17 replies to this topic

[spasticteapot]

I'm currently in a robotics competition called FIRST robotics, in which we have to build a robot to complete a task. This year, it involves launching 8" squishy foam nerf balls (techincally KOOSH brand balls; but they're identical to the Nerf balls anyway) through a hole in a wall, 8 1/2' up in the air. The robot is limited to 5' in height, and the ball cannot move faster than 26 m.p.h.
Does anyone have any good ideas for an automatic or semi-automatic gun that can be mounted on the robot? We've got an air compressor, but it takes a while to build up pressure, and we have to launch ten balls in the first ten seconds, so mechanically based designs are preferred. (We've also got oodles of servomotors, winches, and the like, so we can make mechanisms that can do anything your hands can.) Accuracy is VERY important, perhaps more so than range. Also, remember that the balls must be re-fed into the cannon after every shot; if we could make it possible to get balls off the ground and into the magazine, that would be ideal.
So, any ideas? You all seem to have a much better grasp of foam launching technology than I do, and the engineer working for us is employed by a NERF-free corporation.

[m15399]

Look no further.
Have it use a large short barrel with a magazine (it's a robot, program it to reload). Hook it up to a huge rubber bladder to apply a constant pressure of air. Use the air compressor to fill it up and blast away! Er... program your robot to blast away!

[davidbowie]

Ah, so that's what first is doing this year.

8" balls are more like dodgeballs than nerf balls.

The ball can't move faster than 26 mph? That's like...38 FPS. Nothing.

You could try some kind of pneumatic ram driven cannon, with one big barrel that has a magazine coming off of it, and a rod (should be light) inside it that's driven by a pneumatic ram. When it's retracted, the mag is open, and a ball feeds into the barrel. When the "gun" fires, the rod pushes forward and shoots out the ball. I have not personally built a gun like this, so I don't know the functionality of it. It might be liable to jam with soft projectiles.

It's a shame the ammo is so big. If you were shooting the 2" nerf balls or something, you could just have 10 small pneumatic spudguns. Since the speed is so limited, they could be really small.

On a side note, what makes the gun you linked to "semi-auto". It's bolt action.

Edited by davidbowie, 17 January 2006 - 11:43 PM.

[sporkboyofjustice]

How big is the hole?

How far away is the wall?

Is the playing surface level?

What are your other time limits involved?

On the FIRST site it says you get a set of parts to use, can you use anything other than that set?

Can you list the parts?

Do you need to use the compressor? Since it's slow you could always not use it as there are other forms of propulsion available. I'm thinking low tech but let us know what the rules are.

Can you run the compressor before you begin?

Will the robot have to move into position and be aimed remotely?

[davidbowie]

Could you use regulated CO2?

[Dr Nerf]

On a side note, what makes the gun you linked to "semi-auto". It's bolt action.

Valve - Semi-Auto Homemade

The valve of my homemade is semi-automatic. Hypothetically speaking, if the gun could automatically reload itself, all you would have to do it push the trigger to shoot. The valve only allows a certain volume of air to escape each time. Though the gun itself is only bolt-action, the valve is semi-automatic.

You could modify the design to shoot larger things. I made a barrel out of 2" PVC to shoot water balloons. It worked pretty well. But the balls that you were describing seem pretty big. It could launch them easily because it is pretty powerful, but a repeating design with it would be hard to build. 8" piping of any kind is not usually available readily.

I remember that I once had a ping pong ball shooter. This really reminded me of it. The design was quite simple. A tube ran along the top of the gun and was full of ping pong balls. A spring was at the end to push them forward. Once one got to the end, it fell into the "barrel". At the back end of the barrel was a hammer. Basically, when you pull the trigger, it releases the hammer which hit the ping pong ball, propelling it. Another ping pong ball then falls into place and the action can be repeated. This would seem to be an easy way to build a repeating design (if you can get the parts), but it probably would not be able to launch the balls very far.

[m15399]

On a side note, what makes the gun you linked to "semi-auto". It's bolt action.

(it's a robot, program it to reload)

Let's read the parentheses next time, eh?

I guess you need a barrel of some kind, and whatever it is, I'm sure you'll find a way to attach it to your robot's Nerf gun.

[davidbowie]

I read the parentheses. I fail to see any relation between them and the question I asked. It was just a question, out of curiosity, as to how that gun was semi auto. It has been answered. Let's move on, shall we?

8" pipe, while it is readily available from your nearest irrigation supplier, it's very expensive, and the fittings are much more expensive.

How rigid are these balls?

[spasticteapot]

Ah, so that's what first is doing this year.

