Dual flipped NEO geared at 8.18:1 (10.6 fps free speed), which is a little on the slow side but is around the average speed for a custom dual reduction gearbox I felt.
Uses 3.5" wheels with 1/4" ground clearance due to the spacer I have underneath the frame. I stepped the plates down an 1/8" inch where it meets the frame, since the 1/8" from the 2x1 wall and the plate would allow a bearing to be set in, and I also felt that it would be useful for making sure the plates sit square.
Of course with this type of design, you have to go with Chain-in-tube just to make sure it’s as compact as possible. I included bolts and washers just to see if it would actually work.
The back cover is a 3D Printed cover that I imagine would best be printed in Onyx, though since it doesn’t take any force at all, ABS should also work.
Not sure how I feel about putting a gear reduction on the cantilevered side of a WCD wheel shaft. Without actually doing any calculations, my intuitions says the shaft is liable to bend slightly under the weight of the drivetrain, which would mess with the C-C spacing for the gears. Does anyone have experience with this setup?
Honestly just looking over it I have a feeling it could mess with the mesh and lead to increased wear, but I’d also guess that there should be enough support on the shaft to prevent it from flexing too much, then again, it only takes a couple thousandths of an inch. Also I could maybe recommend some pocketing on the main plate? It may not be worth it, but nothing’s wrong with a suggestion. Otherwise nice job!
The motors are still in the vertical projection of the wheels, so you can’t use 6 inch wheels. Overall I struggle to see how this packaging is better than the conventional flipped neo gearbox designs. You lose the space made up by making the gearbox slimmer by moving the wheel out further, and in addition to higher part count I am concerned about the cantilevered high load shaft messing with the gear CC.
Possibly. Fixes the issue of the gear spacing caused by cantilevered wheels, but it could create a weakness in the rail along the X&Y, mainly taking an impact weird and the rail bending at the slot or slamming an object so hard the frame ballooning outwards if slotted too wide. There’s a good couple ways of how to approach the issue, of course each one will have their problems.
I think you agree with me that this setup is mechanically less robust than putting the gears on the opposite side from the wheel. It introduces various failure methods that you otherwise wouldn’t have. So what do you gain by doing it this way? There are a number of different flipped-NEO gearboxes posted here that don’t use this method. I don’t see what advantage this method has over those. So if it has no advantages and some disadvantages, why use it?
I feel like I am obliged to post this now. There’s a giant /s going over the entire CAD because I did it for fun, and was purely a “what-if” situation. I was bored at my friends D&D game I was sitting in on so I decided to try something a little different. It most definitely doesn’t have a practical purpose.
Is there a reason you did it that way instead of the more traditional flipped-motor gearbox with the output stage gear on the opposite side of the tube from the wheel?
First off, I should note that our robot last season was very light. I don’t remember an exact number, but let’s say 80 lbs with battery and bumpers. Obviously this will result in less deflection of the shaft. Take this as you will.
Not to hijack the thread but to answer your question, our gearbox wasn’t a flipped motor design, but neither was it traditional. Our design was driven by three constraints:
Make a custom, compact gearbox.
No chain/belt-in-tube.
Accommodate up to 8" diameter wheels. (This was designed in the offseason. Ultimately we used 6" wheels).
Pulling the motors over to the first plate, putting the first stage over (and partially in) the tube, and putting the second stage between the outer plate and the wheel accomplished all the above. Both plates were 1/4" thick aluminum.
