DIY 24" bldc motor (just thinking)

[HAL9000v2.0]

Lets say that I have big enough bearing so the 24" rotates freely. Then I put neodyum magnets inside rim and make copy of x5 motor but with bigger diameter and more coils so its outer dia will be inside 24" rim. So now I have 216 coils. Torque is massive but what about speed. Let say that I use kelly bldc 100A controller. Will it be fast enough for such small angle increment (not 5°/f but 1.666°/f)? Where will I put hall sensors?...

[Ypedal]

Not to put a stop to any project like that.. i commend the idea.. but an X5 with 100 amps kicks butt .. as is ! lol.. how fast do you want to go ?

[HAL9000v2.0]

Not to put a stop to any project like that.. i commend the idea.. but an X5 with 100 amps kicks butt .. as is ! lol.. how fast do you want to go ?

I think speed of light is maximum for now?

Just kidding, 50-60 mph flat. But some idea is 35mph on 18% uphill.

[Link]

Okay, who here has not had this idea at one point or another plz raise hand?

You'd need plenty of voltage to get it spinning at a good clip with that many poles/total magnetic field around the coils, but it would be so torquey I wouldn't be surprised if no torque arm in existence would keep it in check. You'd definitely have to secure it far from the axle.

I would love to try to make something like this if I had access to a machine shop. Although you'd probably get a better power ratio out of it if you made it axial flux (like the old Eteks) instead of radial. Finding magnets in the right shape and size for that would be nearly impossible, though.

[TylerDurden]

Didn't we discuss how strong magnets near the rim would attract metallic debris... or do I have the wrong mental-picture?

[Link]

I don't think it would be that big a deal. Big BLDC hub motors don't really attract stuff at all, even when something like a wrench is touched to it. I don't see why this should be any different.

[TylerDurden]

This still has potential... Brett may know more.


http://www.thebackshed.com/Windmill/AssemblyA.asp#3

[Link]

Hmm, lots more room for more windings. Could be useful...

[lawsonuw]

Lets say that I have big enough bearing so the 24" rotates freely. Then I put neodyum magnets inside rim and make copy of x5 motor but with bigger diameter and more coils so its outer dia will be inside 24" rim. So now I have 216 coils. Torque is massive but what about speed. Let say that I use kelly bldc 100A controller. Will it be fast enough for such small angle increment (not 5°/f but 1.666°/f)? Where will I put hall sensors?...

Sounds like a project And don't worry about motor speed, If you're winding your own poles, you can make this monster go any speed you want. Home brew PM generators for wind turbines and custom motor maker for RC would be good people to pump for info.

Marty

[HAL9000v2.0]

This is what I have in mind...

[Dr. Shock]

Sorry to bear bad news, but they're way ahead of you in Slovenia:

http://www.elaphe.si/news.php

[Link]

We know, but they don't have anything available yet. And this would involve DIY and machine work, which is always a plus .

[HAL9000v2.0]

Sorry to bear bad news, but they're way ahead of you in Slovenia:

http://www.elaphe.si/news.php

I know for them, they are less then 100miles from me...

The rumor is that some major auto company is buying them so "Who killed electric car” saga continues...
I am for open source design of electric vehicles; 1. It is my hobby and 2. Which government in Europe will allow implementing 100% of electricity in transportation? They are taxing oil with 100%, and they can not put tax on solar panels or wind generators and call it "road repair tax" or "highway development tax"...

[Dr. Shock]

The rumor is that some major auto company is buying them so "Who killed electric car” saga continues...
I am for open source design of electric vehicles; 1. It is my hobby and 2. Which government in Europe will allow implementing 100% of electricity in transportation? They are taxing oil with 100%, and they can not put tax on solar panels or wind generators and call it "road repair tax" or "highway development tax"...

Wow, we're used to hearing this about the evil US government\military-industrial complex\GeorgeBushVastRightWingConspiracyDickCheney. I confess to a bit of schadenfreude, hearing it goes on in Europe, too.

HAL, I suspect that if wheel motors are going anywhere, they'll have to wind up looking like your design. They need to be powerful enough to run a car, and to be economical, probably the car will only have one powered wheel. Big, fast batteries are what's really needed. Also, better solar panels. Sooner or later, the oil's going to be gone, so the government will have to figure out a way to tax them.

