Prototype Small Wheeled Shooter

If you’re clever, with a custom mount plate and a trimmed down shaft, you can mount a MiniCIM to a VersaPlanetary gearbox.


I’d love to hear more regarding details of your shooter: motors, gear boxes, gear ratios, wheels, voltage to motors, distance between wheels, etc. Thanks.

I can’t believe I didn’t notice it was an 8mm hole until you said that.
I know what we’re playing with next week. (Note: Not for our shooter)

blinks So that’s what that “notch” is in there for. Or so you’d hope. I wouldn’t want to pin my hopes on that set screw in the CIM keyway.

Also, I have a feeling a mini-CIM would really push the limits on the load rating of the VersaPlanetary. And if you swapped a CIM in instead… Well that’d be extremely exciting.

Depends entirely on the reduction, and output shaft configuration. Check out the loading tables at the end of this guide:

I should have pictures up in the next hour or two. I have a student doing it right now.

Instructions are up on our blog

This was a pretty simple conversion. We cut the shaft more than would most likely be recommended but that lets you directly mount it to the Versplanetary with out any adapter plate.

I don’t recall but, do you have any video of shooting with the disc upside down?

Not from this prototype, it needs some sort of cap over the wheels to get long shots with upside down discs (think the C channel from robot in 3 days), otherwise we find that the discs rise up out of the wheels before they leave the shooter.

I stand corrected. If you’re willing to green light a 1/2" hex shaft 300:1 BAG Motor for simple loading, that’s the equivalent of a 1/2" hex shaft 50:1 big CIM motor. Which at 90 ft-lbs stall and nothing to sneeze at.

Are the black spacers for standing off an AM-9015 motor too small to stand off the CIM?

Didn’t think about it but that does work. That will add back a 1/4in to the shaft.

We’re nearly identical to that. BAG motors 1:1 and 1:3 wheels close together and the wall ending at the center of the fast wheel. Compression at about 1/2 inch. You have just verified our settings.

*I’d like to hear from teams – if there are any – who been successful spinning the 2nd wheel approx 30% to 40% faster than the second wheel, instead of 3x as fast.

Do you have any reason to assume that is more optimal? We would like to know.

It’s a hunch, based on engineering intuition; there are too many hard-to-quantify variables to do any accurate calculations. Looking for some experimental corroboration or refutation.

We’re testing a 3-wheel variant tomorrow. Top speeds will be 4300, 6300, and 14000. That’s 3:1 BAG, 3:1 FP, and 1:1 BAG. I know the FP is illegal, but it’s equivalent to the BB 550 while I’m waiting for some 550s to arrive. I’m planning on trying it without the slowest wheel as well to see what differences we get. It’s not your 30% slower, but it’s a little closer.

We are testing our linear shooter today.
We are using one CIM and one mini-CIM, each with a stacked pair PAIR of inline skate wheels - 76mm OD & 75A durometer.
These have dramatically better grip than all the other wheels we evaluated so far.
The disk are captured in the V formed between the two stacked wheels per motor as they spins through.

We have a piece of 80/20 20mm slotted framing opposite the wheels, and we cover the side with adhesive backed, high density urethane foam sheet material.
The friction factor is significant, as we found out with our 180 degree arc 8" wheel, direct mounted to CIM tests.
With the small wheel shooter, we are starting out with lower RPM (surface feet per minute) to avoid melting the softer urethane skate wheels, and we expect that the more optimized friction will make up for lower surface speeds of the smaller wheels.

As we confirm what our 1st stage alone can do, we will initially add the mini-CIM second stage for an extra 1000RPM. If that does not give sufficient exit velocity we will install the BAG 14K RPM motor as the 2nd stage.

We expect that much less compression (~1/4") will be needed with the more resilient and grippy materials contacting the disks. We also expect the groove in the slotted framing will help center the disk as it rolls along the length of the channel. If the slot in the frame is too narrow to keep the disk centered, we will machine a 3/4" radius convex curve into the side rail and it will then be covered with the dense urethane foam sheet.

With the V grip of the wheel pair on one side and the concave face of the rail along the other side, we will be able to keep the disk traveling through the shooter maintaining an air gap below the disk, and only the edge of disks touching the shooter as they accelerate through.

Hopefully nothing will be needed to hold the disk down from above either.

Video soon.

-Dick Ledford


Would love to see the results of this, we are going to do three as well, but if there is no significant difference then we will want to save weight. Thanks!

I’m working on getting some more quantitative results, but our preliminary results are, well, extremely promising.

Stats on our shooter prototype:
Baseplate is 1/4" lexan, unbraced. Shooter is a little bendy as a result. 2-7/8" orange wheels on 3.5" centers. Wall gap is 10.625". Theoretical top speeds are 4300, 6300, and 14000 at 12V. We’re using the bright orange urethane flat belt as our sticky wall material.

For this particular testing setup, we’re firing at a 19.5 deg angle, shooter was 41" off the ground and 25’ from the goal. With three motors, we were nailing our cardboard just above the top of the 2 pt goal. With two motors it was landing about a 6" below the 2 pt goal.

This was testing an integrated solution with a 5 gallon bucket hopper right next to the firing wheels, so we were dragging the frisbee out from under some others. When we were hand feeding a frisbee in, it was throwing them over the top of our cheesy cardboard 2 pt goal and they were still rising. I’d say it would be about 18" above the 3 pt goal.