I’d point out that there are advantages and disadvantage to climb vs conventional milling, and also there are cases where going too slow or conservatively is actually worse for the work-piece and tooling than cutting too aggressively. Therefore I’ll recommend the following CNCcookbook articles:
In general I’d advise against more than a 50-60% stepover where roughing is needed. Depending on the geometry, MeshCAM might leave unmachined “pillars” where there is too little overlap on the roughing passes. That is more a concern in enclosed pockets, rather than the “open” geometry here. Just a heads-up.
Good point Randy. With a ball cutter you need to lower the step over. I should have clarified - if you are using a flat cutter, you can use a high step over value
The thing which I would really like to see is a physics-modeling G-code post-processor (or better still a CAM tool) which would actually calculate the speed at which the spindle was moving at a given point, the endmill’s rotation, and how much material was being cut, how Grbl was accelerating / decelerating the machine at that point and adjust the feedrate and/or spindle speed in the G-code for optimal chip size.
@Darren, you are right, that in parallel ball-end finishing if you want a reasonably smooth surface you should set a stepover of 1/8-1/10 the cutter diameter to minimize the “cusping” (extensive discussions of this over on the MeshCAM forum). But this is a separate effect. See http://grzforum.com/viewtopic.php?f=3&t=15531 for a photo of the “pillars” left by the too-large-roughing-stepover effect.