I'm thinking about creating an ultralight e-bike with an R/C motor that sits right behind the bottom bracket and has a custom 3d printed gear (shaped like a chain) that drives the front chainring in a 1x hardtail. It would be driving the chainring on the back side where the chain is not wrapped around. I've searched online but haven't found anyone who's done it like this. Has anyone seen anything like this? I'm not looking for huge power, rather a small 100w boost for a total system weight under 2KG. I know efficiency won't be great as I'll likely be running significantly lower RPMs than the R/C motor even with a small gearbox. Anyway, if anyone has seen this done or knows of any show stoppers I'd love to hear your feedback.
-
Howdy! we're looking for donations to finish custom knowledgebase software for this forum. Please see our Funding drive thread
Direct driving the chainring with an R/C Motor
- Thread starter jd_3d
- Start date
Unless you use an ultra-low-kv motor, you'll have to use a reduction on the motor's output, because the chain would need to be driven fairly slowly to match the speeds pedalling would make it travel at.
Most RC motors are very high kV, meaning they spin very fast (in the hundreds to thousands of RPM), and if you use them at lower RPMs for the power levels they are made to run at the higer RPMs, they can quickly overheat.
A very basic, simplified version is that: 1kV is one rotation in one minute at one volt. 10kv is 10 RPM at one volt. So if you are using a 6s RC Lipo battery pack, that's about 24v near full charge, and would be about 240RPM at full throttle for a 10kV motor. But most RC motors are much higher kV than that.
Also, the lower the motor's RPM, the lower the power you get out of it, without overheating it, for the same winding and type of motor.
To get higher power with smaller motors you have to spin them at higher RPMs. For RC motors this typically means using a pretty extreme reduction on the output (10x or more).
There are a lot of ways to do this, and there are quite a few threads in the Non Hub Motor Drives forum section showing how it can be and has been done.
Another problem is that a 3d-printed sprocket providing torque against a chain is very likely to wear very quickly, (much faster than the plastic sprockets in a derailer, for instance, which is only guiding a chain).
You're probably going to have to make an adapter to bolt a regular bicycle sprocket to the RC motor shaft (or reduction output's shaft).
Most RC motors are very high kV, meaning they spin very fast (in the hundreds to thousands of RPM), and if you use them at lower RPMs for the power levels they are made to run at the higer RPMs, they can quickly overheat.
A very basic, simplified version is that: 1kV is one rotation in one minute at one volt. 10kv is 10 RPM at one volt. So if you are using a 6s RC Lipo battery pack, that's about 24v near full charge, and would be about 240RPM at full throttle for a 10kV motor. But most RC motors are much higher kV than that.
Also, the lower the motor's RPM, the lower the power you get out of it, without overheating it, for the same winding and type of motor.
To get higher power with smaller motors you have to spin them at higher RPMs. For RC motors this typically means using a pretty extreme reduction on the output (10x or more).
There are a lot of ways to do this, and there are quite a few threads in the Non Hub Motor Drives forum section showing how it can be and has been done.
Another problem is that a 3d-printed sprocket providing torque against a chain is very likely to wear very quickly, (much faster than the plastic sprockets in a derailer, for instance, which is only guiding a chain).
You're probably going to have to make an adapter to bolt a regular bicycle sprocket to the RC motor shaft (or reduction output's shaft).
Thanks amberwolf. I do plan on running some sort of reduction gearbox, just haven't figured out what exactly. The 3d printed parts will be printed in stainless steel so wear shouldn't be too much of an issue. I did find some threads that have been interesting (especially the cycloidal gearbox one). Now I have to figure out the electronics side as that is a big weakness of mine. Quadcopters seem to really be pushing down the price of ESCs, brushless motors, etc so I'm planning to take advantage of as much of that as I can.
Ah--that's not actually 3d-printing; it'd be machined (or lasercut or waterjet cut or plasmajet-cut). AFAIK the only 3d-printed metal is sintered stuff, and I don't know that it is strong enough for what you're after yet. (might be, never used any, but I don't recall stainless sintered)jd_3d said:
Keep in mind that all of those ESCs are designed to start a motor that is not under load--they start an unloaded motor that becomes loaded as the prop speeds up and wind resistance/etc begin to load it.
When you read around you'll find lots of destroyed RC ESCs on ebikes, because they start the motor under load, and the initial inrush currents tend to blow them up. :/ It's not just Castle Cremations either.
The ones used on friction drives aren't as big a problem because some of those start the motor unloaded, then it's own spin kicks or lowers it onto the wheel to engage and load it. Others start the motor unlaoded then a lever or other arrangement is used to engage it to the wheel.
Also, stuff for copters and other flying machines will be really high-kV motors, the opposite of what you need.
ScooterMan101
1 MW
This one perhaps ? ... https://www.kdedirect.com/collections/uas-multi-rotor-brushless-motors/products/kde7208xf-135
Or if it is a Road Bike, it would be easiest to just buy Kepler's latest Friction Drive.
Amberwolf makes a good point about the controllers , would starting from a stop just pedaling without engaging the motor until you get up to a few miles per hour take care of the controller problem ?
Or if it is a Road Bike, it would be easiest to just buy Kepler's latest Friction Drive.
Amberwolf makes a good point about the controllers , would starting from a stop just pedaling without engaging the motor until you get up to a few miles per hour take care of the controller problem ?
amberwolf said:
Actually, I really did mean 3d printed (I work as an engineer in a related field to 3d printing). The 3d printed metal parts are quite strong (slightly higher than cast parts, but lower than forged). I've used several 3d printed stainless steel metal parts on my bikes and they held up to years of abuse.
For the ESC I'm either going to use something like this:
https://hobbyking.com/en_us/turnigy-super-brain-100a-brushless-esc.html
Which seems to be able to limit current, which may help not burn out the motor. Or maybe I just spring for a quality VESC and learn how to set it up (seems a bit complicated, but I have no experience). Any input is welcome.
As for the motor, not sure yet, but there seems to be some nice skateboard wheel motors that would work well.
I didn't know 3d metal printing had gotten so far. 
As far as teh controller/motor/current thing, as I said before, the problem is more of blowing up the controllers, not the motors. (as long as you use the motor within it's limits, or actively cool it, since most of these are made to be cooled by fast propwash)
You can also overheat the motor, but the controller is what has a problem with startups--as noted previously, the controllers usually expect a current demand curve that is opposite to what you see on a bike.
I really suggest reading up on what others doing RC motor drives on bikes have done, so you can avoid any of the mistakes they've made and benefit from their experiences.
As far as teh controller/motor/current thing, as I said before, the problem is more of blowing up the controllers, not the motors. (as long as you use the motor within it's limits, or actively cool it, since most of these are made to be cooled by fast propwash)
You can also overheat the motor, but the controller is what has a problem with startups--as noted previously, the controllers usually expect a current demand curve that is opposite to what you see on a bike.
I really suggest reading up on what others doing RC motor drives on bikes have done, so you can avoid any of the mistakes they've made and benefit from their experiences.
@jd_3d: Did you ever get anywhere with the design on this one, or still researching?
amberwolf said:
Hey Amberwolf,
Since I was new to the electronics side of things and R/C motors in general I decided to start with a friction drive after reading the lengthy Kepler thread. Last weekend I spent a lot of time designing and refining the design and it's working really well so far. I created a build log here:
https://endless-sphere.com/forums/viewtopic.php?f=28&t=91293
I would be happy to get some feedback on it, or let me know if you have questions. It's still a work in progress but its going well so far.
