's enthusiasm for speculation/experimentation, but

This is a phrase that raises my defence every time I hear the "but"
Paul,
The only item you missed is that we are hobbyists,who will build things just for the sake of doing it.
No one really needs an 18 cyclinder radial engine in this day in age especialy in 1/5th scale.
But you find some hand crafted examples made by fine craftsmen all over the world.

The design of electric motors has gone on for at least a 100 years with a huge financial incentive to making the best design. Is it not likely that the adoption of a current design would be more profitable to us than reinventing the wheel?

There is a LOT of design talk happening here that is going way beyond my education & I enjoy that.
& never discount the Inginuity of an individual or group analizing a set of eqautions & therory. I beleive it was Cedric himself that found a better way using tin cans & a freash perspective to beat all the entries form the heavily funded corperations & universities.

so what about asking Cedric (he's a very "alternative" thinker) to produce a motor downscaled for ebike use? What do you think?

I think thats a fine idea, please do & best wishes.
I for one will still be looking for a option that sets a different standard. Maybe not better but definatly more indivdualistic & foward thinking.

(damm you Miles! you beat me to the responce again! you are still winning the internet forum game in my book)

 

[paultrafalgar]

OK, Miles, but I'd still like you to explain why/if that design could not be used in a lower power form. Anybody?

 

[Miles]

Are we willing to sacfrice pole count (raising Kv)

As Kv goes down inversely with field strength and it can be adjusted by the number of turns in the winding, I don't see that this is a great sacrifice.

 

[Miles]

OK, Miles, but I'd still like you to explain why/if that design could not be used in a lower power form. Anybody?

Well, the LEM 130 isn't as impressive as the larger models - I'm not sure I have an explanation, though....

 

[liveforphysics]

The very last thing I want to do is dampen anyone's enthusiasm for speculation/experimentation, but I'm thinking: The design of electric motors has gone on for at least a 100 years with a huge financial incentive to making the best design. Is it not likely that the adoption of a current design would be more profitable to us than reinventing the wheel? I suspect that a lot of the best designs are for higher power motors, so thinking in terms of downsizing an existing design might be the way to go. In that vein, can someone explain to me why an Agni motor couldn't be downsized to give less power? There is the problem of patents, of course, and so what about asking Cedric (he's a very "alternative" thinker) to produce a motor downscaled for ebike use? What do you think?

It's not speculation towards a theoretical goal.

It's been done. The launchpoint motor is 1.5lbs, and delivers 6hp continuously.

Its not mad science or anything crazy. Things to think about:

To be 5hp continuously, a big AC induction motor (or a hub motor) weighs 30lbs.
To be 5hp continuously, a conventional RC motor can be about 5lbs. This is good, but it's also spinning too fast for simple drivetrains on E-bikes. This requires additional weight and complexity and inefficiency before it can drive the wheel.
To be 5hp continuously, a proper axial flux motor can be under 2lbs, and spin at a speed very friendly to driving a wheel.

This is why we seek to make our own proper axial flux motor.

 

[liveforphysics]

OK, Miles, but I'd still like you to explain why/if that design could not be used in a lower power form. Anybody?

Simple, to make power in a small form, you need RPM. His designs all spin the copper windings... Copper is heavy and soft and weak, which all = fail for something you wish to spin fast. You can spin magnets to kingdom come, and they don't care.

Also, we aren't looking to make less power than his motors. We are looking to make the same power of his motors, but in a package 1/4th the weight, and with better cooling. It's not a pipe dream at all, the CISRO motor (and a few others) are examples of it.