[fechter]

Lets say that I have big enough bearing so the 24" rotates freely. Then I put neodyum magnets inside rim and make copy of x5 motor but with bigger diameter and more coils so its outer dia will be inside 24" rim. So now I have 216 coils. Torque is massive but what about speed. Let say that I use kelly bldc 100A controller. Will it be fast enough for such small angle increment (not 5°/f but 1.666°/f)? Where will I put hall sensors?...

The hall sensors go in the middle of 3 adjacent coils.
The coils need to have iron backing. This could be a ring that also supports them.
With the large diameter the BEMF would be great, so you would use very few turns for each coil. By choosing the right number of turns, you can get the desired no-load speed / volt.
The Kelly and most other similar controllers can operate up to 50,000 electrical cycles per second. Divide 50k by the number of number of magnets/3. I don't think this would be a problem.

[HAL9000v2.0]

My intention is to go with high volts an low amps. With this design for same torque you will need only 1/3 of current then for x5. Then I can use much thinner wire with more turns...

[wanders]

This is what I have in mind...

Nice pix, HAL9K.

I gotta be the first to say it - let's rename this thread "RE-INVENTING THE WHEEL"

Seriously, the approach shown by HAL9K is not unlike the approach taken in the late 1990's by CSIRO (although the stator and the rotor are inverted) and the basic design is an axial field, air core stator, dual Halbach magnet array brushless machine. They used it in an electric sunracer and showed 98.5% efficiency, pretty darned impressive. Similar motors are currently being built in the US and elsewhere.

It was even integrated into the wheel (just not a bike wheel).

Let's be practical. If you really want to solve the e-bike problem here, build a more-or-less conventional hub motor but use an axial field, air core stator, dual Halbach magnet array brushless motor and make it appropriately sized (i.e. narrow enough to put a freehub and a disc on, for the rear). Because it does not have an iron core, it will be VERY light. Because it is appropriately sized (both in diameter and width) it will be easy to retrofit and it will be high torque / high power.

It all looks great, right? Where is the Achilles heel?

Cost is the answer, unfortunately. The magnets (especially in a Halbach array) are expensive. For a fixed power output, the cost of the magnets is inversely proportional to the RPM of the motor, and all direct drive hub motors are low RPM.

Food for thought,

Willie

[HAL9000v2.0]

Now you forces me to think again, Wanders.
How about to build motor over inhub gears like rollhof or g-boxx ?!?!? hm..

Another question is: Will motor be more effective if I put coils inside two rows of magnets?

[lawsonuw]

putting the coils between two rows of magnets would allow you to use an air core for the coils and solid steel for the magnet backing. (i.e. no need for laminated steel cores) A Halbach array would save the weight of steel backing but need more magnets.

Marty

[wanders]

putting the coils between two rows of magnets would allow you to use an air core for the coils and solid steel for the magnet backing. (i.e. no need for laminated steel cores) A Halbach array would save the weight of steel backing but need more magnets.
Marty

I could be wrong, but I don't think a steel backing is needed for a Halbach array. Even aluminum would work fine. And while the air-core stator would need to have a fairly sophisticated construction (e.g. very fine stranded copper wires to avoid eddy current losses plus laminated fiber & epoxy for strength) it would be lightweight.

In fact, what lawsonuw has described is essentially the CSIRO design.

Willie

[Drunkskunk]

Seriously, the approach shown by HAL9K is not unlike the approach taken in the late 1990's by CSIRO (although the stator and the rotor are inverted) and the basic design is an axial field, air core stator, dual Halbach magnet array brushless machine. They used it in an electric sunracer and showed 98.5% efficiency, pretty darned impressive. Similar motors are currently being built in the US and elsewhere. Food for thought, Willie

Just thought I'd point this out. The efficancy of bare copper wire is only about 97%. you couldn't build this motor with copper windings. Silver, however is above 99%. I'm pretty sure CSIRO used silver on the sunracer, but copper in the motors they were selling.

[wanders]

The efficancy of bare copper wire is only about 97%. you couldn't build this motor with copper windings. Silver, however is above 99%. I'm pretty sure CSIRO used silver on the sunracer, but copper in the motors they were selling.

Hmm, you are probably right, Drunkskunk. Even so, the CSIRO design is sound, and I would be happy with "only" 96.5% efficiency in the motor. The rest of the system (internal resistance in wiring and batteries, controller, etc) will take off more efficiency as well. There are lots of little issues (like eddy currents in the stator conductors) that must be attended to in order to make a motor very efficient.