Originally Posted by ssaults We have a Milltronic RH30 with a cent. 6 control & a Haas VF6. We like both mills but the Haas is does a better job for us. One problem that we have with the Milltronics control is sometimes it will make a complete circle that cuts through the side of the part instead of an arc. Most of the arcs are big (over 20" RAD) so we try to back plot ever job that we send to the mill. That is my biggest complaint with the Miltonic. |
I have seen this once before. It was found to be related to the step over in a pocket clear routine produced by the cam package. All we changed in the cam parameters was the step over path by .001 or .0001 and that cured it.
I always run the verify screen before letting the machine cut. We do not "back plot" for whatever that is worth. That is where we caught the wild hair when it broke out about half way through verifying the program into a big arc. Aparently it could not resolve the math on its own and went into a lead in, lead out, because it did end up at the same place it started on the screen. No way were we going to run it like that!
I think I would rather have an error under this condition, but it has never been an issue since.
I just recall it being such a small amount, it hardly seemed logical. At least until I worked on some parametric programming that had floating point errors using conditional statements that would only come true if the numbers were accumulated right on. Not one count high or low to the resolution the floating point uses. Which is a lot higher than the encoders display. We had to box in the value using "Greater than or equal to" and "Less than or equal to" which allowed a little grace to handle the minor issues involved with floating point math. I'd imagine Milltronics knows about it. Whether they have an update to correct it, may be worth checking into.
DC