Split-crank Electric Tricycle

[RoboBellman]

Hello all,

I'm a PhD student at the University of Florida with a unique project for which I need your expertise.

I have taken a TerraTrike Rover and have split the bottom bracket assembly to make it a split-crank trike. There are now two cranksets, each independently driving the rear wheel through a customized version of this hub from Staton-Inc with a standard 16T freewheel on the right-hand side (RHS) and one of these left-hand threaded 16T freewheels on the left-hand side (LHS). In addition, each of the two driving cranksets up front have an SRM PowerMeter embedded in them, so I can measure torque applied to the each chainring by its respective crank arm.

Here's a picture and a video of the different ways you can pedal this beast.
http://i.imgur.com/iOwSBl7.jpg
http://youtu.be/3hk_Vkcji1Q

What I'm trying to do now is to motorize both sides of the trike. That is, there's a drive chain on both sides, and I want two electric motors with at least 250 W output each that can assist pedaling for both sides (the idea being that one motor could drive the trike around at a reasonable speed on flat ground). Because of the complexity of the bottom bracket, I don't think I can use a bottom-bracket-mounted kit (I've used a GNG mid-drive kit on another, non-split-crank trike). Because of the complexity of the rear hub and the dual drive feature, I know I can't use a hub motor. That leaves me, I believe, with a mid-drive setup, where there is a motor attached to either side of the frame with its output on the drive chain, between the front crankset and the rear hub*. I think the best place to implement this is behind the seat, which looks like this: http://imgur.com/lvjOLFZ

So, my question to you experts is: How would you motorize this monstrosity, given the constraints I've mentioned? My thought right now is to buy two of these side-mounted kits from GNGelectric and modify as necessary.

What are your thoughts? Suggestions? Comments?

Thanks for your time and any assistance you can provide.

Matt

*EDIT: "flywheel" to "hub"

 

[jag]

Is the purpose of the drive just for aiding with propulsion? Or does it specifically have to drive both the left and right chain and cranks (as you suggest with the proposed application of two GNG drives), e.g. for rehabilitation of a person with lower body disability?

If you tell us a bit more about your application and intended methods we might be able to help better.

For propulsion a number of solutions are possible. You can drive the front wheels with off-the-shelf hub motors intended for- and pre-made with a single sided shaft. At least Ezee and Crystallyte have provided these in the past. Any hub motor can potentially be adapted/used, so you might chose a small and light one.

You can drive the rear wheel through a separate pulley. There are HTD5mm, 9mm width pulleys that will fit on the spoke circle of the 20inch wheel.

However, I'm guessing you have some specific purpose with the split chain drive to measure and drive each leg separately. For independent left - right chain drive you can use the GNG you mention. More noisy, but easier to mount are Cyclone solutions (these are small industrial Transmagnetics brand servo motors attached to planetary gear heads (all metal gears - somewhat noisy).

You can also use hub motors (quieter) as mid-drives, either in-line or with a split chain. You can mount cogs on the freewheel side (obviously) and also on the disk brake side using an adapter, giving many different configuration options. You can either use, bypass or disable the built in clutch. A split chain gives you more gearing choices. You can e.g run 42:12:16 to run the hub at a rpm than the rear wheel and therefore higher torque.

Regards,
Martin

 

[RoboBellman]

Thanks for your reply, Martin.

The purpose of the drive is, as you said, for rehabilitation of a person with lower body disability.

The application is specifically for cycling induced or assisted by functional electrical stimulation (FES), which can be used to artificially induce muscle contractions in paralyzed muscles to yield functional motion (e.g., cycling). Ultimately, the goal is rehabilitation and mobility for people who've had a spinal cord injury, stroke, etc. For this application, I do want the motors to both drive the rear wheel and turn the cranks.

As this is a research project and not for commercial purposes, noise, weight, ugliness, etc. aren't an issue.

I think I'll give Cyclone a call and see what they think. It'd be easier to work with them, since they're based in the US.

The hub motor option is one I hadn't really considered. I'm not sure exactly how I would mount them, but I'll do some more digging in that direction. Do you know of any examples of using a hub motor as mid-drive on a trike that I could look at?

 

[Warren]

Matt,

You have quite a challenge there. Reminds me of PowerCranks,

http://www.powercranks.com/Productdescription.html

only way more complicated. Glad to see somebody working on FES. There was somebody in Europe working on it about a decade ago. I remember one of the researchers was paraplegic. There was some correspondence, I think in the old Bike Culture Quarterly Journal.

http://www.cyclorama.net/blog/cyclorama/encycleopedia-bike-culture-and-bycycle/

It is my understanding that you want the motors to turn the cranks as well as the rear wheel. But do you require the bike to move under motor power with no movement of the cranks, or only when one or both cranks are turning?

