Ball Launcher Test and Thoughts About Launching Methods

Here is a ball launcher I made this morning. It uses the 5" wheels and two high torque motors with a 1:3 ratio, and the ball does not move very far at all. I think even just shooting into the high goal using motors from up close will be difficult at first, but shooting across the field will be quite a challenge. Here are my thoughts about different launching methods:

Motorized launcher: hard to get a lot of distance, but quick and consistent
Pneumatic launcher: quick and good distance, but will likely run out of air too fast to launch very many balls
Slingshot/elastic launcher: good distance if done correctly and consistent, but it would probably not be very fast to wind up a launcher that has enough elastic force to launch very far

Try gearing it a bit more for speed. Much torque shouldn’t be needed, these balls are pretty light from the looks of it. Also, try testing it on an angle after these changes. I assume that you won’t be shooting completely straight, but instead with a bit of an arc. Also, is that the official game ball?

Mind if I offer some advice?

First off, you’re not shooting up. Angle it about 45 degrees, and you’ll get some actual lift.

Second, don’t use Torque. Gear up for speed. You ever seen a pitching machine?

Third, those motors are unsecurely mounted. You’re loosing energy from their wobbling. And it’s not just the motors. That entire assembly looks unsteady. The gears you have on the driven shafts, for example, aren’t secured on the other side. That is very bad.

Fourth, you should compress the ball more. It’s barely being gripped by those wheels. You want it to fly out in a “popping” sort of motion.

You can make a motorized launcher work. But it’s going to take more than 20 minutes.

I appreciate the feedback. I know that this was put together very quickly so the whole thing is wobbly; I mostly wanted to see how easy it would be to launch a bit. I tried angling it up to about 45 degrees and putting it on the floor, and the ball only travelled about a foot before hitting the floor.

I understand that a high quality launcher will take much more time to make. I was simply surprised that the ball went nowhere given how fast the wheels were turning. When I go back to give it another try, I’ll do some actual calculations to see what gearing is necessary to shoot a certain distance.

I used a Clean Sweep small ball.

When I was at worlds, I asked the VEX staff member who was managing the field and he said that these years balls where nearly identical to the clean sweep small balls. I believe that launching balls using wheels turning at extremely fast speeds will be very common at worlds. Possibly as common as the 6 bar lift in toss up.

That will be good. Strategy will be crucial.

Our team was thinking of a similar launching method using flywheels as you showed but instead of having two counter rotating wheels it would have one singular very high rpm flywheel and a slightly curved plate behind it. Balls would be transported vertically up a shaft using treads and then would be pushed between the curved plate and the high speed “flywheel” therefore launching balls hopefully accurately and far.

The launcher would be very similar to those seen in the FRC game “rebound rumble” from the 2012 season.

That literally is the same exact thing we are doing and it has a pre set angle and the ball just rides up a trough and gets pushed info the high speed wheels, pick balls up on one side and gets thrown out the opposite at about 60 degree angle.

Were doing an 84 tooth driven to a 12 which as a 24 connected to it going to a 64 tooth which the wheel is on.

When we made this we used speed motors (turbo gears didn’t exist) and we geared it up 1:7, for a 1120RPM theoretical output speed, which is much faster than you were going. We also used anti slip mat to improve the energy transfer between the wheels and the ball.

I should note that you should build and test these mechanisms with EXTREME caution. Objects dislodged from the wheels rotating at that speed could potentially travel very fast. Proper safety gear should be used (i.e. Safety glasses etc) and the rotating mechanism ideally would be shrouded to prevent people sticking their fingers in it.

It’s also worth adding that because there is such a large amount of energy in the high speed rollers, you should take extreme caution when powering the motors down. If you don’t slow them down in a controlled manor, you run the risk of shredding your gears.

If you are rolling ball against the static plate you are putting part of the energy into the ball’s spin, part into the forward motion and part into heat (by rolling deformation).

I can see that you will have more predictable exit trajectory if ball is rolling off a plate, but I am not sure that you are benefiting that much from having spinning balls in the air or entering the target.

Once again, we really need to test it both ways with actual balls to see which one is better. My feeling is that soft balls traveling slow over a relatively short distance are better off launched by two flywheels with soft uniform cover, like matthew@aura is suggesting. Then you send them through a tube guide. If everything is aligned perfect they will barely touch the tube, otherwise it will straiten them.

Remember to improve accuracy the military’s kept building longer and longer cannons until rocket propelled and guided missile were invented.

I guess safety glasses are a must even in practice this year.

Hmmm. So problems may arise like when the match ends and motors go to 0 instantly in the competition control? A normal match time would be programmable but matches that are shut off in the middle may not go so well.

Hey IFI/Vex, any chance we can get a bike wheel hub as a part? Then they would spin down gracefully.

Yeah, we’ve been thinking that. We used the analog stick to power up / slow down the motors gracefully in our test, but some sort of clutch system to allow the motors to freewheel at the end of a match would be good.

Doesn’t Vex have a part, the clutch, which would allow the connected axles to spin independently above a certain torque?

Or do I have this all backward?

The Clutch was designed for the old VEX 3 Wire motors, which have significantly less torque than a 393 motor, even in speed mode. You would likely just wear out the clutch in normal operation

I believe, the motors will freewheel when match is stopped if they are connected through 29 controller, because it does not brake (short) them when you set power to 0. As for the H-Bridges on ports 1 and 10 I remember reading somewhere on Robot-C forums that they will break or coast depending on the value of bFloatDuringInactiveMotorPWM. (I tried to search their forum now but only found posts related to other controllers, need to test that).

I think you would want the clutch to be one way; so that you would not lose power when you speed it up. You could make a makeshift one by having a idler gear that is only mounted to one gear and then rubber banded into the other gear.

For our high school team we are experimenting with a catapult using a cam and VEX elastic. It definitely has some potential. The elastic will give a consistent throw every time and not be subject to battery drainage or air capacity. The prototype we are working on will have a 2 motor drive for lining up with the goal in our corner (aiming) 3 motors to actuate the cam quickly, 1 motor for the conveyor to feed the balls into the catapult, and 4 motors for the lift to raise our alliance partner. With proper gearing on the lift, and some outriggers I think we can lift most reasonably weighted robots with no problem. Our plan is to not have an intake, but to be able to score 4 preloads in the high goal in autonomous and all of the driver control loads as well.

Our VEX-U robot will be similar except we will have a crazy fast 4 bar feeder robot to play “Baby Bird” so we are able to score bonus balls. With a 12 motor limit we may try and have the 15" robot shoot a few balls in autonomous as well.

You could create a ratcheting mechanism that would spin up the wheels with full torque, but then would slip when the input stopped suddenly and would allow the wheels to slowly lose their momentum.

my recommendation is increasing the compression on the ball, as well as increasing the speed of the fly wheels.