Page 1 of 1

Z Calibration

Posted: Mon May 26, 2014 4:33 am
by cadcam
I have been trying to improve the acccuracy of some of my models.

Using the calibrate function I seem to be able to get really quite accurate xz dimensions, order of +/- 0,2 mm over cm.

However I am getting an error in Z of about 0.2mm/10mm. The error is linear with height so I assume it is a scaling factor. (There is little/no angle error in the test piece) Thus I have been trying to correct the height error by putting 40.2mm in the H box in calibrate. However this doesn't seem to have any effect on the numbers in the blue window surround or on subsequent models.

Has anyone else had the problem and how best to solve?


Re: Z Calibration

Posted: Wed May 28, 2014 5:51 am
by cadcam
To anwer my own question,

pp3dp have kindly got back to me on this one and I understand that there is no software scaling available in Z direction.
GIven the reliability and the consistancy of builds, it seems a shame when you want to make dimensionally correct functional items. (Sometime I will post the statistics I have for build sizes/consistancy with different conditions/material.., it is quite impressive )

However, if you know which orientation you are going to build items it seems that the best policy is to use Meshlab (or something else) to scale the item before loading into the Up! software. Obviously this becomes more awkward if your model needs to be built at an angle. It may be worth reorientating in the source package or Meshlab or a commandline tool to help in loading .

Or does anyone know of an alternative solution?


Re: Z Calibration

Posted: Wed May 28, 2014 2:28 pm
by doraemon
Couldn't you use the Scale function in the UP software? You can scale Z independently from the other axes. You can type in whatever is the appropriate scale factor.

Re: Z Calibration

Posted: Thu May 29, 2014 6:46 am
by cadcam
Dear doraemon,

Thank goodness to the forums, they help point out the obvious, I should read the manual in more detail again.

However, the fact that the code to scale stls in xyz is available it is very strange it isn't included in the calibration, as it could be one fo the last things to be done automatically before the generation of the'g-codes'.,

Thank you once again I will read the manual again.