Once build season starts in Jan, what should I expect?

[emekablue]

Dudee, so I just started cadding in like March. I've got down how to do a lot of things. Our CAD team is learning too. But I'm new to robotics and the challenge.

So we have to CAD a robot....how? where do you start? and how much time does it consume?

[EricH]

Start with the KOP items posted in various places (3Dcontentcentral, for example). That'll take out any KOP work, and you can start now.

Now, CAD any parts that the team needs. If they're COTS parts, see if you can get the CAD for them. Again, saves time and energy.

Put them all together and you have a robot. Might take a few weeks to do, but so does the real robot.

[Joe G.]

There's nothing saying that you have to. However, its a very good idea. CAD is best used for just that, Computer Aided Design. Drawing parts of your robot on the computer first can save you a lot of headaches and wasted time later on.

How long it will take depends on the level of detail you put into it. In order to be a contender for the Autodesk Design award, the extreme detail required can take the whole 6 weeks. But to use CAD simply to figure out critical elements of your robot, the time required can be greatly reduced. Its up to you; the more you put into your CAD, the more you'll get out of it.

As for a starting point, models of components that show up in FRC commonly are avaliable online, in places like here and here. You can use these, and design your own, manufacturable parts, and combine them into your robot design.

[AustinSchuh]

One thing to keep in mind when you are CADing the bot is that even though you can draw something, it might be either impossible to make, or really time consuming to make in the way that it is drawn. So, make sure that you and whoever else is CADing knows what the tools you have access to can do, and thinks about how something is going to be made while CADing it up.

For example, one of 971's sponsors does CNC sheet metal and has a laser cutter. So, it's pretty easy for us to design something that's supposed to be bent up out of sheet metal. We don't have access to, for example, a waterjet, so if we wanted to cut something out of a material that couldn't be cut on the laser-cutter, we'd have to work a lot harder to get it cut.

[DonRotolo]

Some teams approach CAD as a way of documenting what the rest of the team has built. We see their CAD sub-teams carrying rulers and calipers a lot.

Other teams approach CAD as a tool to design parts and perfect* them before spending the time to make it out of metal or plastic.

Still others approach CAD as a way to win an award. I don't think this is a good idea. Do good work and the awards will follow.

When Boeing designed their 787, they designed the whole plane on the computer and then started making parts. That is typical of how it works in the real world. And, my advice is to try to do CAD like that.

The reality is that unless you and your CAD subteam are highly skilled, you will not have enough time to design the entire robot before it is built. While I do encourage you to try (make sure the team is behind you on that) you might consider taking the approach of focusing on a single element of the robot.

For example, an easy project would be the layout of the electrical board (for example, make sure there's room for wires, tools and fingers...). Many teams have a standard drivetrain they build each year, so instead as a more challenging project focus on the game-specific manipulator (such as last year's "kicker") - but you need the mechanical subteam working closely with you on this, and remember you don't have more than a few days.

Once the team decides on how they'll play the game and what capabilities the robot needs (all in the first week after kickoff), another week can be used to design a mechanism in CAD, after which it needs to get built. Ask yourself: Am I good enough to put their design into CAD and prove that it'll work as desired in a week?

If the answer right now is no -then get better at CAD so you can. That will make CAD a huge asset for the team.


*Perfect meaning "good enough to work" not without any flaws.