bridge saw cutter for marble with rotating table

Hello
I am working on a project, to convert an old bridge saw cutter for marble with rotating table for the work piece
to a cnc milling machine. My question is how to configure the kinematics of the 4-axis system.The tool is a disk saw (20cm .. 30cm diameter)
that moves in x,y,z axes and at the same time the work piece table can rotate when making curves.
I am going to use servo drives with analog reference inputs and incremental encoders for feedback.
For the logic interface I will use some mesa cards.
Any suggestion for the appropriate cam processor for the gcode generation of dxf files?
I am already familiar with linuxcnc and mesa cards that i have used in the past on other projects.
Thank you in advance.
George

The axis C is rotating relative to Z axis and coordinated with the other motion.

The tool is a disk saw (20cm .. 30cm diameter)
that moves in x,y,z axes and at the same time the work piece table can rotate when making curves.

I have difficulty picturing this. How are you cutting a curve with a disk saw?


[edit]
Is like at 1:07 in this video (only in your case the work piece rotates)?


If so you don't need any special kinematic in LinuxCNC.

If you want to cut noncircular curves or circular ones that have centers that do not coincide with the axis of the rotary c table then you would need a fairly capable CAM. But I don't see how a custom (ie non-trivial) kinematic could be of any use here.

[edit]
Ok, I maybe kind of get it. For the mentioned cases above. You'd want to rotate the C table so the saw remains tangential to the curve and you'd then need the kinematic to adjust the XY-position according to the C axis angle.
For this you could use for example the included XYZAC-trt kinematic and just ignore the A axis.
However you would need a way to calculate C-axis angles depending on the tools position along the arc.

The machine is similar to the one in the picture.
The table (C axis) is rotating relative to Z axis, which is perpendicular to XY plane.
The disk diameter could be 20cm in minimum.

Like a tangential knife with the C-rotation done by the work piece.

[edit]
Should be possible, I guess. If the calculation of the c-axis angle could be integrated in the XYZC TCP kinematic then you might not even need a CAM. It would just be a regular 3 axis gcode.
I'm not sure I'd want to be the one responsible for writing the c angle calculations though. Just imagine the mayhem if it missed a discontinuity in the math and the resulting angle suddenly changed from positive to negative.

Actually SheetCAM seems to offer a ''Dragknife" module that may take care of the angle calculations:

sheetcam.com/Register/purchase

[edit]

seems this user also uses the SheetCAM solution for this.
www.forum.linuxcnc.org/10-advanced-confi...h-xy?start=50#248422

Regarding the video you attached, the only difference i see is that in my machine,
instead of rotating the disk saw, the table rotates.