G code is not working...

It is possible that you have changed the accuracy level in mm mode, but it has reverted in inch mode?

I have been fiddling round in Excel, and EMC2 is not wrong, those numbers really don't work.

axel88 wrote:

Ok even in Metric and Minimum Resolution =.0001 the error remains.

You might need to also change the number of digits of numbers that get output

As an alternative, you can probably do circular moves in X Y Z R mode, then it is EMC2's job to make the radius work.

So is it a handicap without G2 G3 or is it equal?

Files will be much bigger, and will probably run a lot more slowly as the motion planner has a harder time.

axel88 wrote:

The X,Y,Z format is set as "Cordinate" "5.3" -> XXXXX,XXX but in my *.ngc there are for e.g. "N0110 X-2.6228 Y2.5496" so one position after decimal point more than specified. Thats some kind of strange

The whole thing is rather strange.

Secondly I don't understand why i should make the values much more precise than my mill could handle?

The problem is that the postprocessor is outputting nonsense. If you look at those start-centre-end coordinates they do not define a circle, the centre is nearer one end than the other. EMC2 can't work out what to do about that. I have no idea how it is possible for the postprocessor to get it wrong, but it is doing.

So do you think the error will be solved if I set the Cordinate Format to 5.4?

I would be tempted to set it to 5.8 and the accuracy to 0.00000001 just to see what happens.

Is this an official, paid-for installation of Unigraphics? I think this should be a support call / bug report.

Have you looked at the code and/or checked the line count?

It will be a bigger file than with the lower precision, but only the same number of moves.

Unless the settings are wrong in some way, and G2 and G3 are not in the new file either. Have you looked?

You should be able to ratchet back the resolution and precision to a point that works but is less silly.