So my team is looking to use CAD to build custom parts on the robot next year or even develop a CAD subteam. We figured that veterans on departments with light or close to no workloads during the early weeks of build season could model stuff that can’t easily be built by hand.
Any tips on where to start, especially with what program(s), teaching methods, or how to practice CAD in the postseason?
Solidworks and Autodesk Inventor are both excellent CAD programs to use in robotics. Both have free student versions of their programs that you and your team can use. I personally use Solidworks but they are pretty much identical in most regards.
You can find the signup for Solidworks licenses here.
https://www.solidworks.com/sw/education/robotic-design-resources-students.htm
I’d also recommend you familiarize yourself with GrabCAD and the GrabCAD workbench. GrabCAD Workbench and its app basically lets you share and upload CAD files with other team members. Its kinda like DropBox but specialized in CAD.
As for training? Simbotics has a few good training tutorials you can get started with: https://www.youtube.com/playlist?list=PLG_KOHBuXHNfpa1N23ZJfAHsCVbe3fpXG
How you implement CAD into the team really depends on the machining tools you have available to bring your designs to reality.
We personally use Autodesk Inventor. While I have used Soildworks before, I find Inventor easier to use.
You can grab a free license here: http://www.autodesk.com/education/free-software/inventor-professional
Like waterbott said, get yourself familiar with grabcad. It’s really useful for interactive previews. If you want a simpler method, google drive works just as well.
While I am not aware of any Inventor tutorials, likes simbotics has for Solidworks. (Someone post them if you know of some), I personally learned by just following Inventor’s tutorials (integrated into the app), a bunch of googling, Youtube videos of a specific question, and some creative thinking. After I learned, it was relatively easy to teach just as capable team members.
I would not recommend using Google Drive for version control. You sacrifice a lot of control over what is/is not being updated that you have with GrabCAD or other software actually designed for CAD use.
I’m going to suggest that you look around and try to find mentors before you decide on a CAD package. It will be much, much easier to implement CAD on your team if you have a mentor or someone else familiar with the software to guide you through learning it. If you cannot find a mentor, I would suggest Solidworks, as it’s a good midpoint between being easier to pick up and being actually used outside of education. I’m a big fan of Creo and find it the most intuitive to use, but I know a lot of people don’t share that opinion. You probably want to download and try out each package-- they’re all free to students and mentors in FIRST.
Really, anyone designing parts from your robot should learn CAD. Once you have it learned, it is such a time saver compared to trial and error by hand. Some builders take to it more than others, but if you really want to teach or learn to engineer a robot, CAD is in my mind an integral part of that process.
Oh yes, ofcourse. I am a much bigger fan of grabcad than google drive, was just offering options. And ya, I’m not really a fan of creo, given I didn’t try it out too much, but it appeared to be way more complicated 
And hey OP (or anyone else) if you need any help or anything, just let me know 
Also, in my opinion, I think it is beneficial for at least most of the mechanical subteam of FIRST teams, if not all of them, to at least know how to 2D model stuff in CAD. I have found it is very helpful for students to be able to model their ideas in a 2D sketch and put dimensions on drawings so they can see how things need to fit together or see how big a component will be before it is made.
It’s really good for the students because they can explain their ideas better and how it fits into the robot as well as helps them understand that sometimes components won’t work due to size constrictions that they wouldn’t otherwise be able to visualize.
Anyway, good luck to you guys with your CAD team! Solidworks is my preference, but as others have mentioned, Inventor is a easy to jump into as well.
Thanks for the advice!
So right now, it seems that we’ll definitely be taking advantage of GrabCAD (I remember how useful GitHub was for programming), but we’re still deciding on whether to go with Inventor or Solidworks.
Pick the one that your mentors use most. If none of your mentors use either, and you can’t decide, then flip a coin. They are both great programs.
Ngang,
I think you are in danger of investing a bunch of energy for little return. I might be wrong, but I want to caution you.
