The Halbach Disc Motor Ebike Project

Simplicity

Sometimes you just have to go back and reread your own logic. These forums act as a sort of journal and that allows you to go back and recheck your past thought processes. In retrospect my thinking about Single Phase this last winter was really good. The "push pull" advantage in the wiring is enough to make it a preferred choice over Three Phase. (it literally cuts the kv value in half)

Here's an image of Single Phase. The colors are defined so that anything OVER 1 Tesla is in purple, so you can see how the magnetic flux lines flow:

attachment.php


...0.5 Tesla magnets end up generating a 1 Tesla magnetic flux at the point of interface with the cores and inside the cores the flux rises to 2 Tesla's which happens to be the limit of Silicon Steel cores. (saturation) So if I really want to max out the design the use of Single Phase is in line with optimal results.

The bottom line is that I can pack things a lot tighter with Single Phase.

I was hoping that Three Phase would be equal, but it ends up doubling the kv over Single Phase.

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

It's kind of like deciding between a two stroke and a four stroke gas motor... I know from the past that two strokes dominated, but in the end the benefits (pollution reduction) of four strokes won out. From a pure performance standpoint the Single Phase is like a two stroke.
 

Attachments

  • Magnetic Flux.jpg
    Magnetic Flux.jpg
    180.3 KB · Views: 2,339
Last edited:
Comparisons

Q: What is this Halbach Disc motor best compared to if it's run as a Single Phase motor?

A: A High Pole Count Brushless Servo Motor.

Here's an example:

http://www.microcontechnologies.com/MDM-5000 High Energy Brushless Servo Motor.htm

They describe it this way:

Brushless design provides high performance with zero maintenance
Cut-core or Segmented Stator Technology means high efficiency, high slot-fill windings
Stator molded in thin wall aluminum housing using thermally conductive epoxy provides excellent heat transfer, high torque density and protects the windings and optional Hall board from damage
Rotor coated with polymer shrink wrap which protects magnets from dislodging during operation and prevents accidental damage while handling.
High pole count, low cogging, smooth operation
High torque to inertia ratio ideal for high acceleration/deceleration mode
with rapid start-stop
Low electrical time constant for fast response


Basically a "Low electrical time constant" is saying that it has a low kv value. This seems to confirm my simulation results that using a more servo like design has the benefit of reducing the kv. With a reduced kv value you can wind the motor so that more coils are in parallel and less in series and that's the path to getting the efficiency higher.

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

When it comes to motors you seem to have a choice... either you create a motor with really high rpm and then gear it waaaaay down to make it work in a mechanical way or you focus on the electrical side and increase the pole count and get the kv value waaaaay down. The lower the baseline kv the more parallel coils in a motor and that means better efficiency.

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

In a perfect world there would be a Single Phase controller that already exists and runs well at 1000 watts and incorporates a Hall Sensor as well as current limiting. If I can find one of them it would make my life easier. Otherwise I'm stuck trying to build one on my own.
 
Last edited:
Have I Gone Bipolar?

And in need of medication? Or have I discovered something about the Bipolar motor that might offer a lower kv value that anything else there is to offer?

Here's what wikipedia says:

http://en.wikipedia.org/wiki/Stepper_motor

Bipolar motor

Bipolar motors have a single winding per phase. The current in a winding needs to be reversed in order to reverse a magnetic pole, so the driving circuit must be more complicated, typically with an H-bridge arrangement (however there are several off the shelf driver chips available to make this a simple affair). There are two leads per phase, none are common.

Because windings are better utilized, they are more powerful than a unipolar motor of the same weight. This is due to the physical space occupied by the windings. A unipolar motor has twice the amount of wire in the same space, but only half used at any point in time, hence is 50% efficient (or approximately 70% of the torque output available). Though bipolar is more complicated to drive, the abundance of driver chip means this is much less difficult to achieve.


...this is exactly what the simulation showed. Three Phase and Single Phase produce the same AVERAGE output overall, but the Single Phase using the Bipolar configuration cuts the kv value in half creating a lower rpm. The lower kv allows more parallel coils and that increases overall efficiency.

The primary problem of any type of hub mounted motor is that the wheel rpm is so low compared to what the electrical rpm desires. High pole counts and the Bipolar design enhance the push towards lower and lower kv values. I'm finding that I'm able to achieve a kv of 33 rpm/volt while using lot's of parallel coils. Using Three Phase and it's impossible to get the coils to fit within one pole period without getting rid of the cores and if you go that way it doubles the kv value.

The race is to lower kv values.
 
Last edited:
Absurd Kv Extreme

Okay, so it's all about low kv values because once you capture them you can scale backwards with more parallelism and get the efficiency. The "most extreme" low kv in effect defines the limits of performance of the motor more so than a particular configuration of copper windings.

attachment.php


...using 20AWG wire and ordering them as all series I get an absurd 5 rpm/volt.

