I’ve been working on robot CAD lately but I think I’m missing some points and need a few tips. What are your indispensable CAD tips you can give me?
Use frcdesign.org 
27 Likes
What kind of tips are you looking for? Organization, Techniques, or Mechanical Design?
1 Like
I expect advice from people experienced in technical and mechanical design, such as “If you pay attention to this, your design can be better.”
I’m sorry, but this is still too vague to do much with.
Can you maybe outline what you have struggled with in the past and or post screenshots/links to previous CAD that can be reviewed?
Once you have learned the fundamentals, the only way to improve is practice + review, both self review and review from others.
In general, that line, in itself, is effectively good advice. Pay attention to what you’re doing when you’re working on a design. What is slow? What is fast? Are there things you do that make it harder to work on things later?
When you have the opportunity to build or see one of your designs built, what works? What doesn’t work?
When you’ve worked on robots someone else designed, what worked well, and what didn’t?
Getting good is really about practice, and specifically intentional practice-- looking at where you’re currently deficient and working to become less deficient in those areas.
If that sounds a little too philosophical, here are some concrete pieces of design advice:
- Your drivetrain is more important than your intake is more important than your end scoring mechanism
- You can always use more driver practice and programming time. The robot design is not more important than these things.
- Use what you’ve got. Don’t use what you don’t got.
- If you’re in doubt, model the hardware.
- Fully define, every time.
- If you make a part from a model, make sure that model isn’t lost, so you can remake that part if necessary.
- Try to not be dumb, even if you’re also trying to be smart.
Good luck as you continue on your design journey!
9 Likes
We don’t know what you can do well and what you’re struggling with. We also don’t know what capabilities you want to have.
As Nick.kremer above said, post some examples. Ask for feedback on those examples. State your hidden assumptions and limitations (e.g., “the machines available to make this are a/b/c”, or “this part is being sent out to a sponsor for manufacturing”, or “this part has to be made by rookies”).
Also, go out and read resources. Watch some videos about designing FRC robots or design in general. Learn more about how the parts are manufactured, and what features cause struggles.
Finally, try to motivate the people who you’re asking for help when writing a forum post. Ask questions that generate curiosity and interest. Show (don’t just say) that you’ve undertaken the first step(s). Give us an idea of where you are and where you’re trying to get.
2 Likes
This is kinda a broad question and without knowing more about what Software you use or what you are struggling with, only basic and broad tips can be applied.
My primary broad tip for working on improving team organization: create a spreadsheet that tracks all of the parts you use to create your robot. Assign every part a number, (0001, 0002, etc.); this includes both purchased parts like motors and gearboxes as well your custom/in-house fabricated. (part file/name in your CAD software should look like: 0011 - Compliance Wheel 3in - 0.5 Hex Bore; 0012 - Chassis Front Frame Rail - 1x1x26). Use the same spreadsheet every year.
For purchased parts you can add a column that specifies who manufactures the part and another where you have purchased the part from in the past.
For custom parts, add columns to specify what season robot it goes too as well as what sub assembly (chassis, shooter, intake, etc).
We just started rolling this out this year and it has really helped keep better track of everything we use. Especially when making a parts list on our drawings.
2 Likes
You could be a little nicer and take in advice from anyone who offers it (assuming its true)
1 Like
First of all, since this is my first forum post, I may have opened an open-ended topic, I apologize to all of you for that and thank you for your answers. Now I want to share with you what I did and ask what I should do next.
This is the first robot CAD I’ve made and I did it just to get familiar with assembly.
This is the second robot I have made. Cad taught me how to use the drive train in general, how to design the intake mechanism and the feeder.
This robot is the last CAD I made (even though I’m not completely finished). This CAD taught me information such as turret, reducer and bearing.
Now what I wanted to ask you was this; What can I do after these stages and how can I further improve myself in CAD?
4 Likes
Looks like you’ve so far designed a series of pivoting shooter + under the bumper 2024 bots.
My suggestion to you is to continue to practice by branching out. Choose a game that’s unlike 2024 and design a bot for it. Review CAD for robots for that game for inspiration. I’d suggest a game like 2018 or 2019.
While doing so, you can review your CAD practices and see if there’s any room for improvement in process, organization, ect.
8 Likes
participate in cadathons. FRCDesign.org good but like u should practice with onshape features too use some courses that might benefit u. CADathons most important bc they give you practice that you need real time real stuff.
1 Like
In addition to the excellent advice above (in particular FRCDesign.org !!!), I’d suggest taking another team’s robot (2024 or prior season) with published CAD and see how they built it in CAD. What tricks did they use? How did they organize their documents/parts/assemblies?
4 Likes
The finished product does make for a fancy rendering, but it’s pretty limiting in terms of soliciting feedback. I’ll say what I can.
You clearly understand the common FRC design tropes: parallel plate, plate+tube, using COTS parts, rough sizing concerns.
For the first robot, I see a pair of pivots that are not adequately supported. The axles are cantilevered with stationary support points too close together, and the shooter in the middle is too far from the stationary supports.
