Profile around a countoured pocket

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

  • Profile around a countoured pocket

    Attached please find a sketch of a pocket they want me to measure with a profile called out for all around the pocket. I could evaluate each individual feature (every cylinder segment and flat planes (4 cylinder segments and 4 planes)), but I do not think that's the right way to do this.

    Anybody any suggestions on how to do this????

    By the way, somehow the MMC symbol did not come through on the 0.010 and the datums F and G. But I know how to take care of that. It's just that all around symbol on this pocket that I have trouble with.

    I am running version 4.0.



    Jan.
    Last edited by Jan d.; 08-10-2006, 08:16 AM.
    ***************************
    PC-DMIS/NC 2010MR3; 15 December 2010; running on 18 machine tools.
    Romer Infinite; PC-DMIS 2010 MR3; 15 December 2010.

  • #2
    Let me begin by saying I have not done this. I usually evaluate each individual feature. I know it is not the proper way, but it lets me see which ones or all are off and by how much, which direction, ets. So you may find it benneficial to report both ways.

    As to how to handle it as one profile, I believe in class we were told this could be done with "sets". Have you tried the help menus?

    Sorry, I know this is not a lot of help, but maybe a little. Goodluck and let us know how you got it done.
    sigpic"Hated by Many, Loved by Few" _ A.B. - Stone brewery

    Comment


    • #3
      with cad would be simple to just create points around the profile and create a feature set from them to evaluate the profile callout.

      without cad is somewhat more time consuming...

      I would try to get the eng dept to draw the shape and give you the tangency points and angles of the flats to the radii. translate rotate to those given points and create surface points on the individual surfaces....then recall the profile alignment and create a feature set of all the individual points and dimension the profile of the set
      sigpiccall me "Plum Crazy"....but you only go around once!

      Comment


      • #4
        In my opinion the 2.520 diameter should be a basic dimension since you are checking the walls with a profile. If you had cad would be pretty simple if you don't ???
        sigpic.....Its called golf because all the other 4 letter words were taken

        Comment


        • #5
          All around is simply 4 partial circles & 4 lines. I assume you do not have a model so then you must type in exact point & vector coordinates for the 8 features then create a set from these 8 features & profiles the set.

          Comment


          • #6
            If you don't have the model then i'm not sure this can be done. You need basic dimensions to get to the 2.520 diameter. Then to check the walls without the model the diameter needs to be basic too. That way all dimensions are perfect (Basic). The diagraham you supplied shows the diameter as a non basic dimension which i'm assuming has a +/- callout.
            sigpic.....Its called golf because all the other 4 letter words were taken

            Comment


            • #7
              LINEAR CLOSED

              If you do have a model, I would use a linear closed scan around the perimeter, then dimension the profile.

              Tim

              Comment


              • #8
                Thanks guys. I do have the 3D model, so next I will try to work with the sets, as suggested before. I have never tried that before.


                Thanks, Jan.
                ***************************
                PC-DMIS/NC 2010MR3; 15 December 2010; running on 18 machine tools.
                Romer Infinite; PC-DMIS 2010 MR3; 15 December 2010.

                Comment


                • #9
                  Off subject,
                  But doesn't PC-DMIS construct a center point based on the selected points when constructing a feature set? I have often debated the validity of profile dimensions when using feature sets. Are feature sets a valid way of checking profile callouts or would a scan converted to points with the T value dimensioned be a more accurate representation?

                  Just wandering...
                  Kevin

                  Comment

                  Related Topics

                  Collapse

                  Working...
                  X