Is CAD essential to all teams?

[ASmith1675]

CAD is merely a tool. When used in effective combination with prototyping and effective game analysis it can help to raise your capabilities.

Designing a robot completely within a CAD program will allow you to see issues and interferences in 3d space that you may not have accounted for with a design in your head, or even on paper sometimes. The more detailed the CAD model, the less likely you'll run in to these type of issues while building.

To make full use of these advantages requires a slight change in thinking (in my opinion). While simply building things in a computer may not feel like much is getting done, by properly planning out where everything is going to fit, it reduces the time to put everything together. This might mean you don't have ANY parts for your robot for week three (or later sometimes).

The other thing that needs to be taken into account is the manufacturing capabilities of your team. If you don't have access to a mill (manual or CNC) you may have to account for the lower accuracy with which a part can be manufactured. If you have a sheet metal sponsor, you may be able to have parts cut and bent to very high precision. These issues are the ones that are often most difficult to account for. (Can't tell you how many times I've thought of a design that would be difficult/impossible to manufacture given the facilities and resources of the team that I am working with).

CAD is just a part of the equation, though arguably a big one. Being able to effectively analyze the game (to pick the right mechanisms to build), and prototype effectively (to verify your "theory" designs work in the real world), also play a critical role. Can you build an effective robot without CAD? Definitely. I do believe that becoming more proficient with CAD will tend to increase the consistency of robot builds year after year (meaning fewer of the "our robot didn't do what we wanted" years).

The other posters in this thread have provided some excellent resources to get started. Just digging in and messing around with the programs is a great way to learn.

[cadandcookies]

I'll answer both the title question and the OP's question.

1: Is CAD essential to all teams?

No. Not even close. CAD is a great tool, but you can run a successful FRC team (and build a competitive FRC robot) without CAD.

2: Is CAD the next needed step in order for us to become successful?

Possibly. I think it can be hugely valuable for a team to learn CAD, whether you have mentors in it or not. Be aware that you are in for a slog if you don't have any mentors coming into learning CAD-- when you are learning CAD, you are learning a new language, one that translates your thoughts into shapes on a screen instead of words on a page or audio signals. If you don't have someone who already speaks that language, you are going to figure out a lot of stuff on your own. Like others have mentioned, you're looking at a couple of years to really integrate CAD into your team, especially if not everybody is on board. Don't be afraid to design only parts or subsystems of your robot first-- you don't need to just jump into the deep end.

Best of luck in your future seasons.

Also, ff you choose to learn Creo, feel free to PM me with any questions. While I'm not a professional engineer, I am pretty good at using and teaching Creo for FTC and FRC (by the way, this is a standing offer to just about anyone-- I'm not a complete expert, but designing stuff in Creo is a hobby of mine, so I know a decent bit).

[pmangels17]

Originally Posted by cadandcookies

-snip-
Don't be afraid to design only parts or subsystems of your robot first [in CAD]-- you don't need to just into the deep end.
-snip-

bold added for clarification.

This is also quite true. Designing an entire robot in CAD might not be a feasible or efficient solution for your team. However, stuff like the paper-template-to-make-gussets technique, which requires CAD, could definitely prove useful to a team trying to manufacture more accurately. Also, even rough sketches to find correct geometries or to make sure everything will fit in the space you want can be a huge help to robot construction. Additionally, this will help you get your feet wet and give you real experience with CAD during build season, while at the same time making sure that if CAD is not a solution for your team, you haven't invested so much time in it that other design approaches will no longer work.

[waterbott]

Hey Johns Creek! Its nice to see you guys on Chief Delphi. PM me, I might be able to help you guys out with getting started with CAD.

For us, CAD serves as both a tool and a supplement to design; however, it is important to remember that the quality of design through CAD can only be as good as the users. Our team has been successful without CAD before; however, the quality of design depended highly on having a solid understanding of the mechanics behind the design. In many ways CAD serves a a crutch as it helps us visualize some design problems before they are built; however, without a solid understanding of mechanics, CAD can only help so much.

