Notifications
Clear all

Z axis accuracy  

  RSS
JM
 JM
(@jm-4)
Member
Z axis accuracy

Hi

I am attempting to print a lens mount adapter, so far so good managed to get the lens mounted on the camera and took some photos which turned out sharp.

I do however need to make some modifications, the printed adapter does not conform to the model and Ive found that there is an error in the z axis. The result is an inability to achieve infinity focus so I'd like to improve it.

Some measurements from a printed part include.

Cad model -> Prusa mini+ printed

 1.41 ->1.41

3 -> 3.02

7.24 -> 7.41

All in mm, i.e. printed models are coming out 3% too tall but shorter things are a lot more accurate. Is there a way to program an offset into it? I could remodel it to account for the z axis heights but that doesn't seem like the right thing to do.

Am i just expecting too much? i had the print setting at 0.1mm DETAIL, so I expected dimensions to come out at around +/- 0.05, here I get 0.17mm. Printing with PETG if that makes a difference.

thanks

Opublikowany : 25/04/2023 8:37 pm
Diem
 Diem
(@diem)
Illustrious Member
RE: Z axis accuracy

Part of the design process is adapting to the quirks of the medium. 0.2mm tolerance is about the closest you can get reliably - your first layer has to have freedom to adapt to the irregularities of your print sheet; so scale bearing points, threads and involutes accordingly. Where it really matters design with post-processing in mind. Just as you wouldn't expect a cast metal part to be ready without finishing think of printed plastic as the unfinished penultimate stage.

You can't expect much better than around half the diameter of the standard nozzle, but:

If this is proportional to the dimension then you may just be seeing thermal contraction: is your printer laying down hot plastic accurately - which then contracts as it cools?

Address this with calibration of your original design.

There are other minor distortions, most often of curves, caused by triangulation and rational approximation when translating ideal dimensions to practical printer movements. Trial and error will usually find an optimal value in a couple of iterations.

And the usual advice applies when making dimensionally critical blanks, build slightly oversize and machine to fit.

Cheerio,

Opublikowany : 25/04/2023 9:42 pm
jseyfert3
(@jseyfert3)
Reputable Member
RE: Z axis accuracy

Am i just expecting too much? i had the print setting at 0.1mm DETAIL, so I expected dimensions to come out at around +/- 0.05, here I get 0.17mm. Printing with PETG if that makes a difference.

Where are you measuring from? In addition to all the great things @diem mentioned, there’s a couple other things that get in the way of the +/- 0.5 mm assumption. 

At 0.1 mm layer height, measuring between any two layers leads automatically to +/- 1.0 mm, because you can be off by up to 0.5 mm on one side and 0.5 mm the other way on the other side.

 Measuring from the build plate would in theory be +/- 0.5 mm bit as already mentioned variations in height for build plate non-flatness.

 Both of these in addition to other variations like positioning tolerances and shrinkage.

If you are using supports to support a vertical surface that is dimensionally critical, default support separation is 0.2 mm, at least for 0.2 mm layer height, and this means the surface droops up to this distance, increasing dimensional tolerance issues. 

Opublikowany : 26/04/2023 12:55 am
Share: