View Full Version : Madcam - 4 axis indexed ring machining

08-22-2008, 02:10 AM

I have been practicing more with Madcam, and am going to try to machine a ring in wax using my new rotary axis (taig setup). These operations might seem obvious to a veteran, but for me, I had to think about how best to do this. A couple of options were:

-Machine entirely in 3D and then use a program like WRAPPER that will take the Y axis moves, and re-compute them for the rotary axis (wrapping Y moves around the X axis creating a cylinder). Depending on how good the compute algorithms are, I could see where some distortion could result from the wrap of the flat 3d geometry into a cylinder shape. :confused: So I opted to...

- Machine using indexed 4D. This was done by breaking the ring up into 8 regions all at 45 degree offset in the Rhino Cplane. Because the eight segments are identical, the same machining can be used on each section.

This is done by:

1- Set the clipping plane bottom towards the top of the ring.

2- Draw eight region curves around the ring that slightly overlap.

3- Machine a single region completely in 3D. This lets you simulate and optimize the 3D strategy before moving to the rest of the regions.

4- Implement the same 3D strategy on the rest of the seven regions. I think you can actually copy and rotate the toolpaths using Rhino, but I simpley re-computed after moving the Cplane by 45 degrees.

Note that the four corners of the part box are not roughed off. So options here would be to cut them off before mounting the part, or creating the toolpaths for this operation and running these operations first.

08-22-2008, 02:28 AM
And then you could finish up with a 4th axis finish pass. Not sure if this is needed, but I think it might be useful to save some of the indexed machining. ie, maybe run this and then run pencil z level and pencil tracing at the very end.



08-22-2008, 11:14 AM
It struck me on the way in to work today...

It is not too hard to index the rotary table with a G1 A 45, 90, 135, etc. So it is probably easier to compute the machining for the first segment, and then copy/paste the GCODE or manually re-run on each region after advancing the rotary table.

The indexing is more useful when the part geometry is not replicated.


08-28-2008, 01:54 PM
A final note...run all your roughing passes first, and then your finish passes (all the way around). Saves a lot of manual tool changes.