Here is another True Position question

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Here is another True Position question

    V3.7 MR3

    Sometimes the TP calculation seems to work and other times not.

    Here is an example:

    TP|.002m|D|Em|Fm

    D is a plane with 2 holes in line (E and F)

    My alignment is:
    1: level and origin in Z to a plane on -D-
    2: rotate to a line created from E and F (rotate to +Y)
    3: origin in X and Y on -E-

    Now I have a hole (CYL3) in the X plane about 15 degrees off of perpendicular to -D-

    When I do TP with Datums I choose CYL3, Plane -D-, CYL -E-, CYL -F-
    applying MMC to DF(CYL3), D2(CYL -E-) and D3(CYL -F-)

    The hole is reported out of tolerance. Bonus is displayed for -E- and -F- but apparently not applied.

    I did some research, downloaded and studied the TRUE POSITION - PC-DMIS 3.7MR2 pdf.

    Then I tried an experiment.

    I reversed the order of the datums.

    CYL3, Plane -D-, CYL -F-, CYL -E-

    Now I see results I expect. The best-fit wiggling shows the hole (CYL3) virtually perfect (except for a couple of tenths).

    Is this a bug? Does it have anything to do with the order in which I chose the datums to create the line that I rotated to?
    I have noticed that the TP routine seems to work sometimes and not at other times.
    Do I need to try one order and then the other?

    What a pain. I will hafta run a part before this can be done since offline operation always shows everything perfect.
    Lately, it occurs to me
    What a long, strange trip it's been.

    2017 R1 (Offline programming)

  • #2
    DON'T USE THE DATUMS in the TP callout. You will not see the correct XY locaiton of the hole due to Pcdmis moving the datums around to give you the best values for what you are dimensioning. Just make an alignment to the features in the FCF and then do TP without the datums. You will get many fewer grey hairs.
    sigpic
    Originally posted by AndersI
    I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

    Comment


    • #3
      Originally posted by John Riggins View Post
      V3.7 MR3
      TP|.002m|D|Em|Fm

      My alignment is:
      1: level and origin in Z to a plane on -D-
      2: rotate to a line created from E and F (rotate to +Y)
      3: origin in X and Y on -E-

      I reversed the order of the datums.

      CYL3, Plane -D-, CYL -F-, CYL -E-


      Now I see results I expect. The best-fit wiggling shows the hole (CYL3) virtually perfect (except for a couple of tenths).
      I think your getting this error because your tertiary Datum in your FCF is "F" not "E".

      When you created your alignment your origin is on "E" which is your secondary datum. Which in the second part (rotation) of your alignment is correct.

      Therefore when your True Position Dimension's Datums are "reversed" they will match the alignment that you are currently in. Otherwise it will be looking at the feature in a different alignment (a reversed one in this case).

      For an experiment you should change your alignment by setting your origin to "F" and then building your dimension command as shown on the drawing's FCF. You should get the result your after then.
      Last edited by MrComment; 03-14-2007, 10:39 AM.
      Mr. Comment
      SCIROCCO-NT 13-20-10
      B3C-LC Controller (Leitz Protocol), SP600M, TP200
      PCDMIS CAD++ v4.3 MR1(Build: 12/11/08)
      sigpic

      Comment


      • #4
        Go along with Matt's suggestion.

        Many of us (myself included) have "acquired" our gray hairs through much trial and error.

        Blessings,
        ZydecoPete
        sigpic

        Comment


        • #5
          Originally posted by MrComment View Post
          I think your getting this error because your tertiary Datum in your FCF is "F" not "E".

          When you created your alignment your origin is on "E" which is your secondary datum. Which in the second part (rotation) of your alignment is correct.

          Therefore when your True Position Dimension's Datums are "reversed" they will match the alignment that you are currently in. Otherwise it will be looking at the feature in a different alignment (a reversed one in this case).

          For an experiment you should change your alignment by setting your origin to "F" and then building your dimension command as shown on the drawing's FCF. You should get the result your after then.

          Been there, done that... same result.
          Lately, it occurs to me
          What a long, strange trip it's been.

          2017 R1 (Offline programming)

          Comment

          Related Topics

          Collapse

          Working...
          X