Cannon Design!!!!!! >_<;

So we’re trying to build a cannon to shoot the moon rocks at the trailer, but we’re having difficulty getting the ball up to speed for any sort of range.
So far we’ve tried using two spinning wheels in parallel with the ball being moved between them. Help!?!

-Team 2705

Can you provide us with a picture?

If you have wheels that spin against a ball, it should launch the ball. Unless it there’s too much friction between the spinning wheels and the balls that the ball gets stuck on the wheels and starts spinning.
Depending on the type of wheel you are using the cloth of the ball might be getting stuck.

It would actually also help if you could put up a video of the system working.

From team history, we used a pretty big wheel maybe 8in in diameter to launch a squishy but completely solid ball (unfortunately, I don’t know what type of the top of my head) in the 2006 game Aim High.

Your best option is probably to search chiefdelphi and to look for other shooter designs from the 2006 game.

If there are any other things that I can help you with PM me. I’ll ask my team members who were involved with the 2006 robot, to give you some specifics if you would like.

Good Luck,

Are you starting the ball directly between the wheels? If you are, you shouldn’t be. Give it a little speed coming in.

Also, what motors are you using? Some motors (say, the window motors) are not good for this type of system.

Oh, and popnbrown–the infamous Poof balls (shelled) were the victims in 2006. If you raided your local Walmart in the hunt for the Orbit balls, you may have seen the unskinned version.

We don’t have a video camera/regular camera… whoops…
So far, we’re just prototyping some shooter designs in order to shoot the ball vertically into a chute. The design we have so far is using 2 rubber wheels with a diameter of about 6 and a a wheel tread of about 1 in, so they’re kind of small. We think the problem is that there is too little contact area between the moonrock and the wheel, but don’t know how to fix that when the moon rocks themselves are full of holes. The only solution we have so far is to use 4 wheels and put belts between them to increase contact. We’re just starting the ball at a dead start. Similarly, we arent using any motors so far, we just have some wheels attached to some drills, so we’re not sure how accurate of a representation this is. Sorry if this isnt very descriptive. >_>

It is possible that the wheel is not wide enough to power the moon rocks well, but there are a number of other possibilities.

As Eric mentioned, this type of mechanism works the best when the wheels are already spinning and then the ball is fed into them.

Another potential issue is compression. Based on what teams have been reporting here and what my team has seen in very limited testing you want at least an 1" of compression at the narrowest point between the two wheels meaning they should be 8" apart at the most. You also don’t want more than about 2" of compression so they should be at least 7" apart.

The last potential issue results from the drills. If the drills are not providing enough speed or torque or both your design may not work well. If the wheels are not spinning fast enough they will not transfer enough energy to the ball as it passes through. If the drills don’t have enough torque, the wheels will slow down too much as the ball passes through and the wheels will then not transfer enough energy to the ball.

Wheels are a decent diameter, and using more wheels may help.

Though–let’s look at this closely. You’re shooting vertically into a chute. Straight up or straight down? If it’s straight up, you might want to look into a belt system.

We’ll try out more wheels tonight and tomorrow.

It would shoot vertically into a curved chute. The chute would be on a lazy-suzan type turret - with a wide enough center hole for the balls to pass through unhindered- which could turn to aim in different directions.

You may want a belt system, but if you need to go with a launcher, you want to be able to go horizontally after the vertical launch.

Let’s say you need to go 3 ft. horizontally from the top of the robot. You need enough speed to go that far left over from the upwards launch.

Let’s start with the horizontal portion. Launching at 4 feet, you need to clear 3 feet. So you can’t go more than 1 foot downwards, and you need to go 3 feet horizontally, reliably. You’ll have an upwards trajectory from the launch, but we’ll ignore that for now. We need to go horizontally and vertically, so for vertical motion, s(t) = -1/2gt^2 + vt + hi, where g is gravity, hi is initial height, and v is the initial velocity. hi is 4 ft and v starts at zero. s(t) ends up at -1. So… -1 = -1/232t^2 + 0 + 4, -5 = -16*t^2, 5/16 = t^2, t = 0.559 seconds.

Now, how fast does the ball need to go to go 3 feet in that time?
s(t) = v * t for horizontal motion. s(t) needs to end up being 3 ft, and we have t from above = 0.559 s.

3 = v*.559, v = 3/.559, v = 5.3 f/s coming off the top of the chute.

So, if it has to go 5.3f/s coming out, and it has to climb 3 feet, let’s look at that equation for vertical motion again. s(t) = -1/2gt^2 + vt +hi. We’ll assume you want to lift the ball 4 ft.

