scaling problem on CNC Router

20 Aug 2023 21:09 #278573 by geertd

A while ago, I installed LinuxCNC on a CNC Router and have been using it since then without major problems.

- LinuxCNC 2.7.15, using GMoccapy as GUI
- 3-axis CNC Router, with rack and pinion stepper motors on X and Y and a ballscrew positioner on Z.  Max. X 1500mm, Max Y 2500mm

Until now, I mostly did small projects which generally are not critical regarding dimensions, and some bigger projects, but never experienced any problems regarding dimensions.  I guess I was lucky that also the bigger projects were not very critical.
However, recently I realised that I have a scaling problem: My Y axis seems to be scaling 0.5 % bigger than necessary.  If I cut a square that is 1000mm by 1000mm, the final dimension of the square in X is spot on, but the dimension in Y is 1005 mm.  Cutting other size dimensions confirm that the scaling in Y is consistently 0.5% bigger than it should be.

The X and Y axis have identical stepper motors, identical rack and pinions and transmissions.  They also have identical stepper drivers.  So having the same scaling (steps per mm) in the configuration should give the same results for X and Y, but as mentioned above this is not the case.
First I thought of the stepper motor loosing steps, but in that case, I would expect the dimensions in Y to be shorter, not longer and also not as consistent as they seem to be.  Also, I discovered this error while cutting 3mm thick hardboard with a 6mm endmill on a 1,5kW spindle, so cutting forces are probably not the issue.

I could of course adapt the Y axis scaling until right, but having the same mechanical configurations for both axis, something tells me this should not be OK and the scaling for X and Y should be the same....
Is there any other parameter in LinuxCNC configuration or parameter files that could affect axis scaling that I am not aware of?


Please Log in or Create an account to join the conversation.

20 Aug 2023 21:50 #278575 by andypugh
Your analysis seems correct. I don't know what is going on, but 5mm in 1m is definitely not something you would mis-measure.
I think that I would start by checking the racks, with a tape measure. (ie, compare one to the other over 1m of teeth. Does the same part of the tooth line up with the 1m mark on both axes?

0.5% seems too small to be due to different primary reduction ratios (if you even have that)

Please Log in or Create an account to join the conversation.

Time to create page: 0.061 seconds
Powered by Kunena Forum