View Full Version : Consistent shooter
Is there a consensus forming concerning what type of shooter design provides the most consistent shot from beyond 15 feet?
I'm thinking near the top of the list would be a) metal-spring-loaded catapult or pusher whose springs are stressed well below their yield strength and b) pneumatic firing catapults and pushers,
and near the bottom of the list would be a) motor-powered kickers and b) motor-powered wheel shooters.
I would say almost any spring loaded device to launch the ball would be more consistent than a motorized launcher.
IronicDeadBird
07-03-2014, 17:26
It would be interesting to see actual data comparing the autonomous score of different shooter mechanisms.
It would be interesting to see actual data comparing the autonomous score of different shooter mechanisms.
That would be interesting, but it does not take into account battery drain toward the end of the match.
IronicDeadBird
07-03-2014, 17:36
I am totally out of it but how is battery drain going to effect the score of auto?
I am totally out of it but how is battery drain going to effect the score of auto?
In autonomous, it's not. That's the point.
Lil' Lavery
07-03-2014, 17:40
I am totally out of it but how is battery drain going to effect the score of auto?
He wants to consider battery drain as one of the factors impacting shooter design. It would have a much larger impact on systems which use energy provided from a DC motor (wheeled shooters, motorized catapults, etc) than it would on systems where the energy is obtained by compressed fluids or material deformation.
IronicDeadBird
07-03-2014, 17:40
Sugar and caffeine are wonderful things ill check in later on this form to see if things change when I am more awake...
Our robot uses a type of composite springs most frequently found in bow limbs from Gordon Composites. It's a fairly unique feature of our robot (I haven't seen another one using this technique) and we've found it to be excellent in terms of repeatability and power.
I agree with most of Ether's OP.
Catapult-like devices (i.e. those that start in contact with the ball and accelerate with it) should be more consistent than shooters relying on impulsive contact such as kickers or punching devices.
The trajectory of the latter depends on more variables than the former, for example ball air pressure and exact contact point.
Beyond that it depends at what level we measure consistency: shots from a single location or ability to shoot accurately from a larger area of the field (i.e. robot scoring consistency). In other words we can argue that there is a balance between consistency of the energy source (where springs & pneumatics win) vs controllability where motors have an edge.
DampRobot
07-03-2014, 18:04
Linear punches have consistency limited by two things (as we've found). Besides all the obvious stuff (same pull back same distance, same release, ball in same place, reasonably speced springs, etc) you'll see consistency deteriorate when you don't have a large enough pusher plate and from knuckle.
We found in prototyping and from talking to other teams that large pusher plates give more consistent shots. We had a tennis ball contact the ball in early prototypes and quickly saw that a large plate improved consistency a ton. There's probably a good reason that 118 has a puncher plate the size of a frying pan. The teams with smaller plates generally only go for closer shots, because there consistency doesn't matter as much.
The other thing is knuckle. We aren't putting any spin on these balls, and as they have weird surfaces, they're sensitive to aerodynamic effects. Linear punchers also get the balls going quite fast at first, where they're more subject to aerodynamic effects. We've seen variations of +/- 1.5 ft up and down from knuckle alone. Perhaps someone has come up with a good solution for this, but we haven't.
But all around, I'd say that from what I've seen, high pivot catapults are killer at close range shots, but aren't that great long range shots. Low pivot (team JVN) catapults are quite good at long shots, but are bad at close range. Linear punches (especially at high release points) are pretty good at both types of shots. Strategically, I'd say high pivot and release (think team Boom Done) will continue to dominate the game.
Duncan Macdonald
07-03-2014, 18:05
I realize that 1 robot is not a sufficient counter point but 1114's wheeled shooter has been nice to watch today.
From what I've seen I would separate sprung shooter category into "surgical tubing" and "non surgical tubing"
I realize that 1 robot is not a sufficient counter point but 1114's wheeled shooter has been nice to watch today.
From what I've seen I would separate sprung shooter category into "surgical tubing" and "non surgical tubing"
I dont think 1114 is a wheeled shooter.
s_forbes
07-03-2014, 18:50
We've found that most of the variations in our shots come from ball placement on the mechanism. The balls are only sphere-ish; the fabric stitching has an impact on their shape. If your main contact point on the ball is in the middle of a square patch, vs the vertex of three square patches, the ball seems to react differently (and predictably).
A launching mechanism that is resistant to ball placement would likely fair best. I would expect the large flat-disc linear punchers to be more consistent.
So, how would one add consistency in ball placement to a catapult based system?
Abhishek R
07-03-2014, 20:33
So, how would one add consistency in ball placement to a catapult based system?
Make sure the ball is sitting in the exact same position every time. Any discrepancies will cause variation in flight path. Engineer the catapult so it cradles the ball better.