8" balls are more like dodgeballs than nerf balls.

The ball can't move faster than 26 mph? That's like...38 FPS. Nothing.

You could try some kind of pneumatic ram driven cannon, with one big barrel that has a magazine coming off of it, and a rod (should be light) inside it that's driven by a pneumatic ram. When it's retracted, the mag is open, and a ball feeds into the barrel. When the "gun" fires, the rod pushes forward and shoots out the ball. I have not personally built a gun like this, so I don't know the functionality of it. It might be liable to jam with soft projectiles.

It's a shame the ammo is so big. If you were shooting the 2" nerf balls or something, you could just have 10 small pneumatic spudguns. Since the speed is so limited, they could be really small.

On a side note, what makes the gun you linked to "semi-auto". It's bolt action.

I was thinking something similar, but mechanical. Of course, everyone else is hung up on spinning wheels, with no way to reload.
(we're limited to Bimba pneumatics, sadly; and a high-speed long-throw cylinder would cost a mint.)
That said, a PVC cylinder akin to those you have built might do the trick. One would only need about 10 PSI to fire, too, assuming a 1" diameter rod. (31.4 lbs. of force should do the trick!)
Got any links to anything like this?

[davidbowie]

I've got no links on how to build it, but it should be pretty simple. It would have to be single acting, spring return, because there's no way you could seal up the return side. That said, you would need an outside body (let's say 1 1/4" because it has a great fit around 1" PVC), with a reducer on one end going to the valve (I'd say 3/4" QEV), and a reducer on the other end to center the rod. You'll then need a rod that has a back end with an O-ring.

Another idea I just had is to use a stand-up bike pump. You would need a big return spring... maybe surgical tubing would be best. You could hack off one end and epoxy it into a PVC fitting of good size, attach the return, and then you have a cylinder.

[sporkboyofjustice]

For propulsion I was thinking something far more basic. Instead of working with fittings and pressure why not use a gravity based swing to store energy potential and to deliver propulsion. Set up a rigid swing with a weight, use a motor to bring the weight over and behind the ball and let gravity pull it through hitting a plunger mechanism and away it goes.

The motor will catch at the bottom (might want to lock it at the top of its swing). And it would be ready to fire again. For reloading you might want to look and see how a pitching machine (automated) deals with it. Also if you have an older pachinko machine handy take a look at the loading mechanism, it’s kind of neat.

I realize that this is not very sophisticated but the firing part would be pretty robust and you wouldn’t have to worry about air leaks.

[nerfer34]

Maybe like a Nerf catapult of some sort?

[davidbowie]

No. Catapults would be complicated to automate.

Have any of you seen MXC? Specifically Brass Balls? They have those soccer ball launchers composed of two spinning wheels. All you need to do to fire is push the ball through them. It's like the GyroStrike. So, here's my idea: basically make the gun I was talking about before, except the velocity doesn't need to be high, and the stroke of the cylinder only needs to be a few inches. You would use this to feed balls between two spinning wheels, which would launch them.

[Arcanis]

You're using VEX, right? I have the set. The Gyrostrike idea is feasible, and if you were to do that I'd reccomend you get the tank tread kit's "guide wheels" for the ball. Then again, the basic airtank+tube works just as well, if not better. To be truthful, I wouldn't reccomend you use the gyrostrike idea because the motors don't exactly fly.Therefore, you have several options-

1. Use a motor to activate the launching assembly.
2. Try to activate the launching assembly electrically.

[davidbowie]

VEX!?!

Don't tell me you're using VEX!

AFAIK, First has a "real" division and a mindstorms division. From the sound of it (bimba pneumatics), you're in the "real" division.

I personally hate robotics kits, especially when someone hears that I dabble in robotics, then asks "So, which do you use? VEX or Mindstorms?" Real stuff is the only way to go. It's cheaper, better, and more versatile

[nerfer34]

I'm not positive a catapult would work for this specific project but it isn't very hard at all to make.

[davidbowie]

Catapults can be extremely easy to make, but incredibly hard to automate. Think about every motion you have to make for one firing cycle of a catapult: grab the arm, pull it back, lock it in place, grab a ball, move it into the cup, move the loader out of the way, release, repeat.

Not to toot my own horn, but in my spinning wheel idea, there's only two steps: push and repeat. Not to mention catapults aren't the most accurate.

[Spider-Waffle]

I think the two spinning wheels is the obvious choice. The only other option I see is gas combustion or CO2. these would basically only offer greater accuracy (which might be that important) but two spinning wheels can be quite accurate.

Edited by Spider-Waffle, 16 February 2006 - 01:21 AM.