Following my previous models of flipped-CIM and flipped-MiniCIM gearboxes, I made a flipped NEO gearbox designed to interface with the AM14U-style Kit of Parts chassis. We don’t have the data yet as to how well these motors actually work in drivetrain applications, but I can say that their small form factor makes them package beautifully.
Some specs:
• Extends 2.3" into the chassis (compared to 3.6" for a NEO with the standard TB Mini gearbox, can be lowered to 1.8" by removing the encoder and trimming the motor shafts)
• Weighs 3.24 lbs (1.36 lbs without motors, compared to 1.95 lbs TB Mini)
• 7.67:1 reduction for 16 ft/s adj with the default 6" wheel (or 10.6 ft/s adj with a 4" wheel)
• 2" clearance from the frame perimeter to the motor
One of the things we did with our AM14U in 2016 was to make the center wheel a dead axle (using a bushing in the bearing hole). We did that because we had to stagger our gearboxes for a really narrow chassis, but it had the added benefit of making dropping the middle wheel much easier.
You might consider moving your second reduction stage to inside the frame rail (might have to widen the wheel wells slightly) and replacing the center wheel hex shaft with a dead axle to get the same benefits.
I also think it’s great, and I hope my friends in Kokomo are taking notes.
One question: Is the geared second stage a stock Toughbox stage? That would cut the parts-to-acquire count a bit for teams that get the AM14U3*, and also allow teams to swap ratios around to suit the game.
*Assuming, of course, that we will get the AM14U3 again.
Dead axles are not used to transmit torque, merely provide a support for bearings. The corner wheels of the kit chassis are on dead axles, and the torque is transmitted directly from the belt pulley to the wheel. The axles are not used to transmit torque. Moving the second gear stage inboard and using a dead wheel axle would mean that the gear is directly coupled to the wheel. Dead axles are rather easier to construct than live ones, because you can use standard round bearings and axles. What does confuse me is why it is better to need a live axle for the cluster shaft than for the wheels.
I considered that, but decided against it because it would make the gearbox not a drop-in replacement. It also makes at a good bit harder to assemble and maintain.
Right now the second stage is not a TB reduction because the 36t pulleys only come in 1/2" hex and the TB cluster stage is 3/8". In theory, you could use a 1/2" to 3/8" adapter on the 36t pulley and switch the middle stage bearings and shaft for 3/8" hardware. Then you could use the gears from the kit chassis (16:48 instead of 18:46). It would also slow the robot down a bit, which I was trying to do anyway.
…do you know something we don’t about the AM14U3 for next year?
Exactly. The last bull gear would be mounted to the center wheel bolt circle, rather than to the hex shaft, and the whole center wheel would ride on bearings on a 3/8" (not rotating) bolt.
What does confuse me is why it is better to need a live axle for the cluster shaft than for the wheels.
It’s better not to have a live axle on the wheel because wheels wear out, and it’s way easier to change when the wheel drops out of the bottom of the chassis (after removing the 3/8 bolt) instead of having to take the gearbox or frame apart.
Right, I keep on forgetting to include all of the files so other people can download the model. I added a pack and go zip to the partner space, you should be able to download and open that.
You’re right that it wouldn’t be a drop-in replacement on an AM14U3. The wheel well has to increase by 1/2" to work, so all of the bolts and churro standoffs would need to be replaced with longer ones. But otherwise I think it’s pretty easy to assemble. It doesn’t really matter to me though, as we’re pretty firmly in the WCD camp these days. But for interest’s sake, here’s a quick sketch. (I moved the whole gearbox towards the frame rail and added an extra plate, but I could have left it where it was.)
It’s just Photoview 360 with the settings turned all the way up. Each one took about an hour and a half on my desktop.
The shadows look really good because I added a second very bright light pointing down at the model. It helps illuminate the model so the renders don’t come out as dark as usual, and as a side benefit it makes really cool shadows.
I’m not sure if it’s just me, but the link to your flipped NEO CAD is not working. GrabCAD is giving me a 404 error. Is your link still current? I’m considering a flipped NEO gearbox fall project and yours looks like great example.
Sorry about that, my GrabCAD account got messed up a few months ago and I had to delete it and create a new one. Unfortunately that means all the links I posted before no longer work. Here’s the new link:
I should note that to my knowledge this design has never actually been built, so take it with a grain of salt. This was also designed specifically to work with the AM14U style chassis, so that provided a rather large constraint on packaging and ratio options. If you’re designing yours for a custom tank drive, there are probably a number of places you can optimize further.