Designing a Ball Joint for 135 Degree Flex in 1/4" Brass Tubing

[DaveC426913]

I am building a device out of 1/4" brass tubing. I need to join lengths in such a way as to allow the tubes to flex up to 135 degrees from straight.

I haven't been able to figure out how to get that much rotation from a ball joint without the joint falling apart.See attached diagram below for what I'm trying to do.

The posts fold back on themselves until they are a mere 45 degrees from W.

Note that the joint needs to be entirely symmetrical - i.e. W is not special, it is identical to X Y and Z. (I could take the joint by itself, toss it in the air and catch it and not know the difference in orientation.)

The one solution that might do it would be to make them all universal joints, like in the drive shaft of a car, like this:http://upload.wikimedia.org/wikipedia/commons/b/b5/Universal_joint.gif
but I don't have the room for 4 universal joints in that little space.

Looking for alternate ideas. Would accept off-the shelf parts or build by hand.

Thoughts?

 

[Danger]

Hey there, pal.
You didn't mention what sort of stress, strain, and assorted insults this thing might be subjected to.
Would connecting the tubes with a coil spring work? I've used that in place of U-joints several times in my lack-of-career, and it works pretty well.

 

[DaveC426913]

Hey there, pal.
You didn't mention what sort of stress, strain, and assorted insults this thing might be subjected to.
Would connecting the tubes with a coil spring work? I've used that in place of U-joints several times in my lack-of-career, and it works pretty well.

Ideally, very little stress.

I just stumbled on to "bendy" type solutions myself. I would work, yes. The only downside is that springs and other bendy things have a certain amount of "play" associated with them. i.e. the joint, if under even moderate torque or compression, could collapse on itself.

But yeah. Thanks I'm, going to mull over designs.

 

[Danger]

springs and other bendy things have a certain amount of "play" associated with them. i.e. the joint, if under even moderate torque or compression, could collapse on itself.

Oh! Hey! What about braided stainless tubing, as we use in automotive fluid lines?

 

[pantaz]

Look around some R/C (radio control) hobby shops. I don't have a specific recommendation, but I've seen some pretty clever component designs.

 

[DaveC426913]

Well, first attempt.

The core is a tetrahedral carbon atom left over from one of those tinker-toy-chemistry sets*. The joints are heat shrink tubing.

It works great. The problem with it is that it's too big. The whole joint is about 3/4 of an inch in radius. The sets a lower limit on the scale of what I'm building.

*or a caltrop if your geek flag tilts toward nasty antipersonnel weaponry

 

[Danger]

Hmmm...

 

[256bits]

The only thing I can come up with is that eye-glasses and sunglasses have tiny joints which perhaps you could make use of.

I also thought of an umbrella configuration - at a certain partial opening 3 of the ribs would be 90 degrees from each other and 45 degress from the stem ( W ) , but W would be unique.

 

[DaveC426913]

The only thing I can come up with is that eye-glasses and sunglasses have tiny joints which perhaps you could make use of.

I also thought of an umbrella configuration - at a certain partial opening 3 of the ribs would be 90 degrees from each other and 45 degress from the stem ( W ) , but W would be unique.

Yes. Sorry it is critical that the joint is symmetrical.

 

[DaveC426913]

Well, attempt #2 is a fail. I thought of the brilliant idea of making my own polyhedral core by soldering 4 very short lengths of my brass tubing to a ball bearing. It would have been perfect.

Alas, whatever these ball bearings are made of, it isn't something that solder sticks to.

 

[Danger]

Okay, Dave... let's take another shot at this. How about...
Drill some holes through the ball bearings and scuff up the surface with a file. That should give the solder something to cling to.

 

[Ranger Mike]

http://www.bocabearings.com/bearing-types/spherical-ball-bushings-features

could be solution

 

[DaveC426913]

Okay, Dave... let's take another shot at this. How about...
Drill some holes through the ball bearings and scuff up the surface with a file. That should give the solder something to cling to.

Yeah. That'll be my next thing. But really, the solder didn't even pretend to stick. (Of course, it's possible that I didn't get all the grease off it. )

Another thing I'm going to try is some BBs I have left over from a slingshot. Not sure what they're made of, but we'll see.

 

[DaveC426913]

http://www.bocabearings.com/bearing-types/spherical-ball-bushings-features

could be solution

Cool solution but
1] nothing small enough
2] at 10 bucks a pop, my device will cost 240 bucks just for joints

 

[AlephZero]

Would a countersunk ring magnet like these
http://www.aliexpress.com/product-fm/550870918-10-pack-super-Powerful-n35-NdFeB-Neodymium-Disc-Countersunk-Ring-Magnets-D20-5mm-hole-5mm-wholesalers.html
with the countersink forming the "socket" part of a ball-joint work? The ball would have to be big relative to the magnet, to give the 135 degree movement.

Those magnets are too big but you might be able to find smaller ones. There is probably some new age cult that uses magnets as beads...

 

[DaveC426913]

Would a countersunk ring magnet like these
http://www.aliexpress.com/product-fm/550870918-10-pack-super-Powerful-n35-NdFeB-Neodymium-Disc-Countersunk-Ring-Magnets-D20-5mm-hole-5mm-wholesalers.html
with the countersink forming the "socket" part of a ball-joint work? The ball would have to be big relative to the magnet, to give the 135 degree movement.

Those magnets are too big but you might be able to find smaller ones. There is probably some new age cult that uses magnets as beads...

Interesting. A creative solution. But I can see that having a bunch of magnets in this thing would be disastrous.

 

[DaveC426913]

Attempt #3 is a success!

The solder is sticking to the BBs perfectly. I can now make my own arms out of brass tube. And I've made an improvement. I can solder a very thin flange at the end of the arm (a 1mm slice of tube one size larger in diameter than the arms) and now my heat shrink tubing will be locked on instead of holding by friction.

The fact that I can lock it on also means that I don't need any more heat shrink coverage than a mm, so the arms can be arbitrarily short. It is really critical that these joints be as physically small as possible. The size of the joints sets a lower limt on the size of the structure I'm building (which is otherwise arbitrarily scalable but it's already larger than I want it). So any shortening I can do on the arms is bonus.

(The next big problem is how to solder four pieces on so close together without each of them falling apart as I do the next one. I'm not fast with a soldering iron...)