 

[RoboBellman]

Warren,

I almost bought a set of the PowerCranks for this project, but they aren't "backdrivable," i.e., a motor couldn't drive the cranks.

Regarding FES-cycling, most of the work that's been done recently has been by researchers in Europe, like Ken Hunt, Nick Donaldson, and some others. There are two companies, BerkelBike and Hasomed, both in Europe, that sell mobile FES-cycling devices, though there doesn't appear to be a great effort to market them. Thank you for the link to Cyclorama -- it's something I'll keep an eye on.

You are correct in your understanding that I want the motors to both turn the cranks and drive the rear wheel. I do not require that the bike move under motor power without movement of the cranks. If the LHS motor turns, I want the LHS crank to turn and the rear wheel to propel. Same for the RHS. On the flip-side, if either of the cranks are moving, the rear wheel will propel, but the motor will not need to be engaged (nor do I want the cranks to have to drive the motor, which is easily done by making the motor output a freewheeling cog, I think).

 

[amberwolf]

Do you know of any examples of using a hub motor as mid-drive on a trike that I could look at?

There are many, some well-documented here on ES, though they may not do it the way you need you might adapt them.

I don't have a link but off the top of my head I know Rassy has a 9C hub middrive on his tadpole.

 

[Warren]

Matt,

Amberwolf is right. Razzy's trike's

http://www.endless-sphere.com/forums/download/file.php?id=54774

and the new Stokemonkey design

http://www.ebikes.ca/media/wysiwyg/RSD_Kit_Installed.jpg

use a direct drive hub motor with a standard freewheel to drive the chain, wrapped around an idler, to get correct rotation.

At the speeds I assume you would want in a low gear, even 250 watt motors will be overkill. Those GNG Currie units should work OK, but its going to end up a bit more cluttered than the above mentioned method to drive those chains. I would suggest using the smallest front wheel, geared hub motors you can find,

http://www.ebikes.ca/shop/conversion-kits/geared/front-mini-geared-kit-advanced-pas.html

and disc cogs

http://www.velosolo.co.uk/

to run a setup like those above.

 

[Drum]

Hi Robo.
Using two small geared hub motors with freewheels to drive the chains in the Stokemonkey arrangement sounds like a good option. If you wanted the installation to be symmetrical you would need one left-hand and one right-hand freewheel and the motors would rotate in "opposite directions" because they would be facing in opposite directions but both driving forwards.
With this setup, the chainline is not altered.. the same chain runs around the sprocket on the motor and continues to the back wheel.

Another possibility would be to use two geared motors of any type but mount a freewheel on the output shaft of each, with two sprockets side by side on each freewheel. Each motor shaft would act as a jackshaft.. the pedal cranks would drive a chain going back to one of the sprockets on the freewheel, and the wheel would be driven by a second chain on the other freewheel sprocket.
When pedaling only, the front chain would drive the rear chain through the linked sprockets, the motor shaft would be stationary and the freewheel would ratchet.
When the motor started rotating, the freewheel would engage, driving both the pedal crank and the back wheel.
This would interrupt the chainline as the sprockets would be beside each other, but on drive chains as long as a recumbant trike, the small sideways shift would probably not be significant. If it was an issue, the cranks could be moved slightly further apart to allow for it.

The physical size of the hub motor and the required mounting arrangements might be the deciding factor in which setup would be easiest to fit.

Sounds like a really interesting project in several ways. Good luck with it.
Dave

 

[Warren]

Dave,

Actually, you can have two small geared hub motors sit symmetrically, say with the motors near the centerline, with the disc cogs on the outside. They will be running in opposite directions relative to the direction of travel. But this only affects whether you pull the chain from the top run, or from the bottom run, after it has rounded the chainring.

 

[Drum]

Quite right Warren..
Obviously I hadn't been awake very long when I posted this morning!!
Yes, just run the chain over one and under the other. Will still need freewheels to allow pedaling without having to turn the motors, but allow the motors to drive the chain when powered up, if that isn't a standard feature.
The geared hub motors will probably suit the application well as their output rpm and sprocket size will probably match the application without modification.
Dave

 

[Warren]

Dave,

"Obviously I hadn't been awake very long when I posted this morning!!"

Don't worry. I correct myself on the interweb all the time.

"Will still need freewheels to allow pedaling without having to turn the motors'

The freewheel is already built into a geared hub motor, and it's better than the external ones. They typically use either silent sprag or roller bearing clutches, rather than ratchet and pawl. The outer hub shell will turn, but not the planetary gears or motor. I find the weight of the shell unnoticeable when pedaling unassisted.