Shop OBEX P1 Docs P2 Docs Learn Events
Worm Gear Box Versus Regular Gearbox When Power is Cut — Parallax Forums

Worm Gear Box Versus Regular Gearbox When Power is Cut

Martin_HMartin_H Posts: 4,051
edited 2011-08-13 16:52 in Robotics
With a regular gear box the output shaft can turn the input shaft. So cutting motor power results in the momentum of the robot being transferred back through the gearbox to the motor. Useful if your H-bridge supports dynamic braking as you can bring you robot to a speedy halt.

However a worm gear box only allows a one way transfer of power. So I'd assume that if you cut the motor power the output shaft should lock up once the residual momentum within the motor is consumed. So at this point what happens to the robot?

Comments

  • ercoerco Posts: 20,255
    edited 2011-07-30 17:42
    Robot stops quicker with a worm, no coasting. But worms are evil and inefficient. Big losses, faster wear. Spur gears roll, work gears slide on each other. Worms are your last choice for power transmission, and they need to be bathed in grease.

    The only good things I can say about worms is they are compact, use fewer parts and are self-locking.
  • RonPRonP Posts: 384
    edited 2011-07-30 17:52
    My project uses a worm gear box for the movement of the shovel. It doesn't stop instantly but quicker than a standard gear box I think. My skill saw is worm drive and it doesn't come to a complete stop right away, also the blade can be turned by hand in either direction by hand when stopped(if I remember right). So I think it depends on the quality of the gear box. My next project might include and arm and I was thinking of using worm drive boxes(Tamiya) because like you said when stopped hey wont turn.
    In my project video you can kinda make out when it is coasting.

    Ron
  • W9GFOW9GFO Posts: 4,010
    edited 2011-07-30 18:07
    The Parallax motors use worm gears, as do most wheelchair motors. They both can be back driven very easily.

    Rich H
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-07-30 19:34
    I put one of these in a toy car. When I was experiencing with it (all the gears turn a worm gear which meshes with a mitre gear on the axle) the resistance that I sensed when trying to turn the axle gave me the feeling that something bad would happen if I forced it.

    IMG_1372.jpg
    IMG_1373.jpg
    IMG_1375.jpg
    640 x 480 - 48K
    640 x 480 - 45K
    640 x 480 - 49K
  • RonPRonP Posts: 384
    edited 2011-07-30 19:44
    PJ Allen wrote: »
    I put one of these in a toy car.

    That's the same one I have. I was thinking they would be good for an arm because they wont move without power. Unless something bad happened.
  • Martin_HMartin_H Posts: 4,051
    edited 2011-07-31 06:38
    FYI, the genesis of this question is this project from the RBB companion web site: http://www.robotoid.com/appnotes/project-roverbot.html

    It uses the the Tamiya worm gear motors instead of the double gearbox which is commonly used. So I began to wonder if the locking of the worm gear box provided some advantage. Particularly because some low end h-bridges only have a enable and direction line which means they can't brake the motor.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-07-31 18:32
    Martin_H wrote: »
    So I began to wonder if the locking of the worm gear box provided some advantage.

    Yes, these motors will provide some braking, and they will lock. These motors are capable of a little more torque than the twin motor gearbox and similar designs. They use a beefier RE-260 size motor, and the gears are thicker. Of Tamiya's high torque motor kits the worm gearbox has the distinction of being the most slender -- the main advantage of worm motors, after all. Most of the others are wider, so they influence the minimum size of the base.

    -- Gordon
  • Martin_HMartin_H Posts: 4,051
    edited 2011-08-01 07:59
    Gordon, simple question. I assume you intended the drive wheels to be the front of the RoverBot? Generally the widest part of the platform is the front, but usually the drive wheels are as well. But on this project the widest part is over the castor, so it's a bit ambiguous to me.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-08-01 09:37
    Martin_H wrote: »
    Gordon, simple question. I assume you intended the drive wheels to be the front of the RoverBot? Generally the widest part of the platform is the front, but usually the drive wheels are as well. But on this project the widest part is over the castor, so it's a bit ambiguous to me.

    The width of the robot is the same front and back, as the motors and wheels are inset to provide a straight contour. it just looks wider on the caster end because of the perspective of the photograph.

    In general robots with a single caster on one will track (travel in a straight line) better when the caster is in the rear, so it acts as a tailwheel. (Nosewheels tend to act as steering wheels if they don't re-center well, though this isn't usually a major problem with casters that are 1/2" wide.) Wide or narrow base front or back shouldn't make much of a difference. Extremes notwithstanding, you can alter the shape of the base to provide a little visual distinction to your bot. Rover was made to be reproduced with simple straight cuts, and is the type of base you'd make with a budding robo-builder.

