Loading ...
Sorry, an error occurred while loading the content.

Re: Problems in Timing Belt drive system in 6WD robot

Expand Messages
  • dpa_io
    oops...
    Message 1 of 41 , Jan 25, 2013
    • 0 Attachment
      oops...


      --- In SeattleRobotics@yahoogroups.com, twcarroll@... wrote:
      >
      >
      >
      > In a message dated 1/25/2013 2:56:58 P.M. Pacific Standard Time,
      > davida@... writes:
      >
      > Alan,
      >
      > I have not observed this behavior with jBot. Basically the timing
      > belts/pulleys/motors are strong enough to twist up the original plastic
      > driveshafts without the timing belts jumping teeth in the pulleys:
      >
      > http://www.geology.smu.edu/dpa-www/robo/jbot/jbot_twisted_driveshaft.jpg
      >
      > I eventually replaced the drive shafts with steel CV joints. I don't
      > think I have ever seen the timing belts jump teeth on the pulleys. Perhaps
      > your pulleys are too small?
      >
      > Relatedly, my experience is that lowering the center wheels is not needed
      > with this platform design, and actually will decrease the performance of
      > the platform. YMMV.
      >
      > If you check out the "first movement" video of jBot spinning in place you
      > can see that the wheels on the front and rear are sliding sideways quite a
      > bit. But the center wheels are tracking nicely around a circle, no
      > sideways movement, like you might expect from a traditional 3-wheel platform. I
      > believe this is opposite from what Tom was saying:
      >
      > The center two wheels will track smoothly as they have the greatest
      > traction as they are virtually going straight forward or backward with very
      > little side slip. The front and back pairs are sliding more sideways as
      > they are further away from the center of the turns. However, they (the front
      > and back 2 pairs) have the advantage of double the number of wheels
      > turning and thus, twice as much ground friction as they would have if only 2
      > wheels. Hmmm. Go figure.
      > Tom
      >
      >
      >
      > http://www.geology.smu.edu/dpa-www/robo/jbot/jbot_spins.mpg
      >
      > Can you grab the wheels and rotate them by hand and force the timing belts
      > to slip while holding the motors fixed? I cannot on jBot. The tires
      > begin to slip on the wheels before that. (BTW, standard practice is to
      > super-glue the tires to the wheels to prevent them from slipping, if you haven't
      > already done that. I found it was necessary on jBot)
      >
      > The only other difference that I see in your setup is that jBot uses a
      > reduction between the motor and the first pulley, while you are using a
      > gear-up. Not sure what difference that might make. The belts on jBot are taut
      > but not tight (they don't "sing" when plucked). Again, perhaps the pulleys
      > you are using on the axles are too small? (i.e., too tight of a radius
      > for the torque required). I don't have jBot here with me but as memory
      > served I used 2 inch pulleys on the axles.
      >
      > looking good!
      > dpa
      >
      >
      > --- In SeattleRobotics@yahoogroups.com, twcarroll@ wrote:
      > >
      > > Just as Robert mentioned, looking at the photograph, it looks as if
      > you're
      > > using three solid shafts with the toothed pulley / wheels freely
      > spinning
      > > on each, and no differential. 6-wheeled robots have a problem with this
      >
      > > arrangement. During turns, the robots basically tries to turn around a
      > point
      > > in the center. The two center wheels turn the slowest during turns,
      > > whereas the four end wheels are turning faster as they are further away
      > from the
      > > center of the turn. If they are digging in the grass or pavement, the
      > > force on the shaft and pulley is enough of a difference between front
      > and
      > > middle to cause the belt to slip off the pulley a few notches or more.
      > >
      > > Quite frankly, if you're 'under the gun' to get the robot completed in
      > > time for the RoboMagellan contest, I'd remove the center wheels
      > entirely.
      > > You're really not going to gain enough in traction with 6 wheels over 4
      > wheels
      > > to make any difference at all in the contest.
      > >
      > > Ackermann steering is the best for this type of robot but that requires
      > a
      > > whole new set-up in your steering and control. Six wheels look 'cool'
      > but
      > > are best used with differentials in the sets of wheels. Differential
      > > steering, or as some say- "skid steering," with more than 2 wheels eats
      > up a lot
      > > of power when making turns. That's why it is called skid steering.
      > > You'll never see a car with this type.
      > >
      > > Good Luck,
      > > Tom C.
      > >
      > >
      > > Maybe it's not the belts, but somewhere else in your drive train, say
      > > between the pulleys and the tires? I'm not familiar with your setup. Do
      > you
      > > have differentials in the axles that turn? Are you sure the "popping"
      > isn't
      > > caused by the loading of the drive train due to the difference in path
      > length
      > > between the inner and out tires when turning? Have you tried it on a
      > > slippery surface such as ice? That would at least let each tire spin as
      > it
      > > wanted to when turning.
      > >
      > >
      > > Robert
      > >
      > > ____________________________________
      > > From: "KM6VV"
      > > Sent: Friday, January 25, 2013 12:51 PM
      > >
      > > I'm having problems with the timing belt drive system in a 6WD robot I'm
      >
      > > building for Robomagellan competition in the upcoming Robogames.
      > > (Mechanical design courtesy DPA).
      > >
      > > Pictures in LM forum:
      > >
      > > _http://www.lynxmotion.net/viewtopic.php?f=9&t=7839&hilit=6WD&start=45_
      > > (http://www.lynxmotion.net/viewtopic.php?f=9&t=7839&hilit=6WD&start=45)
      > >
      > > The belt system appears to run fine in straight runs, grass, concrete,
      > > etc., even when moving quite fast. The problem arises when the 'bot
      > > starts to make a turn (more torque required). I start hearing loud
      > > "popping" sounds, which I take to be the belt(s) jumping teeth.
      > >
      >
      >
      >
      >
      > ------------------------------------
      >
      > Visit the SRS Website at http://www.seattlerobotics.orgYahoo! Groups Links
      >
    • KM6VV
      David, Thanks. I m still not finding the exact motor, http://electromate.com/db_support/downloads/PittmanExpress2011Catalog.pdf but I get the idea. 45 oz-in
      Message 41 of 41 , Jan 26, 2013
      • 0 Attachment
        David,

        Thanks. I'm still not finding the exact motor,

        http://electromate.com/db_support/downloads/PittmanExpress2011Catalog.pdf

        but I get the idea. 45 oz-in 142 RPM ??

        my 10 kgf-cm = 138.873866 oz-in. So maybe I've just got too much
        torque? I like my ground speed so far, should be 6-7 MPH with the
        1:1.533 gears I have. 1:1 gears would give me about 4.5 MPH.

        Alan KM6VV


        On 1/26/2013 4:13 PM, dpa_io wrote:
        >
        >
        >
        > Alan, yes motors are 24 volt Pittman GM9413J820 (or something like that) with 20:1 gear reduction and 500 CPR encoder.

        Drive pulley is 15 tooth and main pulley 30 tooth.
        >
        >
      Your message has been successfully submitted and would be delivered to recipients shortly.