I’m still brainstorming, but I feel that this game puts teams in a corner when making a manipulator. My team is probably going to do an arm and I’ve been doing research on how to make it move. Does anyone have any recommendations or experience for making a ball and socket joint?
Do you need the range of motion of a ball and socket? That’s like your shoulder - it can move in multiple directions. Or do you just need rotation in a single direct, around a single axis?
Short answer: don’t make a ball and socket joint.
Long answer:
If you are talking about a driven ball and socket, it’s possible (essentially wheels driving the ball) but it almost never done in FRC, industrial automation or anything else I’ve seen. If it looks like something has a driven ball and socket it is probably 2-3 simple rotations close together.
If you are just looking to build a ball and socket, you are probably better off buying something off McMaster-Carr or using simple rotating joints. If you can avoid joints with that many degrees of freedom I would highly recommend doing so.
The reason, I’m trying to accomplish this is because it can move in multiple directions.
ok thanks, do u have a link for McMaster- Carr so I can look into it?
McMaster is your friend, they have everything mechanical you might want that you can’t get through a First vendor, faster than Amazon shipping, and a wealth of knowledge.
I personally think there are simpler ways to allign with the nodes, but if you really need 2 axes of motion, it is far easier to split it up into two serial 1-axis joints, like a gimbal.
This might be a good place to start for ball and socket type joints, but like I said if you can think of a way to use simple rotating joints, do that.
Consider the complexity here. The cost of that range of motion in your shoulder is 8 different, distinct muscles working together. Contrast that with single axis of rotation, like your elbow (mostly, ignoring the small stabilization muscles) - two ,muscles working together to move it.
Ball and socket by itself isn’t difficult. Whenever we use a gas spring, it usually has ball and sockets on either end to make connecting it easier. The problem comes with controlling that motion along multiple axis - that is a very difficult problem.
Unless you’re trying to turn cones upright, you probably don’t need a spherical joint.
If you really need the range of motion offered by a spherical joint, I would reccomend using a sort of wrist-elbow-wrist setup of simple joints. All the joint axes go through a single point, so you have basically the same range of motion. The upside is that it’s much easier to actually control the joint–a ball-and-socket would need at least 3 actuators, and it’s not obvious how those would connect and interact.
Further reading: google “spherical wrist”
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.