Halbach Axial Flux Motor

[safe]

I think you are referring to an axial motor as a Halbach. Halbach only refers to a positioning of magnets to get a directional flux.

The so-called "pancake" motor can be either "standard" or "Halbach", the "Halbach" not needing iron.

I always knew about the "pancake" motors... but the ironless axial Halbach "pancake" motors were new to me as of this last saturday.

The big advantage of the ironless for the homebuilder is that you have less to worry about in the design. All you need is magnets and wire... and some fiberglass... and you can build a motor.

 

[safe]

It Was An Induction/Halbach Hybrid

The Field Weakening pdf was actually of an AC Induction motor that was sort of a hybrid with Halbach magnets injected into the rotor. This explains why they ended up with a very AC Induction like powerband.

This pdf talks about all the "attempts" to achieve Field Weakening with "permanent" magnets.

Looks like Field Weakening is "out" for the pure simple Halbach.

-------------------------

A review of the classic ebike issues:

Geardowns are undesireable and should be eliminated if possible.

Multispeed gearing is undesireable and should be eliminated if possible.

Weight is undesireable and should be eliminated if possible.

Power needs to attain 1 hp from 0 mph to 20 mph and then be open for extension beyond that level to 30 mph (most states) or unlimited.

...based on these criteria the Halbach knocks off items #1, "maybe" #2, #3, and "maybe" #4. The use of multispeed gearing combined with 1 hp will easily break the law, so there are major issues with using them.

 

[safe]

Sensorless Control Without Cogging?

If sensorless brushless controllers watch for a "bump" in the EMF and that "bump" is normally the cogging force then those types of controllers wouldn't work with a Halbach motor.

Or would they?

 

[johnrobholmes]

Im not sure that the halbach always results in zero cog. Sensorless can control it just fine anyway, it isn't like back EMF is eliminated- just smoothed to a nice sine if wanted. If EMF was eliminated, it couldn't produce power.

 

[safe]

Well what is "Field Weakening" then?

From what I understand the whole idea of "Field Weakening" is to reduce the backEMF. (at least that's what that pdf file suggested) They go to all these weird extremes with PM motors where they create reverse acting fields to counteract the backEMF and allow the motor to spin faster. (actually adding a separate set of coils for reverse backEMF)

Anyway... I guess you are saying that Halbach motors have been tried with existing RC controllers and have worked.

So maybe I might buy an ESC one day.

(however, I still would need to build a current limiting circuit and a current sensor)

 

[safe]

Greedy Massive Power

What I'm thinking is that since the Halbach design allows for a lot of power packed into a thin and lightweight disc and that since there is plenty of room to build a big disc on the rear wheel that I could design the motor to be two or three (or four) times what I need.

So maybe I design with the goal of 2-4 hp and then scale back as needed so that it would run under whatever limitations I need for legal purposes.

The disc could be 8" in diameter... the bigger the disk the more leverage and since efficiency is all about the wire alone it should work. Without the iron to worry about the disc size can grow quite large.

A 48 tooth aluminum bicycle chainring should be strong enough to be the foundation and then you glue the magnets onto it and then seal it all up in fiberglass.

http://www.bikepartsusa.com/bikepar...-10H-OR8-48T-110130-SIL-18&category=chainring

 

[johnrobholmes]

Field weakening could occur with a winding interacting with magnets or increasing airgap, but you would be reducing efficiency either way. It does speed up a motor by screwing with the effective KV and reducing back EMF (and reducing torque per amp). Back EMF cannot be eliminated with a PM motor, or else you don't have a motor anymore.


The Halbach array doesn't change the basics of PM motor driving, it just tweaks the flux a bit. The axial flux motor is better suited for pancake style, as you eliminate much of the dead area and non useful "end" wire that plagues thin tall motors.

 

[safe]

The Halbach array doesn't change the basics of PM motor driving, it just tweaks the flux a bit. The axial flux motor is better suited for pancake style, as you eliminate much of the dead area and non useful "end" wire that plagues thin tall motors.

Normally in order to increase inductance you increase the number of turns of wire in your coils. In the solar racer they apparently realized that they had all the copper they needed and were running short on inductance so they added "extra" inductance to boost things up to what they need.

I like that idea....

You can then control the top speed by changing the "extra" inductors to match what you want. This might be a way to satisfy the 20 mph rule... for a 20 mph ebike you use inductors with a very high rating, for 30 mph you use a lower inductor rating, and for unlimited you use a very low or non-existant inductor.

BackEMF is apparently less of a problem because it takes less copper to achieve the same torque since the pancake disc "leverage effect" magnifies the strength of the forces actually being used.

Total Force = Force of Coil Magnetism * Radius from the Center

F1D1 = F2D2

More research is needed (especially to get all the formulas clear in my mind and in a simulation somehow) but it looks like a good path.

The fact these Halbach motors are ironless is very good and the fact they can be made efficient at very low rpm is even better.

 

[johnrobholmes]

A halbach can be built into any motor design, are you still specifying axial construction too?

I don't like the idea of adding inductance, it makes a motor that does not want to change power levels very fast. You can get the same effect with amp limiting.

 

[safe]

I don't like the idea of adding inductance, it makes a motor that does not want to change power levels very fast. You can get the same effect with amp limiting.

Don't forget that my goals include the ability to restrict the motor to satisfy the 20 mph rule.

Try reading the section in the CSIRO "Solar_motor_ kit_ notes.pdf" where they say:

9. Due to the low inductance of the motor, most motor controllers will require additional series inductance (3 will be supplied).

(they explain fully on page 23)

...I think that they need so little copper to do the job that their problem is that in order to restrict the motor so that it doesn't overheat they need to insert "extra" inductors.

Most motors have so much inductance that they need to be "encouraged" with higher voltage to go faster. With these Halbach motors they apparently are naturally able to flow current really easily.

A halbach can be built into any motor design, are you still specifying axial construction too?

Not sure yet. Leaning towards an Outrunner design with the aluminum sprocket holding the magnets and having it all embedded in fiberglass. Max speed of the motor is wheel speed, so 800 rpm is 60 mph, so the centrifugal forces should be small. Road bumps will bash the motor more than motor speed.