View Single Post
  #16   Spotlight this post!  
Unread 28-03-2008, 07:52
ayeckley's Avatar
ayeckley ayeckley is offline
Registered User
AKA: Alex Yeckley
FRC #2252 (Mavericks)
Team Role: Mentor
 
Join Date: Apr 2007
Rookie Year: 2007
Location: Sandusky, OH
Posts: 268
ayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond reputeayeckley has a reputation beyond repute
Re: Cim breakers tripping?

It sounds like you (the O.P.) have a very similar drivetrain to ours this year, except that we used two CIMs per wheel. Based on our experiences, I suspect that you've got a combination of issues:

With the low reduction ratio you've got, the robot is probably nominally operating at 50% or less of the no-load speed of the motors. In that scenario, you are dumping a lot of current (more than 40A) into each motor in order to quickly accelerate the robot. If you were to nominally run at a higher fraction of the motor no-load speed (most experienced teams seem to recommend about 75%-80%) then the robot would reach it's top speed quickly, causing the motor input current to drop quickly as well. Less time would be spent drawing over 40A. The drawback is that the top speed isn't as high. In our situation, we found that we could never use the top speed we'd planned on because of the small playfield size. Translation: gear it down if you can, or add a second pair of motors (as others have suggested).

The other possible issue is your control scheme. At one point for us, the encoders which measured our drive wheel velocity failed. That resulted in our velocity-based PID control loops getting very, very confused, and as a result our PWM values would essentially change from 0 to 254 (full rev/full fwd/repeat) at the slightest change in joystick position. You couldn't tell by looking at the robot, since it still accelerated and decelerated normally, but in reality the motors were being asked to change direction 38 times per second. As a result they saw average currents much, much higher than 40A. In our situation, the breakers didn't trip, but the main battery overheated and dropped to about 4V (according to the IFI rep's logged data). Once we switched back to open-loop control, the PWM values changed in a more controlled manner, and we never overheated anything again. If your control system is doing something similar to your PWM values then the problem your are experiencing might be fixable with just a software (or driving style) change. Hope that helps...
__________________