New friction drive concept...

Discussion in 'Friction Drive' started by loquin, Oct 8, 2009.

  1. loquin

    loquin Active Member

    thought about this a week or so ago, and have been tossing it around in my head... (yeah, I know - that's NOT a good place to keep ideas!!!:ack2:)

    Gates and other belt manufacturers make toothed belts that are toothed on BOTH surfaces. What if you build a friction drive with a toothed drive pulley and a driven, idler pulley about 6 inches apart. Use a 2 inch wide (approximately) belt, with the tension JUST snug enough to keep the belt from coming off.

    Lower the assembly down to the tire, so that the outer toothed belt surface contacts the tire. As you increase the pressure, the belt deflects some, allowing a much larger surface to be in contact with the tire surface than with a single roller. (and, the belt tension goes up, keeping the belt on the rollers, under load.) There will be less deflection of the tire, resulting in lower internal power loss. (standard friction drive drive tires essentially have double the 'rolling friction.') The system should perform better when it's wet, also. And, the pulleys should have minimal wear, so you could use lighter weight, less expensive extruded aluminum pulley stock, rather than steel.

    Speed calcs would be the same as any other friction drive, but the diameter used in the calc would be the diameter at the center of the belt.

    Ref the attached sketch.


    Attached Files:

    Last edited: Oct 8, 2009

  2. Seems like a great idea to me. Time for you to get to work and build one to test it out. I would do it myself but I've just finished a GEBE belt drive bike and I'm busy working the bugs out. Good luck. Let us know how it works and when you start your new business building and selling them.
  3. eastwoodo4

    eastwoodo4 Member

    i think your gonna lose to much power.the bigger surface area of the belt will create more drag than the steel drive roller being pressed into the tire.the belt will slip to unless its pressed against the tire pretty the tensoin of the belt on the pulleys is gonna cause drag to.i say no go but would love to see you prove me wrong.your probably gonna have to special order that small belt.
  4. loquin

    loquin Active Member

    But, it's not the surface area that causes the drag - it's the deflection of the tire. Flexing the tire surface takes some work. If your air pressure is up, and you can press down with little deflection, the drag is less.

    Where the increased surface area should help is that it would provide more 'grip' between the motor and the tire, with less downward pressure leading to less deflection.
  5. moondog

    moondog Member

    I like it. You could use two motors. 2 electric or 2 gas or 1 gas and 1 electric motor.

    Also, the pulleys would not have to be the same size. The 2 motors could have different gearing.
    Last edited: Oct 8, 2009
  6. ibdennyak

    ibdennyak Guest

    Wow, thanks Lou. I shall have to shamelessly steal that idea from you. In Juneau Ak with its 320 days of rain, friction drive is pretty hopeless unless you run a large pneumatic tire as the drive roller. It works, but is pretty cumbersome from a size point of view. The belt idea is worth pursuing. Kind of like putting tracks on a snowmobile.

  7. spad4me

    spad4me Member

    That looks nice!
    Last edited: Oct 9, 2009
  8. Mountainman

    Mountainman Active Member

    going to be a lot of loss of HP driving that thing
    but if you have a little extra HP to wast
    may just be a good unit
  9. moondog

    moondog Member

    You could put 4 motors on it. 4 electric motors. Each one with a clutch. Power up 1, 2, 3, or all four of them for max power !
  10. loquin

    loquin Active Member

    The cog belts are almost as efficient as a chain in transferring power, MM, and the two extra bearings won't rob much at all... There should be a net increase in power when compared to a standard friction drive roller pressing into the tire.

    While the power loss in a standard friction drive is proportional to the amount of tire deflection, the amount of torque available to push you along is proportional to the amount of surface area between tire and roller. In a standard friction drive, these two goals oppose each other. In order to reduce slippage when wet, you have to deflect the tire as much as possible, in order to increase the amount of surface area (and increase the pressure between roller and rubber, both of which act to increase static friction between tire/roller) BUT, this also increases frictional losses due to tire deflection/flexing, and resultant heat in the tire.) On the other hand, in order to reduce losses and increase performance, you reduce the tire deflection, but, this raises the possibility of roller slippage.

    After a while, we get used to 'setting' the roller pressure - we press down on the drive frame and lock the drive in place. If it's going to be wet, we press down harder. If it's dry, we don't press down so much. But, learning how much pressure we need can be a learning experience, that can cost us in eaten-up tires.

