Quote:
Originally Posted by thefro526
Fast forward to Lunch Break before Eliminations, when one of the other mentors tells us that one of the PWMs to the left side of the drive was unplugged... We plugged it in and it worked perfect.
|
Heh, this reminds me once again of 2008. We wanted to make the robot maintain a straight heading in autonomous mode. So I added a gyro and a PI loop, which worked very well. So well, in fact, that we decided to use it in teleop as well. It did an excellent job of keeping our four-CIM, six-wheel drive train going in the right direction.
So we went through VCU with this drive system, and for the most part, it worked wonderfully. There was just one problem: our driver kept complaining that the drive was slower than usual. He said it didn't seem as powerful as it used to be. When asked if he had any difficulty turning, he said no: it just didn't feel as powerful as before. We figured the motors and bearings were just wearing down and kept playing without any changes. We made it to the quarter-finals, I believe.
Fast-forward to Chesapeake, and we had completely forgotten about the drive train. That is, until we were testing some sort of autonomous routine that required me to print out the actual voltages the PI drive loop was sending to the motors. I instantly knew something was wrong: the software was sending half the power to the right motors that it was sending to the left motors, and yet the robot was driving perfectly straight! We investigated the problem, and found that one of the left motors was completely dead. No one knows how long it hadn't been working; it was probably dead before ship. It could even have been dead since before we installed it in the robot! The PI loop was so good at keeping the robot straight, it was able to compensate for a dead motor well enough to make the problem virtually invisible. Unless, of course, we tried to push another robot with all four of its motors working.