5 rpm / volt!

So at 24 volts that's only 120 rpm and for a 24" wheel only 9mph. Silly thing to do in practice, but it says how much room there is the play with this way... assuming Single Phase.

attachment.php


Yesterday I made this which is a replacement base for the Halbach Disc motor. I had started with a 160mm brake disc and found that it flexed too much, then added a second 160mm brake disc and found that it weighed too much. So this is going backwards to some degree, but I was wasting some excess weight the other way and it's better to get it right the first time. Today I fiberglassed the new base into a disc and will eventually graft it onto the stuff I have so far.

The holes both lighten the new base and also give something for the fiberglass to grab onto when it's put together. Don't want the fiberglass slipping away from the base. (the new base is about the same weight as just one 160mm disc and extremely rigid)
 

Attachments

  • Absurd Kv Extreme.jpg
    Absurd Kv Extreme.jpg
    137.6 KB · Views: 2,141
  • New Base Disc.jpg
    New Base Disc.jpg
    143 KB · Views: 6,843
Last edited:
attachment.php


This is the new base disc replacing the old one. I figure that everything you see weighs about a pound. Add the magnets and the weight goes to about 3-4 lbs. Add the coils and the cores and you are at about 5-6 lbs. The performance should be about equal to a motor of double or triple it's weight by comparison.

Radius of 7". (Diameter 14")

No geardowns needed.

No heavy weight to disrupt handling.

Ideal for 1000 watt racing. (upper limit probably 5000 watts)

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

I've been looking at some Thundersky batteries to use on this bike. Two 12V Thunderskies would fit perfectly into the place for my present NiCads. I had figured that I didn't have any competition and could make due with NiCads, but since it's looking like in the next few years 1000 watt racing has some chance to get going this bike will be ideal for it. (starting to get a little worried that next years Arizona race isn't going to happen for 1000 watt bikes though)
 

Attachments

  • New Base Disc 2.jpg
    New Base Disc 2.jpg
    128.9 KB · Views: 2,066
Last edited:
attachment.php


Before I have to take the rim off the hub so that I can spin the Halbach Disc and not have anything in the way I figured I'd post this to show the general size of this motor. It's big... not heavy, because it's made of fiberglass... but the leverage you get from a pancake this size is very good and allows for smaller (cheaper) magnets of lessor peak Tesla's to achieve a positive outcome.

The actual magnets will add another 3/4" to the radius, so the total size will be a little larger. Even the largest hub motors I've seen aren't this big in diameter. (and they weigh a ton)

Ideally this would be built with carbon fiber. (one day... one day...)
 

Attachments

  • Halbach Build Progress 001.jpg
    Halbach Build Progress 001.jpg
    109.2 KB · Views: 6,466
Last edited:
More Progress...

attachment.php


...the day started off with a thunderstorm that cooled everything off. That prompted me to get out there and do some grinding on this, but once the sun came out the heat shot up to over 90 degrees with high humidity. It's a hot summer... much hotter than last year which was a really good moderate heat year.

One of the grooves is done (the side you see) but the other side isn't.

It still doesn't weigh much...
 

Attachments

  • Halbach Build Progress 003.jpg
    Halbach Build Progress 003.jpg
    207.6 KB · Views: 2,047
Last edited:
dunno

good work !

what's the latest position ?

pk


Simplicity

Sometimes you just have to go back and reread your own logic. These forums act as a sort of journal and that allows you to go back and recheck your past thought processes. In retrospect my thinking about Single Phase this last winter was really good. The "push pull" advantage in the wiring is enough to make it a preferred choice over Three Phase. (it literally cuts the kv value in half)

Here's an image of Single Phase. The colors are defined so that anything OVER 1 Tesla is in purple, so you can see how the magnetic flux lines flow:

attachment.php


...0.5 Tesla magnets end up generating a 1 Tesla magnetic flux at the point of interface with the cores and inside the cores the flux rises to 2 Tesla's which happens to be the limit of Silicon Steel cores. (saturation) So if I really want to max out the design the use of Single Phase is in line with optimal results.

The bottom line is that I can pack things a lot tighter with Single Phase.

I was hoping that Three Phase would be equal, but it ends up doubling the kv over Single Phase.

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

It's kind of like deciding between a two stroke and a four stroke gas motor... I know from the past that two strokes dominated, but in the end the benefits (pollution reduction) of four strokes won out. From a pure performance standpoint the Single Phase is like a two stroke.
 
Dec 28, 2013.... Safe, after all of your research and construction effort, why did this this thread/project suddenly end? Anyone as interested as I am in seeing it come to fruition? Or have you gone into production?? Anyone?, Anyone?? Buehler? Cheers, Joe
 
far out.

erm. death by various causes, change of employment and cash flow, it didnt work, patent infringement, um... theres plenty of reasons why a good thread can come to an abrupt halt!

sad. 3 and a half years :(
 
Back
Top