hexagon logo

Making 2 Part Programs into 1. One Graphic, One report

I have a top half part program, and a bottom half. Is there a way I can merge the two? I want to run the top half, program A, then input a comment to the operator for them to flip the part, touch off, then measure the bottom half. All using one graphic, one report, one model. I don't want the features measured on the bottom half, program B, to be in space. I want the features to be pulled from the same model used in program A. Then dimension all 500 items in one report.

Can an equate alignment work? I thought those were used if the part moves laterally in any direction. But in this case, I need to flip the part to measure the bottom half.

Any advice?
  • Flipping parts is NOT what Equate is used for. The reason is exactly as you've discocered: Ya can't hit the same surfaces and acquire all the features.
    There's one way to do it right, elevate the part.


    I'll respectfully disagree. Flipping the part is a great example of equate. You have to do it right though.

    I get what you're saying about elevating the part and going underneath with star/hook probes. Would always prefer this myself.but sometimes it just isnt possible. Depends on the part.So either write 2 programs or use equate in these situations. I make the decision based on program size. If it's getting on the large side with long run time I create a 2nd program.


  • I'll respectfully disagree. Flipping the part is a great example of equate. You have to do it right though.
    I get what you're saying about elevating the part and going underneath with star/hook probes. Would always prefer this myself.but sometimes it just isnt possible. Depends on the part.So either write 2 programs or use equate in these situations. I make the decision based on program size. If it's getting on the large side with long run time I create a 2nd program.


    This is the first I've heard of this... Can you walk me through that alignment - 'equate a flipped part' process a little...
    Are you saying you have used Equate Alignment with the Equated rotation points
    vectoring in the opposite direction of the alignment it is based on??

    If saay your alignment rotation is in Y and you flip the part (about Y) that's what has to happen
    in order to re-acquire the alignment, the same rotation surface now facing the opposite direction???

    Are you sure of this?

  • My understanding of Equate is that it is a work-around for parts that are too long to
    fit into the machine's measurement envelope.
    ...that it can only be used to slide the part in Y. (IOW, the alignment surface vectors must match)

    I can't imagine how you would use Equate for a part that fits on the machine.
  • Equate should work well for both sliding and flipping, as long as you are aware of what's happening. All that's mathematically needed are two alignments: one before, one after the move, where the resulting alignments have the same zero points and the same axes in the same direction in relation to the part.

    If you can't measure the same features before and after, you will get more uncertainty in the result, then it might be better to create special alignments just for the move, so you can touch the same surfaces.

    If you cant find surfaces/features with tolerance fine enough for this, you can attach three magnetic spheres on the part (and make sure they don't move when you flip the part) and do the equate alignments on them.
  • Equate on a flipped part works fine we use it to measure the distance between bearing faces on large turned/ground bearing housings. The first alignment is Z+, on the flipped part it is Z- using the same features, results correlate very well compared with hard gauge methods limit is +/- 0.02 mm
  • They might be opposite in MACHINE coordinates, as measured, but that doesn't matter. You could stick on 3 tooling balls to use before and after flip. Or use the Y+ face with 2 holes in that face.

    Yes, that's exactly what equate is for.
  • For a part that you can't get to the bottom of, like we've been talking about. :P
  • Most of the parts I work on have equate.
    Some just flip over, some flip and rotate 90 degrees.
    Main issue is making sure your alignments have the same origin, direction in relation to the part itself regardless of how it is set on the cmm, and fit the part back into the identical position.
    Most of the time i have an extra set of points verifying setups are identical.
  • Tanks peeples. I learned something here today.
  • I'm trying to use Equate as described above, spinning the part 180 degrees (fixture plate, repeatable locations) in X/Y.
    ...Nice flat bottom part.

    I finish the 1st half, spin the part, and re-probe the same ABC features. (2 planes constructed from V-pts and a circle)
    Then I build the new alignment and hit Equate.... And nothing happens... that I can see.

    I am expecting to see the change in the graphics window. Can one of you Equate gurus tell me what I'm doing wrong.

    And a side question... Is a switch into manual mode necessary?