T-shirt bot mechanical design requirements

[cxcad]

My team is working on building a t-shirt bot during the semester. I am in charge of the project and I came up with a few design requirements for the project. I would like some feedback on the checklist and some advice on how to divide out the work. The team I am working with is mostly freshman and sophomores, many of whom have no CAD or design experience.

[asid61]

If they have no CAD or design experience, I would hold off on shifting gearboxes. They are complex and the design work is better left to experienced gearbox makers IMO. A single-speed gearbox would have a much shallower learning curve. It's hard to beat the weight and size of WCP or Vex gearboxes anyway (excepting 192, whose gearbox is the best IMO).
If you do decide to go with a shifting design, I would get it checked by another team or somebody who has done work on shifting gearboxes before.

EDIT:
1. 8lb chassis? What does that include? Wheels, bearing blocks, etc?
2. I recommend hex axles. Saves time and stress in my experience.

[DampRobot]

Ok, a few things to suggest.

First, go hex. It's 2014, no one should ever have to use keys again (except maybe in CIMs).

I challenge that you'll be able to build a lot of this stuff better than COTs. The COTs wheel trucks (WCP and VersaBlocks) are already really well thought out, and even if you have the resources to make them for less than you can buy them for, I'm not sure you can really make them "better." For gearboxes, (again, depending on resources) you can probably make them cheaper and maybe lighter than COTs options, but no way will they be easier to machine or faster to make, and they probably will be less efficient as well. Well, except if you're 1717.

It sounds like you basically want to copy the poofs machine. That's cool, you'll definitely learn something from it, and the kids definitely will too. But there are other ways of throwing t-shirts that might be worth investigating. Could you catapult them, or maybe slingshot them?

But, other than that, it sounds like you know what you want. Good luck with your project, and have fun.

[RunawayEngineer]

Is there another document that describes the performance requirements of the machine? Top speed, range of shot, sweeping angle of shooter, trajectory angle(s) of shooter, maneuverability, etc. I don't see how you can design without at least a couple of these. "Driving places the fastest" doesn't narrow things down.
What is the reason that you are barring using chains/belts outside the drivetrain? I am assuming that it is to make the machine more reliable - and therefore to get practice in designing a competition machine that is reliable. Personally, I find it counter-productive to ban solutions to problems because examining different solutions is a key part of the engineering process. I know that you have to draw lines somewhere - but belts/chains are not only extremely useful in FIRST, but in industry, as well. The off-season is the perfect time to test out your assumptions.
That being said - I think that it is a great idea to use a fun project to teach/practice DFMA.

[asid61]

Why not use belts or chain in the drivetrian, then require them to use the one they didn't use in teh drivetrain for the shooter? So chain drive, belt shooter or vice versa.

[cxcad]

Thanks everyone for all the input,
asid We are kind of low on budget right now, so we are seeing what we can do by ourselves, though probably for the season we will buy something COTS. 8 lb chassis is just the metal framing. Thanks for the recommendation on hex shafts.

DampRobot For the regular season, we are probably going to use a lot more COTS components. I feel that by going custom, the students will be more involved in the design process; we also want to try out some possible sponsors. I do not see how a custom gearbox will be any less efficient than a COTS gearbox; we have a fabrication sponsor that can nail a four ten-thousandth of an inch in tolerance.
I hate the fact that we are pretty much copying the poofs machine. During the summer we did some research, and thought that going with a tried and true method would be better. Any suggestions to differentiate the robots.
Runaway Better than the poofs . I will put some thought into those constraints. I'm trying to limit chain, because I do not want to dealing with chain tensioning.
aside Good idea. Will make those changes soon.

Anyone have any suggestions for teaching how to make this robot? Also what do you guys think about skipping the single barrel design and going straight to 254 style gun?

[asid61]

If you want your robot to be different, why not make an auto-loading mechanism for the t-shirts? That's something I wished was on the Poof's cannon.

[artdutra04]

Originally Posted by cxcad

Anyone have any suggestions for teaching how to make this robot? Also what do you guys think about skipping the single barrel design and going straight to 254 style gun?

Have you built any tee shirt cannons yet?

If not, you probably want to experiment and make a prototype single barrel shooter first. Experiment with air pressures, air discharge volumes, etc. Try to increase your efficiency of shots - make sure you can shoot eight tee shirts in a reasonable time before making an eight barrel launcher.

If you're making a multi-barrel shooter, most of the "coolness" from it just stems from the fact that either A) multi-barrels looks visually cool or B) you can rapid fire. Perhaps you only need four barrels, or maybe you can have a breach loading magazine and achieve the same level of coolness with a simpler/smaller mechanism.

[cxcad]

Originally Posted by artdutra04

Have you built any tee shirt cannons yet?

If not, you probably want to experiment and make a prototype single barrel shooter first. Experiment with air pressures, air discharge volumes, etc. Try to increase your efficiency of shots - make sure you can shoot eight tee shirts in a reasonable time before making an eight barrel launcher.

If you're making a multi-barrel shooter, most of the "coolness" from it just stems from the fact that either A) multi-barrels looks visually cool or B) you can rapid fire. Perhaps you only need four barrels, or maybe you can have a breach loading magazine and achieve the same level of coolness with a simpler/smaller mechanism.

We have someone working on developing the pneumatic system right now. I'm not sure how a four barrel design is any simpler. And from the research, I have done, it seems that breach loading is difficult with t-shirts. I'll give this some more thought

asid Any suggestions for auto loading? I'll look into it

[Caleb Sykes]

Gear ratios must be optimized for driving places the fastest

What kind of places?

[asid61]

Originally Posted by cxcad

asid Any suggestions for auto loading? I'll look into it

One team pre-loaded the t-shirts into pcv pipe, then the pvc pipe was ejected like a spend shell after firing. You could also prepare the shirts for shooting by rolling them, then putting that thin, clear tape around it to hold it in position (or use rubber bands).