Re: [SherlineCNC] Re: #25 chain and sprockets
- You might call to a few local motorcycle shops to look for a chain breaker.
On 7/26/2013 3:02 PM, KM6VV wrote:
> Hi Ceal,
> Yes, I'd like to see the design for the chain tensioner!
> I haven't found the chain breaker on Andy-Mark yet. I'm about to order
> some #25 sprockets and chain. Andy-Mark says made in USA.
> I don't like the idea of back-driving my 'bot, I suspect the Traxx drive
> shafts to the wheels are a possible future weak point. DPA had to swap
> his out for metal ones, I'd like to put off that exercise until I get a
> RoboMagellan race or two in first (if that happens).
> Granger looks pretty pricy for the chain breaker, might look elsewhere
> for that. I didn't remember Tim using chains.
> Alan KM6VV
> On 7/26/2013 9:29 AM, Ceal wrote:
> > Alan, I see Tim already provided the info about the chain
> > breaker/relinker with master links. We purchased one of those at
> > Grainger years ago. Not sure if that is comparable price-wise or not. I
> > have an account at Grainger by us in Berryessa (San Jose) if you need
> > one from them. I even used it, so if I can use it, most folks can. :)
> > Also, Tim designed this neat chain tensioner that we used on several
> > robots in years past for FIRST Robotics Competition (FRC). Sort of like
> > an aileron on a plane, as he told me. But composed of parts
> > off-the-shelf at OSH.
> > The students on FRC teams who use chains learnd that chain tensioners
> > are an important element in a robot chain-drive system. Alternatively
> > the length of chain driven is so small and the wrap so much, it may not
> > be as important. And of course FRC robots do get pushed around a bunch
> > on the field. Your robot may not undergo that kind of stress. Andy-Mark
> > sells one too.
> > Tim has photographs of the design if you are interested. Some flat
> > corner brackets and something else I'm drawing a blank on.
> > ceal
- Hi David,
Yes, first stage belt drive from motor to 1st axle. I put the drive
aside for a while and worked on the sonar array and camera mountings.
I agree with your analysis. And yes, our designs are very similar, as
far as I've been able to tell. Hard to improve on your design!
I never did get a good idea of the torque from your Pittman motors. I
calculated 8 ft-lb at 240 RPM from my 24 V IG42 SuperDroid motors. Any
idea what your motors put out?
I first did a search for different belt materials; I have another set of
belts to try as well. Hard to find the "Road Rage" you mentioned.
I'm heavier, what with the Mini-ITX CPU and the sensor head.
Good videos! I've studied them over and over.
Thanks for the comments.
P.S. Let's schedule a play date for Jbot and HexaTraxx Moose once I get
On 7/27/2013 8:24 AM, dpa wrote:
> Hi Alan,
> I'm trying to get up to speed on the problem you are attempting to solve. Seems like the starting point was
that the timing belt driving the first stage of the robot's drive train
is jumping teeth when you attempt to
turn/rotate the robot. Since this is not a problem I have encountered
with jBot, a very similar design,
(http://www.geology.smu.edu/dpa-www/robo/jbot/) I started wondering why.
> I think the answer might be as follows. The motors you are using can produce more torque than the Pittman's
on jBot (I don't have the numbers at hand but I believe this is true.)
Since rotating a 6 wheel skid steered
vehicle requires scrubbing the front and rear tires sideways, there is a
lot of friction involved, much like
> To that end, jBot uses "Road Rage" tires with very little tread, as we have discussed, which are relatively
easy to scrub sideways. jBot, by design, also has very light wheel
loading, on the order of 3+ pounds per wheel.
I believe your robot is heavier, and also uses the stock EMAXX tires
which are, by design, very hard to scrub
sideways. Hence much more torque is required to rotate the bot that is
required to rotate jBot.
> Since jBot is both lighter, has smoother tires, and weaker motors, the robot either rotates successfully or
the motors stall, long before the point of jumping teeth on the timing
belt. The stall condition is monitored
by the microprocessor and handled as an error condition.
> Check out the video of jBot getting itself unstuck from various situations here:
> and look at the escape around 1:30 into the video. You can see the motors stall and the robot react to that
>>From your description I believe that your robot would, in that same situation, not stall, but rather apply
enough torque to jump the timing belt teeth. Too much torque for the
condition, or so it would seem.
> Not sure how much any of this helps but it looks like what you are describing is a total systems design
problem rather than a chain vs. timing belt or tensioning problem,
having to do with available torque, your
robot's tires and performance requirements, and the controlling software.
> You might try going to smoother tread wheels, lighten the robot if possible, and see'f you can monitor the
motors to trade off torque vs. stall current and prevent the timing belt
> In any event, good luck. jBot is looking forward to having a mate to play with...
> best regards,