Converting G2 & G3 Moves to Rotary Axis?

[skibbey]

In the past I've used Excel to convert purely coordinate-based g-code to cut shapes on my rotary axis. The spreadsheets simply convert the length of the each linear move into an equivalent angular move for a given circumference.

However, I now have an eliptical cut that has a significant number of G3-based moves. Is there an easy way to convert those moves too, or should I change the way the initial g-code is being generated to eliminate the G3 moves in favor of a much larger number of G1 moves?

Thanks.

[dcoupar]

I take it you don't have the cylindrical interpolation option? That would make life easier.

[skibbey]

I do have it now, but I didn't in the past.

The problem is that I am struggling to figure out a way to mill the part in the attached image, and the methods recommended so far end up creating blades with a wedge-shaped cross section.

The blades were originally modeled in Rhino by (i) wrapping an eliptical arc on a cylinder, (ii) extruding the arc normal to the cylinder surface, (iii) offsetting the extruded surface .05" to create a solid, and then (iv) using polar copy to create 10 blades.

I want to mill the blade surfaces using a constant z and continuous 4th-axis motion to create an efficient tool path and a good finish. If, as has been recommended, I model either a single blade or a pocket between two blades as though they were on a flat surface, then wrap multiple copies of the tool path on a cylinder, the g-code runs fine but creates blades with the wedge-shaped cross section.

The concave blade surface is no problem, but I need to come up with a way to cut the convex surface so that it is parallel to the concave blade surface rather than normal to the surface of the core.

Does this make sense, and do you have any thoughts?

Thanks.