[OpenSCAD] Joining parts

A. Craig West acraigwest at gmail.com
Fri Jul 19 07:47:23 EDT 2019

The best way to find out how the snap pins work is to print Emmet's gear
cube https://www.thingiverse.com/thing:213946
There are two versions of the hole inset, one which allows rotation and one
which is fixed. They hold very solidly, but can be pulled apart with enough
effort. The smallest I have been able to make useful pins is 4mm long with
a 2.5mm diameter, anything smaller than that is too fragile to use.
The fully customisable pins are at https://www.thingiverse.com/thing:3218332,
and the output options include a test socket you can print to try it... The
code generates an STL for the pin, which is printed flat, and an STL for
the socket, which is difference'd from the surface you want to embed the
pin in.

On Fri, 19 Jul 2019, 07:26 Steven Dick, <kg4ydw at gmail.com> wrote:

> I actually designed a part to answer exactly this question.
> https://www.thingiverse.com/thing:3579313
> If you print male and female parts that are exactly the same size,
> they are unlikely to fit together.  If you taper one, you can press
> them together up to a point at which they get stuck.  Where they get
> stuck depends on the accuracy of your machine and to some extent the
> type of plastic you are using.  How much they gets stuck depends on
> the smoothness of the part and the slipperiness of the plastic.
> The idea is that you print this part which has a taper, then you slide
> the sides together, and measure where they get stuck and calculate a
> tolerance from that.  In retrospect, this part only measures your XY
> tolerance;  the Z tolerance will be different.  I suppose you could
> try to rotate the part and print it, but if I had designed it for
> that, I might not have made it so long and skinny.
> Measuring the tolerance from this part only gives you a starting
> point.  There are other factors that might change the tolerance over
> the life of a printed part.  ABS stretches.  PLA stretches a bit less
> but also rubs and changes shape.  Some of the size of the tolerance
> comes from wobble in the printer, so if there are bumps in your PLA
> part, some of those bumps are smoothed out a bit each time you slide
> them together.   TPU does both but not permanently; I thought I could
> get away with zero tolerance for TPU but found that isn't true, it
> also needs some tolerance to make parts fit.
> For sliding parts, my original Makerbot Cupcake with ABS needed
> 0.25mm.   My Ender 3 with PLA needs a bit more than 0.1mm.
> For a friction fit, using the bumps in the Z axis isn't a bad idea,
> assuming the part doesn't delaminate under pressure and fall apart
> afterwards.
> ABS friction fits nicely.  PLA seems to bind to itself less, so I'd
> not try a XY friction fit there, as others have mentioned, I'd want to
> make a peg that snaps in, although that works well in ABS too.
> I'm not sure how legos do it; I would imagine they have an extremely
> small dovetail taper or something.  Weather or not they use a taper,
> they do use extremely tight tolerances.
> _______________________________________________
> OpenSCAD mailing list
> Discuss at lists.openscad.org
> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openscad.org/pipermail/discuss_lists.openscad.org/attachments/20190719/500a4140/attachment.html>

More information about the Discuss mailing list