hexagon logo

PC-DMIS 2021.1 SP2 - Possible bug with linear open scans & findnoms?

When I create a linear open scan with “findnoms”, after I close the command and open it again it changes to “master”.
After I changed it a second time to "findnoms" it keeps that setting.

It doesn’t happen every time and I can’t reproduce it but it happened enough times,
that a coworker and I both noticed it independently on two different CMMs.

Does anybody have the same issue or maybe knows why this could happen?

I will continue to try and reproduce it but I’m happy for your input.
Parents
  • Sorry for the delay, let me try to clarify a bit more. What JDevaraj said is correct, in fact it came from the team that works on this area.

    FINDNOMS is one way to teach the routine what the nominal locations are. It is designed to be a LEARN mode function only. In other words, it finds the nominals the first time, when you are learning (or teaching, writing) the measurement routine. After that, if you have the execute controls set to NORMAL it will continue to take those same hits, one at a time, in the traditional fashion of a touch trigger probe.

    With FINDNOMS as the Nominals method, but this time you have the EXEC CONTROLS set to DEFINED, it will take those original (found) nominals and scan the surface on a path defined by those nominals. Both the Exec Controls of NORMAL and DEFINED are true execute mode functions.

    The difference here is when you have Nominals Method set to FINDNOMS and the Exec Control set to RELEARN, that is saying that you are "Mastering" the part the first time. Once you've done that, no need to learn those points over again. In fact, the Nominals Method control is pretty much ignored after the first time. There is no need to find the nominal values or vectors as you've already done that, and it is somewhat time consuming. So instead, this control switches to MASTER to basically tell it to look for the values you "found" the first time.

    Admittedly the terms used, and the dialog itself, are not all that intuitive. It has been tweaked a few times over the years to try to make it better but obviously could still benefit from some improvements. For now, I will see if I can get some documentation/help improvements anyway.

    nb: For my own future reference, PCD-218223, PCD-49948
Reply
  • Sorry for the delay, let me try to clarify a bit more. What JDevaraj said is correct, in fact it came from the team that works on this area.

    FINDNOMS is one way to teach the routine what the nominal locations are. It is designed to be a LEARN mode function only. In other words, it finds the nominals the first time, when you are learning (or teaching, writing) the measurement routine. After that, if you have the execute controls set to NORMAL it will continue to take those same hits, one at a time, in the traditional fashion of a touch trigger probe.

    With FINDNOMS as the Nominals method, but this time you have the EXEC CONTROLS set to DEFINED, it will take those original (found) nominals and scan the surface on a path defined by those nominals. Both the Exec Controls of NORMAL and DEFINED are true execute mode functions.

    The difference here is when you have Nominals Method set to FINDNOMS and the Exec Control set to RELEARN, that is saying that you are "Mastering" the part the first time. Once you've done that, no need to learn those points over again. In fact, the Nominals Method control is pretty much ignored after the first time. There is no need to find the nominal values or vectors as you've already done that, and it is somewhat time consuming. So instead, this control switches to MASTER to basically tell it to look for the values you "found" the first time.

    Admittedly the terms used, and the dialog itself, are not all that intuitive. It has been tweaked a few times over the years to try to make it better but obviously could still benefit from some improvements. For now, I will see if I can get some documentation/help improvements anyway.

    nb: For my own future reference, PCD-218223, PCD-49948
Children
No Data