I think that you can get a crazy fast, maybe even the fastest possible, 4 motor launcher with a puncher system, if it’s built right. However, once you get to a certain speed, the limiting factor becomes how quickly you can feed balls, and even how quickly they can come to rest in the right position, which will slow you down. That is why within an entire application single flywheels are probably always faster because they can be fed so much easier.
I think the fastest launcher is a single flywheel. It can use velocity control to speed up launching and can be loaded easily. It is better than the puncher because it doesn’t cause your bot to jump. Our team built a puncher and it jumps around crazily after each shot. It would probably cause the robot to get misaligned after every couple shots
I’d have to say single flywheels as well. Easy to build if you have straight axles and such and know how to make things squared (at 90 degree angles). Speed controllers are a bit on the advanced side depending on the complexity and tuning, but Take Back Half is very easy to get working right and very easy to tune.The only thing I would definitely say you want to make different from some flywheels is you need to add ratchets. I broke a few motors this year because I didn’t have them, but there are ones which can be built without ratchet gears and ones that fit on 60 tooth gears. For the ratio, I would stick to between 1:28 and 1:40, and I would stay low unless you know exactly what you are doing.
I completely agree with this. One of the motors on our single flywheel had its internal gears become stripped within 2-3 days of testing. Now I have a ratchet and the motors are fine. Also, the ratchet prevents the motors from overheating as quickly because when the wheels stop moving, their momentum goes straight back into the motor and heats it up. With the ratchet, the motors are able to stop spinning.
One of my teams had a single flywheel shooter (8059A-esque) that did not use a ratchet but it did have a speed controller. They used set of global variables held the actual motor speeds and commands were sent to functions to change the speed up or down with delays. When a Stop (0) was called then the delay was increased so it slowly slowed the motors to a stop.
I had exactly this. Speed control is crucial and I’ve seen a lot of good built robots perform terribly because they are lacking good code. Look at Jpearman’s TBH( of which mine is heavily based off of) and velocity control codes to get the launcher shooting more accurately and quicker. Also don’t forget to add slew rate control. Slew rate is change in voltage per change in time, and you need to add a limit so the motors don’t overheat instantly, and this is crucial for spinning up and spinning down to prevent motor over heats and motor breaks.
In theory, I think the fastest shooter’s probably a double flywheel with an insane amount of mass. This is considering the full court shot. The reason I’d say double flywheel is because there is no limitation to how fast you can feed them (there is no feed motor so you can just the drop the ball straight into the flywheel). If your flywheel has tons of mass, the balls going through start to make a negligible difference to the flywheel speed even though they would take forever to start up. At this point, it would be impossible to feed at that rate but…
Are you going for a single launcher or multiple? I have seen teams with double punchers and a flywheel on the robot, and if they didn’t care for wheels (because we are going for the fastest possible), they could easily get a 4-5 bps flywheel and 8 linear punchers (using one motor per puncher and a 4 motor flywheel). The problem then becomes actually loading the balls that fast. Beyond 4-5 bps it is practically impossible to load at full speed.