One of the aspects my team is hoping to improve next year is our wiring. Particularly the amount of time it takes to wire and the speed of which.
We need to allow our electrical team to work in parallel along the mechanical team building the robot to allow it to be wired more effiently.
I’d be curious if anyone uses any computer aided design software to help manage wiring and design schematics. Our wiring planning is particularly poor
You can typically plan out mounting holes in your bellypan for pdh/rio. Work with your electrical team (or learn electrical) and try to plan it out mentally while adding access holes in box tubing and whatnot. Also model battery placement/mounting (with strap holes!). As a bonus, you could use the wiring featurescript if you use onshape to design the power wires from battery to breaker to pdh. It doesn’t take very long but it’s just annoying (though it looks cool).
Having a pocketed bellypan “layer” is also nice for ziptie placement which is good for wire organization.
Are you sure more extensive prior planning will get you the results you expect?
When I got my first chance to go to worlds I was surprised to see how many high performing teams have a “stick controls where it fits when everything else is done” mentality.
Opportunistically designing generic packaging space like open predrilled belly pans and flat surfaces for stick-on wire routing solutions has been pretty effective for long-term flexibility and on the spot routing opportunities in my experience.
Thats basically what we do haha. I’d just like getting our eletrical team doing more in parallel, thats all
Battery placement is usually baked into the design pretty early on since it affects mechanical access as well as center of gravity. From there we typically layout components roughly in CAD, starting with the PDP/PDH and breaker since those needs to be in relatively close proximity to the battery and have stiffer cables. Then we’ll place motor controllers near the PDH, then RoboRio, VRM, PCM etc.
Once we have a rough layout that looks in CAD and some of the mechanical parts of the robot are prototyped we will lay out our physical components on a board on the robot and make sure that still makes sense, and then turn that into our final electrical panel by fine tuning the CAD before making final parts. The last step isn’t necessary, but does make for a more polished electrical panel.
That addresses the bulk of the electrical components, but during design the mechanical team is also looking at wire routing down into our drivetrain, up into any payload, through superstructure etc. Since this moves pretty quickly we are generally doing things like popping large access holes, and doing best guess, not really modeling the wiring. Wiring lengths and final routing are determined when we hook up our panel, and refined during testing leading up to competition (s).
We always book 2-3 days (10-15 hours) for wiring once the robot is “initially mechanically complete.” It’s not something you want to rush. If we manage to do it faster than that, great! If not, that’s fine.
But making sure everything is properly crimped or soldered, properly seated (tug tests on all joints, junctions, connections, etc), with strain relief and labels, and continuity tested…this takes time to do right.
And doing it right is much, much better than doing it slipshod and ending up with some intermittent electrical failure on the field that you can’t reproduce in the pits.
Rough layout in CAD that we tested and lived with for a few weeks, then panel as finally fabricated.
THIS is really cool
Totally right. Actual wiring is pretty time consuming, especially when you’re labeling, testing connections, redoing that one that failed, etc.
It amazes me how many teams don’t label, either adequately or at all. We label both ends of every wire, and both terminations of every air hose, as well as whatever those things are plugged into. If something happens and things get pulled out, we don’t want to have to guess what goes where, nor to take the time to trace wires/tubing–we want to just be able to put it back where it goes.
We are very guilty of this. Usually we slap the electronics in “temporarily”, and then it just becomes permanent. ¯_(ツ)_/¯
Hopefully that changes with our offseason swerve chassis.
There’s only one way to stop doing that… STOP DOING THAT! shakes fist
Just imagine a gray-bearded man dressed like a grape glowering down at you, arms crossed, nodding in approval as you label your wires or tsking in disappointment if you don’t. You’re welcome.
Our method is actually pretty easy! The electrical team gives a cursory wiring to let us programmers do testing and we don’t let go off the robot until it is competition time.
What worked well for several of the teams I have mentored is to install the control system components (electrical and pneumatic) on sub-panels. This allows construction work on the mechanical structure and the control system to proceed in parallel.
Once space has been set aside in the design for the control system, the sub-panel(s) are cut out. The layout of the components is developed, allowing space for wiring channels. The components are installed and connected. This allows the control system to be pre-tested before being installed. The pre-testing helps eliminate silly errors like motors turning in the wrong direction.
Since the detail design and construction of the electrical sub-panels occurs in parallel with the detailed mechanical design and the construction, more time is available for the electrical design and construction.
This parallel construction methodology is what a lot of manufacturers of industrial electrical equipment use.
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.
