Controls! Robot wiring is not easy to get right. All of our problems with wiring were ultimately caused by a lack of supervision and training by experienced students.
So what did we do wrong?
Here we have a
CAN bus wire that was not tested in the full range of motion of our arm, the XT-30 is getting pulled out and the wire is taut. We will remake an extension cable of the correct length, and strain relief this wire.
These wires are too loose, and are likely going to get caught in something. We try to avoid exposed wiring as much as possible, especially smaller wires as shown here. That
PWM cable is for a proximity sensor, and it is spliced into the black cable shown. That bend radius also scares me a little bit, but the worse part is that it can just be pulled out. This will al be run inside corrugated tubing.
This wiring is not very protected at all and is even touching the chain above it. This part wasn’t super concerning, but It could be done better. The mounting of this spark max hides our logo, and makes the encoder harder to access if we need to service it. Also the port on top should be covered. We will remount this motor controller better, strain relief the wires, and route everything through corrugated tubing.
This one isn’t so bad. This is 22awg wire taken out of a 4 conductor cable. We will be rewiring this inside the tube, and leaving the jacket on the cable, as it adds extra protection.
These wires will interfere with our bumper mounting. Also no strain relief, and wire bundles that can get caught in things. Will be rerun and strain relieved.
These wires are hanging out in the intake, not strain relieved, and not bundled. These will be secured down and strain relieved.
Remember that prox sensor cable from earlier? This is the same one. It was run with the wires below it, but didn’t actually go through any of the cable ties. Like all of our wiring, we will run this inside corrugated tubing, and strain relief it.
If this bundle of
CAN wire were to get caught and pulled, it would tighten on itself and cause even more problems for us. We’ll be bundling this properly, inside some corrugated tubing or somewhere more protected.
These 12awg bonded wires were run out the end of a tube (good!) with no grommet protecting them (bad). This is an example of somewhere where we were on the right track, but better supervision from myself and our leads would have prevented these issues. Also while it is run in front of our radio, we will be remounting the radio elsewhere.
An unmounted pigeon? Not only will that get caught in something, but it won’t work if it’s moving around as
CTRE has yet to implement magic in their products. This is an example of poor training. A student was instructed to do something, without understanding what they were doing and why.
Just generally messy. We’ll be cleaning all of this up.
That explains our radio issues. Like
CTRE, REV also unfortunately has not implemented magic. You can also see some exposed wire behind some of the heatshrink on those connectors.
We’ll be replacing all of our 6awg with 4awg so this isn’t a huge concern, but it’s still good to be mindful of bend radii.
Not all of the strands made it into this ferrule.
Wire stripped too long, exposing the conductors.
A broken XT-30.
Missing strain relief, and exposed wiring. This is bound to get caught on something.
Oh boy.
Overall, this will be a learning experience for everybody. This was a failure of our team leadership. Students were not properly trained or supervised to wire a robot. These are the things we need to remember for next time:
- Protect everything
- Strain relief everything
- Do things right, the first time (no shortcuts!)
I hope this will be helpful to someone else, and I want to emphasize that nobody who worked on this wiring is at fault here. There are things about wiring that you can only learn from having to fix broken wiring at a competition after it got chewed up in a pulley or caught in a drive wheel.
