20-05-2009 19:37

Rick TYler

Re: pic: Modified vex(Read: Drive Train)

Have you tried driving this robot yet? With this gearing you might find out that it won't turn very well, and won't achieve anything like its theoretical maximum speed. You'll get a lot better performance by replacing that 12-tooth gear with a 36.

20-05-2009 19:45

bobwrit

Re: pic: Modified vex(Read: Drive Train)

I have tried to drive, but I haven't changed the code that was origonaly in the controler so it's only 2WD and I haven't changed that yet. It still drives... kind of.

22-05-2009 05:00

Mr. Freeman

Re: pic: Modified vex(Read: Drive Train)

Doesn't the fact that you're using gears make this not a direct-drive system?

22-05-2009 08:29

Andrew Schreiber

Re: pic: Modified vex(Read: Drive Train)

Im going to second the notion of swapping your gears or going with a smaller ratio, right now it is at 3:1 which means every 1 rotation of the motor output shaft you get 3 rotations of the wheel. When you first see this you think, "Sweet I want to go fast that sounds great" the catch is you will not be able to accelerate quickly. This stems from the fact that when you increase speed you decrease torque. Force = torque /(radius*sin(theta)) Using that and F=MA we can calculate your max acceleration to be A = motor_torque/(wheel_radius*robot_mass*sin(theta)) For the sake of simplifying we can assume that sin(theta) = 1. Using that our equation goes to A = motor_torque/(wheel_radius*robot_mass)

Here is the my issue with the way you have this system geared, you may have a top speed of 3*motor_speed but you will more than likely never see that unless you run down a long hallway. Recall v = v_0+a*t, assuming v_0 is 0 (starting from stand still) you could calculate how long it would take you to reach any speed.

As for the turning issue, I'll let someone else show you how to work out the math for that.

22-05-2009 19:39

bobwrit

Re: pic: Modified vex(Read: Drive Train)

Yeah, I've ran through the math on this before, so I recognize the problems with it. My own way of getting around the acceleration problem is just pick it up, because it's a demo robot anyway(hold it while demoing it, it still works.).



And yes, it is still direct drive, only modified a little bit.

28-05-2009 11:10

RoboTIP

Re: pic: Modified vex(Read: Drive Train)

Keep the math coming! Even if bobwrit has already "done the math," I think it is an excellent idea to highlight the math for others. So, how about it? Who wants to "do [share] the math" for turning?

28-05-2009 11:20

Andrew Schreiber

Re: pic: Modified vex(Read: Drive Train)

I would work it out for you but honestly I am only ~50% sure how to do it. I will try to figure it out when I get home later tonight if no one else has posted it.

28-05-2009 11:44

Collin Fultz

Re: pic: Modified vex(Read: Drive Train)

http://www.chiefdelphi.com/media/papers/1443

This is a white paper by Chris Hibner that cannot be pimped enough. It will walk you through why it won't turn based on wheel selection/position and motor/gearing selection.

Teach a man to fish...

28-05-2009 11:56

Andrew Schreiber

Re: pic: Modified vex(Read: Drive Train)

Thanks, Im glad to see my thoughts on how to do this were pretty much right on. Chris does a much better job explaining than I can though.

Also, fishing? I'll pass, not a fan of fish, I'll keep my vegetables thank you very much!

28-05-2009 12:08

Collin Fultz

Re: pic: Modified vex(Read: Drive Train)

Ok, fine.

Teach a man to plant...