Setting aside CAD classes and vendors, “CAD” is a tool and not a product. That means saying you want to “Implement CAD” is sort of like saying you want to implement screwdrivers, or that you want to implement bandsaws.
Like you said in your post, what you have in mind is building stuff; and because CAD is one way to specify/model stuff, and because good specifications/models help human avoid mistakes; you are thinking that investing in CAD might pay off.
Let’s think about the big picture for just a minute. Unlike bacon, CAD doesn’t always make everything better. 
Producing your system(s) (Robot, Driver Controls, Pit setup, etc.) is a real, tangible goal that has explicit boundaries, constraints, start/stops, etc. On the other hand, “learning CAD” is an open-ended notion that can turn into a lifetime of (satisfying) learning.
Also, there are alternative tools. Napkins and crayons can help you produce those systems. Pencil/pen and paper/vellum drafting can also help you produce those systems. Physical models made from simple materials can also help you produce those systems. Lots of talking and gesturing can help you produce those systems.
Before you pick a tool, remember that no tool is going to be helpful if you aren’t clear about what the tool is supposed to help you do. So, before choosing to learn any specification/modeling tool (like a CAD tool), learn what information must be specified/modeled in order to successfully answer questions about the thing you want to produce, and THEN pick a method/tool to use recording/communicating that information.
In your case I think you need to think through these questions:
**Who is going to consume the specification/model your chosen tool produces?
What questions will those people have about the thing you are producing (your custom part)?
What is the easiest way to create a specification/model that encodes and conveys the answers to those questions?**
Bottom lines:
Learning a CAD tool is good, if it’s a tool you need, and/or an investment in your future.
Learning a first CAD tool; becoming proficient with it; and then using it to record the hundreds (at least)
of decisions/data needed to describe an FRC robot, or the tens of decisions/data describing how one simple custom part integrates with the rest of an FRC robot, is a fairly big undertaking.
Definitely take the plunge, but A) Learn what needs to be specified/recorded, B) Plan to walk before you run, and C) Use CAD when it’s the right tool for the job.
Blake
PS: SketchUp is an OK modeling/specification tool to use, especially if you want fast results, from an easily obtained/installed tool, for a simple task. On the other hand, it’s something of a dead-end; and isn’t a CAD tool you are likely to want to hang a career on. I use it because it’s the right low-overhead tool for many of my jobs.
I would suggest picking up Solidworks because some of the higher-level teams ie 254, 973, 1114 have VERY good tutorials on both basic and advanced topics. Usually Solidworks and Inventor are toe-to-toe when it comes to features, but I feel that FRC-specific help is more prominent with Solidworks.
Ngang, think about what gblake said. While I would 100 percent recommend cad, what gblake said is also true. If you don’t feel that your guys have the focus and power to learn cad, and even have the slightest inclination that they might get frustrated and quit half way through, don’t do it. It will only mess you up.
On the other hand, if you know you have the focus and resources to do it, than forge ahead.
I can’t stress this enough, every free second you get before the next build season, practice, practice, and practice. DO NOT wait until the build season to start learning how to cad… it is a huge steep learning curve. My advice is (and what I did, and make new guys do) once you learn the basics, CAD up your bot from this year, before build season next year. Learn how to CAD every component. This way, once you get done, you will almost certainly be ready for any part that you might have to CAD up next year.
We personally use inventor, but for me (since I was the first one on my team to ever cad (no mentors)), it took a long time to learn everything, months. But now that I know, I’m able to teach the next generation and hopefully that will keep going in the future.
Since all the FRC tutorials use Solidworks, you might be compelled to use Solidworks. But I honestly can’t say. If you do decide to use Inventor however. You can message me anytime asking for help or advice, and I will be glad to help.
Yeah our team just really started to incorporate cad into our team. We use Solid works and love it. I have personally worked a little in autodesk inventor as well and find that Solid works can be more intuitive. Our team has really benefited from it and hopefully your team will too!