EDIT: We actually added a bar to the back of the catapult to somewhat simulate a curved shooter to hold the ball better.
Brandon_L
07-03-2014, 21:44
We've been nailing 10pt shots from around the 18ft mark consistently with a spring launcher. Battery drain only effects our reload time. What kind of data are you looking for, specifically?
Jeffrafa
08-03-2014, 09:56
and near the bottom of the list would be a) motor-powered kickers and b) motor-powered wheel shooters.
The sample size for wheeled shooters seems to be pretty small this year. We have a wheeled shooter consisting of 2 8" Colson wheels in a vertical orientation, shooting from fairly low. Each has independent closed-loop pwm control, to make sure the speed is consistent no matter the battery state (on the practice field we are making the same auto shot for a battery just used in a match as a fresh one). Also having the wheels above and below the ball lets us control the amount of back/top spin the ball has (currently tuned for backspin, although the effects of top spin on the ball trajectory are interesting and not quite intuitive - I am fairly confident we could increase our sweet spot range with top spin, but only the longest shot distance would improve. We have only done enough testing to demonstrate the spin effects I calculated, but not enough to justify a change.)
So far in our 1st day of qualifying matches we are 8/10 in auto mode (shooting from behind the white line), and have the 2nd highest cumulative auto score in our district (2nd to a 2-ball auto, but not by a lot). The 2 misses could likely be attributed to human error, as auto wheel speed was being read from an analog knob on the drivers station, now it's a set calibrated value.
Scoring high in teleop has not been as consistent as auto, but I am confident that will improve with more driver practice. We will see how things pan out in day 2 of this district, and at our second district in 2 weeks.
on the practice field we are making the same auto shot for a battery just used in a match as a fresh one
What about ball inflation?
Have you tried making the exact same shot with a slightly overinflated ball and a slightly underinflated one?
We have only done enough testing to demonstrate the spin effects I calculated...
Could you explain what calculations you did, and what testing you did to confirm those calculations?
I dont think 1114 is a wheeled shooter.
And to what would you base this assumption?
And to what would you base this assumption?
On that when you watch them in their matches it looks like they have a pneumatic puncher.
On that when you watch them in their matches it looks like they have a pneumatic puncher.
:)
That definitely looks like a gatherer to me.
That definitely looks like a gatherer to me.
The red and white roller is their intake. I wonder what the two black wheels do...
NickTosta
08-03-2014, 16:02
(image)
I see two pneumatic cylinders directly behind the ball. Granted, it's hard to say exactly what they're for, but they appear to be positioned correctly to "punch" the ball.
edit: To be honest, it looks like those small cylinders punch the ball into the black rollers. So, it's probably both.
mman1506
08-03-2014, 17:19
I see two pneumatic cylinders directly behind the ball. Granted, it's hard to say exactly what they're for, but they appear to be positioned correctly to "punch" the ball.
edit: To be honest, it looks like those small cylinders punch the ball into the black rollers. So, it's probably both.
We talked to them, it's a combination of wheeled and pneumatic but a the majority of the work is done by the wheels.
Duncan Macdonald
08-03-2014, 21:27
Alright, lets get this thread back on track....
I agree with Ether's original rough sorting of shooter types. Additionally, waiting for the ball to settle in your shooter varies by design (SS style being the best I've seen) and effects your in game shot percentage by giving defences extra time to close in.
For teams using surgical tubing, what have you done to improve consistency? I built a surgical tubing catapult in 2008 and it was never exactly the same between matches. This didn't matter much when throwing a ball over a bar. Does the variation come into play when aiming for a goal?
We were shooting from 8-11 feet not very accurately with a motor driven catapult at 100% power today we I bagged for our slotted time we are allowed and lowers the speed to 84% and now we are shooting at about 98% from three feet off the wall( just behind the low goal) we use 100% speed to shoot over truss
For teams using surgical tubing, what have you done to improve consistency? I built a surgical tubing catapult in 2008 and it was never exactly the same between matches. This didn't matter much when throwing a ball over a bar. Does the variation come into play when aiming for a goal?
Our team didn't end up using surgical tubing but I had thought up a way early in the season to ensure consistency between matches with surgical tubing.
If you wind up the surgical tubing on two drums, then unwind it just enough to attach it to the plunger or whatever it was propelling, you could then add a load sensor to the surgical tubing or the motor such that it measured the amount of tension in the tubing. Using two motors (Bag motors would would fine for this) you can have software that causes the drums the tubing is wound on to tighten or loosen the tubing depending on how far you want to shoot and how much the tubing has stretched. This keeps the tubing at the same tension for every match and also allows for lower powers settings in order to pass or shoot from a different distance.
