Analyzing a calibration piece

[xonegon 0]

So i'm printing with an ultimaker + ultimaker filament and using Cura 12.10 on the mac.

The belts are tensioned well as in the vid. of Erik , the bed is leveled so that the first layer sticks well.

You can see the printed pieces underneath, the final result is the best I can do...

Printed with layer thickness of 0,1mm @75mm/s

wall thickness 1,2mm

fill density 30% (lines)

bottom / top thickness 0,6mm

temp. 220°C

filament Dia. 2,89 with fill density set at 97%

initial layer thickness: +0,3mm

extra wall thickness bottom top : +0,2mm

infill overlap 20%

This Square has a dimension of 40x40mm x4mm thick

Inside there is a square of 15x15mm , a rectangle of 20x10mm and a circle of Dia. 8mm

These are the results:

Horizontal dimensions:

40 --> 39,98 (ok)

15 --> 14,8 +0,2

10 --> 9,4 +0,4

vertical dimensions:

40 --> 39,6 +0,4

20 --> 19,6 +0,4

15 --> 14,15 +0,42

thickness = 4,1mm + curling up ont the sides

Can anyone help me on how to get my printer calibrated?

What is acceptable shrinkage, what is to be changed in the software, what is to be changed on the printer?

Thanks a million to those who can help me with this...

[alaris2 1]

for an UM that's an excellent piece. i'd be surprised if you can do better.

how accurate were you hoping for?

[Daid 306]

It's a bit hard to see on your photo, but it looks like the wall lines of open print are not connecting in the X direction. Could indicate a backlash problem, usually small belts with too little tension.

[jsamuels 0]

It's a bit hard to see on your photo, but it looks like the wall lines of open print are not connecting in the X direction. Could indicate a backlash problem, usually small belts with too little tension.

Agreed, I had the same problem with some early prints. Also I would check that the pulley on the motor is mounted correctly such that it's not rubbing on the motor or that the belt is rubbing on the frame.

[billdempsey 0]

For the warping issue, set the skirt to 5-10 with distance of 0 in Cura. Then print with the fan turned off. (min and max speed 0 in expert settings) This works well for me for parts under 100mm. Once the parts get over 100mm, I'm still working out how to prevent the curl. If I figure something out, I'll post it. A heated bed would solve it, from what I've read online, but they are a pain in other ways. For now, I've been working around it by making big parts in segments which snap together.

[snowygrouch 1]

The problem here is not "accuracy", which is at any rate fixed by the machine because of the belts

and pulleys. Only fiddling with the tension can optimise that.

What you are mostly dealing with is shrinkage. The plastic is going from 220 Deg C to 20 Deg C, although

PLA has a fairly low coefficient of thermal expansion its still going to shrink as it cools.

So all you can do is to measure what shrinkage you get by printing a very long test piece then scale

UP your STL files by whatever the shrinkage is. I think for me about +0.5% worked well.

Unfortunatley its very much like metal casting, the shrinkage also depends on the shape of the parts

and how they are constrained. Large solid areas will shrink alot, open areas which are mostly infill will not

shrink very much, also varies by the infill density and so on.

The very best you can do (realistically), is to work out your scaling factor and work to that. The shrinkage is what you get from cooling plastic and has nothing whatsoever to do with the mechanics of the printer or the software.

C.