Here are the cliff notes for making a shooter than I use after my various experiences making FRC shooters.
More grip - use the stickiest/grippiest material that can survive your shooter design requirements
Long engagement time - if the wheel and note match speeds then the release speed of the note will be consistent
Gentle engagement and release - gradually increasing and decreasing the compression of the note will improve engagement and release consistency, especially as game pieces age over a competition
Enough spin - a shooter can only add some maximum amount of energy, if that energy is spent spinning the note then it won’t go into moving it down field. But if there’s no spin then the note is likely to become unstable during flight and susceptible to air currents, resulting in a loss of consistency. Sometimes this means two-sided shooters, or sometimes the fixed side of a shooter gets a grippy or slippery material to tune what you want.
Enough inertia - if the roller speed drops or changes too much as notes are fired then you’ll see inconsistencies as notes age and change during competition, but too much inertia might take too much time to spool up to shoot effectively
Consistent feed position - make sure the note is in the same place and moving at the same speed every time it is fed into the shooter. Depending on shooter design this can be a really big deal, e.g. in 2022 we could see multiple-foot shifts in landing location if we moved the feed point by <1in
Reposting our shooter pic from above to provide easier context for its details.
Grip - 35A stealth wheels and either 60A or 80A colson wheels, both are reasonably grippy shooter wheels. Both have been shooter prototypes for us in the past.
Engagement - the over-under wheel or roller approach offers a wonderfully long engagement time/distance
The shooter wheels being notably bigger than the note cross section lends itself to a relatively gentle engagement
Enough spin - we guessed that the 5in (left) to 4in (right) ratio of free speeds would be ‘enough’ spin. The one-side-wheel shooters (effectively 1:0 spin ratio) seem to have way more spin than is needed, by eye anyway.
Enough inertia - 10x heavy wheels will do for a prototype
Shooter setup is probably:
4-5krpm, 2x CIM
1.5in gap/0.5in compression (I think this isn’t quite right, but it’s what we aimed for)
35A stealth wheel, 5in
Colson, 4in
20% spin
16° angle from horizontal
28’ flight, 2’ elevation gain
One aspect of testing for shooters I like to take into consideration is make sure each game piece I use is numbered/labeled uniquely so I have data points for each piece. Its nice to see if that variation you see is correlated to game piece variation.
The other thing here is that you’re still hand feeding the shooter. I know you know this already so I’m saying this more for anyone reading, but I’d imagine you results will differ once you remove the manual loading in your testing.
A quick update on shooter testing for tonight. More on robot packaging in the near future.
Here is the 28’ full-power shooting video.
The last thing we shot was a pool noodle cut and taped into a 14in diameter torus, which has been a pretty okay stand-in for the note.
Test master Nate after finishing a second batch of shots to dial in a ~11ft shot aimed roughly at the Speaker elevation from roughly the robot elevation. This angle, ~34°, will help guide initial CAD work.
At this shorter range and with more care taken to feed the note consistently we are getting good shot groupings. Errors in the range of ±6-8in per axis, with nice normal distribution, feels pretty good for a prototype with zero speed control or mechanical synchronizing.
We are very quickly chewing up our note. There are particles of it all over the floor in front of the shooter.
Yeah…
Highly encourage teams that have multiples to stash one or two fresh ones away to test in their final robot tuning efforts.
James this prototype looks awesome! The different diameter wheels is a creative and smart way to put some spin on the ring. The image makes it look like the shafts are also angled? Or is that just an optical illusion from the wheels being different diameters?
EDIT: Read back a couple posts and found out that yes they are not parallel**
Perhaps there’s a limitation on your current network or another issue? The video is working on my computer outside of the team’s youtube account, traffic looks good, and there are no restrictions put up on our end.
Yep I’ll try again once I get home, my school network might be preventing me from watching the vid, although it usually shows some videos as restricted, not unavailable.
One kinda sneaky way that sometimes works is linking the youtube video as an embed somewhere else. I’ve only tested it by posting it in a private discord server with myself, but I also face your problem and this fixed it – maybe other platforms like slack also do it. Just like you, it normally says Restricted but this one just says unavailable (even when I copy the link and paste it into a new Google tab).
We put the axles+wheels together and put them through the bearing plates. Long and short is we stuck a 1.5in spacer (i.e. a hunk of 2x4) between a wheel set, clamped the plates together, match-drilled and riveted them. Repeat on other side. Very crude, but very fast, and I have no idea what the angle is
This is ingenious. My biggest gripe about an under-the-bumper intake was the loss in width.
Could you share how the mounting is setup behind the bumpers? I see a plate on either side of the new frame tube, but where is that mounted to in relation to the swerve module?
