pic: NanoShifter



Just this last render and I’ll quit. After doing the AM shifter, I wondered if it could be packaged in a Toughbox Nano size housing. Here it is - it’s actually not too bad. I’ve shown it with a 6" Mecanum wheel and adjusted the ratios accordingly. The speeds should be about 8 and 12 fps. The high speed could be made higher but it would be difficult to make the low speed much lower.

This render is amazing

AMAZING! :smiley: That is just (insert word to majorly increase awesomeness)ing awesome!

How much does it weigh?
How stable is it?
When and where will I be able to get one?!

If you could make the final ratio faster I imagine it’d be a cool / compact way to do a 4" wheel transmission. Gotta get rid of that mecanum wheel first though.

If you went with 4" wheels, the standard AM shifter gears would work. I’m not sure it would make sense to do it though, if you did a 6 wheel drive you’d need 6 of these. May only make sense with Mecanums.

The way I see it, depending on the weight of these things, and the do look light, one could use 6 of them, one per wheel, or at least 4 of them, 2 on each side, making a very powerful drivetrain.

Excellent gearboxes for mecanums, however I see a lot more use in these than mecanum wheels.

If you used six, you would have to have six motors in your drivetrain. Also, if you used 2 on each side for a non-omidirectional drivetrain, you would save weight by simply using one transmission. The power would be the same since the motors supplying the power are the same as they would be with two gearboxes.

I agree that there are more uses for them than mecanum, they could be useful in swerve and omni-drives. But they could also be useful in non-drivetrain applications like shooters.

If 2 of those weigh less than one of the standard transmissions, it’ll be less weight no matter what. Plus, you gotta admit: Just by seeing it you wanna use it on a robot. :slight_smile:

A standard Nano weighs 4.8 lbs without motors. I can’t imagine shifting ones weigh too much more.

A SuperShifter weighs 4 lbs. without motors and pneumatics.
AM Shifter weighs 3.4 lbs. without motors and Pneumatics.

I would just stick to the standard SuperShifters or AM Shifter.

Fine, don’t use the really cool greatly rendered awesome-sauce gearboxes. :expressionless:

A ToughBox Nano weighs ~1.7 lbs…?

Were you thinking of a NanoTube?

If designed right the additional weight going from a single to two motor gearbox is a little more plate, and an additional pinion gear; far less than a second set of gears.

Haha. Good catch. I was indeed thinking of the Nano Tube

First, amazing render. I don’t think anyone here will ever get tired of these types of ideas, so don’t feel the need to ‘quit’. They’re creative and are a spring board for a final product.

Any way the encoder can be re-incorporated into the gearbox?

It also appears that the inner races of the bearings on the last stage touch the inner races of the bearings in the housing. Would this cause a problem due to the difference in speed of each shifting gear?

You have a good eye for detail. If this was real, I’d put a thin washer between the inner races like the bearings on the other side of the same shaft have.

The easiest place to put the encoder would be on the cluster shaft. This could cause a few problems - getting it to clear the CIM might be an issue, the speed might be too high for the encoder and you need to know for sure which way you were shifted.

That brings up another challenge - if 3 of the positions shifted and the 4th didn’t you’d have a mismatch. If the shift position was detectable you could compensate the CIM speeds and get a “limp home” mode.

If the encoder didn’t work on the clutster shaft, I’d probably try to change the shift cylinder setup so I had access to the end of the shaft (not sure what that would look like.)

As for an encoder:
-It could go on the other end of the wheel shaft
-You could use a code wheel on the wheel itself
-You could use a gear tooth sensor on one of the gears - This does not know post-shift speed

As for not all shifting:
-How would this issue be any different than that of using (2) Dewalt transmissions per side, except that pneumatic shifters are faster than servos?
-I think, using the highly reliable Andy Baker shifting transmission design, and properly protecting the pneumatic system from damage, you should be OK if you don’t explicitly detect and handle the issue of one transmission not shifting.

Another issue I may have missed: How do you get everything into the box? It looks like it is impossible to assemble, specifically the shift shaft.

That’s the question I’ve been waiting for. It’s bit like building a ship in a bottle. I think I can be done but I haven’t thought through every step. This would be a good excersize for someone - explain the exact sequence of putting the shift shaft together.

please to give us the CAD file?

Yes! Please to give to everyone! :smiley:

Step 1: Press 3/8" bearing

Step 2: Insert small shifter gear into tube, align and insert shift shaft with dog assembled.

and

Brick wall…

even if you flipped the last bearing there is no way to get the other gear on the shaft. It may have some way out there solution but I don’t see it. The hex boss on the shifter shaft creates a mechanical limit.

May I suggest putting a hole large enough for the gear (and entire shifter assembly) through and use 1/8" plate to patch the hole. To determine alignment use pins and threaded inserts for fastening in such a compact space. You may have to do this to insert all shafts and gears. Using the tube to determine spacing and fit, Swiss cheesing the outer wall and making a single face plate to close off the assembly.