Why not friction drive?

I have a dual motor friction drive that I've yet to mount as it's too wide and heavy with a pair of wheelchair motors. I took 2 of the EV Warrior rollers and inserted them into one larger roller, so the clutch bearings in the EVW roller allow either, both, or neither motor to be in use. I want to swap the motors out for a lighter smaller unit, by using either 2 Kollmorgens with voltage boosted to 48V, or a pair of RC motors. I really appreciate the input in the thread, because now I know over 40mph is ok with friction, and I need to do some homework regarding a rig for variable pressure on the tire.

Going with dual RC motors, it might make more sense to use 2 of those EV Warrior rollers instead of the route I took, since their 1.3" diameter would put you at 38mph at 10rpms. The rollers themselves are some type of hard metal and have a diamond pattern I could never equal.

Regarding wet conditions, I remember 1 member who swears by friction drives and has used a friction drive wet and dry for years and still has the same tire. I think all that's needed is the right roller, though muddy conditions would obviously be a problem.

John
 
You can just parallel both controllers into a single servo tester/pulse width modulator and they will drive fine. No amplification needed. :)

You will definately want to run one roller with two motors for best efficiency.

Matt
 
John in CR said:
2 Kollmorgens with voltage boosted to 48V, or a pair of RC motors.

John, I've talked to a couple of people about dual Kollmorgens but I would personally just go with higher and higher voltages until I was happy with the power. 2 Kollmorgens both with external controllers would start to get expensive. I will say though, Kollmorgens do have a nice amount of torque and 2 would really be a kick.

Matt, good to hear that you can drive 2 controllers off of one servo tester. In that case I'll probably go with another cheap hall throttle with a voltage regulator to power it.
 
OK, I'm pulling the trigger.
I'm getting two of these: http://www.castlecreations.com/products/phoenix_hv_series.html

Two of these:http://cgi.ebay.com/RC-Airplane-Mod...photoQQcmdZViewItemQQ_trksidZp1742.m153.l1262

Now, Do I need two seperate batteries, or will one split to both controllers?

I think I will try the single roller, should I mount those motors directly to the roller, or use a freewheel setup between motor and roller in case the motors are not in sync? Should I be concerned about this?
 
EVTodd said:
John in CR said:
2 Kollmorgens with voltage boosted to 48V, or a pair of RC motors.

John, I've talked to a couple of people about dual Kollmorgens but I would personally just go with higher and higher voltages until I was happy with the power. 2 Kollmorgens both with external controllers would start to get expensive. I will say though, Kollmorgens do have a nice amount of torque and 2 would really be a kick.

Matt, good to hear that you can drive 2 controllers off of one servo tester. In that case I'll probably go with another cheap hall throttle with a voltage regulator to power it.

Todd

I already have 3 Koll's doing nothing, so it's just a matter of two 48V controllers that can handle the commutation rate, so for me it's about the cheapest thing I could do, especially since I already have the drive roller and housing that I made easily adaptable to accept most any motor pair.

John
 
Etard, I have to admit, you've lost me now.

The real reason I want to try two motors and controllers is because I'm cheap and I'm hoping to spread the amp load to avoid blowing one up. If you're going to get the castle creations controller (or two of them for big $$$) why not just use one motor? I'm sure that controller can handle the power from what Matt has said.

John, I hear ya. I tend to use whatever I have sitting around the garage too. If you do ever try the dual koll setup please let me know how it works.
 
For me the money isnt so much the issue, and all I hear is castle creations, nothing else works right. I want dependability.

Anywayz, I am having problems with mounting the roller setup because of pedal clearance. My remedy:

Attach the motor assembly to 2 carbon fiber rods that run to a mount at the rear dropout. The system will free float on top of the tire. I will use a conical roller (fat at edges, skinny in the middle) to keep it from walking side to side. I can set the axis where ever I want, but if I put it outside the axis it will act as Matts leading link idea and draw the motor into the tire for more traction. If you have ever seen a Softride bicycle you know that carbon fiber is torsionally stiff, light and strong. And the best part is that I will be able to remove the whole system (besides the mounts) with four screws in 1 minute.

