In one of our matches at the LA regional last weekend, my team experienced a drop in comms after hitting the wall a little too hard. The robot did not reconnect until after the match ended (~20 seconds). The FTA who helped us mentioned that both the roboRIO and the radio rebooted, which he guessed was because of a power issue. Interestingly, our human player states he didn’t see anything flicker – including our blinding camera light (hard to miss).
Back in the pits, we stress tested the power, and were only able to replicate the power drop to the radio through flicks to the body of the main circuit breaker (not the red button). This caused only the radio’s power to flicker, nothing else. Even pounding other parts of the robot much harder did nothing, which causes me to doubt it’s a moving loose wire (another FTA also helpfully yanked on some likely connections – nothing). The problem persisted even after replacing the main circuit breaker with a new one from spare parts.
Essentially, I’m wondering if anyone can offer an alternate explanation to the match’s events, and/or a way to fix this problem. Could the FTA have been mistaken about the RIO also power cycling (which would make the breaker the main issue)? What can we do to prevent this?
If the radio is lost/reboots during a match, the field monitor that the FTAs use will show both the radio and the roboRIO being disconnected. This does not necessarily indicate that the roboRIO rebooted as well, it may have just been the radio. If it was at the main breaker, there is a good change the roboRIO rebooted as well but without more data it is hard to tell.
As far as your human player not noticing anything flicker, the power drop was likely quick enough that a human would not be able to perceived any lights turning off.
It’s possible that it was just your radio that rebooted. The field logs indicate whether the field can connect to your radio and RIO, both of which will appear to go down even if just your radio goes down. The driver station logs should be more helpful in determining whether your RIO actually rebooted (since they should include the time since the RIO last rebooted when it comes online). Then again, the FTA should know this, so it’s possible that they noticed your RIO status lights change on the field, or something else.
For the human player thing – that’s definitely possible. Thinking back to our tests in the pits, I’m not sure I actually saw the radio lights flicker – only that several of them were lost and took some time to come back (comms i believe?) – the actual power light may have been too fast to see. I may be able to try the test again tomorrow and report back.
Are the driver station logs saved for an extended time? Since this was over the weekend (I believe Saturday), everything that could have been restarted likely has been. If they are more permanent, I can attempt to retrieve them tomorrow at our meeting. Our electronics plate is in our withholding allowance.
Figuring out when this occurred could be a bit difficult – posted match schedules were quite different from the real time, and our robot got rebooted quite frequently. Then again, I would probably notice about 10-20 reboots in fast succession if it occurred while we tested the problem.
Driver Station logs are saved forever.
If you haven’t cleaned them out yourself, then you can still see the very first time you ever connected this laptop Driver Station to the robot.
You should really clean the logs out (in Public Documents/FRC/Log Files) at least once a year to keep the clutter down.
What you describe is a typical loose fuse on the end of the PDP. Make sure they are fully seated. It takes a lot to push them in. It would help if you had a picture of the main breaker mounting. It is possible that your frame is somehow contacting the back of the breaker in some way that is giving a frame fault. Anytime both the radio and Rio reboot, you will be down for at least 20-30 seconds, the RSL will go off at the beginning and then come on full until reboot and field connect occur. At that time the RSL will start to flash and you will have full control.
Our team had a similar problem at CHS Greater DC - it turned out to be the connection from the battery to the main breaker and the connection from the battery wires to the battery poles. When we tightened these, the issues disappeared.