$50 Experiment - ball & socket

Discussion in 'Technologies and Hardware' started by Whystler, Nov 21, 2008.

  1. Whystler
    Whystler New Member
    I printed out a ball and socket in different materials, and with different clearance distances. Here are my results.


    *Strong/White/Flexible - Test failed. There is no movement whatsoever. The ball is completed fused within the socket.

    *White Detail - Test passed. There is full movement. The ball is completely free of the socket and loose. The connection is so loose that it will not hold it's position for posing (ie. posable figures). This material version was actually sent to me in error, but I'm glad to have it.


    *Strong/White Flexible - Test failed. Ball and socket fused as above.

    *White Detail - Test without data. I had ordered this piece, but for some reason the above version in this material (which was not ordered) was sent instead. So I have no data on this.


    *Strong/White Flexible - Test failed. Ball and socket fused as above.

    *Transparent Detail - Test failed. No movement as ball and socket are also fused. The interesting thing to note on this model is the fact that when peering through the transparent material, there is no indication of the ball. It is as if the ball is part of the socket material. You can see right through.


    *Strong/White/Flexible - Test failed. Ball and socket fused as above.

    *Transparent Detail - Test failed. Ball and socket fused as above.

    *White Detail - Test failed. Ball and socket fused.

    *Black Detail - Test failed. Ball and socket fused.


    Last edited: Nov 21, 2008
  2. daddymack
    daddymack New Member
    nice one whystler, got any pics?
  3. Whystler
    Whystler New Member
    You bet :) I was just in the process of adding it!

  4. Fingers
    Fingers New Member
    Funny, I was just designing something that I'm thinking would benefit from ball & socket joints. Did you print all of these with the ball inserted into the socket? Have you considered printing the parts separated and assembling the joint post-print? This should allow a tighter fit without fusing, but it'll be tricky to design the socket so that it'll flex a bit to accept the ball.
  5. Whystler
    Whystler New Member
    Hi there fingers :)

    These pieces in particular are designed with the ball printed within the socket.

  6. bvicarious
    bvicarious New Member
    Thanks for the info on this whystler! I'm working on something that COULD benefit from poseable joints, not sure if I want to spend the time/money experimenting with clearances and stuff though.

    I guess you could also create it in two parts that you glue/screw together? Although whats the point of 3d printing if you can't print things already assembled :confused

  7. pete
    pete Shapeways Employee CEO & Co-Founder
    Hi Whystler,

    nice experiment. The results partly match our data, which is that anything closer than .1mm fuses and above should work.

    Therefor I am surprised the 0.5mm test with WS&F failed.
    Could you point me to the model used? Are you sure it is fused? It can be residual powder that is blocking movement.

    I have a model here, which can be seen in our Christmas page here (the tree) that has 0.1-0.2 clearance and can move nicely.

    Let's figure this one out together if you want!
    Again thanks for all your efforts.

    kind regards,
  8. Whystler
    Whystler New Member
    Hi there Peter,

    Yes, the ball and socket of the model are completely fused. There is no gap whatsoever between the two. I even tried to poke a needle, and there is no space. The ball is attached and part of the socket.

    Here is one of the products in question that I ordered that was fused with Strong White and Flexible:

    http://www.shapeways.com/model/8416/ball_joint___large_hollo w.html

    Note that this same model was printed with White detail and was not fused.

  9. Nice experiment and design. :)

    There is only one thing missing in this trail. Build direction is an important issue when using layered manufacturing techniques. Especially with the white, strong & flexible material.
    There is a clear difference between up and down facing surfaces due to the "z-bonus" and the layerthickness of 0,1 - 0,15 mm. That is also why holes and axels build in vertical direction might be a bit oval shaped. When build in horizontal direction they are perfectly circular.

    This clearence. Is it 0,5 mm onto the radius or diameter?