Now, will someone send me a link for carbon fiber?
 
etard said:
I will use a conical roller (fat at edges, skinny in the middle) to keep it from walking side to side.

I like that idea a lot actually, You could even machine it to kind of match the shape of the tire. The only problem I can see is that it would be impossible to knurl a shape like that (at least I don't know how to do it).

You might be able to find some carbon fiber stuff at http://www.mcmaster.com. I get sheets of it there from time to time.

etard said:
For me the money isnt so much the issue, and all I hear is castle creations, nothing else works right. I want dependability.

Gotcha. But I still don't see the reason for two motors if money is no object. Some single outrunners can put out crazy power by themselves.
 
The only problem with a conical roller is friction. If it too closely matches the tire shape, you will have huge friction. I would make it flat with side dams (large washers) as flanges to keep it on the tire.

For CF tube, go with McLean Quality Composites. For flat panels, go with DA Graphite.

I have been working with carbon fiber for years. I have gone through tens of thousands of dollars worth. Those are the best suppliers I have ever found.

Oh, pulltruded CF tube is not torsionally strong. It twists alot. I use it for helicopter tail booms. You really need woven CF tube for torsional rigidity. However, it is very difficult to find woven CF tube in small diameters.

If you have any specific CF questions, PM me and we can talk. I know how to wet layup (without vacuum bag), I do CNC CF cutting, and I know alot about joining CF with various materials.

Matt
 
Evtodd,
See thats what I was thinking too, just piggyback a single motor with either a belt drive (ha, I get another free reduction in gearing!), or with direct gears (noisy?).

Then I thought about the loss of efficiency of any drive......
Will the extra reduction make up for it in torque?
Or is a dual motor system already in theory less efficient because of syncing issues?

But now I think about how the contour of one of the AXI outrunners between those of the roller I could make it look like a shark fin. Beware, a Landshark!! Haha!

Matt,
Thats good news, I want to do simple flat bar, that operates like the leaf spring in a car. Will you recommend a thickness and width for use on 26 inch wheel?
Remember my setup is detachable, so I am not worried about freewheel friction. With that in mind, does your previous statement concerning friction still stand?
 
recumpence said:
The only problem with a conical roller is friction. If it too closely matches the tire shape, you will have huge friction. I would make it flat with side dams (large washers) as flanges to keep it on the tire.

That is a good point about additional friction. I will also say that I've never had a roller even try to come off of a tire yet. I have braces from my mount to my frame but like I said before, that's just to make sure it won't move over a large bump or pot hole. My first try at friction drive was mounted to a cheap seatpost rack with no extra support and it never moved.

If you use an EV Warrior roller and leave a little room for movement it shouldn't be an issue since it floats on a roller bearing it pretty much self centers.

I say start with something simple and see how it works for you. I think you'll be surprised (yes, I know I keep saying that) at how good it works without getting overly complicated with the setup.
 
I would say 4mm would be a good thickness. Maybe even 5mm. But, anything less than 4mm thick would be wet noodle at a foot long unless you made the leaf really wide.

I love the friction drive idea. I think you may find lot of things that work well and a few that won't. Just keep working at it.

Matt
 
So I have one RC SMART 560 KV motor, 80amp ESC, and I've contacted DA Graphite about what I need strengthwise (carbon fiber is relatively cheap). I have some gnarled bmx pegs coming, not sure of the diameter. I went ahead and also ordered some Goped drive spindles which come in sizes .600 inch to .800 inch. Those are available with grit, but I just got normal (on sale for $5). I also got a Maxxis training tire(for stationary training rollers) that is supposed to be long lasting.

My friend that works for a titanium medical instrument manufacturer says he can CAD and waterjet both the axle mounts and the motor housing (I wonder how much that would cost in titanium). The beta will be aluminium, but its wide open after I get good results.

