Help With Offseason Cannon!!!

Our school has asked us to build a t-shirt shooting robot to use at different local school district sporting events in the Sacramento area. Our team is looking to build an offseason t-shirt cannon. We are a big fan of team 254s and team 4910s t-shirt cannon. Does anyone know of an easy to build t-shirt cannon robot, that takes minimal materials? We are starting from scratch and need something easy and quick to build. If someone can help us out with a list of materials and some guidance on instructions that would be AWESOME. Thanks!

Ours is probably as easy as it gets. We used some 2" steel threaded piping and fittings for a reservoir, four 1" sprinkler solenoid valves (each run by a motor controller), and four 3" ABS barrels. The reservoir is fed from a 15 lb CO2 tank that is good for about 80 shots on a $20 refill. The pipe, CO2 tank, and valves cost a few hundred total.

Incidentally, we had some success with a wheeled t-shirt cannon, which is even simpler. The catch is you really have to spin the wheels. We got a ~50 yard shot out of it, only after we threw four cims, 1:1 gearing, and 8" wheels at the problem :slight_smile:

If I had to create a t-shirt cannon, I would lean more on the side of a motorized or pneumatic catapult. The stuff on cannons like 254’s is for many reasons beyond the knowledge of FRC students just cause its different tech. By creating a catapult with standard FRC stuff your kids can get more training on stuff they would typically use on an FRC robot, plus it will be simpler to troubleshoot stuff your kids and mentors have experience with. Designed well and placed on a preexisting/cots drive base like the Vex drive in a day or Andymark kitbot drive, you could feasibly build a t-shirt launcher in a day or two.

Insert argument about PVC here

Here’s a few past ones:

What did you use to connect the PVC pipes to the CO2 Tank and how you actually engage the cannon and shoot the t-shirts, was that air pressure or did you build a mechanism inside to push the t-shirts out? Sorry, I am new to building t-shirt cannons. I love the design, you did a great job!

I like that idea, would you use pneumatics to activate the catapult or would you use strong rubber bands to release and a 775 pro motor to pull back the arm to engage it? I have seen catapults done both ways, so I am just wondering which idea you were thinking of or if you had a different idea in mind.

It’s not PVC. PVC, as linked up-thread, has a brittle failure behaviour that could lead to dangerous shrapnel when used in compressible gas service. The reservoir (bottom right of the robot) is galvanized steel pipe. The barrels (which shouldn’t hold pressure anyway) are ABS, which is a lot more ductile than PVC.

The CO2 tank is mounted in a strong vehicle mount and connected to the reservoir through a regulator (set to 40-100 psi, depending on range), and flexible tubing. The reservoir also has a relief valve and vent valve which we use to depressurize the system when loading t-shirts. The reservoir reduces down to a manifold made of 1" steel pipe (in a Menorah sort of shape). Each barrel is fed with one of these sprinkler valves. These valves probably worry me most since they do have PVC bodies; I hope to replace them with a brass model soon.

It’s a really useful outreach tool. It does have some unique risks (most FRC robots don’t store this kind of energy) so we have some pretty strict operating rules for safety. Be careful and it can be really be fun!

Yeah, it looks great. That makes more sense that it is ABS, I must have missed that. You had mentioned that you had made one with a flywheel shooter. Would you have to make the t-shirts into a ball shape? Which one would you say would be an easier robot to build if we have never built a t-shirt cannon before?

We were able to fire the same folded and rolled t-shirt burritos out of the wheeled shooter as the pneumatic one. Here’s a photo of it, in a very rough “first iteration” version:

The wheeled shooter took, and will still take, more engineering to get right (and to add a magazine / hopper), but it might have been cheaper had we gone that route. The pneumatic one was easier to get together for a pep-rally deadline.We might return to the wheeled shooter version soon though. We’re planning to convert it to shoot stress balls, which are cheaper ammo than t-shirts.

You could do either pretty easily. I’d steer clear from a pneumatic catapuld, because then it requires you to have a whole pneumatics setup for just one function (assuming you dont have pneumatics anywhere else). Beyond that, you could use a winch type thing, a cam, a choo choo? or like one team in Carver (I forget who) did just attatch some motors (through a gearbox) straight to the catapult arm. The hard part is getting the system to work consistently without failure (a broken mechanism is the worst thing to happen at a demo), there are many options for how to fire the catapult, its all about your team’s manufacturing and design expertise. Good examples of non-pneumatic catapults include 1836 2016 (we used a cam), the 179 2016 offseason bot, or one of many many 2014 bots who used this type of thing. The Vex Build Blitz series also gives some good insight on how a choo choo catapult works if you want to go down that road. Do some prototypes to find a ballpark spring tension with surgical tubing (probably fine cause no accuracy is really needed) or springs (we got them from mcmaster) and with good prototyping finding a decent way to pull back the catapult will be the easy part

That is a great idea. Do you know what wheels you used on that? I like the stress ball idea, that is cool to shoot out too. Would you add the hopper to store the balls/shirts so you don’t have to refill the robot each time? Thanks so much!

That is a great idea, we will have to discuss it and see which one is the most feasible for us. Thanks so much!!!

They look like KoP wheels from 2016/2017 as the first stage and 2010as the second.

If you do use pneumatics, the key to working at pressures similar to those allowed in FRC is to have 3/4" or larger lines and valves from your tank to the barrels.


All three years used Hi-Grip 6". Based on what we had in inventory at the start of 2017, I’m pretty sure the 2012 wheels were translucent, not opaque white.

Thanks, I think you are right! I really appreciate your knowledge!

The demo wheeled shooter was more like a “mortar”. We feed t-shirt burritos in one at a time by hand (pushed in with a safety stick). If we return to this design, we’ll add an auto feeding magazine of some description. For stress balls, it may end up looking like a mini SteamWorks bot :grinning:

You’d be right for the 2010 ones, I think. The others are from 2012 if I’m not mistaken, look like 6" Hi-Grip (white).

Awesome! Thanks so much!

I have one more question. When you were feeding the t-shirts in that picture, was that just a plain rolled up t-shirt or did you have some tube or did you wrap it in something to feed it through?

Happy to answer. They were plain rolled up t-shirts without any wrapping. We either use the inside out sleeve trick or added a bit of painters tape to keep them rolled. The diameter has to be pretty perfect for the pneumatic cannon, but it’s less critical for the wheeled shooter.

Sounds Awesome! Thanks!