An Electric Bicycle

[ivansgarage]

Here is a great place for dc motors.  http://www.robotmarketplace.com/products/npc_ungeared_motors.html

I have been running the black maxx on my bike for the last 6 years.
http://ivanbennett.com/e-cycle.html

[vtsteam]

The heat load if the above calculation is correct will be ~60 watts at full power. So think of a 60-75 watt light bulb.

On the plus side for cooling this is a relatively open frame motor made largely of aluminum with an outrunner spinning around, and the bike is traveling, so in early tests just put a hand on the motor perodically -- the best temperature gauge for this kind of thing. And if it feels hot, add a fan.

Steve you mentioned questions about motor torque. Given the RPM and assumed power, seems it should be possible to solve for torque. Looking it up on the internet I found the equation:

power = torque x 2pi x rotational speed

so

torque = power(watts)/ (revs per sec * 2pi)

so

torque = 500/ (5700/60 * 6.28)
torque = .84 Newton meters (or .62 foot pounds)

A 26" wheel at 30 mph is doing 388 rpm (if I also got that right --- better check that, too) so your final reduction is maybe 5700/388  or 14.7 to 1.

If you had a 3 to 1 sprocket ratio on the chain drive, your gearbox ratio would be 14.7 / 3 or about 5 to 1.

So torque out of the gear box would be about 5 x .84 = 4.2 Nm or 3.1 ft lbs at that speed.


Could be a lot of incorrect assumptions or outright units blunders here.....  take with a grain of salt!     

[S. Heslop]

Here is a great place for dc motors.  http://www.robotmarketplace.com/products/npc_ungeared_motors.html

I have been running the black maxx on my bike for the last 6 years.
http://ivanbennett.com/e-cycle.html

Hey I remember finding that website when looking electric bike stuff up. It looks cool as heck but i'm not sure on the legality of a purely motor powered bike in the UK. In the UK Segways are illegal to ride on both paths and roads (not that i'd ever want a segway).

I was looking at youtube again last night and found this video. He doesn't show much but it looks awfully similar to what i'm planning. I did think about mounting the motor there originally, but I didn't like the look of the motor being so exposed.

Looking a bit further this website looks like it could have some good information on it: http://endless-sphere.com/forums/viewtopic.php?f=28&t=29531


Keeping the motor cool was something I was concerned about. Another thing that bothers me is maybe getting my parachute pants tangled around the spinning outer case, so maybe a shroud around the motor won't be a bad idea.

If I figured it out correctly the free wheel should work with it the right way round. It still shouldn't be too much of a problem to flip it though, i'd just need to cut grooves to help tighten it.

[S. Heslop]

The heat load if the above calculation is correct will be ~60 watts at full power. So think of a 60-75 watt light bulb.

On the plus side for cooling this is a relatively open frame motor made largely of aluminum with an outrunner spinning around, and the bike is traveling, so in early tests just put a hand on the motor perodically -- the best temperature gauge for this kind of thing. And if it feels hot, add a fan.

Steve you mentioned questions about motor torque. Given the RPM and assumed power, seems it should be possible to solve for torque. Looking it up on the internet I found the equation:

power = torque x 2pi x rotational speed

so

torque = power(watts)/ (revs per sec * 2pi)

so

torque = 500/ (5700/60 * 6.28)
torque = .84 Newton meters (or .62 foot pounds)

A 26" wheel at 30 mph is doing 388 rpm (if I also got that right --- better check that, too) so your final reduction is maybe 5700/388  or 14.7 to 1.

If you had a 3 to 1 sprocket ratio on the chain drive, your gearbox ratio would be 14.7 / 3 or about 5 to 1.

So torque out of the gear box would be about 5 x .84 = 4.2 Nm or 3.1 ft lbs at that speed.


Could be a lot of incorrect assumptions or outright units blunders here.....  take with a grain of salt!     

Thanks for doing the calculations. To be honest though i've got no real frame of reference for those values, and Google isn't giving me any useful information. Yeah though with this gear reduction, and the motor running at 6000rpm, I got about 370 rpm on the 26 inch wheel.

[S. Heslop]

I mocked up the placement of the motor today with a block of wood, and it turns out it was in the way of my legs. So i'll have to mount the whole thing elsewhere.

Eventually I came up with this over complicated monstrosity.



It has a v-belt that might serve as a clutch and reduces the rpm of the wheel down to 250, but it might not be so efficient.

[tom osselton]

Could you make a expandable pulley to vary the ratio? that way you would not need a transmision.

[S. Heslop]

Could you make a expandable pulley to vary the ratio? that way you would not need a transmision.

They're too expensive but it's just made me wonder why I don't just replace the sprocket on the rear wheel with a huge pulley so I won't have to bother with gears.

[vtsteam]

Belts waste a lot of power, and there isn't a lot to spare. Chain and sprocket would be better.

[vtsteam]

But if clutching is a necessity, I also think that the friction drive against tire or hub has got to be the simplest of all, with a really good reduction ratio -- probably a 2" dia drive would get you in the ballpark, but easily changed-- maybe from a boat rubber trailer roller.

And this has been done on bicycles a lot -- I think I remember them in France. Aftermarket adaptations of standard bicycles I believe.

[S. Heslop]

I also remembered why I gave up on pulleys earlier. I don't have the space between the wheel and frame! Or anywhere good to mount the motor.

     

Found this from googling a bit. I didn't think about using a motor on each side but that seems expensive. Although maybe not much more than spending a fortune on metal...

Can't find any videos of the thing actually pulling a person though.

[vtsteam]

Steve, what about an idler on the other side? You'd still get the pinch action....

[S. Heslop]

Steve, what about an idler on the other side? You'd still get the pinch action....

