CAD for “other” Mechanisms

[Aurgia]

So I now know how to make gearboxes and have successfully made a custom WCD that will be posted shortly but am wondering how experienced CADers go about making other mechanisms like lifts and shooters. Are they just based on specifics for that season or are there general methods to making these components?

[gellnick]

“Steal from the best, invent the rest”
I tend to find ideas from previous years, prototype it with any modifications I need, and then CAD it.
That’s how most mechanisms on our 2018 bot worked, at least.

[ClayTownR]

In general, you’re trying to handle an object in the simplest and most efficient way possible. Most game piece handling in FRC is already a “solved problem;” while there are exceptions (for instance, 2019’s hatches were relatively new for everyone), there’s almost always a previous game with a very similar game piece.

Nominally, the overarching design begins outside of CAD - your team will likely adapt previous concepts to making a new set of manipulators for handling this year’s game piece. However, when it comes to new ideas, you have to be clever at every phase, including CAD.

Certain designs have sort of funneled themselves into a very small solution space. For instance, you mentioned elevators. Very few teams uniquely prototype an elevator design; instead, they’re almost all very similar forms of the same architecture. How many teams built their elevator out of 1x2" almunimum tubing, using a set of roller bearings to reduce friction, powered by either a rope or chain run?

Often, design begins from either a bottom-up method or a top-down method. In bottom-up, you typically start with the elements which are directly interacting with the game pieces and figure out how to put the pieces together. In essence, you begin with whatever your known elements of design are from the prototyping phase and fill in the blanks as you go. This can sometimes result in some rather janky integrations, though.

In a top-down method, the overall path and movement of the game piece is outlined first, and then you work towards procedurally improving detail. For FRC, this usually means starting with one master sketch highlighting all the movements of the game piece, and then designing based off that master sketch. The difficulty here is that you have to make assumptions at the beginning which may not be correct; you could accidentally allocate too little space or weight and realize halfway through that it’s not quite fitting as you’d like it to. You can mitigate these problems through experience (as you get a better idea of how ideas tend to crystallize) and by efficiently parametrizing your model so that you can quickly make changes to the top-level layout.

As you grow as a designer, less and less will seem to be “new” and a lot of problems seem to be the same thing, but only slightly different. That’s part of the process. The way you build up that sense is entirely through practice - the more you design, and the more you understand the flaws and strengths in your designs, the better you become.