Any tips for designing two pieces that fit together?

[Adam Haile]

I'm trying to design a small board case the that is recessed on one side and the other side fits into it. But found that when I made the inner width of one 24.9mm and the outer width of the other 24.9 mm, they do not actually fit together. When printed off, the outter width one was correct at 24.9mm but the recessed side was 24.7mm.

I'm assuming this is something to do with the minimum X/Y width being 0.4 (I'm using a Replicator 2) but is there some general rule I can follow for making sure that the printed parts will actually fit together in the end?

[DHeadrick]

Even when not 3D printing, you rarely make too pieces that are meant to fit together the same dimension (called a zero-zero fit).  This is only done if they are meant to be press-fit together and even then the overlap of the parts depend on the flexibility of the materials used and the length of the perimeter you wish to interfere.

Do you want these parts to never come apart?  Or do you want them to just slide together nicely.  I have found that 0.01" (0.25mm) gap is good if you want 3D printed parts to slide inside each other.

[Adam Haile]

Makes sense... but what about the fact that both parts didn't actually come out the same size? Is there any way to account for that. Hard to know what size to make it if you are unsure what size it will actually be in the end.

[DHeadrick]

It will depend on the size of the feature and the mass of plastic near it as to what shrinkage might occur.  I usually make the parts 0.01" different in size.

However based on your previous post, take the measurement you got and just make the outer part larger by a little more than the differnce and try again.

[Eighty]

Tolerances are important, even on 3d printers.  I usually use a 0.2 mm gap for tight fits.

 

As far as the part not being the size you drew:  Sometimes you have to fiddle with your design to end-game the slicer or printer (or both).  Have you opened the gcode to see what XY dimensions were sliced?

 

I've noticed that a 20x20 calibration cube is ALWAYS smaller in the Y dimension, albeit by only a few hundredths of a millimeter.  I've been meaning to test larger pieces to see if the difference is absolute, or if it scales with size.  There's not much you can do about the first, but the second can be compensated for (by adjusting the stepspermm).

[David Celento]

That's very interesting, Eighty. Y is smaller regardless of the slicer used?

If you test this further, keep us posted!

[Adam Haile]

I can confirm this. I made a 1cm cube to calibrate and it's 10.10mm in the x and 10.02mm in the y. Weird.

On Mon, Apr 29, 2013 at 9:31 AM, David Celento <dcel...@gmail.com> wrote:

That's very interesting, Eighty. Y is smaller regardless of the slicer used?

If you test this further, keep us posted!

--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/m15CuQyxqbQ/unsubscribe?hl=en-US.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Joseph Chiu]

How fast are you printing this at?

Depending on the particular machine, you could get some overshoot that could lead to slight "outward" growth. In the case of the recess, that outward growth could narrow the size of the recess.

Filleting your corners might help with the print being closer to the modeled size, and also help with the fitting of the pieces.

If you don't need the pieces to be "airtight",you might want to model your design so that you have clearance between the pieces, with small fingers / tabs used to lock the pieces to each other. (If you open most electronic devices, you will see that is a Common technique.) I have done this with a lid that had about ~ 0.5mm of clearance to the box and then had little fitting bumps that squeezed against the corners of the box.

I'm trying to design a small board case the that is recessed on one side and the other side fits into it. But found that when I made the inner width of one 24.9mm and the outer width of the other 24.9 mm, they do not actually fit together. When printed off, the outter width one was correct at 24.9mm but the recessed side was 24.7mm.

I'm assuming this is something to do with the minimum X/Y width being 0.4 (I'm using a Replicator 2) but is there some general rule I can follow for making sure that the printed parts will actually fit together in the end?

--
You received this message because you are subscribed to the Google Groups "MakerBot Operators" group.
To unsubscribe from this group and stop receiving emails from it, send an email to makerbot+u...@googlegroups.com.

[Adam Haile]

Printing it at the default speed in MakerWare... 100mm/s I believe. 

What do you mean by filleting the corners?

--
You received this message because you are subscribed to a topic in the Google Groups "MakerBot Operators" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/makerbot/m15CuQyxqbQ/unsubscribe?hl=en-US.
To unsubscribe from this group and all its topics, send an email to makerbot+u...@googlegroups.com.

[Joseph Chiu]

Sorry for the chicken scratch, but this drawing should explain...

(Having fun with my new phone- Galaxy Note II ! )

[DHeadrick]

A fillet is a radius.  What you drew is a chamfer.

[DHeadrick]

Another thing you could do if you are unsure if the parts will fit, is put a draft on the surfaces so that they will slide together for sure (taper the walls by about 1 degree).  The only problem with this is that the parts will slide in a different amount before they bottom out (depending on the actual dimensions that end up being printed).

[Dan Newman]

On 29 Apr 2013 , at 7:56 AM, Joseph Chiu wrote:

How fast are you printing this at?

Depending on the particular machine, you could get some overshoot that
could lead to slight "outward" growth.

And, on a Makerbot, which axis has the most mass moving and thus the

most momentum / kinetic energy in need of braking?  The Y axis.  So,

if overshoot owing to momentum were a factor, then, all other factors

equal, it would be most pronounced along the Y axis.  I know it is on

a Thing-o-Matic.  In the attached picture, I used a technical pen attached

to the extruder/Z carriage and drafting film on the build platform.  (On a ToM

the build platform moves in the XY plane and the extruder carriage only

moves up and down along the Z axis.)  Each of the diagrams was at the

same feedrate and only the max XY accelerations were changed: 500, 1000,

2000, and 4000 mm/s^2.  X axis travel is horizontal in the picture; Y

axis travel is vertical.  You can see more overshoot along the Y axis

which moves more than twice the mass as the X axis.

Dan

[Dan Newman]

Oh and that picture just shows the corners with overshoot. If you
have infill and only one exterior shell, then it may be possible for
long infill strokes to overshoot enough to cause the exterior dimension
to change across the length of a side.

Dan