For your second robot, I see belt runs with inadequate wrap on some of the pulleys, which can lead to skipping and damage to the teeth. However, the full-width intake will make it much easier to pick up game pieces than the narrow one on the first robot.
There’s a lot more going on with your third robot, so it’s harder to understand. I’d be concerned about the rigidity of the turret; the two vertical plates that support the shooter’s pitch mechanism don’t look particularly rigid. I’m not sure if the gears at the right side of the robot are to pitch that frame, but if they are, I’d be concerned about them failing.
In order to give more meaningful feedback, it would be helpful to understand what the constraints or starting guidance for each of the three robots was. What did you set out to construct for each one? How close did you get? Are features missing? Do the designs do more than was specified?
If you’re wondering how good you are at CAD, part of that answer is dependent on your practices – how you define your parts, how your sketches are constructed, your organization. It’s kind of like posting some screenshots of running software that you wrote; we don’t know how well it functions, nor what practices you used in creating its source code.
Ultimately, the goal of CAD is to manufacture, test, and iterate on designs. If you design in isolation, you’ll miss out on a lot of important feedback from your team – whether the parts are easy to manufacture, how well the parts go together, controllability, and serviceability, to name a few items.
1 Like
Practice, Practice, Practice, and never stop learning. These are in not particular order.
Learning
- Many teams publish their CAD, design binders, and much more of Chief Delphi. Look through them and try understanding how they are designing systems. You can also check out Spectrum’s list of robots.
- Watch conferences / talks. Some are linked below, but there are many more on YouTube.
Effective FIRST Strategies
Citrus Circuits
Jack in the Bot - Watch reveal videos. These usually show a good view of game objects interacting with the robot’s mechanisms. Just keep in mind that these only show the robot in the most ideal circumstances and could only be one of many takes.
- Watch onboard match video. There isn’t a ton of this out there, but It is indispensable for showing the different robots in real match situations.
Robustness & Reparability
- Design for reparability. It isn’t a question of if, but when and how your mechanism will break. Think about how you remove every piece of a mechanism. Are bolts obstructed? Can you remove an entire subassembly? Are spacers finnicky to get in place?
- DFMEA (Design Failure Mode and Effects Analysis) / design reviews. Consider how your mechanisms interact with the field, other robots, and the game objects. How can things go wrong and when they do what are the effects. Will any particular failure put you out of the match? How easy are they to repair? What spare parts / mechanisms do you premanufacture? This is important to do as a self check and as a group.
- Pay special attention to anything that extends outside of the frame perimeter. These are most likely to be damaged. I would avoid mechanisms that cannot be retracted during full field sprints.
- Use Polycarbonate (Lexan). Polycarbonate usually elastically deforms (will bend back to it’s original shape) in low and medium speed collisions making it a great choice for intakes.
- Think about how force is transferred through your robot. Your intake may be super strong, but that just means the forces are passed through to the mounts and then the frame.
Iterate
Your first design is almost never the best design. Continually reconsider whether your current mechanism can be improved. This may be a slight design change or starting from the ground up. Don’t get super attached to a certain mechanism or way of thinking. Be prepared to scrap a week of work If isn’t working how you would like it to.
Use this to inform how you design mechanisms. Make sketches easy to modify and understand when you look back in 6 weeks. Additionally, a fully integrated robot looks great, but can be difficult to make changes to. For example, if you wanted to change the shooter angle or compression you would would have to remanufacture the entire side plate as well as disassemble and reassemble a significant portion of the robot.
Miscellaneous Tips and Tricks
- Use COTS parts. In addition, look through vendors catalogs so you know what COTS parts are available when designing.
- Use simplified models for swerve modules, electronics, and gears.
- Consider how mechanisms mount together
- Use design calculators to calculate gear ratios. Onshape4FRC.com has a good list of calculators.
- Consider how wires are routed through the robot. You may need to make special cable trays or tiedown points. Pay special attention any time a wire wires move.
- Design mechanisms in place.
- Create design goals and continually reevaluate if the mechanism you are designing match your goals. Discuss if the mechanism or goals need to change.
- Consider how / if you team can manufacture the parts you design.
- Consider if your team can program what you design.
- Reduce degrees of freedom.
2 Likes
What has helped me most in the past is doing exactly what you are doing now - finding examples of mechanisms from past teams and making models of them yourself. The exercise of creating the mechanism while thinking about manufacturability, function, and repairability, will really help in the future if you want to re-use the concept.
Another thing that can help is laying out your plan and goals before you start detailing your assemblies. What do you need the mechanism to do, what space does it have to fit into, what COTS do you plan on using? Starting with a layout sketch can help keep things organized and ensure different parts of the robots can function together.
Can’t answer about the first two, but I believe it was based on 987’s 2024 robot.
First of all, thank you for your suggestions.
@KH987 As you said, I tried to design my last robot inspired by the 987 robot. Even though I am not as professional as them, I thought it would be useful to try to design such a robot based on my own CAD knowledge. I hope that one day I will have the pride of designing a robot and seeing it go into the field.
1 Like