[FTC]: Shooter Wheel Safety

I’m a veteran of Aim High in 2006, where we built a robot shooting the 7" foam balls with 8-inch wheels running on a direct-drive CIM motor. What is that, 3,600 rpm? It worked well.

Our FTC team has built a shooter mechanism for this year’s game with a pair of 4-inch wheels spinning at an estimated 5,000 rpm using multiple motors and a three-stage gear train. This scares the whee out of me. The wheels go so fast that the rubber tread is visibly lifting off the plastic wheel as it spins.

I would like some thoughts from VERY experienced robot folks here. I am concerned about the safety of this mechanism, especially since these are axles running in bronze bushings – not bearings – using wheels that I am concerned were not designed for these forces. Since these are shooter wheels, there is no way to fully enclose the mechanism, either. Are the axle shafts designed for this? The wheels? Are the gears going to survive the heat build up without spectacular failure?

These aren’t being used on a big FRC field with the drivers behind Lexan shields, either. Drivers are standing a couple of feet from these robots with no protection except for their eyes.

Am I right to be concerned here, or were these parts engineered with these forces in mind? I don’t know if this has really been thought through and I would appreciate some discussion of this.

One thing I would like to add would be: Do you need 5k rpms?

Wiffle Balls are a lot lighter and should go far with half that, well atleast cover the 40 some inches.


Aim High had a m/s speed limit on moving parts, didn’t it? you would probably be wise to keep moving stuff below that speed. Small diameter things spinning fast are not as dangerous as large diameter things spinning at the same speed. But it is definitely something you should be concerned about.

A “three-stage gear train” with the gears we have should give you a 1:27 gear ratio.

In all the tests that i have done, I’ve seen that even a 1:23.xx ration using the 3" wheel was enough to project the ball beyond the 10ft shot distance limitation in the game manual. Not sure what you final release of the ball is, but you may want to re-design that so that you can slow down your shooting wheels.

Besides all of the safety risks…from your own observations you are already greatly increasing your own reliability risk. I would suggest you look at a re-design.

I just wanted to mention that I am lead mentor for a program that has more than 70 students, 12 robotics teams and 10 mentors. I wasn’t involved in the design of this robot, another mentor was.

Thanks for the advice and comments thus far.

I’m in a similar boat as you coming from FRC. Last year, team 2228, we used two sets of rollers, similar to your wheel rows, to fire the moon rocks and I’m trying to go with the same approach, and had for a time a 16:1 ratio, but have toned that down to 8:1 as a result of the burning out of a motor (in my opinion due to faulty mechanical work not an inadequacy of torque) and I now hearing that you have sustained a 27:1 ratio, according to grampashades, I’d love to know what are some of the things you guys did to better secure the gears.

Are you sure you need a 1:27 ratio?

Because we have a 1:9 ratio on a 4 inch wheel powered by 2 separate motors and it shoots far over 10 feet I could only imagine how far that one goes, as for heat build up we are allowed to use a kind of lubricant on the gears so that will eliminate some of your problem. Try to zip tie the rubber on the wheels onto the wheel thats what we did and there is no problems with it. For shielding the system you could shield half of the wheel depending on your design and amount of materials available.

just my .2 cents


And remember robot rule <R16>:

<R16> Game elements launched by Robots should not be launched with velocity greater than that required to reach a maximum of four (4) feet above the Playing Field surface, nor travel a horizontal distance greater than ten (10) feet.

Why would you need to zip tie rubber onto a 1:9 ratio shooter wheel?

And how much grip does the wheel have on the ball when it’s zip ties touching it and not rubber?

From an inspector standpoint, speed that produces even small failures in the parts should be avoided. If the shooter is running fast enough to lift the tread off the wheel, then I think you are looking at something that has crossed the safety threshold.

Rick, I would recommend that the mentor in question redesign the bot. Here are some things to bring to the table:

Is the shooter single-sided? In other words, is one side the spinning wheel and the other a firm backing? If so, a team can cut the RPM need in half if they replace that backing with wheels. One of my teams had this design until they realized they wouldn’t have the needed space to add the necessary gearing. The solution was to add an additional spinning wheel to the top that the balls hit once they get to the proper angle in the shooter circle.

Even at 1368RPM (1:9 ratio, which is what most of my teams are going with I think), the shooter’s contact time with the ball is so small that even a large amount of motor torque applied to it will not accelerate it as much as you think – these are wiffle balls with alot of slip. There are ways to solve that though:

  • Increase contact time via an hourglass-shaped design between two tetrix wheels
  • Try different wheels. I’m not 100% certain these are the same LEGO wheels a student brought in that are 3.25" in diameter, but they definitely look similar. Attaching these to the TETRIX axles is easy via cunning use of the allowed plastics as a makeshift hub. The tread attachment is a little flimsy, but we’ve definitely see an improvement in contact time.
  • Increase the input velocity into the shooter. If you notice successful '06 designs from the videos have very speedy ball manipulation via belts, fast helixes, or other wheels that pre-accelerate the ball before it hits the shooter. It’s like the merry-go-round as a kid – once you get it started, it’s easy to slap it as it goes by quickly and give it enough juice to spin faster. Actual contact time was nearly infinitesimal, yet due to its previous velocity you were able to boost it to speeds that made the onboard kids very very dizzy. Everything you need to know you learned in Kindergarten!