true position with and with out legacy dimension

  • Filter
  • Time
  • Show
Clear All
new posts

  • true position with and with out legacy dimension

    I am getting extremely different results for my true position when I use legacy dimensions compared to when I am not. I have a bolt circle on a zplus face that has the holes called out at TP.005 BFC Datum B is an OD, Datum F is the face the holes are in and Datum C is an ID bore perpendicular to Datum B and F. Datum B is an OD that is only .150 long so I can't get a reading on it to make a cylinder in order to control any degrees of freedom other than X and Y. I made my alignment this way Datum F face was used to level and be my ZPLN and control Xaxis and Yaxis rotation, Datum B was used to control X and Y and datum C was used to control my Zaxis rotation so all 6 degree of freedom are contained. When I do that I get a tru position of .002 but when I do a feature control frame that same hole is at .0124. I am at a loss as to how these can be so extremely different.

  • #2
    When you use Legacy, you have to create the alignment you want and then recall it when you want to use it (like reporting a hole pattern back to it).

    XACT is designed to ignore your current alignment & PC DMIS builds the alignment you set up based off of the feature control frame you created and uses THAT. So, using XACT and selecting BFC would cause PC DMIS to report that TP while leveled off of -B- and rotated -F- ABOUT -B-. Very different then leveling off of -F- and rotating through other features.

    You're difference is due to the above.

    All of the above being said... your datum scheme is what it is. Following GD&T rules, you SHOULD be leveled off of -B-, rotated through -C-, XYori = -B-, Zori = -E-.

    I totally hear you about the short cylinder...any form error over that small surface will extrapolate through your alignment and will f*ck you over. As the programmer, you can only measure what the print says. Not engineer it yourself!
    Last edited by DAN_M; 11-27-2019, 11:38 AM.
    SF7107(PCD), SF454(PCD), 152614(Quindos), 9159(Quindos), 7107(Quindos), B&S Manual, M&M Gear Checker


    • #3
      Thanks for the clarification. The only thing I could think of was that it was leveling off of Datum B per the print even though it doesn't make sense being such a short cylinder. Thanks again for the info.


      • #4
        I used to see this all the time (at RR no less!).

        Large parts 1m+ in diameter with a 'primary' datum of a 10mm deep cylinder, with the plane as the secondary.

        I'd argue in cases like this (bearing Can-May-Must) rule in mind, that , the short cylinder CAN only (reliably) control the x and y origin and not the level.

        I'd do a couple of things.

        1) Build the xact FCF as FBC and verify that doing it that way you're getting the same answers as using legacy.

        2) CYA and get clarification, and if it's agreed that it should be FBC but they're unwilling to change the drawing, the measure B with sample hits (or level to F first before measuring B so it has the same vector)

        Automettech - Automated Metrology Technology


        Related Topics