Question about rotary toolpath display.

Hi to all,
I have a 4 axis CNC milling machine and I can not configure Linuxcnc to display rotary toolpath. When joging A axis the visualisation of the tool is corrrect but if I load a G code with some rotary toolpath, it isn't displayed correct, I only see a flat blue lines and the tool moves in flat surface instead rotary traectory. If I load a 3axys G code I see the entire model with white lines and the tool follow exact path but in rotary don't. Is this bug in AXIS or my configuration is wrong?

www.linuxcnc.org/index.php/english/forum...xcnc-need-help#58080

Thank you for the answer, ArcEye
I have set my INI file to GEOMETRY = XYZA. I see the tool cone rotate around X axis when jog it, but don't see the toolpath like a 3D object. Do you mean that this is impossible in Linuxcnc and the rotary toolpath is displayed always wrapped?

There is a limit to what you can get a 2D display to show

Did you try this one?
www.linuxcnc.org/index.php/english/forum...s-on-cylinders#40497

regards

Not yet, I will try this. I must change INI to 9axix GEOMETRY if I understand correct?

Ah, OK, Thanks.
I will try this .

Update - the AXYZ configuration works, but now have the following bug: When open G code and the machine is set in zero machine coordinate, everything looks fine. When do touchoff and set G54 or G55 or other user specified position, the visualisation of the toolpath start deforming. Like a sample, the ring with 30mm diameter looks like a ring with 120mm diameter. I tried AXYZ, XAYZ and XYAZ, always the same result.

Any solutions this?
Hello all, I have the same issue "the visualisation of the toolpath start deforming".It happens when we works in non-zero absolute coordinates, but I think the physical aspect works fine: this issue appears only on visualisation window.

Anyway to define the rotation axis of the displayed tool cone? A least it should be possible to define it in G54 abs coordinates? in my setup the axis for A is parallel to x (ok) but shifted to z=limit_z > 0 which is weird. I would expect rotation about local (y,z)=0, respecting G92 offsets.