[Cory]

I think it's great for people to learn FEA, but you'll likely run into a number of issues:

1) garbage in, garbage out. If you've never done this before, your input parameters will likely be wrong. You will be fed a result that shows you some pretty colors and you will be tempted to believe the answers are true. This is even easier to fall victim to if you don't have intuitive experience/a way to do some hand calcs to verify you're in the realm of possibility. It's best to learn on simpler bodies with simple force setups so that you can verify what you're seeing is what you should be getting, via a hand calc.

2) You can't accurately model the effects of gameplay, because we can't quantify all the forces the robot sees. As a result you'll need a huge factor of safety, which ties into...

3) It's really just not worth it from a time and effort standpoint to do FEA on something like a gearbox plate during build. Just build it beefier. Some things might be worth it, like sizing a shaft or other simple bodies with limited forces being directly applied to them, but a gearbox plate gives so little relative savings by being more aggressive with pocketing that it's not worth it.