Hey guys: A client called me up with a weird problem. His machine is an OB based design, running GRBL1.1F. When he engraves a circle, it consistently is about .020" smaller in diameter than it is supposed to be. The diameter of the circle doesn't matter, both a 2" diameter circle and a 4" diameter circle are .020 undersized. This makes me think it is not calibration, it was checked and found to be bang on. There is no play in the machine, and the circles are perfectly round and measure the same side to side as they do top to bottom. This pretty much eliminates lost steps or slop in the carriage. He has tried different diameter bits, with no change. Pretty much at wits end here. Only thing I thought of is that the CAM program he is using has some type of compensation setting or other that is incorrectly set. Any ideas?
He uses Aspire, 2 flute flat nose spiral bit. Didn't say anything about squares. He's going to try some various sized squares in the next day or two. MG
One thing to consider, runout. I have an Incra fence for my woodworking router table and was having trouble getting box joints to be tight. There was a 1/64" gap - tiny but visible and the joints were loose enough to rattle. It turns out I had inserted my straight bit to the "min insert" line on the bit's shank. When I inserted the bit just shy of bottoming out, the gap went away. Have him measure the outside cut to see it it's 0.02" too big. Another thing to consider - is his measuring tool accurate?
I was considering runout, but that will always cause a toolpath that is wider than specified - ie for example: using a 1/8" bit results in a .135 wide kerf. It is not possible for runout to result in a kerf narrower than the bit. In this case the circle diameter is less than it is supposed to be. This is an outside measurement - the bit cuts a circle that is the width of the bit, and the outside diameter is smaller than it should be. Actually, it's basically an inlay, cutting a circular flat area to inlay a clock face into. Measurement error does not come into play - the clock face is what it is, and the cutouts are consistently smaller than the clock face diameter. MG
But a hole undersized is different from a circle undersized... By the way, has your client actually measured the clock faces? There can be some significant variability in manufactured components. Though, 0.02"/0.5mm is pretty big.
How is it different? It's just a matter of degree. A hole is a just series of concentric circles... Only the outside one matters... Anyway, this has nothing to do with the problem. The point is, the machine is making holes the wrong size. It has nothing to do with what goes in them or the semantics of the definition of a hole. MG
What David said but I believe its called "Allowance Offset" in Aspire. That's really the only thing that makes sense given he's tried different diameter holes/circles. Might even have them make a test circle using OB CAM just to rule out Aspire.
