I know that I’ve seen teams articulate tubes to a desired angle, but they used 2 separate motors. Does anyone know of any way to accomplish the same thing but using only one motor?
Thanks!
Yes, You connect the motor and one set of rollers to one output of a bevel gear differential and the other set of rollers to the other output. By locking the differential case in place, the rollers spin in opposite directions rotating the tube within the claw, by locking the two outputs together, the rollers spin in the same direction to intake or release the tube. Now the real challenge is working out how to lock the outputs to each other and the case to the claw.
An idea that we played with, but never really got working, was to have the top and bottom roller both connected to the same motor, using pulleys, but having the top roller geared somewhat faster. This makes it so you can suck the tube in, then continue to suck in and it will rotate up. Put a backstop so the tube can only rotate so far, and all you have to do is suck in till the tube hits the stop.
11’s roller claw worked this way. Big wheels on top slightly tilted the tubes upward.
You’re probably definitely going to need 2 motors to articulate a tube.
I think it could be done with one. It wouldn’t be the most elegant solution, or the lightest, but I think it could be done.
What I’m thinking is something like a dog shifter, with a “forward” mode, a “reverse” mode with a second gear to reverse the belt, and a “neutral” mode (in between forward and reverse) for slower articulation. Alternatively, you could just have the “neutral” mode.
Now, if you want to count any servos used as motors, then yeah, that is 2 motors.
Well let’s look at this. You want to be able to achieve 4 states:
------R
| <- (Pull Tube In)
-----R
------F
| (Push Tube Out) ->
-----F
------F
| (Rotate tube counter clockwise)
-----R
------R
| (Rotate tube Clockwise)
-----F
With a motor and a standard gearbox you have only two possible states Forward and Reverse. Similarly you have pneumatics with two states In and Out. In order to get 4 states using those devices you need at least two of them (2x2 = 4), many people on this thread have suggested novel ways of utilizing a servo or pneumatic in combination with a motor, however none are simpler than just using two motors.
The ONLY way you could do what you want using just one motor and nothing else would be to increase the number of possible states the motor can have. This could be done by creating a centrifugally shifting version of the gearbox that EricH recommended, for example. Such a gearbox would allow intake and output at high speeds and change its output to allow rotation at low speed. This is probably the WORST solution suggested to date as it is crazy complicated, but it is the only way to achieve what you want with just one motor.
tl;dr Use two motors!
- Alex
It sounds like a much more fun challenge if you limit the mechanism to using only one degree of actuation! I can think of a few approaches, but they all inevitably end up using more hardware than simply hooking up two motors to drive the rollers.
This was a good lesson we took from last year; if you minimize the degrees of freedom of a manipulator you can make a system that is in theory simpler than one that uses more actuators, but the final result typically ends up being more complicated than simply using multiple motors. Still, that won’t stop us from trying… manipulators that utilize minimal actuators have a nice appeal.
The lightest simplest design is to use 2 motors in this case. Adding a linkage or switching system adds complexity and failure points. This is a case of bad engineering practice. Remember just because you can make something doesn’t mean you should. Complex mechanisms like that have been the downfall of more FRC robots than anything else over the years.
If you want to arbitrarily orient the tube, then two motors is the simplest option.
But I would argue that there is not a lot of value in being able to arbitrarily rotate the tube, in and of itself. Typically, there is one orientation you want those tubes to be in when you go to score. So being able to rotate the tube is only useful in that you can bring any tube you pick up or are carrying to that orientation.
If getting the tube you picked up into your desired scoring orientation is your goal, there are several one-motor ways to make it happen. You can design your roller claw such that the geometry forces tubes to be sucked in to the correct orientation.
340’s roller claw this year oriented tubes pretty consistently from what I saw at Chesapeake. They used one motor (I think) and a series of rollers to suck tubes in while also turning them to vertical. Here’s a picture for reference: http://www.chiefdelphi.com/media/photos/36576