A Pedal-Pushing Drive Build

First of all, that's an awesome concept and 100 points for executing it.

Too many talkers in the world and not enough doers. Love it.

I have to ask though, why use a PMG 132? Was it a question of torque and relative gearing?

Those motors are capable of a peak of 14kw and have a stall torque of 200Nm..... Is there a bigger plan?

I'd love to see that thing run at full power.

If you need a test rider, I've got my hand up.

Either way it's totally awesome the way it is.
 
Greyhair said:
I've noticed that the twist-grip method can be a little tricky to control well during acceleration surges or hard braking. Has anyone settled on a throttle control that works well but that doesn't cause sudden inrushes of breath at tricky moments?
There are at least two kinds of throttles; with potentiometer types commonly used on things like the Alltrax (don't know what model you have, though) consisting of linear and logarithmic versions.

Often a log-taper pot will get you a better response, with little acceleration in the first half, but LOTS of power if you "floor" it. You just have to make sure it's taper is in the right direction for your throttle direction.

If using a hall-based throttle, there are different versions of those, too, and probably some that offer similar responses to the above taper types.

There is also sometimes a delay or startup curve you can program or set with an adjustment pot in the controller itself, so that any throttle input takes a moment to be fully implemented. Some have independent accelleration vs decelleration curves, while others like my 2QD use the same adjustment for both (it's just a resistor-capacitor circuit, easy to tailor to a specific use).
 
Greyhair said:
Hello All,
This is my first post, having discovered this forum about halfway through my project. The craftsmanship and ingenuity of the folks here is stunning. After watching the many creative and well-executed solutions posted here I hesitated to share my own "non-hub drive" with such a crowd of master builders.

OMG are you serious!!! you have absolutely nothing to be hesitant for, your work is inspirational and HUGELY impressive...fabulous job...Thats a Perm 132 or 80 driving Joules?

Glad you found the forum BiG WelCome to ES from me too :)

KiM
 
Let me belatedly welcome you, Greyhair to ES. Your work is truly mind boggling. It's technically and aesthetically brilliant, a true work of art. Joules deserves design and engineering awards! Bravo!

Ambrose
 
Very cool alternative connection to the bottom bracket. Saw your bike on www.gizmodo.com before I even saw you on here. Very amazing stuff!
 
Do a Google search on "Joules tandem" :)

You can track almost all of them back to the slightly defective link, to Carl's post, that I sent to the Bicycle Design Blog....
 
getadirtbike said:
I have to ask though, why use a PMG 132? Was it a question of torque and relative gearing?

The PMG-132 motor selection was a bit of over-planning on my part. I didn't want power issues to be the limiting factor in the design. As it stands, it's 'way overdone. Even with the Alltrax controller set to limit at 60 Amps, Joules can exert about 400 lbs on the pedals from a stop, or input about 3KW with the 48 volt system. The acceleration is startling. That's only a small fraction of what the motor can do, but I suspect that things would start to come apart (on the bike, Joules, or me) if it were set for more. If I had found you guys sooner, an RC type system would have been more sane and a whole lot lighter. I have to admit, though, that the prospect of wheelies on a tandem would be neat if the mechanics would take it and if my medical deductible was lower...

Carl

P.S: I made a website for the project with a few more pictures: http://www.rainyisland.org/joules/Joules_Home_Page.html
 
Beefing up the bike should not be that hard ( considering how you managed with Joules !! ) with freewheeling cranks for the human.. Would self ballancing without the training wheels be possible ? or too dangerous ? ( a guy on here by the title " Methods " would certainly have a go at 400 amps and 72v if you are looking for a pilot !! :mrgreen: )
 
Ypedal said:
Beefing up the bike should not be that hard ( considering how you managed with Joules !! ) with freewheeling cranks for the human.. Would self ballancing without the training wheels be possible ? or too dangerous ? ( a guy on here by the title " Methods " would certainly have a go at 400 amps and 72v if you are looking for a pilot !! :mrgreen: )

The first ride with Joules didn't have the training wheels. It was impossible to get on or off of the front seat without someone else to help keep the whole thing upright. At a stop, just turning around was very tricky alone due to the high C.G. and long wheelbase. The training wheels made solo rides possible and less scary. (a great product from fatwheels.com).

Joking aside, I doubt if I will beef up the bike for more performance-- the main objective of Joules was just to show local kids that thinking things up and making them is really fun. The training wheels makes it easy to demo and to run in place on blocks for that purpose.

None of the local kids seemed to ever make anything with their hands, much less go into engineering, and it was bugging me.

Carl
 
A pedicab trike, with Joules as the driver, would be cool. You could add the mechatronics to his arms, to do the steering, and control everything from the comfort of the passengers seat :)
 
Carl... fabulous work Sir! Yer thought to inspire young folk... wonderful.

Greyhair said:
I've noticed that the twist-grip method can be a little tricky to control well during acceleration surges or hard braking. Has anyone settled on a throttle control that works well but that doesn't cause sudden inrushes of breath at tricky moments?

