Second Print run always a looser fit - same stl file, looser tolerances

Discussion in 'Technologies and Hardware' started by Old_Git, Jun 5, 2013.

  1. Old_Git
    Old_Git New Member
    OK, I've been dipping my toe into this 3D printing this for a few weeks now and I love it. As a traditional modeller who used to carve everything by hand this is so much easier, added to the fact that I've lost the fine motor skills in my hands and 3D printing is just what I needed. just when I needed it. I'm sold!!!

    However, I've noticed some strange behaviour in the printing process. It goes like this, I upload my STL file and it's accepted, the model is printed and shipped to me and it's perfect. In this instance we're talking about 3 separate pieces for a wheel in 1/6th scale. Firstly there's the wheel, then there's a tyre for the rim and finally an axle bar. The model arrives and everything is perfect. The fit of the tyre on the rim is fantastic, it slides on and holds itself there in a nice tight compression fit. The axle bar goes through the hole and again a nice tight fit but still allows the wheel to spin. Exactly as designed, I couldn't be happier! Everything is perfect and I do not need to change a thing. So I order a second set. The second set arrives (remember it's the same stl file, nothing changed) and this time the fit is rubbish, the tyre will not hold itself to the rim and there's a lot of wobble between the axle bar and hole in the wheel hub. The problem seems to be with all three pieces, if I try any of the 2nd pieces on the first set they do fit better but nowhere near as good as the first set all fits together.

    It's not the first time this has happened either. The very first items I had printed were some links that should have gone well together, when they arrived they were too tight and needed rubbing down with a needle file. Even then they were a tight fit. I figured it's a tolerance issue and resigned myself to the fact that I would need to do this. I order a second batch and when they arrived they were perfect, went together without any need for rubbing down and moved freely together as they ought to. Again, the stil file was not modded in anyway, same file as before.

    I'm thinking that it's different machines set to different tolerance levels, or a belt drive issue on the stepper motors? Has anyone else seen this? Can anyone explain why it happens?
    Last edited: Jun 5, 2013
  2. mkroeker
    mkroeker Well-Known Member
    As far as I know, they have two different printer models for the strong&flexible nylon material - same manufacturer, but
    different size and tolerances. Print orientation may also play a role - perhaps the first set happened to have the greatest
    deviation in a direction where it would not affect the fit ? Neither print orientation nor choice of printer can be influenced
    currently. (Still there is the off chance of a badly misaligned printer inadvertently being used - hard to tell from only two samples).
  3. Youknowwho4eva
    Youknowwho4eva Shapeways Employee Community Team
    There are several things that can effect the size of pieces. Which material did you print in? Check the material pages for the individual tolerances on the design pages.

    Also, if the pieces are all the same material, do you have them in separate files, or all one file?
    Last edited: Jun 6, 2013
  4. Thre are a few factors that will affect the fit of printed parts:
    - The printing machines can "wander" within their tolerance range.
    - Environmental factors during printing - including humidity - can affect the final dimensions.
    - No two machines will ever be identically callibrated.
    - The materials themselves have tolerances in chemical make-up.
    - Murphy's Law. :D

    I recently printed two different sets of threaded parts - both approximately Ø1 inch pitch diameter, and not two prints of the same models - and had interesting results. The prints were done in Alumide. One set was modelled with 0.005 inch clearance between mating surfaces and threaded together with, I would guess, about 0.001 to 0.002 inch clearance. (I had to clean off a fine dusting of excess powder for them to fit together at all!) The other set was modelled with 0.002 inch clearance and only threads together about 30° of the first pitch. I'll be doing a reprint of the latter set shortly and will post the results when I have them.