  10. pete
    pete Shapeways Employee CEO & Co-Founder
    Hi Whystler,

    have a look at this model from Bart: Tree

    The axels are 0.15mm clear from the sockets and the thing really moves nicely. The difference between your design and the axels/sockets in Bart's design is that the support material could be removed much easier from Tree.

    What software did you use for your joined model?
    Perhaps we could do some experiments together until we make it work?

  11. Whystler
    Whystler New Member
    That sounds great Pete :)

    The clearance between the ball and socket was .5mm. So, the ball's radius is .5 mm less that the socket's radius. to be clear, I will also say that the ball's diameter is 1mm less than the socket's diameter.

    The item was created in 3dStudio Max, and exported as an .obj file.

    I then imported it into Accutrans, where I:

    -checked to make sure it was "watertight"
    -clicked on "extrude 2d surface" (as I have with all of my models, which is suggested in your tutorial)
    -exported it as a .dae file

    And then I uploaded it into Shapeways.

    Important to note: The same model was printed in White Detail, and this works perfectly. There is proper clearance of about .5mm on the White Detail model, as shown in the picture.

    I can send you the max file, the .obj file, and the .dae files if you like.

  12. AbcAbcwebd
    AbcAbcwebd New Member
    Thanks, that its very interesting.
  13. pete
    pete Shapeways Employee CEO & Co-Founder
    check out Bart's forum post here
    and our tutorial here

  14. Whystler
    Whystler New Member

    Hi pete,

    So, is the solution to provide extra air gaps for the powder to be removed from my ball and socket?

    I have a theory that perhaps the support material in the .5 mm gap somehow it got fused. Is that possible if it were trapped there? Would it have been subjected to suficient heat?

  15. pete
    pete Shapeways Employee CEO & Co-Founder
    Hi Whystler,

    a few days ago I had an interesting discussion with someone who made a nice design with Shapeways (unfortunately not public) and he indeed pointed out that adding air-gaps makes it possible to go to tighter fits. Indeed the Star model in our Christmas page made by me has at some points only 0.15mm clearance, but with airgaps.

    Perhaps that is the solution.

    I will try and make a tutorial or blogpost on the topic.

    Last edited: Dec 3, 2008
  16. svenpb
    svenpb New Member
    I'm not sure if you're willing to do that, but what happens if you cut it in half? Still fused then?
    Last edited: Dec 7, 2008
  17. Tercero
    Tercero New Member
    Does anyone know if the White Strong and Flexible material has enough flex to allow for ball/socket joints not modeled together to be popped together? Also, any idea on clearances that should be left if this is done?
  18. Drawn_Steel_Hero
    Drawn_Steel_Hero New Member
    It absolutely does; that's what I'm using for my transforming figures. I don't know how they work with Detail material yet.

    For WSF prints, I've been using 4mm diameter balls with 3mm diameter shafts, and I've found that with no clearance between the ball and socket (4mm ball, 4mm socket), it grips nicely while still allowing you to move the pieces. However, I have realised that's it's helpful to allow a little clearance around the shaft for ease of movement (I'm using 3.2mm diameter channels for the 3mm shafts).

    Here's a demo of the joints I've been using. The socket's completely open on one side (i.e. 4mm, same as the rest of the socket) to let the ball into the socket. There are two little nubs either side that you have to push the ball past, but once through, they hold the ball in place.
    (Click thumbnails for larger pics)

    Same image again with dimensions.

    Hope that helps.

  19. Youknowwho4eva
    Youknowwho4eva Shapeways Employee Community Team
    I also did this in stainless. I had a 3 pronged socket, and the ball (with some force) would snap in. It could also be pulled out with some force but only after hours of playing with the little thing. It was a very fun twittle toy. *what did I do with it?*
  20. polyoptics
    polyoptics New Member
    Heyo, I would LOVE to see this stainless ball joint. If you can't find the file anymore, could you give a more detailed description (drawing?) of what it was... How was the final product?
    I was under the assumption that stainless was fairly brittle and
    would not flex or give enough for 'snap together' designs. Is this false?

    Last edited: Nov 1, 2010