Team 2014 can shoot from beyond 18 feet. At GTRE they had an extra volunteer to stand further back because we could launch the balls so far. We went to a full practice field just before the end of build season putting out springs to the second highest setting and made a moving 10 pt goal shot from about 45 feet. We do have massive power draw from out batteries as we run 2 rs 775s for our reload and also have a 6 cim drive. During our first matches it guys were telling us our voltage dropped to 4 volts and even under 4 once.
One problem we have run into though is the fact the the inflation of the balls can be different. if we use the harder balls on the field we usually shoot to high because the ball doesn't absorb as much energy.
Team 314 is using a "spear gun" tubing catapult and we find it effective for about 700-900 shots before we have to replace it which when done is cut to correct length then tested with scale to make sure of the same force. We have changed it 3 times through build season and our first district this weekend which was kettering. I dont know what all of you consider accurate or a "good" launcher but if you want to judge go to youtube and FIRSTinmichigan and you will find archived matches. Ill leave this one up to you guys to decide.:cool:
Teams with wheeled shooters:
Have you tried making the exact same shot with a slightly overinflated ball and a slightly underinflated one?
Speaking of adventures in consistency...
We couldn't for the life of us figure out why we had about a 75% accuracy rate in autonomous mode on the field as opposed to near 100% on the practice field. We controlled for starting position, angle, battery voltage, air pressure, etc. We even added a sizable delay between our "charge" forward and when our catapult fired in order to let the ball fully settle.
One of our four batteries had a terminal lug that was perpendicular to the rest. As a result, the battery power cable protruded *ever so slightly* into the catapult bucket, preventing the ball from seating properly in the bucket. Thus, every few matches (namely when the aforementioned battery was installed), we would miss our auto shot.
Needless to say, the fix was applied the moment we got back and is ready for Mount Vernon this weekend.
Kevin Leonard
11-03-2014, 17:24
From what I've seen of matches thus far this season, a high-pivot catapult with springs resistant to wear seems the most consistent.
We have a low-pivot catapult, and from our testing, we've discovered a few different things affect the variability of our shot (some controllable, some not).
Where the ball seats itself in the catapult
How far we pull the catapult back
The thickness of our catcher panels (we generally shoot with our side-panels for catching retracted. This helps to:
a)center the ball, and
b)avoid breaking our catchers when they are under defense. They are filled with a padding to help deaden the ball when we catch, but the padding slides down our catchers over time, changing the effect of the catchers)
Ball Inflation and Ball Condition-
This is a very small effect, but it can change our accuracy if we're shooting from either end of our sweetspot. If the zipper of the ball sits right on one of the two prongs of the catapult, it can also change our shot.
Our springs were custom made to be extremely resistant to wear. So even after hours and hours of drive practice, we still shoot very well.
Andrew Schreiber
11-03-2014, 17:32
Alright, lets get this thread back on track....
I agree with Ether's original rough sorting of shooter types. Additionally, waiting for the ball to settle in your shooter varies by design (SS style being the best I've seen) and effects your in game shot percentage by giving defences extra time to close in.
For teams using surgical tubing, what have you done to improve consistency? I built a surgical tubing catapult in 2008 and it was never exactly the same between matches. This didn't matter much when throwing a ball over a bar. Does the variation come into play when aiming for a goal?
We found small clamps actuating into position when the ball is in the cradle helped eliminate almost 100% of the settle time.
Mark Sheridan
11-03-2014, 17:36
We found small clamps actuating into position when the ball is in the cradle helped eliminate almost 100% of the settle time.
I second this, we just made a very crude one and it does wonders to the settling time.
In my teams shooter we incorporated a wheeled shooter along with a pneumatic launcher, the way it works is that when the pneumatic launcher is engaged it lifts the ball right into the rectangular formation of rotating wheels and gets shot out of it, the distance is controlled by the drivers, who control the rpm of the wheels. We ended up winning the innovation in design award at gtrw for it.:cool:
Answer42
14-03-2014, 02:11
During our first matches it guys were telling us our voltage dropped to 4 volts and even under 4 once.
Under 4? I would expect the radio to reboot way before that. You guys have any brownout/ main breakout problems?
Tim Lehmann4967
14-03-2014, 07:58
4967 has a extremely accurate and consistent shooter, but we need to be about a foot away from the front of the low goal. We missed 5 shots total during the Gull Lake Regional. 33 has a very powerful springloaded shooter, but it is very easy to push them out of shooting alignment due to their use of omni-wheels.
I'm not sure about wheeled or linear shooters, but we have done a lot of prototyping catapult shooters. It's all about holding the ball consistently. High pivot, low pivot, springs, pneumatics, close shoot, distance shot does not matter. Can it position the ball the same everytime?
