I'm trying to find a suitable brushless outrunner motor to run a e-bike. The problem is that a bike wheel turns very slowly (approx 250 RPM for a 26' wheel to go 20MPH), and electric motors like to run fast. Even with a slow motor, you need significant gearing down, which is undesirable because it adds complexity to the design.

I was wondering if it would be possible to take a giant motor, then rewind it to get it around 100KV, but I don't know how realistic this is.

Any thoughts or brilliant ideas, besides a hell of a lot of turns on the stator?

There are hub motors designed for this purpose...

Trying to do it with a RC motor might be difficult if you want efficiency, the stator and magnet layouts are likely not optimal.

Reduction gearing is where the efficiency comes into things, having the motor running under a practical load and at an efficient RPM driving a larger gear. Simpler is simply not practical unless you can find a stator and magnet housing that are about the size of the bicycle wheel.

In the attached the planetary gears will rotate in a circle, if you immobilize the planetary gears from rotating and attach the outer ring gear to the hub, the internal gear will drive the hub around.

In the image the gear teeth are very poorly drawn, good spur gear teeth would not look like those.

Jack

Originally Posted by pia32

Any thoughts or brilliant ideas, besides a hell of a lot of turns on the stator?

Rewinding for lower Kv won't do much good because it raises winding resistance, which increases power loss at high current and limits torque. It will not increase the maximum torque and power that the motor can deliver at low rpm.

Rewinding should only be used to match the motor to the supply voltage. You can rewind for lower Kv if you need to drive a heavier load, but only at lower power and/or worse efficiency. The same result could be achieved by just using a lower voltage.

Using a gearbox allows the motor to spin at higher rpm and deliver more power, which the gearbox translates into more torque at lower rpm. Although the gearbox wastes some power, motor efficiency can be maintained at higher power. The limiting factors are usually cooling (even though high motor efficiency is maintained, there is still more heat to get of) mechanical issues (centrifugal forces on rotor, bearings, gear wear) and whether the ESC can handle the higher commutation frequency.

The other way to increase torque at low rpm is to have more poles and stator slots, but that requires a redesign of the whole motor (not just rewinding). There is a limit to how much this can get you in a particular size, as the space for each winding becomes smaller. Therefore to get the best from a direct drive setup you need to increase the diameter of the motor. If your 'giant' motor is not up to the task in its existing Kv, rewinding for lower Kv won't help. What you would need is a motor that's bigger than giant-size!

Have you looked at any of the ready-made hub wheels for E-bikes that are on the market? Someone already figured all the technical stuff out for you and put it in a nice drop-in package.

20mph is peanuts for some of them.

I'm trying to avoid using a prebuilt solution. I'm a masochistic DIYer

I guess I'll just use 2 stage belt reduction

Thanks for the help

Originally Posted by pia32

I was wondering if it would be possible to take a giant motor, then rewind it to get it around 100KV, but I don't know how realistic this is.

Any thoughts or brilliant ideas, besides a hell of a lot of turns on the stator?

As pointed out, rewinding a motor does not change it's torque-RPM curves. All it does is change the battery voltage and current.

A ring gear as shown is a good solution for a gear reduction. These are used in many starter motors, electric drills, winches, etc. They are as common as dirt. By adding a second ring gear on the outside they can have a reduction of 25,000 to 1. These high ratio reductions are common in power rear view mirror on cars. Winches may also use these very high reductions gears.

However the simplest reduction is sprockets and roller chains. Roller chain sprockets are easy to make, even with hand tools and 410 chain is cheap.

Pia32, one other issue you'll bump up against is that as you increase winds, inductance increases which at a certain point will limit the max current that can develop at high RPM in higher pole motors. You can readily see this on a scope. You can readily emulate this too: Take a favorite low-KV motor, run it up to expected load at max RPM. And then double the phase inductance by winding an air-wound inductor out of 12g with the same value as the phase inductance at put it in series with the controller. You'll very likely see top end power fall off appreciably due to the extra inductance.

The second issue you'll encounter is saturation. As you double the number of turns, you halve the current you can put through the turns. I2R is indeed one reason. But more importantly, Amp*turns must remain constant for a given cross section of steel. If you exceed this, you'll see the Kt of the motor start to flat line. Worse, at some point, the steel will become completely saturated and the inductor value collapse and suddenly what was an inductor will appear as a wire. And the nice sawtooth that you did have going into the inductor will suddenly ramp skyward, just as a wire would. And if your controller isn't smart enough to figure this out, then it will die on the spot.

If you really want the full DIY experience on this, your starting point on this is to pin down your desired acceleration and gearing. That will tell you the torque you require from the motor. Then you can easily derive the torque constant (and Kv) for your desired battery and currents.

And then you'll know what motor to go after. At high power levels, rewinding won't work magic. You can nudge a bit one way or the other. But you won't transform a 3KW motor into a 6 KW motor via rewinding, unless the 3 KW motor was woefully underspec' ed in the first place.

PS. Plan on wye. Delta issues at higher power levels related to circulating currents at higher harmonics, and these will rob efficiency. Plus, wye will give you a lower KV.