There has been some discussion over the high number of dead bots in the playoffs at champs and even among high profile teams. I am curious as to why so many? I feel speed over the Charge Station may have been hard on electronics this year but there are other years that had tough terrain. So I thought I would ask a few questions about dead bots?
Did your robot die on the field this season (not just at champs)?
Did you identify a cause?
Did you lose main power?
Did your rio reboot? if so was it a 1 or 2
Did your radio reboot? (how was it powered if so)
Did you lose CAN?
Did your CAN freak out with a bunch of error messages? if so what controllers did you have? Were you on a CANivore
Our robot never outright died, but there were multiple times in our matches on Johnson where we and our alliance partners felt delay in robot control, and one time where our driver input continued for about 2 or 3 seconds after we stopped moving the bot, pushing our arm down into the grid.
From my memory, we had five matches (including practice) where our robot couldn’t drive for part of the match.
Two were because our robot tipped over.
One was because of a loose connection at one of the main breaker stud connectors causing a roborio reboot.
Two were because the anderson crimp was not pushed all the way into the housing of our SB-50 connector, causing intermittent loss of main power and roborio reboots.
.
Not necessarily that it died, but both swerve modules on the left side of our robot in Hopper sf-13 started going haywire (overspinning and making the robot undrivable) at the end of the match. We took a look at it in the pits, looked innocent to us then with no issues, so we threw it back out on the field, where, wouldn’t you know it, the left side of our bot was still bad. After taking the loss in finals, we took a look at everything and found that a CAN wire fell out of a wago.
We used zip ties instead of metal fasteners for the bracket and the battery took out the CAN wiring to one swerve module.
Also we tipped over. Ramp rate limiter nearly eliminated that possibility but our driver decided after some matches to be more careful than to have the limiter change the robot response. A limiter for the narrow angle of travel that it was needed could have been useful but harder to program.
Turning on before setting robot down giving bad IMU calibration didn’t kill it but made it hard to drive.
Shattering plastic pneumatic fittings and losing air to gripper replaced with metal fittings and eventual 90 degree turn.
No outright electrical or CAN failures this year using inline WAGOs and clipped PWM CAN connectors. And plenty of zip ties and hot glue on JST connectors.
There were a few instances in division playoffs and on Einstein where our drivetrain’s CAN network briefly dropped for an instant, but came back up. We still don’t know the root cause as we went through every single CAN wire and did a tug test. It usually happened after a hard impact on the field.
Speaking of hard impacts, at one point on Einstein we noticed that on one side of the robot our bellypan was sagging by an entire inch. All of the rivets attaching it to the frame on one side were sheared… it was a miracle our bellypan didn’t fall out from under us.
Between the two teams I was on, Spark max wired backwards, Improperly assembled andersons continually disconnecting, a loose piece of a threaded wire going into an Anderson making contact, & someone using a random connector they found in a bin to connect CAN lines in a hurry. Oh and twenty smashed limit switches.
We were powering our radio with the RPM and the ethernet cable was only half plugged in. It made it most of the way through autonomous but died after going over the bump.
Our Ethernet connecting our radio power module to the Roborio got slightly disconnected.
This was in the beginning of the season so whenever we brought out robot onto the field after that I always gave each Ethernet a little snug to make sure it was seated properly
P.S - the roborio Ethernet lights are YOUR FRIEND!! I learned that this year.
Orange light shows number of transfer data going from roborio and radio
A few seconds after beginning teleop, another robot’s arm hit our main breaker as both robots collided in the middle of the field. The main breaker was protected by a lexan guard, but the lexan shattered. We replaced the main breaker guard with an aluminum one for the next match. There was no red card because the ref said our robot was the only one initiating contact.