Plan B might be a pedal-assist only. No hand accelerator at all. Then you just dial in more or less how much Joules adds to the riders pedal effort. If the rider doesn't pedal, neither does Joules.

tks
Loc

ps... In my jurisdiction Joules would not be permitted as the current legislation bans passengers on ebikes <hehe>
 
Miles said:
A pedicab trike, with Joules as the driver, would be cool. You could add the mechatronics to his arms, to do the steering, and control everything from the comfort of the passengers seat :)
Brilliant!

Or... Perhaps just add a rickshaw-style trailer and keep the tandem captain-stoker arrangement (technically, it is not an e-bike, it's a bike with a mechanical rider. 8) )
 
Very cool. I was hoping on seeing the robot do a massive sprint, but I guess thats not going to happen. Maybe it could sprint on blocks at like 200rpm? :)
 
Greyhair said:
I've noticed that the twist-grip method can be a little tricky to control well during acceleration surges or hard braking. Has anyone settled on a throttle control that works well but that doesn't cause sudden inrushes of breath at tricky moments?
This is a characteristic of the Alltrax controller, which uses speed control rather than torque control, so it tries to achieve a certain speed as soon as you twist the throttle. I have an Alltrax/Etek on my motorcycle and the takeoff can be exciting if you're not careful. You can dial the throttle down a little using the Alltrax software, but the response is still not ideal. I've heard that Kelly controllers are better suited for applications like this as they use torque control, which gives much more progressive throttle control.

I know exactly what you mean about trying to get kids interested in actually building things. That was what got me back into tinkering myself – building go-karts for my kids and their friends to play on. Soon they were coming up with their own wacky creations. Joules is a league or two beyond my own metal-bashing, and I'm sure he'll sow a lot of seeds.

Another vote for Miles' idea of Joules as rickshaw driver :)
 
I think I misunderstood the throttle control problem. :( Malcolm, I think you're right about why it's happening. I'm going to take a long nap. :oops:
 
Good ideas on throttle control from Amberwolf and Malcolm.

Your comments inspired me to look at the Alltrax throttle control program. By dialing the ramp-up rate (starting) to min, and the ramp-down rate (stopping) to max, the control is less hair-raising. I'll experiment with the curve that sets throttle potentiometer angle vs speed response, which may help as well. Thanks for the great suggestions.

Are there better hand throttle controls than the twist-grip? Needing to work the hand brake lever while assuring that the hand-grip twist throttle is fully off can be tricky. I'm sure that a good throttle solution ought to stop if you let go, but maybe a trigger-pull of some sort would work better with the brake?

Carl
 
Greyhair said:
'm sure that a good throttle solution ought to stop if you let go, but maybe a trigger-pull of some sort would work better with the brake?
All the twist grip throttles I've come across have been spring loaded so that they automatically return to off when released. Does yours not work this way?
 
Malcolm said:
Greyhair said:
'm sure that a good throttle solution ought to stop if you let go, but maybe a trigger-pull of some sort would work better with the brake?
All the twist grip throttles I've come across have been spring loaded so that they automatically return to off when released. Does yours not work this way?

Yes, it returns to off if you release it. What I've noticed is that when squeezing the brake handle it's easy to accidentally roll the throttle grip, since it's what you are also squeezing against. Perhaps a twist throttle grip that is not the full width of the handgrip would help a lot.

I also like Mile's idea of a controller cutoff switch on the brake levers.

Carl
 
Alternately, you could make a second old brake lever parallel with the brake to control the throttle. A light spring-return-to-zero could be easily built onto it, either a pull-spring from where the brake cable used to be or a push spring from the handlebar side onto the back of the lever. Just use a hall sensor and magnet to actually do the throttle control.

It'd be on the bottom of the pair, so squeezing it with fingertips is easy, and then easy to either let go of or to continue to hold as you squeeze the brake above it. Your option as to which one is closer to the center of the bars; for me that'd've been the throttle, as I only need to gently squeeze the tip of it, while the brake must be able to be fully gripped and squeezed sometimes. :)

Simply add a reed switch and magnet on the actual brake handle so that if you start to squeeze the brake lever it disengages the throttle.

That's one of the throttle controls I considered for DayGlo Avenger, and later for CrazyBike2, because it's easily done with parts already laying around. Only reason I've not used it yet is because I keep intending to use a chain-tension throttle control on the CrazyBike2, and am always "a little bit away" from working out the details (since I have to compensate for my bent-up offset chainrings and whatnot), and have a working "thumb" pot throttle of sorts.
 
Whoa!

That is really a nice piece of work! Would it work with 2 wheels only if you didn't have the problem of a high center of gravity?
 
Wayyy too much to say to express my enthusiasm, and how impressed I am with the workmanship on this project, Carl!

I'm a latecomer to discovering this thread (I somehow manage to overlook the cool stuff going on in the non-hub motor section of ES), but posted a link to Joules elsewhere when I noticed that Make! Magazine had picked up on you.

Mezmerizing to watch in action, and just the ultimate piece of engineered ART!

You could go "on the road" with that thing in more ways than one. It truly does inspire.
 
Back
Top