What prompted you to go with, what I’m amusing are, pockets for lock-nuts? I would think that just drilling and tapping holes in motor plates would be a FAR easier and more ridged way to build something like this.
Otherwise I like the design, definitely one of the more compact I’ve seen. Personally I’d swap the chains for spur gears for the sake of maintainability, but to each their own. It’s not clear from the pictures, but is the idea that other wheels would be driven by chain-in-tube?
Depending on hardware size, a drill-and-tap might be rather “interesting” in the end of a plate. You could significantly weaken the material if your screw was too large for the plate’s thickness. If the plate is too thick, pocketing becomes required to meet the FRC weight limit.
OTOH, doing pockets for locknuts like that means you lose your side-to-side stiffness, and if it weren’t for a rather small notch at each joint it’d be relatively easy to get the plates apart when you don’t want them to be apart… May want to increase that little notch’s depth and put more plate into it.
It’s a tough tradeoff.
We ran a similar setup this year. Our solution to assembling the plates was an interlocking slot and tab in the perpendicular plates. It was inspired by 192’s 2012 gearboxes. All of this was achieved on our CNC mill, however could have also been done on our Bridgeport (albeit, at the cost of a little more time).
As EricH said, it would leave the material too thin. It’s a #10 bolt going into .25" thick plate, which would only leave .026 or so plate left on either side of the tapped hole.
Thank you! A picture of the side view should answer your question- if I use gears, the CIM will go lower, which will hit the 2x1. That picture also shows chain-in-tube.
Yeah, everything becomes tradeoffs when size is the final goal. The side view linked above shows why these notches can’t go any deeper, unless I’m willing to file nuts into shape.
Hopefully, the grooves I’ve put into the design will help keep the gearbox together, sort of like not-full-depth slots/tabs, better shown in the top view.
Then use #8 screws? 1/4" is plenty thick enough to drill and tap into the ends of, my team has done it dozens of times with zero issues (even into polycarbonate). Nothing against your solution if you can get it to work, it just looks like a TON of unnecessary machining.
If you’re concerned about rigidity and overcomplicating these parts, why not just use a 3 by 3 tube or something to that effect? Helps reduce part count (though it would be a pretty complex single part), and just generally being one piece makes assembly and life much easier.
[strike]Ease of[/strike]Lack of Access (for assembly and maintenance etc.)
I’m not sure about you but I really wouldn’t like trying to get the CIMs onto that tube! You could get away with it if one side was open, I think, but it’s not going to be fun getting the motors installed properly when you have to work the allen wrench on the far side of the CIM shaft, and then you get to install the other shaft with the gear on it…
So I notice you’ve put thrust bearings behind the pinions on the cim shaft. It’s good that you have them but right now they aren’t against anything that would support an axial load, they have to be up against the plate not just floating on the cim shaft. You should also have one behind the bevel gear, that one will actually have much higher axial loads than the bevel pinions.
I’m also not confident in putting that much side load on the cim shafts, probably not much more than having a spur gear that far out but I would advice testing it a lot before putting it on a competition robot.
I second this suggestion. Even the weakest 8-32 screw will provide over 500 lbs of clamping force, more than enough for an FRC gearbox. Also the cutouts you have right now create a large stress concentration at the sharp 90 degree corner inside them especially when under tension from the bolt. Over time you’re likely to see cracks developing in that area.