But we’re interested in v(t), so we take the derivative of s(t). v(t) = -g*t+v.

5.3 = -32*t+v; t = (5.3 - v)/-32.

Substitute into the s(t) and you get 4 =-1/232((5.3-v)/-32)^2 + v(5.3-v)/-32 + 1; 3 = -16(28.09 -10.6v+v^2)/1024 -0.165v+(v^2)/32

3 = -0.438 + 0.165v +.0156v^2 - .0165v + 0.0313v^2
3.438 = 0.0469v^2
v = 8.56 f/s launch speed minimum. 10 f/s would be better to start the balls at.

I’m sure that you can run the numbers to figure out what speed the wheels need to spin to impart that speed; wider wheels may help as well to avoid digging into the holes in the ball.

I too think that is your problem there.

Back in 2006 we were prototyping shooters and ran into the same problem. The balls just spat out instead of shooting them and after moving the wheels closer together they went flying.

Paddle wheel is my solution, instead of compressing the ball it pushes it from behind, and since it’s a circle it can reload faster then a pneumatic ram.

With wheels on the side you need at least 2, one for each side, if your using belts you can get away with one since the ball will be in contact for a longer time. The main issue with the wheels is friction (again!), compressing the balls increases the force of friction (which your going to want maxed out), so you might want to rap the wheels in hocky tape or velcro to increase friction between ball and wheel

No, you don’t. 1 will do just fine, if you have a back plate to compress against. 2006 showed that. Contact time has little or nothing to do with it.

I was thinking a bit (dangerous, I know), and had a thought: What if most of the chute didn’t rotate, but just the top part did and had the launching wheels close to it? You could run a belt up to the top and have a slower launch speed because you no longer have to fight gravity for the full distance. You also have the ball moving when it comes in, so there is less motor speed loss, and you can shoot faster.

This sounds a lot like what we’re planning on doing. We used the 8" andymark traction wheels to shoot it up, with what I’m thinking is about 1 inch or so of compression. We’re driving it with a 32 teeth drive sprocket and the sprockets on the wheels are whatever they come standard with. The gearbox is a banebots 4:1 planetary. I’m are trying to get a bigger (more teeth) drive sprocket, but I haven’t found anything yet. I don’t even know where we got these 32 tooth ones. We haven’t tried shooting them up a chute yet, as we don’t have material yet to prototype that. We got to shoot it vertically about 8 feet or so. Seems to be fairly consistent. Hopefully we can find something to prototype the chute with Saturday, because I’m worried about being able to actually get the ball out of the chute. I don’t think we have quite enough power yet behind the launcher.

BTW, we also direct drove them from the cim motors. Don’t do that. The speed is ridiculous, and it could probably throw the orbit ball a long way, but I bet the ball would get destroyed in the process. It already looks pretty battered.

I was thinking about this as well, but mounting the motors and getting it all situated seems like it would be somewhat more complicated. If you just use the chute to direct the orbit balls, it can be extremely light, make of thin aluminum that could save you a decent amount of weight compared to mounting it on top of the chute, which would have to be made out of much stronger material.


Try using 4 of the previous years traction wheels. Two by two really keeps the ball cruising and helps pick them up out of whatever hopper you have to store these things. I agree that the CIMs should not be used directly, as we had some trouble today. A geared or belt system running both sets of wheels will keep them spinning at the same speed and at reasonable speed. Direct drive with the CIM motors could clear 15 feet in our test, but that just bounces of the center post in the trailer. Good luck!!:smiley:

One problem you can encounter is if you try to fire the ball from a stopped position, but if you give it some speed before the ball meets the rollers it should work a lot better. For instance a gravity driven shoot to the rollers or a start-up roller/wheel.

From a Physics standpoint I’m not sure that argument holds water. With all things being equal, the amount of time that I apply a given force will directly affect the velocity of an object. The more time I have to apply the force the faster the object will travel. For example: compare the muzzle velocities of a pistol vs. that of a rifle.

True. However, we are dealing with relatively short range, low velocity objects. As 2006 showed, teams that don’t have a lot of contact time are about equal to teams that have more for this type of application. Especially this year…

From a practicality standpoint, it doesn’t really matter. Use a rifle at 10 ft and a pistol at 10 ft. Sure, the rifle gives more speed. But at that range, the pistol is just as effective. Extend the range to a mile and the rifle wins, but we aren’t talking that range here, we’re talking a few feet.