Webcam-Holder with BallJoints for Anycubic Vyper and c270-Webcam

Since I wanted to move my camera into different positions depending on what I was printing, this mount was created.

I also wanted to be able to use the camera in other positions without having to make major changes.

After many unsuccessful attempts with "The Perfect 3D-Printed Ball Joint" (https://www.printables.com/de/model/288411-the-perfect-3d-printed-ball-joint/), which were simply too loose due to the design or became loose over time, as PLA yields a lot under tension, I decided to design something similar, which is (hopefully) a bit more stable.

The "Holder-X-Axis" needs support, everything else is printable without it.
I recommend to make a test print of the "Holder-X-Axis" first (a few layers) to check if it fits over the belt tensioner, or if you have to scale it...).

To assemble, just slide it onto the belt tensioner.

The "Sphere_Base" has to be fixed with a M5 screw (up to a length of 20mm). The printed thread fits a printed screw, but experience has taught me that it doesn't really hold (breaks too easily; integrating the ball into the holder is not the best idea either, as I had to learn painfully; that's why this became two parts of its own ;-) ). Therefore I strongly recommend a real screw! However, the thread may have to be recut...
(I have uploaded the screws anyway, but advise against using them!)

For the "V-Slot-Front-Base" another "Sphere_Base" has to be printed and mounted with a M5 countersunk screw.

It fits for the front side of the 40x40 profile as well as in the V-slot.

Finally, there is the tripod adapter. It has a common 1/4″ UNC thread.

The camera is mounted with an M3x20 screw and nut.
The camera cable can be clipped into the clamps of the "BallJoint", adding to the stability.

I have also uploaded the FreeCAD file in case anyone wants to make changes. I am a FreeCAD beginner, so please bear with me... ;-)

Update V2:

After the balljoints still became loose relatively quickly, I have now lengthened the lower part by 1mm to better enclose the ball. (…looks like it remains a work in progress…) :-/

Possible solutions:

After further tests I found two solutions that seem to work quite well:

1. Printing with Cura's "Fuzzy Skin - Outside Only - Thickness 0,1mm" creates the desired friction of the surfaces and is quite stable.

2. Printing with TPU (95A) also results in a rougher surface (here!) and works quite well. Due to the flexibility, the whole thing is of course a bit wobbly, but this does not matter in practice due to the small movement of the axle. (For TPU I have a new STL file without the cut in the sphere which is not necessary for this material).

Translated with DeepL