Is that drum being balanced through some means I’m not seeing? Or is there going to be a second one of those?
How is the tooth going to be held in?
The new Avatar is a giveaway:
+1 on balancing necessities, looks neat. I assume you’re going to weld that bar in?
Still deciding between tooth on other side or balancing by removing material. The tooth is getting welded and bolted in.
Looks like someone used the wrong sized keystock. 
Nope! Keyway was manual machined by me 
What weight class? There are a couple of issues I think you’ll run into if your doing serious combat.
Featherweight.
I suggest having an odd number of teeth greater than one. Odd number of spokes naturally balance better than even numbers. (check out wind turbines, car wheels, etc.)
My fear with going from single tooth to three-tooth is that with the intended weapon rpm (10,000-11,000), the opposing robot would ride up the wedge just fine, but the teeth would “grind” down on the other robot instead of hooking and throwing/breaking. I could lessen the weapon rpm, but I’m wanting to go with maximum impact. Unless someone knows this wouldn’t be an issue.
Single tooth is the way to go if possible. There are no effective modern spinners with more than two teeth AFAIK.
I’d be concerned about the hardness of your tooth. Most combat robots use AR500 (pre-hardened) or hardened S7 teeth (54 HRC). Even my AR400 beetleweight single tooth disk was pretty much useless after a couple of matches. Also the way the tooth is inserted into the drum creates a large stress concentration. It’s probably going to end up cracking at the inside corners.
If you look closely, wind turbines, car wheels, and such tend to have prime numbers of spokes. This reduces the number of resonant frequencies to worry about providing damping. Likely not a significant effect for this application.
While I’ve never done any combat robotics, I do agree that rounding those corners (especially on the hub side) would distribute the stress and improve the longevity of this if (as I’m understanding) that key is intended to be hitting things with a significant delta-velocity. I would also consider compliant ways to keep/return the key in the keyway as an alternative to a hard keeper.
Edit: OBTW, I’m not sure if everyone sees this, but on my computer, the OP photo looks like this device is stuck to a wall over a doorway rather than resting on a kitchen counter; it’s rotated 90 degrees counter-clockwise from the direction I now assume to be vertical. Perhaps this perspective will make my earlier post make more sense…
Banana for scale?
What are the dimensions,weight and how are you planning on driving it? That axle hole looks pretty small for a featherweight drum…( not that I can really tell the size).
That’s why the drum itself is softer than the tooth. I was worried about wear
Sorry Gus but I don’t quite understand the question…
3.25" dia. x 3.4" length, about 8lbs, propelled using 2 v-belts and a brushless motor. Hole is 5/8"
That makes sense. If you only want one tooth, I would place a denser material where you would put two of three equally spaced teeth to get the center of mass back in the middle of the shaft. It will make balancing the drum easier than just putting weight directly opposite the one tooth.
Edit:
Or just provide mounting points on the outside of the drum to add mass for balancing, but I think with machined components you can have a designed drum that’s balanced with out messing around with adding weights to it. The single tooth unbalanced will drive harmonics into the robot.
Tungsten Rods are available that have 2.5 x the density of iron. You could mill/drill one or two lengthwise holes inside opposite of the Key Feature and force fit them into the cylinder to fully balance it.
You would need calculate the mass moments of inertia, to find a solution that is balanced around the rotational axis. I could help with the calculation if you have the existing materials and full dimensions.
Another thought is to just machine out material 60 degrees on either side of the tooth. That would have the same effect as adding mass 120 degrees equally spaced around the drum.
Not exactly. It doesn’t have to be a prime number. The odd number disconnects the beam elements and essentially doubles the stiffness of those beams because they only cross half the diameter instead of all the way across. This would double the resonant frequency in those beams hopefully driving them above operating frequencies.
Stiffness is directly related to length, and natural frequency is the root of stiffness over mass.
The odd numbered spokes or mass placement creates a trigonometric geometry that is more forgiving to the center of mass allowing for more error in the manufacturing. It’s slight, but slight helps at high rpms.
Another reference, see fidget spinners. I don’t recommend an even numbered module spinner to help your focus… :rolleyes:
So basically, because math.
That’s what I’m planning on doing, I just need some serious help with the math :yikes: