How do you incorporate manufacturing concerns into the design?

How does your design (CAD) team make a robot that incorporates manufacturing/pit members’ concerns? Or, how do you merge the manufacturing and design (CAD) teams?

1 Like

The best way I’ve seen to manage this is to not have a CAD team. Instead have members of your mechanical team who know CAD, but also know the mechanical stuff. So when they’re designing the robot in CAD they take all that stuff into account, because they’ll also be the people manufacturing the parts and fixing it in the pits. The mechanical members who don’t know CAD should also be right there when the robot is being designed. Even if they aren’t holding the mouse, they should be an integral part of designing the robot. Same goes with electrical and (perhaps to a lesser extent) programming.

10 Likes

One way is to envision how the parts will be assembled, what tools will be needed and what access those tools will need.

Frequent design reviews with people “who have built a lot of stuff” would also be beneficial.

1 Like

This, this, and more THIS.

I generally recommend that engineers-in-training take a machine shop course, or have machine shop experience, at some point in their early education. Say about now. That way they may learn how to allow room for the machines to get in on the parts, and how to THINK about accesses and the like. if you can’t get the tool in to do the job, you need to redesign (or place an extra hole).

Also: The smartest engineer isn’t the one with the highest-level degree. The smartest engineer is the one who goes to the people that have to build something and says “I am looking to do THIS, and I was thinking of doing it THAT way. What issues do you see?” and then listens to their feedback on why THAT way is a bad idea and THE OTHER way will work.

I have a, er, number of examples. Suffice it to say that I try to at least design the stuff so it CAN be built, and ideally I’m the one getting the pain if needed…

Back on track: The best way is to make sure that the lines of communication are open. I’ve sent drawings back to CAD for missing dimensions on many occasions, or gone and talked to them about what they mean by X, and worked with them. I don’t do much CAD on the team myself–I train shop tools and some CAD–but if you make a mistake I’ll be helping you to not make it in the future.

7 Likes

Personally my team has the build captain help co-design so that they can work together to make sure that they’re aren’t any major build concerns and then the design captain co-builds the robot so they can make sure the design aspects are coming together correctly. Although it used to be just the build captain telling the concerns to the design captain and them just working together to make sure they communicate possible problems.

We don’t.
/s
We build it in CAD, then either CAD lead or someone from mechanics checks it. If it’s good, it gets built/made. Personally, I’ve been known to make some REALLY annoying parts for mechanics to make. Some of them actually did get made, so I guess they can be made.
Cues in custom swerve drive with no tolerances whatsoever

2 Likes

EricH
I got involved in robotics courtesy of one of my sons. This was before FIRST but the same concept, mixing the virtual and the physical, the academic and the practical. Way smarter than me he never took a single AP class in High School. He did however take every tech ed class available. His ultimate destiny clearly was to be an engineer but when he graduated ( early, HS had had enough of him and vice versa!) he went to tech school and got a 2 year degree in precision machining. He then worked a couple of years before admitting that I was right…that he knew more than the people running his company. Then and only then it was off to college for a manufacturing engineering degree. He now designs robotic assembly equipment and is one of those incredibly rare people who can CAD it up…then go across the hall and fab the prototype.
He also helps mentor our FIRST team. I tell the students they should always question my ideas. But to my son the best response is: “My Liege, it shall be done.”
Ah well, long winded reply but that’s how I ended up here…
TW

4 Likes

Precisely. Teach either the CAD folks how to do mechanical stuff, or teach the mechanical team how to do CAD. Or both. CAD is pointless unless there is some understanding of how it all works in reality.

832 is a small team, and so by necessity almost everyone knows at least a little of Inventor, while the team leads are all quite fluent.

3 Likes

I’m with AriMB on this. Of course, our team is small enough that we really couldn’t have a separate CAD sub-team anyway, so it’s kind of moot. Our mechanical team does the limited CADing that we do. Most of them have CAD training from classes at school, so it’s really just about using the skills they have to aid in the design and build process.

Not having a separate CAD team as @AriMB suggested promotes a much more collaborative design and development process.

Sometimes, when the CAD team is a separate group, the CAD team takes on the attitude of “I designed it, it’s your problem to figure out how to build it” and/or “my CAD is perfect”. I have seen this at several workplaces and on a few FRC teams I know well. It often takes a lot of time and effort to clean up after designers with that attitude and to rebuild the trust of the manufacturing people.

1 Like

Using exploded views is a great way for your team to experiment with how things will actually go together while still playing in CAD land.

3 Likes

This is a great idea that I’ve never thought about before. Thanks!

2 Likes

Summarizing several of the above, it comes down to three words:

  1. Communication
  2. Communication
  3. Communication

The best product design and engineering always involve communicating with all the stakeholders, whether you’re designing a bookshelf, a killer app, a fire truck*, or a robot. You can’t meet every need with any one product, but you’re certain to miss a big one if you don’t know and understand it, or at least communicate with someone who does.

More specifically, meeting manufacturing and maintenance concerns is a matter of looking at each assembly and sub-assembly and thinking about:

  • What order does this go together initially?
  • How easy will this come apart and go back together if needed? Can the tools go in? Will the parts line up automatically, or will it require jiggling? Will this take three (or more) hands? What if I drop a screw - where will it go?
  • How often will this have to come apart and go back together (if it moves or goes outside the frame perimeter, this is nowhere near zero).

* As I understand it, nearly every fire truck is a one-off, designed to meet the specific needs of the neighborhood it will work, and the firefighters who will rely on it.

2 Likes

I completely disagree with this. At Least for my team as a larger team it would not be practical or efficient to have our manufacturing team and cad team to be one and the same. We have probably around 20 manufactures and 10 designers which would lead to a large group of 30 people who have a little bit of cad experience and a little bit of manufacturing skills, rather then having a very strong cad team and a very strong manufacturing team. So the way we deal with this problem is we have what we call a critical design review, this is where we set a date for the design to be finished then we sit people from every subdivision and look over the designers work. So fab tells us if we can or can not make something and assemblers make sure that it can be put together. This I believe is the best way to deal with this problem rather than giving everyone a little bit of skill in everything.

I believe that this is not a problem if you communicate correctly between your teams and if you set out certain rules that the designers follow, we have a few and they are all unspoken but they allow us to not clash between the teams. As the design lead I teach this to the new caders from the beginning.

This is however a very time consuming process when you cad your whole robot.

100% of the robot does not need exploded views however compact, critical and complex assemblies all benefit greatly from the design team being more consensus and thorough.

Every team should strive to CAD their whole robot. Also remember to include fasteners in your CAD, fastener interferences will get you every time

Fully understand that, we have had very little to no interferences between fasteners because we cad them in and around our parts. If you read my other posts about what we use (the critical design review) we have the cader explain what they did and what they are not sure is doable when explaining the cad to the audience.

You are correct if the people all want to communicate and accept feedback. Experience has shown that this does not always happen and that some designers have the attitude that their only responsibility is to crank out designs and implementing the designs is not their job. Merging the design and manufacturing groups into one forces such designers to confront and deal with their errors rather than letting it be someone else’s problem.

What you are saying can also be fixed with proper teaching, if you have strong teaching that teaches them how to work with other subteams and that the lead of this group understands it very well that they are able to make sure the subteam members are not being rude about it.