Ball and Socket

by Coder-Tronics Jan 28, 2017
Download All Files

Thing Apps Enabled

Please Login to Comment

This is cool and almost exactly what I need. My Spigen mount fell apart so the thread and collar are what I need, along with a Richter mount plate that I found elsewhere. The ball on the mount is 17mm and, so far, for the life of me I can't get the remix, in Tinkercad, or scaling, in S3D, just right. The end result is either too loose or too tight to screw the tightening ring on. Any tips on how to tweak to get it to work for a 17mm ball would be great. I think 17mm is a Garmin standard, but that hasn't helped me one bit ;o)


I would say I could make a 17mm version but I just don't have the time at the moment.

I would approach it as follows:

1) Determine the X,Y and Z tolerances of you printer by printing a 10Z20Y20X solid cube (it can be different size if you want but this is what I normally use, Ideally needs to be solid as your using more material so any errors due to under or over extrusion may come into play).

2) 17mm is 12.5% larger than the current 15mm ball and socket ± your printers deviation on the X, Y and Z axis.

I have used this method before and worked for me on a more primitive part, the ball and socket is a fairly complex part and the tolerances are quite small 0.3mm between the ball and socket, so may take few tries.

No worries, I wasn't looking to put you ut and get you to rework it. Just needed a pointer and....
Doh, simple maths kills me again - so close to the problem i couldn't seethe solution for the trees. Will give the sensible scaling calc a try ;o)


No problem, I didn't think your were, I wish I had the skills to make a parametric version of this as that would be really cool.

Hope the scaling works out and feel free to share the results.

Will do, thanks again!

Very cool! Might I suggest a ball with a mounting flange? just a suggestion! thanks for sharing!

This is awesome! Any chance you could share the fusion 360 files or assist in designing an adapter that would add a socket as the top piece to https://www.thingiverse.com/thing:2491366?

CR 10 BED camera mount , c270,action cam, gopro, picam

hey coder
as achichon asks can you plzzz add the 3d files for fusion 360 ?
so we can learn and remix ?

this is great, any chance you could add the 3D files so we can remix it as we require (working with stl's on fusion 360 is a bit of a nightmare)? if not any chance you could add a ball joint with a gopro adaptor?

Exactly what I was looking for! This will be perfect for many many projects! Well done!

Cool hope it works well for you and please share your final prints and application.

I think your design is great, i am going to use it for a project i am making (a robot) only thing i found out is that the ball joint is a little bit to small for the hole and its difficult to get the ball locked tight... i read its 15 mm and the hole 15,3 mm so i think i am gonna make the ball a little bit wider....
i let you know when i have remix it for my project , first i need to learn fusion 360 :-P

Hi Martin,

Thanks and remix away, another option you should be able to reduce the size of the ball in the slicer e.g. by 1% say for a test and see how that comes out.


Could you drill through the ends and add rods to make it a universal joint of sorts?

Hi Silveradored,

Not entirely sure what you mean, is this for strengthening or to enables wires to be passed through?

It's possible the design could be made to print with hole, therefore removing the need to drill or making it easier in any case.

No, it would be to turn them into an axle of sorts

Arh ok, then probably not. You could take some inspiration from these designs and make something that better suits your motor or rod needs:



Universal Joint with minibearings
by jbeale
CNC / 3D printer coupling / Universal Joint

From what I can see in the clip the locking system works by forcing the four sides of the socket slightly inward when the threaded ring upper surface meets the cone shaped portion of the socket at the top. I am thinking of making something somewhat similar, without the slotted sockets, with shorter threads on the base and having the threaded ring actually grip on the top of the ball, perhaps with an o-ring in it or in the base to work as a friction point. The main issue I can see with my approach is that the threaded ring would not slip past the ball, for example for a double ended rod, though for my specific application the ball would be a cap over a bolt head. For some uses it could just be printed in place, after all 3D printers are made for impossible geometry.

Yep thats how it works, pretty simple but took me a few prototypes to tweak it so it worked every time.
I think for your application you have identified the main issue and 2 possible solutions:

1) Print the ball and locking ring as one piece, probably need support to raise the locking ring above the ball level.
2) Print the parts separately, but you need some mechanism to join a ball-ball joint. One example could be multiple pieces of 1.75mm filament pushed into the joint and heated into place.

I did a similar design in OpenSCAD a few months back but gave up on it. My socket prints fine but I could never get the ball to print round enough! I like your solution. It solves the issue I was having. Just print 60-75% of the ball sideways supported on the bed. Excellent..

What fraction of the ball do you print?

I was working on a way to articulate a solar panel. This is the design I was working on: http://www.thingiverse.com/thing:2051608

Portable solar panel tripod mount
by rasarmg


You always have to compromise with 3D printers, so I try and incorporate that into my designs as well as avoiding support if possible. The ball with the flat top is obviously the best option to give you the most complete ball shape, with very little surface area loss as far as ball to socket, this is actually the same system Manfrotto use. However, the neck of the ball is a weak point due to the way the layers are laid down, so the horizontal is my preferred choice.

I didn't actually calculate how much of the ball is cut away for the flat surface, just eyeballed it :-) all the dimensions are there though so easy to replicate and make different variations or tweak the ball size to meet your printers tolerances.

Thats quite a detailed construction for the solar panel mount, do you use that as portable power? I experiment with solar power myself have a 4 part tutorial on my site about a solar Maximum Power Point Tracker (MPPT) http://coder-tronics.com/c2000-solar-mppt-tutorial-pt1/

That is a very nice write-up.

Were you able to convince yourself that that power you gained with optimized positioning sufficiently exceeds that consumed by the tracking and/or positioning system? I think your approach is a clear winner for larger panels of permanent installations that track the sun's changing trajectory over the seasons and produce orders of magnitude more power than the computation and control circuitry.

As for a small temporary power setup, I would love to see a purely mechanical wind-up approach that repositions a small panel over a 6-8 hour period. Just wind it up and align with the expected trajectory for the day... Effectively dealing with nonlinearities would be a challenge.

Your project execution and presentation are awesome!

I didn't use any active mechanical tracking system to track the suns movement across the horizon, the MPPT is purely used to track the maximum power point of the photovoltaic cells power curve.

Active tracking of the sun is also a good way to increase performance, but its a trade off compared to fix systems as they have greater initial costs and require more maintenance over their life cycle. They are still commercially viable on larger systems though and really depends on the geographical location, the UK for example it is far more cost effective to have fixed systems pointing south at ~35 deg.

A wind up system would definitely be a cheap robust solution for a small portable system.

Very nice! Did you print in PLA or ABS?

Hi, thanks just PLA not used ABS yet due to the fumes.