Let's Make Robots!

How do you make a two axis pivot?

How would you make a pivot that would allow motion in an arc around a tube of a larger diameter by a tube of a smaller diameter? The pivot must be able to be moved along the larger tube for adjustment.

Some starting ideas:

  1. Nested hollow spheres. The inner sphere is split and drilled to fit around the larger conduit. It is held together by recessed cap screws. The outer sphere is drilled and hollow so that it can pivot around the large tube. Perpendicular to the hole there are two pivot axles. Mounted on the axles are two C channels with with bearings which form a box around both spheres. The second smaller tube is attached to the non-bearing sides of the channel-box. -[o]-

  2. Replace the spheres with a ring captured by two clamps which are tightly around the larger tube and enough room between the clamps to permit the ring to pivot around the large tube. The ring has two pivot axles. The remainder of the pivot is similar to idea #1 above.


pivot_side_1_zoom.JPG17.91 KB
pivot_2.JPG16.23 KB
pivot_side1.JPG18.12 KB
pivot_top1.JPG12.86 KB

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Metal ball, drill press, vice, and a drill bit?

Thank you for the comment. When I read your suggestion, "Could you instead offset the smaller axis to one side of the larger axis? If so, you could put a movable ball joint on the large axis and then connect the small axis to a pair of rings that could capture the ball." I imagined a ball on a shaft connected to the side of the large axis with a socket holding the small axis. The small axis would have to be bent around the large so that it ended up by being centered on the large axis. This would work. The difficulty is finding or making the parts.

However, this is not what you were suggesting. Your suggestion is similar to my first idea. You replaced the larger slotted ball with the two rings. This too would work.

However, I don't see how the rings are held together or to the small axis. Did you have something in mind to accomplish these tasks?

could be accomplished by drilling 4 equadistant holes and running bolts through then adding a nut to the other side. As for the small axis, JB Weld comes to mind or epoxy of some nature.

Good idea.  And the small tube could be attached with two rods. Each rod has two flats to fit between the rings and with a hole for one of the bolts.  The tube could slip over the rod and be bolted to the rod as well as the rod bolted to the rings. Very nice idea.

Now, how do I make the ball for the big tube?

required to be in the same plane as the large axis? Could you instead offset the smaller axis to one side of the larger axis? If so, you could put a movable ball joint on the large axis and then connect the small axis to a pair of rings that could capture the ball.dual axis ideo

replacing the rings with two sheets of delrin.

How about replacing the ball with a ring?

imagine how you would do that. If you can see it and make it work. More power to you.

"You have a large axis, with a smaller perpendicular axis that is bisected by the larger." YES
"The smaller axis can pivot 360 degrees freely around the larger axis." yes but 120 degrees
"The small axis can be fixed to a point along the length of the large axis." YES
"The small axis can also be adjusted and slid to a different position along the length of the large axis." YES
"The large axis is free to spin without affecting the small axis." In the real world the large axis is not going to move but in this discussion it could spin 120 degrees
"If the large axis moves (the two end points are moved relative to each other), it will cause the small axis to shift." YES (I think)
"The small axis cannot spin freely, because it is fixed to either side of the joint. Spinning the small access will cause the large axis to rotate." YES
However, the drawing does not show the tilting of the small axis along the length of the large one.
Here is something that moves like what I need. A joystick. It can be pushed forward and pulled back. It can be tilted from side to side. Now, imagine that it is mounted on a horizontal pole and the joystick handle has another handle on the opposite side of the pole that moves in the other direction of the one on top. Tilt the stick along the axis of the pole and the bottom stick tilts towards you. Tilt the stick towards you and the bottom tilts away. Tilt it to the right and the bottom tilts left and so forth.
Now, imagine that this entire assembly can be slid along the pole and locked in a new position.
There are now some more sketches attached to the first post.

[How do you put the neat drawings in your comments? (I don't see the option to insert images.)]

There are three motions of the small tube in relation to the large one.

1. tilting with the pivot point centered on the large tube \|/ along the axis of the large tube
2. tilting \o/ across the axis of the large tube by rotating around the large tube
3. <-|-> sliding along the axis of the large tube (but this is just an adjustment, normally it is locked in one position on the large tube)