For our team, we started recently started the process of designing with CAD first before building it. We've had a mixture of successes and failures from this approach. For example, one thing we noticed is that because many of the members dedicated to CAD had little background working in fabrication and assembly, we weren't "designing for manufacturability", meaning that we created designs that were either too complicated to fabricate or difficult to assemble. The convenience of CAD made us overlook some of these problems. One of the successes we had with CAD is that it let us lay out the design before it was built and archive an analyze and changes we make. It proved to be an invaluable tool for visualization as well as easing the fabrication process.

Conclusion to this rambling? Having a successful robot primarily predicates on having a foundational understanding of what you are building, CAD serves as a tool that can only be as good as the person who wields it.

[GreyingJay]

The real answer to your question is: to take your team to the "next level", you need to start working through a proper engineering process.

You can build a robot by assembling the Kit of Parts and then spending the next six weeks trying to improve it. "Ooh, I have an idea!" "Well, that didn't work." "Ooh, I have another idea!"

However, this is not engineering. This is hacking.

Engineering is a disciplined process. There are various definitions, but one I found on Wikipedia lists these steps: research, conceptualization, feasibility assessment, establishing design requirements, preliminary design, detailed design, production planning and tool design, and production.

For FRC, that would look like:

Research/Conceptualization - what is the game? what are all the possible things a robot could do? What are the offensive actions? What are the defensive actions? What are the point values of each? What are the cycle times of each action? What can be done in a 2-minute match? In a 3-robot alliance, how could each robot contribute?

Feasibility Assessment - Of all these things that could be done, what are the easy things? What is harder? Is a harder task worth it? This starts to take into account your team's level of ability and available resources. For example, climbing is a hard task. Maybe it takes 20 seconds. It's worth 10 points. Are there other actions you could do in 20 seconds that could score the same, such as scoring two high goals or damaging the fifth defense?

Design Requirements - Based on all the above, choose what you are going to focus on. Lock this down early to prevent scope creep. For example, we decided early on that our robot would be low bar capable, would shoot low goals, use vision assist for an autonomous low goal mode, be FAST, and be able to handle all the ground obstacles. It was tempting at various stages to try to add a climber, high goal capability. a sally port door opener, etc. But we knew we didn't have the resources to build a Swiss Army Knife on wheels. We kept our focus on the design requirements we agreed on, knowing that if we had time later, we could go back and iterate.

Preliminary design - brainstorming, napkin sketches, prototyping. Now you have a rough idea of exactly what you want to build.

Detailed design - now you nail down exactly how the mechanisms are going to be built, their dimensions, weight, cost, etc.

Where does CAD fit into all this? It helps you with the design stages, and it forces you to actually DO the design stages so you're not tempted to jump right into the shop to start cutting metal without first having a plan. Plus, it is a tool to help you with those design stages so you can play with dimensions and designs without committing to building anything first.

[protoserge]

Originally Posted by GreyingJay

The real answer to your question is: to take your team to the "next level", you need to start working through a proper engineering process.

...

Design Requirements - Based on all the above, choose what you are going to focus on. Lock this down early to prevent scope creep. For example, we decided early on that our robot would be low bar capable, would shoot low goals, use vision assist for an autonomous low goal mode, be FAST, and be able to handle all the ground obstacles. It was tempting at various stages to try to add a climber, high goal capability. a sally port door opener, etc. But we knew we didn't have the resources to build a Swiss Army Knife on wheels. We kept our focus on the design requirements we agreed on, knowing that if we had time later, we could go back and iterate.

If there is anything to take away from this thread, this statement is it. Understand how to set requirements and design to those requirements.

That's good engineering.

[Boltman]

Not essential..can help

We uses CAD mainly to figure out overall bot dimensions for what we wanted to do and bumper placement in first week or so. It was something "to do" until the parts came in.

