View Full Version : Passive Joints

11-09-2013, 01:09 AM

I've been working on the design for my quad's legs, specifically the feet, and I had a few ideas.
Now my quad is going to have 3dof ax-12 legs, so nothing radical there, and I was deciding between pointed feet (like the TRC quad) and flat feet, to provide a more stable platform for walking on. From prior experience, the main problem with flat feet is that they do not touch the ground at a constant angle with 3dof, and rotate (forcing them to slide along the ground). This could be solved by a fourth degree of freedom (the constant angle part anyways), however the about 500g of extra servos just isn't viable.

So basically, my idea is to have a passive joint at the end of the leg, which allows free rotation along two axis. One perpendicular to the ground, and the other perpendicular to the plane of the leg.

Do you think this is a viable option for adding stability?
Any ideas for actually implementing something like this? I poked around McMaster Carr a bit but nothing really looked suited to this application.

11-09-2013, 04:52 AM
I seem to remember a previous thread about something similar, but now I'm thinking it was just rubber cones with threaded inserts.

Maybe something like these (swiveling versions):

Could always make your own out of a ball stud, a piece of plastic with a hemispherical recess, and a couple small springs or elastic bands to attach the plate to the stud.

11-09-2013, 07:52 AM
I considered it for a while (and the 3 DOF issue you noted is a reason that Emissary has 5 DOF legs). An issue that I kept coming across was the range of motion that spherical joints provided. I had issues finding such a joint that could move +/- 45 deg, or even +/- 30 deg.

The other main option was a gimbal of some sort, but that proved to be rather bulky. I'd be very interested if you found something though!

11-09-2013, 06:56 PM
After considering these things, I ended up going with pointed feet rather than flat/articulating feet.
Because you rotate, you want ball joints, not universal joints (beacuse of hose those joints are generally constructed.)
And, as stated above, finding ball joints with large degrees of articulation is nigh impossible.