Halbach Axial Flux Motor

[safe]

In The News...

http://www.greencarcongress.com/2009/10/whp-2009029.html

....the magnetic particles in the composite are magnetized as a Halbach Array after the rotor is manufactured avoiding the need for backing iron to direct the flux. As the magnets in an MLC flywheel are comprised of tiny particles and there is no additional metal in the structure, the eddy current losses of the machine are significantly reduced. This can result in one-way efficiencies of up to 99%. The ultra-high efficiency means thermal management of the system is easier and it can be continuously cycled with no detriment to performance or reduction in life.

So apparently this is being used with F1 for the KERS systems.

 

[safe]

The Voltage Formula

Figure 4 is a schematic of the electric machine. It is to
be interpreted on a per phase basis for machines having
multiple phases. Note that this representation contains
only linear elements, and also note that the value of
magnetic field Bo, that is to be used in calculating the
induced voltage, is that of the Halbach array. In this
ironless system, and for all feasible values of the winding
currents, there is no "back reaction" between the stator
windings and the inducing magnetic field. There is, of
course, an effect of the winding inductance on the out
put voltage as well as the usual resistive drop. However,
since the system is ironless, inductances are low, and
with good design, the resistive drops are also low. As will
be shown, not only is the power output very high, but
the efficiency is also typically much higher than that of
an iron core machine of comparable physical size.

One wonders how the RC motor ESC is going to be able to "sense" the backEMF.

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

How this appears to work is the top motor speed is actually determined by the copper in the windings. If you add more copper, then you add more speed. (and weight) This is assuming that you increase the thickness or add Litz wire as the copper increases and not just increase the number of turns. Ideally you want to discover just the right amount of copper to achieve whatever you want... but it seems pretty flexible.

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

The Power Formula

If the rms output voltage of each of the 3 phase windings
is V0, the winding inductance is Lo, and its resistance
is Ro, then the output power of each phase into a
resistive load with resistance RL is given by the equation.

...I'm going to have to plot this to see what it looks like.

 

[johnrobholmes]

If there was no EMF and you spun the motor up, this would imply that no EMF would be present on the leads. Therefore it would not be a motor or generator. Similar to an alternator before you energize the rotor.

Can you cite the source?

 

[johnrobholmes]

Check out p108, it shows the sinusoidal flux density http://books.google.com/books?id=je...Q6AEwBQ#v=onepage&q=halbach array emf&f=false

 

[safe]

Amazing

When you plot the power formula you can get just about ANYTHING depending on what values you put in.

It appears that you can control the powerband shape itself to a large degree since power is mostly a reflection of how much current you pass through your coils.

The answer to:

"What does the powerband look like?"

is:

"It depends."

...the images below are from plotting the formula and then using different values for the variables. (X-axis is w... angular speed which is either rpm, or radians per second however you want to look at it)

There is no "field weakening" going on (because it's a synchronous motor) but depending on how you configure the inductance you can get a powerband that approximates an AC Induction motor.

The fourth chart (the flat one) is when you use very, very low inductance.

 

[safe]

If there was no EMF and you spun the motor up, this would imply that no EMF would be present on the leads. Therefore it would not be a motor or generator. Similar to an alternator before you energize the rotor.

The backEMF is going to be smooth, no cogging. You are probably right that an RC ESC can properly sense the backEMF, but it's just something to want to know as a fact before spending $$$ for it.

 

[johnrobholmes]

Ask/search on RC groups. I know it has been done, I have seen pics there.

 

[safe]

The Halbach Power and Torque Ideal

Power = Torque * Rpm

...so if you plot using the Halbach power equation and include torque and then use variables that would create the ideal result it does appear that you can get close to attaining the ideal motor output for a limited power input situation.

If the goal is some limit (say a constant 1 hp) it "appears" that the Halbach is able to do it.

Obviously things like heating limits might make this unrealistic... but as they say "the devil is in the details" so I'll learn that when I get there.

The spreadsheet is in an old "Works 3.0" format... sorry if you can't read it.

 

[safe]

Inrunner Design

I think I got the name wrong before... the design will be an "Inrunner" because the magnets will be attached to the chaingring and then cemented together with epoxy. Outside of the chainring would be the coils and the idea for that is that I'll build the coils individually by winding them around some pipe. Once the shape is set I then apply a layer of fiberglass epoxy resin to the wires. Let harden. Then remove the solenoid from the pipe. Now apply a layer of resin on the inside. Repeat again and again.

You now have all the poles...

Take the poles and set them up on a flat board (like AussieJester suggested)and then build the stator by applying fiberglass and cloth to the outside of the coils but leave the holes inside the coils.

The idea is that the holes in the coils would act as air cooling vents.

The outside of the coils now form the frame of the stator and you then come off the frame with the mounts. (will require some welding on the frame) These mounts are designed so that you can adjust the stator with precision so that everything spins smoothly and is aligned right. This means that in order to remove the rear wheel you need to remove the motor... but it might be possible to figure out some sort of quick release mechanism for it. (will have to think about the mounting)

Parts List:

Rotor

Magnets (either 3*8*2=48, 3*9*2=54, 3*10*2=60, 3*11*2=66, 3*12*2=72)
Aluminum Chainring
Fiberglass Resin to bond everything together

Stator

Copper magnet wire coiled into 24, 27, 30, 33, or 36 poles (8, 9, 10, 11, 12 pole count)
Fiberglass resin to bond everything together
Fiberglass cloth to add extra strength (it's non-magnetic)
Maybe some aluminum backing for extra strength

 

[johnrobholmes]

Which way will the pole faces point? Towards the sky/road, or towards the left/right?