In the interest of keeping other threads clean(er) here’s a thread dedicated to discussion about catapult designs based off 7682E’s original work and 7682’s subsequent full court variant.
Some other teams in New Zealand have been experimenting with catapults with varying degrees of success but in general it’s a pretty easy mechanism to get working for short shots and more attention to detail required for full court.
Please don’t go asking for a complete Wingus & Dingus Catapult how-to guide. It defeats the purpose of your participation in VRC. My favourite quote from a related post “So I suppose we actually have to engineer…” sums it up nicely. You should be engineering - that’s why you’re here! Time is at a premium in the Wingus & Dingus labs and October/November is exam time so apologies in advance if questions don’t get answered straight away.](http://kiwibots.nallen.me/events/view/147?p=results&v=1,50)
From about 4 hours of searching through modern revivals of catapults, and how they use elastics to launch, I found a couple of interesting methods, however, through prototyping, I found that very few worked well. I was wondering how you were mounting your elastics.
The three methods I tried were by attaching the elastics from the top of the launcher tower, and linking them to the end of the lever formed by the catapult arm. The second one I tried was by putting rubber bands on the inactive side to see what would happen, and my third design, which showed the most promise was a vertical linkage from the closest point to the rotational axis.
6 Torque Motors Stalled out, and didn’t have enough speed.
6 Turbo Motors Stalled out, and couldn’t hold for long enough.
I may be building very incorrectly, but I can’t see where I’ve gone wrong.
I’ve already found that extending my arm helped with greatly increasing my angle, due to the physics behind a lever.
I myself have not built one but a friend at my school has built a catapult himself and it works decently well. He is using 4 torque motors I believe, two per side and is using a radial release mechanism. He is using rubber bands as the launching force behind it. It works quite well but it was designed to really only shoot from half field or closer.
Has anyone been able to achieve a long range shot with a 4 motor catapult, because otherwise I think that the catapult design might be dead since long range shots require 6 motors and close range shots are much slower than a flywheel.
Did turbo speed, high speed, or high torque internal gearing work best for the catapult? We tried both turbo and high speed and we don’t know which to chose. The high speed requires more rubber bands to shot, so I would think that the turbo speed would be better because the gears have less pressure being applied at the top, but I would like to know your opinion on this.
An elastic assisted motorized catapult is definantly an interesting idea that seams to be working well. I’m currently in the process of building a catapult powered by elastics with a pnumatic gearbox to winch back and fire. I’ll report back on how it compares to yours.
We had the same problem, and we solved it with the potentiometer. We have it set to so that after the arm goes back a certain distance, the motors slow down to like 10 speed (somewhere between 1-30). This helps stop stalling because if you just push the catapult down and hold it there, the motors are running at 127 speed but are not moving anywhere, but all that power is still going into the motors causing them to stall.
My junior team has managed to build a 2 motor full field catapult, it could do with a re-design as their are a few aspects that are not perfect, but it works. They are using quite a nice wind back mechanism, but a little slow, so you are probably better off with at least 4 motors for speed, 2 has enough power though, all about gearing.
After a month of testing, my team was able to create a full court catapult using two high torque motors. Based on the potential energy of surgical tubing (Not rubber bands) we able to get to about 95% accuracy in 3 days of practice. We went to Winward Tournament last sunday and won with the design. Instead of pulling the surgical tubing vertically, our design called for linear motion through the usage of cams.
If you guys are willing to share I’m interested to know the firing rates of these launchers, and also how long can they keep firing at their maximum rate without tripping the motor PTCs? A 2-motor full-field catapult is intriguing because it’s a low enough motor cost that you could put two or three catapults on the same robot, or you could add the mechanism to a robot that is mainly designed for some other role (possibly as a modular mechanism that you add only when you need it, for example in qualification matches with a partner who can’t score well from the starting tile).
We will be posting about our robot very soon (in the next few days). It will include pictures and a reveal video.
To answer questions about speed: Because we had only 3 days before our first tournament to fix some issues, we did not change the torque value on our catapult. At the moment its a 3:1 external with a high torque motor. We actually found you can run the external gearing at half of what it is (instead 3:1, 1.67:1). In all it is not as fast as some of the flywheels that we have seen (at winward we were against 404s who had a faster shooter), but we have been more accurate with our shot than most others. We have a design for a multi-tensioner in case we are the only scoring robot in an alliance that will allow us to score from all places on the field.
Interesting idea, although it may be more practical ( rather than multiple catapults on a single robot ) to create a catapult capable of launching multiple balls at once. If someone could accurately launch 4 balls with a 6 motor catapult every 1.5-2 seconds they could do really well.
I saw that vid a while ago. We need faster with more balls! xD Remember there was a 3-5 second stop between each shot, which is quite large considering it was only 2 balls.
I’m a bit skeptical of multi-ball catapults being better than single ball ones because the aerodynamics are likely to be much more chaotic and a larger “projectile” means more accuracy is required (the “projectile” in this case being all the balls you launch when you fire). I also think it might be tricky to get the a catapult to fire one ball accurately, and also two balls accurately, and also three balls accurately (up to four, potentially).
But this is before actually trying it. It definitely deserves to be tried, and I would love to see someone pull off a full court multi-ball catapult. I do think it can be done.
When we first started to brainstorm launching ideas at the beginning of the season, 8000 tried develop a mechanism that launches four balls with the same catapult. We noticed that shooting multiple balls with the same launcher led to only making, on average, 3 balls out of 4, mainly because the balls interfered with each other in an uncontrollable way, leading us to stick with shooting one ball at a time. Granted, we only spent about a few hours on this mechanism that shoots four balls at a time, so I am not saying that it is impossible to shoot multiple balls with the same catapult accurately, but I am saying that 8000 has not had much success with it so far.
We actually only use 7 motors total on our robot plus pneumatics. We use 1 low speed motor for the long launcher, which shoots about once every 1.5 seconds and is very accurate (the match in the video isn’t our most accurate run because our robot was not aligned correctly prior to the match). We definitely plan to speed this up by the next competition. Also by the next competition, we plan to use all 10 motors plus pneumatics, most likely by redesigning most of our current robot. We may also post a video about our skills run once we get that robot working.
Great match! Had to re-wind and check what was happening at 1:02 though. - YouTube
There has been much discussion here from day 1 about parallel ball launching and it’s nice to see something implemented. Ultimately it comes down to the complete collect / score cycle time which no 1 mechanism is solely responsible for.
I would imagine 4 balls launched along the same axis would diverge slightly as they displace the air around them. Any variation in ball size, weight and density would put 1 or more behind the leader(s) and leaving a disturbed airflow in their wake. I do wonder if these effects would be negligible though given the launch velocity.
The massive downside to multiple ball launching is of course there’s no way to stop and correct at ball 2, 3 or 4 after the first few miss. All or nothing.