I can't really understand what you mean with the mechanical bits without some drawings, but rest sounds pretty solid in theory!
One of the main problems with hub motors is that they add a lot of unsprung mass which adversly affects the suspension action.
The first picture is a less mature concept, but it is a more complete drawing.
http://publicarium.com/motor9.jpgYou're looking at a rubber wheel, with ordinary spokes connecting the wheel to the rotor. This rotor has magnets permanently fixed to it, which are placed very near, but not touching, the coils (that little bit of yellow). The inner "spoked" piece of steel holds the coils stationary. The coils, inner spokes and the axle do not rotate at all. This is a fairly "massive" solution, and it was replaced by the next picture...
http://publicarium.com/motor10.jpgThe second picture is a more elegant solution, both in terms of machining and unsprung weight. If the wheel spokes were shown, then the picture would include all things spinning - and the plates as shown are unnecessarily thick. What isn't shown there is that the stationary coils are permanently fixed to the axle and sit between the magnet plates. Again - nothing "rubs" on anything; there is just a very large ID bearing in the center.
I see what you mean about unsprung mass... I was misunderstanding the issue, and now I realize that even the stationary bits are an issue because the suspension will have more trouble dynamically "pushing" the massive wheel into the road. Aside from the steel plate holding the magnets and the bearing, all other components can be made of aluminum or carbon fiber.
No core?
Yes, a "core" in motor terms is usually insulated layers of metal (called laminations) that the wire is wrapped around. This metal helps to conduct and hold the electromagnetic field. The lower the frequency of the controller, the more massive the laminations need to be, and the hotter they'll get. They're usually about half of the motor's overall weight. As the controller frequency approaches 150KHz, the need for laminations diminishes to almost zero - helping to eliminate a huge amount of unsprung mass and heat losses.
Do the magnets and copper pass each other like wear points on clutch plates?
Yes, almost exactly, except without the wear.
When people mount hub motors on bicycles they use a torque arm attached to the frame of the bike. Kinda like a coaster brake for a bicycle. Kind of a similar torque on the axle I imagine.
Excellent explanation... Thanks for the tip - I'll see what I can do to achieve this.
Putting even a 30 pound motor in the wheel would be pretty extreme (like that Harley guy).
I actually have a video of my controller in action. I'm using an off-the-shelf 9HP DC Brushless motor intended for giant-scale remote-controlled airplanes. Here are the videos:
DC Brushless Motor Controller - #2DC Brushless Motor Controller - #3That motor weighs 3 pounds, but has a substantial core. The larger size I'm aiming for, minus the core, ought to weigh around 10 pounds.
How to do the brakes? A basic concept for bikes is to get as much weight as possible near the center of gravity and certainly as little in unsprung weight in the wheels.
The batteries will be adding that center-weight. As far as brakes - I'll start with a front-only mechanical brake. The motor is a regenerative servo drive, so I have full control over the motor's speed, including braking... no extra parts.
With bikes, a constant development of HP has been through faster engine revs. Everything stays the same but you manage to spin faster.
It ought to be a piece of cake to dynamically balance a brushless electric motor for that and greater speed. That's the way to go. Any reasons why not?
Electric motors have a similar efficiency story - the faster they can electrically commutate, the more HP you can get out of them... The trick is, a conventional motor rotates once per electrical commutation, while this motor design requires more than 100 electrical commutations for each single wheel rotation. This will give it tremendous low-end torque, and it should peak-out at around 70MPH on flat ground.
Also, electric motors have a much flatter torque curve, especially DC brushless motors, meaning that I don't need to worry about switching gears in order to stay where the torque is.
As far as dynamic adjustment, the speed/torque ratio can be adjusted by increasing the gap between the magnets and the coils; the results are just like shifting gears. While I don't have any way to do this while riding the bike, I can fine-tune the motor's gap as needed. I'm hoping that the motor's characteristics don't require me to adjust it more than once.
- Kipp
Author
Topic: I'm engineering an electric motorcycle drivetrain (Read 2307 times)