    I really don't have the time to work on it now, so, anyone who wants to, can use the idea. (IF you start making money with the idea though, please build one for me!!!)
  11. loquin

    loquin Active Member

    There would be complexities when you want to change the gearing, though.

    If you make the drive cog bigger, then
    • the idler pulley would have to be smaller, so that the same number of teeth would be engaged by the belt, or
    • the belt size would have to change, or,
    • the idle pulley position would have to shift towards the drive cog.
    Having a large adjustment on the idle pulley probably makes the most sense.
  12. augidog

    augidog New Member

    interesting concept, lou...i see possible trouble with maintaining constant equal pressure in the gap between rollers, i think it may turn out that the only working friction will be the two tangents of the rollers...however...imo that's not a bad thing: two rollers working in conjunction, tied by a belt. simple tensioning-adjustment allows for gear changes.

    sketch this one out, a bit complicated but still simple: lay a vertical shaft engine on a triangular plate over the back wheel. two rollers (geared for opposing rotation, or one could simply be an idler for proper pressure) on each side of the tire. they wouldn't have to be knurled metal or treated surfaces, rc/car or small kart tires would work, i don't think there'd be any wet slippage problems with that. gear changes would easily happen at the clutch output.

    anyways, that's just a crazy idea that's been in me brain for a while now... :ack2:
    Last edited: Oct 9, 2009
  13. moondog

    moondog Member

    We got a kit from Boygofast that had an 1 1/2 steel roller. It has grooves across but not knurled like the others.

    It does real well in the rain. I think larger rollers work better but you need a big enough motor to turn it.

    Gearing it down with sprockets would be a way to use a smaller motor with a larger roller. Deacon uses a scooter tire for his roller but it is geared down with sprockets.

    Your two rollers could be just a few inches apart and both apply pressure to the tire and the belt would push against the tire.

    Seems like this would be max friction drive traction.

    I went to sleep thinking about this and woke up thinking about it !
    Last edited: Oct 9, 2009
  14. liscos439

    liscos439 New Member

    I've got a Sinclair Zeta that works like that.
    It's a bit naff but I think it's because of the weedy motor rather than the belt.
    Oh, and the side of the belt that touches the tyre is smooth.
    I can't say how good the grip is in the wet because there's not a lot protecting the motor from the elements.
  15. augidog

    augidog New Member

  16. Rats, we will have to get permission from Sir Clive Sinclair to use his patented belt drive.
  17. loquin

    loquin Active Member

    If it were patented in the U.S., you would if you were building them for sale. Now, if the patent has expired, anyone could make them. Sinclair's Zetas were manufactured beginning 12-13 years ago, and it probably took several years from issuance of patent until manufacturing began. In the U.S., there's a 17 year life on patents, so, in all probability, either the patent has expired, or, it is due to expire shortly.

    I had no idea that anyone else had come up with something similar, though. Hats off to Sir Clive.
  18. moondog

    moondog Member

    Yes, thanks for showing that ! I am learning about friction drive.

    Like it a lot.

    A friction drive section would be nice. Is there one anywhere else ?

    I am looking for a small lathe to make parts for friction drive tests. :idea:

    great thread here.
    Last edited: Oct 12, 2009
  19. Tinker1980

    Tinker1980 Guest

    Not too crazy an idea. It's been done. Have a look. I'm nearly done building such a creation myself, I'll post pictures when it's done IF I can avoid getting flak about a couple of amatuer-ish welds :dunce:

  20. loquin

    loquin Active Member

    The problem with the vertical drive roller is this: The roller makes contact with a chunk of the tire rubber, vertically. The distance from the axle to the top of the contact area (call this diameter d1) is greater than the distance from the axle to the bottom of the contact area (call this diameter d2).

    So, as the wheel rotates, the distance 'traveled' by d1 with every tire rotation is greater than that 'traveled' by d2. Let's suppose that this drive roller is 1 inch high, and, with a 26 inch wheel, d1 could be 12.5 inches, and d2 could be 11.5 inches. This means that d1 is 8.7 percent larger than d2, and the tire sidewalls will be generating friction and heat at the sidewalls at the top and bottom of the drive roller, and the captive roller.

    Sidewalls are thinner than the face of the tire, so, it won't take as long for the friction (and scuffing) to cause problems. (sidewall blowouts, anyone?)

    Attached Files:

    Last edited: Oct 12, 2009