IronicDeadBird
26-03-2014, 13:48
Teams with wheeled shooters:
Have you tried making the exact same shot with a slightly overinflated ball and a slightly underinflated one?
Are you talking about making a shot with one angle from one point as being the same. Or are you talking about getting the ball to target consistently?
Are you talking about making a shot with one angle from one point as being the same. Or are you talking about getting the ball to target consistently?
The former.
IronicDeadBird
26-03-2014, 19:43
The former.
I will ask the team I work with (1339) I don't believe we tested different ball size but I will ask and see if we have any information at all on it. Another team that might have the info is Robotknights they have a pretty intense wheeled shooter so if I can get a hold of them I'll see if I can find anything out.
Mike Norton
26-03-2014, 20:13
How about 100% in auto mode. 14-0 at WPI and 18-0 at Bryant. We use surgical tubing cross bow style.
We use extrude aluminum 1"x1" as the launcher with two rollers at the very top. This give the ball a slight backward spin to keep it floating.
Johnnybukkel
26-03-2014, 21:36
Our robot doesn't fit any of the shooter catagories... We are very consistent in the high goal (when our pit checks are done right).
Our robot doesn't fit any of the shooter catagories... We are very consistent in the high goal (when our pit checks are done right).
Your category of shooter is Scary.
Johnnybukkel
26-03-2014, 21:50
Your category of shooter is Scary.
Nah, Howie's great! Definitely a robot that the kids can play with.
Kevin Leonard
26-03-2014, 22:16
Your category of shooter is Scary.
Quoted for truth. And spotlighted.
See you guys tomorrow!
#repeat?
At the LA Regional, our pneumatic catapult shooter had a 100% scoring percentage in both autonomous and teleop from our 2nd through 10th qual round and had 1 miss in elims amongst heavy defense (scored 6 in elims QF).
Note in the first round we missed shots due to a pneumatics leak which was fixed shortly after.
I was impressed with how consistent our shooter ended up performing.
We shoot between 6-8 feet away when stopped, and 8-10 feet while in motion (we use sensors, indicator LED lights, and telemetry to determine shot attempt distance)
DampRobot
26-03-2014, 22:34
Your category of shooter is Scary.
Uhh... So how exactly does it work?
Uhh... So how exactly does it work?
You know those tanks from Monsters Inc....
Kevin Leonard
26-03-2014, 22:40
Uhh... So how exactly does it work?
It's a giant, super-scary powerful slingshot-thing.
Basically it could probably shoot from one field to another at championships. Legitimately.
Johnnybukkel
26-03-2014, 23:26
It's a giant, super-scary powerful slingshot-thing.
Basically it could probably shoot from one field to another at championships. Legitimately.
Well, to be honest, we can only shoot 19 feet... Vertically...
And #Repeat is possible.
IronicDeadBird
30-03-2014, 01:25
Teams with wheeled shooters:
Have you tried making the exact same shot with a slightly overinflated ball and a slightly underinflated one?
Man this took me a long time to get any solid data on our team doing anything related to this until today (so sorry for the wait). As of today we did scrims with another team who had 2 balls. One was under inflated and one was reasonably inflated. The level of inflation had a major impact on the distance we got out of it. Other then that everything else was consistent we generally just lose distance as we lose compression on the ball. Now that I think about it I might see if it is within the rules to let our arm driver feel how inflated the balls are pre match to see how to compensate for aim...
Right with ironicdeadbird on this one. The under inflated one was VERY underinflated, and was actually a yoga ball from Target inside the official cover. sitting on it compressed it to 1/2 diameter. Shots were about 1/2 normal distance. We've been practicing lately with an overinflated ball similar to what we found on the field in Utah so that our shooter would be set up to handle it.
Right with ironicdeadbird on this one. The under inflated one was VERY underinflated, and was actually a yoga ball from Target inside the official cover.
Just to be sure I'm reading this correctly: are you saying that the underinflated ball IronicDeadBird used for his testing is the "very underinflated" one with a Target yoga ball inside?
If so, does anyone have any test results with an "official" "regulation" ball that is somewhat (but not "very") underinflated?
Ironicdeadbird is one of the mentors on my team, and we're talking about the same ball. A little more background:
Prior to Utah, we practiced with a ball that we deemed properly inflated. In Utah, the balls were slightly more inflated than we assumed they would be, and they became tighter as the event went on, so they seemed "overinflated" to me. Since returning, we've inflated our ball more to simulate field conditions. Yesterday we played with two balls that were not ours: one was very slightly under what we had been using prior to Utah, the other was the aforementioned yoga ball.
We shot from specific points on the field yesterday that we had previously tested. The distances were similar for the ball with "normal" inflation and very short with the other ball. Consistency is very hard to measure with our design, however, as ball shape, location in the shooter, and battery charge all effect it as well.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.