Modifications to FIRST Gray Wheels for Better Consistency

On this years robot we are using the FIRST Gray wheels for our shooter. Does anyone know of any modifications that were made to these wheels that increased shooter consistency. I have seen…

-Ball Compression - What is your optimal compression?
-Filling in the wheels with foam - How do you do this?
-Filing down the rubber to an angle - Method?
-Friction Tape on the Wheels - We tried this without a substantial reward

Does anyone have anything to contribute?

Our compression is 3", which is waaaaaay too much, but it’s too late now to change that. What we are trying to implement is a compression testing system (somewhat like what 1218 has), that tests each ball before it enters our shooter, so we can snipe like crazy.

We used 6" FIRST wheels with a 2.5-3" squish factor. That’s what our team named it, it’s stuck around since week 1.

We had a similar issue with consistency on leaving the shooter. However, we improved this by funneling the balls more precisely into the shooter, not modifying the wheels. Another way you could do it with wheels if your design allows for it is making it four wheels (if it’s already not.) We used two wheels and couldn’t change ours but we added two pieces at the top of our turret to ensure that balls entered the wheels at the same spot since before there was a good 3-4" wiggle room that they could enter variably at.

I don’t know how your shooter is designed but if there’s anyway you can make the balls enter more consistently I think that will be much more helpful in improving your shooter if you have a good squish on the ball.

-Anna

The 8" solid wheels from 2010 (and 2011)? How would you fill those with foam? The best way to improve consistency is to have a system that works with a large margin of error. For example, our shooter puts sufficient backspin on the ball that it essentially drops straight down as soon as it hits something. This results rather lax requirements for the height at which the ball must reach the basket in order to score. Also, we have a shooter hood that has adjustable compression and angle. It is a bit late, but designing things like this into your mechanisms can save effort when you have to adjust things.

EDIT:

QFT

In testing we noticed that if the ball was too far forward in the feed mechanism it would contact the shooter wheels just a bit longer than if it were farther back. This was enough to slow the wheel and ball down resulting in a significant loss of range and height. To fix this we adjusted the funnel to get more consistent feed to the shooter wheels. The consistency of the feed to the shooter is related to the consistency of its output.

Team 3322 uses 2 2010 8" Rubber treaded FIRST wheels with 1.5" of compression. We tested compression up to 2.5", which did not work well at all. Our shooter is in the style of an “S,” so we have a larger period of contact between the ball and the shooter wheel. We also feed balls with a 4" foam wheel coupled to a window motor. The slow speed of the window motor gives time for the RPM of the shooter wheel to reapproach the assigned speed (with PID).

Mechanically, our feed wheel is attached to a spike. Within our code, the feed wheel will not run unless the speed of the shooter is at the set value.

We are using 2 8" FIRST wheels running on a common 1/2" steel keyed shaft. Our compression is about 2". We found that by spacing the wheels apart by about 2" will yield less compression, but more consistent shots. We later replaced the spacers with a 6 inch diameter flywheel to increase the inertial mass, which seemed to yield some additional consistency. The flywheel also had the added benefit of giving us a better mounting surface for the shooter wheels and made for a cleaner looking shooter.

Did you test anything before and after adding the flywheel, to determine how much improvement it provided?

I’m very interested in flywheels and how much benefit they provide. I started a thread on it:
Flywheels and Shooter Wheels

To answer the OP’s question, the only special thing we do is clean our shooter wheels with alcohol pads between matches.

We have a surprisingly precise shooting system, even at high power, and I think our secret is our conduit guides. Most teams I’ve seen have either two sets of rotating wheels, which is inherently inaccurate because of the difficulty of syncing two motors, or a flat hood to guide the ball on the top. Rather than one hood, we have two pieces of conduit that roughly follow the curve of the wheels spaced between 3 and 4 inches apart. We have two shooting wheels on the same axle, between 2 and 3 inches apart, and the combination of those keeps the ball exiting the shooter very consistently.

Another thing we were careful about is avoiding any interference on the ball near the top of the shooter. The conduit is attached securely at the bottom so we don’t need any attachments at the top that might contact the ball inconsistently and so throw off its trajectory.

Other specs:
-$@#2" squish factor

  • 6" traction wheels, covered with duct tape because we were shredding balls
  • 5000 rpm and ~150 oz-in of torque on the shooter (the torque is on the low side of what we would like)
  • consistent feeding from a belt

A few tricks that I’ve picked up with the help of some friends:

  1. Turn the ‘Crowns’ off of the wheels. If you look at a fresh FIRST wheel, the tread isn’t flat but actually ‘U’ Shaped. I turned the crown down by putting the wheel in a lathe and taking a file to it.

  2. Make sure the balls are entering the shooter consistently. Some systems I’ve seen have allowed the ball to move around left to right before hitting the shooter wheels. In a perfect world, the ball enters on the same center as your shooter wheels 100% of the time.

As far as compression goes, it’s really dependent on your style of shooter.

Our team, in between competitions, simply fabircated our own custom 9" wheels out of 1" thick aluminum stock. All the weight is now on the edge, and they have a custom chamfer now. Those turned out really nice :slight_smile: We love them now.
However, before we did this, we were using the 8" rubber tread wheels from the KOP a few years ago. What we did to somewhat improve consistancy was chamfering them, albeit rather crudely. The quickest way to chamfer them would probably be just spin them up and hold a file to them. (And cover anything below them! ::safety:: )

We padded the side walls on either side of the ball from before it gets fed into the shooter all the way out. We originally had about 3/8 inch on either side.

Our original “hood” was a 1 1/2 inch wide piece of hardwood cut on a bandsaw to get the required curve. At Alamo, we discovered that the cut was not perpendicular to the wood, causing a tilt to one side when the ball was launched. We were able to straighten it out with several layers of duct tape on one side. Our new hood is adjustable for tilt, compression and curvature.

We may have had some inconsistency when the wheels were not rigidly attached to the shaft. We are using two of the 8 inch gray wheels on AndyMark hubs and a 1/2 inch smooth (no keyway) shaft. We drilled and tapped the hubs for two set screws each. We also drilled into the shaft, through the set screw holes, so that the sets screws have more “bite”. Sometimes the set screws loosen and the wheels will wobble on the shaft even though they do not spin.

Using an encoder to enforce a constant speed on the wheel will help, if you are not already doing so. Also, allow time for the wheel to come back up to speed. I have seen several instances where one of the top teams fires off several shots in rapid succession and the later ones all fall short, by more and more.

Probably the most important thing is to make sure your test method is consistent. We were able to see variations in the trajectory of the ball as it left the shooter by holding up a yardstick and comparing the trajectories of a series of shots.