hexagon logo

Surface Profile - Iterate and Repierce function

Can someone give me a rundown of what the iterate and repierce function does? I turn it on and all of a sudden my surface profile improves significantly. Is that real?

On a similar topic, has anyone else had trouble with vector least square best fit causing a best fit error? I have had several parts now that had profile dimensions set to vector best fit, which runs fine offline, that have given me best fit errors at the end of the program. If I hunt down the offending dimension and change it to least squares the error goes away immediately. I have also had a similar problem if I set the iterate and repierce function with too small of a tolerance. I'm using an .039 (1mm) ruby and have found I need to set the repierce tolerance to at least .025 to reliably run through the program without any best fit errors. This seems to be related to features with very large radiuses of contour (almost flat, but not quite). My assumption is that the math runs out to too many decimal places and it can't handle it.

Thanks for any advice. I am inspecting a part that is just one giant contour, no flats at all, and need to have all of my ducks in a row around how all of the surface profile options work.
Parents
  • Ah, thank you. I was worried that without checking this box the calculation for best fit was only performed for the initial part. Yes, there should not be an issue with unpredictable probe radius compensation in my pattern. Engineering has been "convinced" that there is something wrong with my program since some parts measure fine and others measure out. This is one of the more complex parts this shop casts, so it's possible there's enough variation in the process that some parts actually are slightly out. So, one more box I can check off on what's not wrong with my program. Thank you very much for the clarification.
Reply
  • Ah, thank you. I was worried that without checking this box the calculation for best fit was only performed for the initial part. Yes, there should not be an issue with unpredictable probe radius compensation in my pattern. Engineering has been "convinced" that there is something wrong with my program since some parts measure fine and others measure out. This is one of the more complex parts this shop casts, so it's possible there's enough variation in the process that some parts actually are slightly out. So, one more box I can check off on what's not wrong with my program. Thank you very much for the clarification.
Children
No Data