I feel silly for not thinking of that myself.

Yeah this seems like a better idea now. It's gotta be easier to build than the whole stupid system I came up with.

[vtsteam]

This looks like maybe a tire drive:

[vtsteam]

A little more searching -- this is what I remember (and from the pic, looks about the right time, too:



Velosolex

[S. Heslop]

Looking on youtube it looks like people even produce and sell kits to turn chainsaw motors into tyre drives. Seems like it'd be noisy to ride with though.

Just measured the front forks of my bike. The second-hand bike I bought has some real huge aftermarket forks.

[vtsteam]

Here's walkaround and a ride on a Solex.

I detect a reluctance to stop for any reason once moving.....

Does look like fun though!

[dsquire]

Here's walkaround and a ride on a Solex.

I detect a reluctance to stop for any reason once moving.....
.
.
Does look like fun though!

Steve

I think the only reason he ever slowed down was to prevent  a "T" Bone. 

I am enjoying following along on this one. 

Cheers 

Don

[tom osselton]

I think the driving friction would wear out the sidewall too fast it's better to have it ride on the tread,  depending on what tire you use some have fairly deep distinct treads offering possible tire/drive roller engagement. A buddy at work had a weedwacker motor on his bike and drove it from Calgary through the mountains to Vancouver bc from there he started heading north up into the Charlotes.  It would be a hell of a holiday!

[tom osselton]

Well it looks fun but I was waiting for a car door to open    he was cutting it close to most vehicles or so it seemed  but could just be the camera.  Might be a candidate for the Darwin award!  One thing to remember is they were not designed to stop at any great speed!

[vtsteam]

Yes Tom, Don, reminded me of of The French Connection in a sort of 20 mph bicycle way, if I can put those two things together....!

I kept pulling my right leg in to prevent hitting a parked video car!

Definitely not sidewall drive as a possibility vs tread, Tom -- I was thinking of the metal rim and a rubber drive.

Steve, Just as a gut feeling I think 500 watts isn't going to do 30 mph, and so the ratio should be aimed at slower speeds w/ motor running ~ 6000 rpm. If you optimized for say 15 mph @ 6000 rpm, you could run a killowatt for higher speeds without changing ESC or motor at your projected 24 V batt supply. That would be about 40 amps x 24 V. Your ESC is a 60 amp controller.

So say bike wheel is turning somewhere around 185 RPM and motor 6000 rpm, and work out the ratio from that, or even lower speed.

I think if you do wheel drive the rubber driver is going to end up somewhere closer to an inch plus in diameter then -- but I haven't done the math.

[S. Heslop]

I bought a 100A controller but i'd still like to keep the amps as low as possible, with it being designed for RC planes most likely. So yeah 15mph might be a good bet.

A nice thing about this drive idea is that I can easily just switch the contact wheel out for something smaller if needs be. To reach 185 rpm on the outer wheel the contact wheel would be 40mm diameter. Although at that size the motor would be getting uncomfortably close to the spokes so i'm going to try 50mm, giving the bike wheel an rpm of 230.

One thing i'm wondering is how do things work in terms of using the speed controller to run the motor slower than 6000rpm? As in, would it put more or less strain on the controller/ motor?

[vtsteam]

Steve, the ESC does run the motor slower than 6000 RPM when the throttle is reduced -- the ESC changes its timing to create the reduction in speed.

The main strain on the ESC is if the amperage exceeds its rating.
It takes about 24 volts to turn this motor at 6000 RPM.
How much current it takes depends on the load at a specific RPM
It might never reach 6000 RPM if the load is too great before that point.

An ammeter and a fuse are important to have in line while testing. Check your current draw with the particular setup as you test. Gradually increase throttle while watching the ammeter. Stop when you get to 20 amps (or whatever you decide) This will give you a maximum operating speed for the current limit that you want to specify.

Optimizing your setup by test would would mean that, if the maximum safe operating speed of the motor, by your specified current limit, is considerably below your 6000 RPM target, your gear ratio should be increased until they correspond. This will reduce the top speed of your bike until it matches your power specifications (500 watts @ 24 volts).

Any further speed increase will require more power.

If you are running under 6000 RPM at spec'd amps you are not sending effective full battery voltage to the motor, and are therefore running less than 500 watts. You are running at whatever amperage you show times the effective voltage.

ps. that's my understanding -- could be a mistake in it -- others chime in, please!

[vtsteam]

Let me just simplify that:

Under test with typical load on the bike with a 24V system, approx 500 watt desired, 24 amp spec limit on a 250 kV motor:

Test 1: Ammeter reaches 24 amps at 2000 rpm. Servo controller knob (throttle) is not at max.

Conclusion: gear ratio too low, wattage less than 500 watts

Solution: reduce diameter of rubber spindle on wheel. Bike speed should actually increase (even though gear ratio is increased) because you are absorbing more power.


Test2: Ammeter reaches 24 amps at 6000 rpm. Servo controller knob is at max (full throttle).

Conclusion: gear ratio correct for wattage. Full 500 watts available. But bike is still too slow for satisfactory performance.

Solution: Increase spec for amperage and wattage. Since ESC servo controller is dialed all the way up in this case and you have no more throttle range, you will need to decrease the gear ratio.

If you do that, 6000 RPM (and full throttle on servo controller) will then be reached at a higher figure than 24 amps.

[S. Heslop]

Ah, thanks for clearing that up. Shame the speed controller is coming all the way from Hong Kong, that could take a good while yet.

[kayzed1]

I have a pair of motorcycle sprockets here if you want them, not new but good, 420 chain size.
Lyn. 
PS: may even have a new DID chain as well..