Inspection Stories

[ksafin]

This just happened at the South Florida Regional this thursday to us:

First a little bit of background:
During build season, we were focusing on building a launcher. About week 3 or 4 we decided to make a backup dumper bot in case the launcher wasn't ready or was inaccurate. Needless to say, the dumper bot was far from done when build day came around, and the launcher bot was untested.. So we ended up working from stop build day to the competition day on our 30 lbs of withholdings to finish up our bagged bot.

The story:
Thursday we come to the tournament, weight our witholdings (which was pretty much the entire robot frame) to be about 25 lbs, and we get to building. We spend about 2-3 hours attaching all the witholdings, and reworking the mechanical kinks, tightening, etc. Then we start mounting all our electronics.. This took about another hour. After this all, we turned on the robot to start testing. We were glad, because we had been done before practice rounds started. We got through about 3/4 of inspection before the electronics were on, the remaining 1/4 being some minor bumper issues, as well as power-up and pneumatic checks.

So, we started it up and began testing the program. Reason being, we stuck our pneumatic cylinders randomly into solenoids, so we needed to run the program to check which were plugged in wrong, and fix accordingly. After turning it on, our RSL didn't go on, which was odd. Next odd thing was that we couldn't drive, our conveyor didn't work, etc. Neither did solenoids. Point is, NO motors worked. And as mentioned, neither did the RSL.

We were perplexed, but thought it'd be a simple fix so we got down to it. We couldn't figure out the problem, so we called an event volunteer from the Help Desk they had set up. He came and began to help us. We did many things: tried to use a second digital sidecar, used a PWM tester to check for bad PWN cables, using motor controllers in different modes, switched wiring to various amp fuses, used several batteries, as well as use about 20 different code deployments.

We had been looking for the problem for about 3 hours with this volunteer when another robot inspector came up, as well as mentors from another team (801). So, we had about 4 robot inspectors and 3 team mentors looking at the problem. We did pretty much the same exact thing as we had before, with some small achievements:

With some new code, we managed to get our drivetrain to turn on, which was powered by 4 jaguars. (4 CIM drive). After that worked, we tried to turn on our other motors, which used Victors. They turned on for a split second, spun one rotation, and stopped. So, we switched one victor for a jaguar. BAM! It worked. We were ecstatic. So team 801 graciously lent us jaguars to exchange our victors with. We tested each individually, and everything turned on!

Not so fast. As soon as we tried to turn on all the motors, nothing worked again. Dead silence. We were about on the verge of collapse, working on the same problem since 12pm, missing every single practice round, never having really drived the final bot whatsoever, and it was 7 pm. We weren't even done with 1/4 of inspection, not including sizing or weighing.

Then came Kowalski. He was a Robot Inspector. He looked at the digital sidecar and noticed that the 5V light was on, but 6V wasn't. He thought it was odd, and asked where the DS was plugged into for power. We checked for him, and told him it was plugged into the 5V slot on the PDB. Wow. That was the entire problem we had for the past 6 hours. We changed our entire code 20 times over, used another digital sidecar, 30 different PWM cables, and even switched our victors for jaguars. But the only problem was that the DS wasn't in 12V.. ugh.

We should have figured it out earlier, the symptoms were obvious, and i'll tell you why. A PWM signal is about 5V, which is why we were able to get our drive to work. The Jaguar's are very power efficient, able to run on the meager 5V. A Victor, however, is less efficient, and uses slightly more, primarily, 6V. This is the reason for a 5V and 6V light on the DS. So, Jaguars would run, but Victors wouldn't. Even then, when using 7 jaguar's, the 5V wasn't enough juice apparently. The RSL should have been another indicator. It was working the day before, but suddenly it wasn't. It also needed 12V.

However, Kowalski saved us, and we got the robot to work. With the help of about 5 inspectors, we were speedily rushed through a thorough inspection in about 45 minutes. They even helped us fix all the problems we ran into, if any.