[Kevin Leonard]

I'll join the chorus here and say it's not essential, but useful.

I'll also add that my use of CAD in FRC has helped me immensely in college. It's a phenomenal tool for solving tight packaging problems (which this year definitely had).

[messer5740]

Is someone willing to direct me to tutorial videos on CAD or to send me an email about how I can incorporate CAD into next year's design?

[indieFan]

CAD is a great tool provided the people using it understand that just because something may be designed in CAD doesn't mean it may be manufactured.

My example for students is always: You can draw a solid steel sphere with a hollow core, but how are you going to make it with no seams? As far as I know, you can't.

The other example is that a student drew a wheel knowing we had access to a CNC machine. What the student didn't know/understand was that it was a 3-axis machine and couldn't do the angled portion of the wheel that he was expecting to be done.

[protoserge]

Originally Posted by indieFan

CAD is a great tool provided the people using it understand that just because something may be designed in CAD doesn't mean it may be manufactured.

My example for students is always: You can draw a solid steel sphere with a hollow core, but how are you going to make it with no seams? As far as I know, you can't.

The other example is that a student drew a wheel knowing we had access to a CNC machine. What the student didn't know/understand was that it was a 3-axis machine and couldn't do the angled portion of the wheel that he was expecting to be done.

Out of curiousity, would you mind posting the wheel drawing in a new thread or upload an image? It might not be impossible on a 3 axis machine and would generate some good discussion.

[indieFan]

The wheel was designed back in 2003, I believe. I do not have any copies of the design. Is it possible? Probably, but not cleanly on a 3-axis machine with the tools that we had available per the Professor whose lab it was in.

[SteveGarward]

CAD is a tool, like pencil and paper. We use it extensively, but we have experience in doing so.

As others have said, work on your process. To me, more important than CAD are, in this order:
1) A solid understanding of your means with which to build, and how to use them effectively
2) A solid game strategy
3) A solid, decisive list of tasks you want your robot to achieve (this comes from working out your gameplay strategy, and does not have to be everything in the game!)
4) Building effective, rapid prototypes of the mechanisms you want to build for the robot (good enough to take final measurements from, to draw up/build real robot assemblies from)
5) Get some of the basics down, and don't reinvent them every year (for example, work out a good drive train, stick with it a few years, so you can focus elsewhere. Then try more complex stuff when you have more experience)
6) Keeping things simple

Once you have that down, you can more easily start to introduce CAD as a tool to help with the design of assemblies and the robot. If you can build working prototypes, you can build better final assemblies, whether you use CAD to help with that or not. If you plan these around readily available components like versa frame and gussets, you can still build well, without CAD.

For example, here is our second prototype of our shooter this year, both built in 1.5 weeks, along with all the others. The first was very, very rough, made from plywood. After some sketches and working out geometry and measurements on paper/whiteboards, we built this which was a much closer-to-final working model. From this, when we had it working just right, we measured it up, then drew up that as a sketch in CAD, to use in final modeling of parts.

Obviously if you have a sponsor cutting parts for you, CAD will be essential. But to me, that comes in time. My list of 'things that come before CAD' is kinda like getting your house in order first.

This also doesn't mean don't learn it - it takes time. Start now. But that doesn't mean it's the solution to every problem. If you are too slow using CAD for build season but are relying on it, you may not have a robot at all..

[waialua359]

CAD is essential to our team because I hate wasting material and eyeball engineering.

[Michael Blake]

Originally Posted by waialua359

CAD is essential to our team because I hate wasting material and eyeball engineering.

AND it certainly eliminates the "insert magic" phase of building a bot.

Example: "How we gonna mount the motor/gearbox to move that turret?" "We'll just see how it looks after we build the tower and TA-DA 'insert magic'"

NOPE. Doesn't work.

LOL ;-)

--Michael Blake