Hi. We competed this weekend and quickly discovered that our vertical horizontal shooter is not (currently) optimal for this game. Needless to say, our whole serialization/indexing system is dependent on this type of shooter, and I believe we’ll find more success reiterating our shooter design then scrapping for next comp. Has any bone had any success with this type of shooter? We ran into issues mainly with accuracy, and were wondering if any teams had been able to iterate a similar mechanism successfully. Thanks in advance.
Can you provide a picture?
I’d be happy to in ~20 min. I realized I should have included a little bit more information. We are powering via two neos that are geared up using belts to approximately 1:2. We use two 4” colsons.
2 Likes
You could add two more identical wheels and motors on the horizontal axis, then spin up the lower wheel to a slightly higher RPM than the others. That would give you a bit of backspin but (probably) keep the accuracy you seek and not force you to radically redesign your shooter mechanism. Of course, this all depends on you having the weight, available PDP slots, etc. to pull this off.
1 Like
So we initially also were prototyping a vertical shooter and we ran into accuracy issues. We were running 2 4" Colsons on the bottom and 2 4" Colsons on the top. The colsons were 1.5" Wide. We were direct driving from 2 Neos. So one neo drove the bottom wheels and one drove the top. We were able to shoot pretty far but our shots ended up quite inconsistent.
We found that when we gapped the top wheels further apart (so there was a larger gap between the wheels) that our accuracy increased tremondously. We then moved on to a hooded shooter which gave us much better results in terms of accuracy.
If you notice 4414’s and 1678’s shooters from LA North, they also were essentially vertical wheel shooters. The difference they had is that the top wasn’t exactly wheels but more or less a roller. Similar to how 254 does their intake rollers (see 2017 and 2014). The other big difference I noticed with their shooters compared to how we did ours is that there weren’t any gaps. Since we used Colsons, no matter how far you pushed them together, there would remain a gap due to them being wider at the hub. This probably helped 1678 and 4414 achieve much more consistent shots.
The final difference I noticed between theirs and ours was that they used fairlane wheels. Our team has never used them before but that may have made a difference since there is hype around fairlane wheels.
So that’s what my team did and that’s my thought process into why theirs worked and ours didn’t. Also, 4414 and 1678 don’t say specifically what wheels they used but looking at their shooters, the bottom wheels appear to be these 4" 60A Urethane Rollers. 1678 appears to have used these wheels for their top roller and 4414 appears to have used these wheels for their top rollers. I may be wrong since I am not on either one of those teams but that’s my assumption based on looking at their shooters.
If we were to re-attempt doing a vertical wheeled shooter design, we’d definetly take design cues from 1678 and 4414. You’d still need to do testing to figure out the optimal compression for what you’re trying to achieve and any changes to match your robot.
So yes, I’d say that there are people that have had success with this type of shooter although you’d need to spend more time testing to figure out why your team’s shooter is having accuracy issues.
If @Michael_Corsetto or @jjsessa want to chime in and tell us how they did they’re shooters that’d be sweet but seeing how they want to stay competitive, I’d understand if they didn’t want to share any information.
2 Likes
See the rest of my post but backspin didn’t increase our accuracy tremendously. It did a little but it was still pretty bad.
Be ready to spend a lot of money on fairlanes, those solid rollers get expensive.
1 Like
Thanks for the well thought out response, it means a lot. I’ll definitely look into those designs. Theoretically, I think we could improve tremendously. Those bots are perfect representations of our end goal.
No problem. Also, I forgot to mention this, but in the picture you posted, it appears as though the top shaft is bent. That would cause a lot of inconsistencies. It might just be because of the angle though.
Honestly, I would not be surprised. Haven’t looked too much since this comp tho, but I’ll make sure we’re squared up this week. Thanks.
Big issue I see here. It may not make sense as I explain but look up how crowned rollers work after reading my explaination. We are using top and bottom 4" coulsons directly driven by neos, so quite similar setup. I would suggest changing to a single centered coulson on each shaft. Additionally play with the neo rpms to get spin, we are currently shooting at 5500 bottom and 3500 top to generate backspin. This give a more predictable airflow around the ball. If shooting at the same speeds top and bottom you are likely getting knuckleball effects.
As you look at a coulson you will notice that they are crowned, which means it has a slightly larger diameter in the center of the wheel compared to the edge of the wheel. Crowned rollers automatically center things that are touching them. Let’s pretend the wheel is a .1 of an inch crowned at the center so the outside has a diameter of 4" and inside 4.1". In one revolution a point on the wheel has moved πd. Therefore a point on the outside has moved 4π and center has moved 4.1π. the higher velocity point on The wheel is pulling the ball to that point, naturally centering it. with two coulsons on a single shaft the ball is probably attempting to center to one wheel or the other. That would also explain why when you spread them apart it gets more consistent, if they get far enough apart that you’re not hitting the crowning on the roller you’ve avoided that effect, but probably aren’t getting much compression and/contact time with the ball.
5 Likes
Thanks for the explanation and you’re right. When we spread our Colsons apart, although we were a lot more consistent, we also had very little compression. This didn’t allow our shots to go very far.
That makes so much sense, thanks for the insight. I will definitely look into this. Season went by fast, and we didn’t consider shooter design extensively. Since you have a similar design, would you mind sharing how much you’re compressing between the two horizontal wheels?
Edit: I only ask because it was an issue this weekend. We’re currently running an inch, which I assume is not ooptimal in comparison to the ~2 in. number I’ve heard thrown around quite often this season.
For reference, we were running 1.5" with our setup. I would definitely try more.
Awesome, I’ve been speculating that may be an issue for some time. Our max shot distance currently is probably ~20 feet without an ideal arc.
I believe it’s 6inches, so 1 inch of compressing from us.
Ok thanks for the info.
If you have access to a milling machine, I’d recommend milling down the inside face of the colson wheels so that they sit flush together. This gap between the wheels could result in some inconsistent behavior.
1 Like
This is true of belts running on crowned rollers under tension. Is this really true for rollers and something other than tensioned belts?