    -- Gordon
  • Martin_HMartin_H Posts: 4,051
    edited 2011-08-01 10:25
    Gordon, thanks for the reply.
  • Martin_HMartin_H Posts: 4,051
    edited 2011-08-09 06:43
    I've purchased the Tamiya worm gearbox and will assemble them soon and build the RoverBot. The motors support two gear ratios: 216:1 or 336:1 which are both fairly large amounts of reduction and I'm not sure which to use. I plan to over volt these motors with four NiMH cells, so they'll be running off 4.8 volts instead of 3 volts. I'm guessing that will make them run faster, so perhaps the 336:1 would be a better choice?

    I know that over volting will reduce motor life, but I can just buy new ones for a few bucks, so I'm not concerned.
  • ercoerco Posts: 20,255
    edited 2011-08-09 09:50
    OK, OK, I wash my hands. Use your worms if you must. But please, never on a wooden platform. :)

    But kindly note that I'm getting excellent control using CR servos with SPUR gears here! (Never mind the controller) http://www.youtube.com/watch?v=PfMnl4oGzs8
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-08-09 16:28
    erco wrote: »
    OK, OK, I wash my hands. Use your worms if you must. But please, never on a wooden platform. :)

    Since the worm gear dates back to Archimedes, and he made all his robot in wood, I think the two go together.

    Nice proof-of-concept with the servos, BTW. I'm too lazy to hack into servos to that degree, but it's neat to see that it can be done.

    @Martin: 336:1 seems awfully slow to me. However, you'll probably be able to haul around your cat on top of the bot.

    -- Gordon
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2011-08-09 19:11
    I know that over volting will reduce motor life, but I can just buy new ones for a few bucks, so I'm not concerned.
    The data show that "over volting" reduces motor life significantly.

    [At Hack-A-Day it'd be: The data show that "over volting" reduces motor life quite significantly.]

    Since you're rolling in cash, why don't you just shell out for the 6V-ready versions already?
  • Martin_HMartin_H Posts: 4,051
    edited 2011-08-10 06:35
    PJ, I read the analysis of the Mabuchi FA-130 motor on the Pololu web site. The author ran that motor for sixteen continuous hours for the 4-5 volt range. That translates to a lot of use for a hobby robot. So I'm hoping the Mabuchi RE-260 motor responds similarly to the FA-130 motor. I already have working spares that I harvested out of dead toys as well.
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-08-10 12:15
    Let me know if you come across better 6V-12V versions of the RE-260 motor that don't consume 3-5 amps at stall. That would be a nice find, but so far I haven't located any convenience sources. Ideally we're looking for something similar to the high efficiency FA-130s.

    -- Gordon
  • ercoerco Posts: 20,255
    edited 2011-08-10 13:18
    Just because two motors look the same doesn't mean they are the same. Like everything else, nothing is as simple as it seems AFA PMDC motor choices are concerned. Mabuchi is the "gold standard" of economy motors, and there are numerous Chinese knockoffs that range in quality.

    The case size & shape are just the beginning of motor variations. RE-class economy motors look just like RC-class motors, but the brushes are vastly different (RE=stamped metal, RC=carbon brush, heavy duty), several different windings may be inside, some have oriented magnets or different commutator timing for better performance in one direction.


    @Gordon, the RC-260 RA comes in several flavor windings:

    2670: Istall@4.5V=6.8A 70 turns of 0.26 mm magnet wire
    18130:Istall@6.0V=2.61A 130 turns of 0.18 mm magnet wire
    10350:Istall@12V=0.67A 350 turns of 0.10 mm wire
    08450:Istall@12V=0.33A 450 turnsof 0.08 mm wire
  • ercoerco Posts: 20,255
    edited 2011-08-10 13:23
    And here's a link to an RE-260RA/18130: http://www.mabuchi-motor.co.jp/cgi-bin/catalog/e_catalog.cgi?CAT_ID=re_260ra

    Should have about the same stall current as the RC unit above at 6V, ~2.61A

    Browse THIS! http://www.mabuchi-motor.co.jp/en_US/product/p_0303.html
  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-08-10 16:23
    Yes, I know about the Mabuchi pages, but you can't buy them. You need some importer to have bought the zillion minimum order that Mabuchi requires.

    You'll find other RE-260s on Alibaba, but it's the same thing. "Minimum quantity 10,000 units." I only need two.

    -- Gordon
  • RonPRonP Posts: 384
    edited 2011-08-13 16:52
    My next bot might make use of 260 size motors so I thought I would look around for a higher voltage source and found these at Jameco.

    Just thought I would share the find. :smile:

    -Ron
Sign In or Register to comment.