I dont know how to calculate the belt drive because I dont know what roller will work in which diameter it will be available in. I was wondering how a belt drive acts above 10,000 RPM, does anybody know if this is too fast or noisy for belts? Also, I'm not sure this motor will put out enough power for what I am looking for. Since that controller only accepts >30 volts, I am thinking about ordering two Milwaukee (28 volt) batteries, with 2 chargers that would give me an hour long recharge, with a total system weight of under 10 pounds. It should be one of the first entries into the bantum weight class some are talking about.
 
Most toothed belts run fine up to 20,000 RPM. I run mine at 6,000 RPM. Typical rule of thumb is 20,000 rpm for fine pitch, 10,000 rpm for course pitch.

Belts are generally silent unless they are really wide.

Matt
 
Etard,

If you get a chance check out the EV Warrior roller. It's by far the easiest way to get a freewheeling setup (and you really have to have a freewheeling roller imo). You can pick them up on Ebay 2 at a time for very little money.
 
EVTodd said:
Etard,

If you get a chance check out the EV Warrior roller. It's by far the easiest way to get a freewheeling setup (and you really have to have a freewheeling roller imo). You can pick them up on Ebay 2 at a time for very little money.

Better yet, go straight to the source http://www.industrialliquidators.com/wesell/index.html . I've bought stuff direct from them several times, and will again I'm sure. All of the EV Warrior parts are solid quality, and their white nicad batts are workhorses. The panasonics are old stock (which I don't believe affects nicads greatly) and they quoted me $1 each for the 110 that were left the other day. At only $0.29/wh shipped, the weight that costs so dearly for my forward shipping from FL to Costa Rica, was the only thing that prevented me from snapping them up. I read about 1 guy still using some nicads he bought back in the 80's.

John
 
John,

I have also ordered directly from their site. I've always had good luck with them. I do wish they would ship the rollers via USPS instead of UPS since it would be cheaper but you can't have everything you want :D . I've also purchased the ev warrior thumb throttle in the past. Great price if you need a pot throttle.

I'll have to check out the nicads. I never ran across them on their site.
 
Thanks for the connect on DA Graphite, he was very helpful. Its nice to talk to someone interested in your project rather than focusing only on the big fish with tall coin to spend. We had an interesting talk about how, for a relatively cheap substance, carbon fiber is a seldom used material. Oh well, its perfect for my application, and its gonna make my project look so much better to serious (read LYCRA) bikers. Anyway, he is sending some samples this week, I decided to go as thin and small as he felt necesary, then I can just double it up if I need more stiffness.

Can anybody point me to some math calculators for the belt drive. My wheel is 26", if I use the EV Warrior roller its1.3", and my motor is 560KV at 28 volts. What 2 pulleys do I need? I'm going fine pitch.
 
etard said:
Can anybody point me to some math calculators for the belt drive. My wheel is 26", if I use the EV Warrior roller its1.3", and my motor is 560KV at 28 volts. What 2 pulleys do I need? I'm going fine pitch.
Calculators are available to download from the Gates and Contitech sites. There's also MITcalc: http://www.mitcalc.com/en/pr_tbelts.htm
 
John in CR said:
[

Better yet, go straight to the source http://www.industrialliquidators.com/wesell/index.html .


The Super twin gear drive looks interesting... 4 motor setup anyone?
 
AussieJester said:
John in CR said:
[

Better yet, go straight to the source http://www.industrialliquidators.com/wesell/index.html .


The Super twin gear drive looks interesting... 4 motor setup anyone?

Jeez, don't tempt me. I was just looking at the Sparc hub motor and wondering if 4 Mabuchi 540 motors would put out much power with friction drive. Not sure how you would control it though, could you just run potentiometers for speed control?
 
540 motors are good for 180 watts each. There would be a bit of loss through the gears. Also, you do not get a direct 1 to 1 power output when you double or quadruple the motor count. Still, I bet you could get 500 watts from 4 of them. The best way to control them is to parrallel them into one brushed controller. that would not cost much.

Man, for $20, that is one nice gearbox!

Matt
 
Back
Top