Profile dilema

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

  • Profile dilema

    I am calling out a profile of a plane, and it is showing out of spec, but all of the individual hits are in spec. Now if I take individual POINTS, and construct a plane from the individual points, then call out THAT planes profile it shows in...


    WTF!?!? I am soooooooo stressed!
    CMM Programmer
    Jackson Michigan
    Mistral 7.7.5
    4.3MR2

  • #2
    You doing FORM ONLY or FORM AND LOCATION?
    sigpic
    Originally posted by AndersI
    I've got one from September 2006 (bug ticket) which has finally been fixed in 2013.

    Comment


    • #3
      .

      Sorry, yeah Forma and Location.

      And when I said that the individual points from my original plane were good, that was a mistake. What is happening is if I look at the hits from my first plane (created in auto features) the Z height is whacked, BUT...if I go to the exact same X&Y location and create points (using the Z location pulled from the model by our engineering dept) and then construct a plane from those points, the profile dimension is in spec.

      But if I ask for profile using my auto features plane it is out by like 2 millimeters!

      I am so confused right now. lol.
      CMM Programmer
      Jackson Michigan
      Mistral 7.7.5
      4.3MR2

      Comment


      • #4
        Ah...profile.

        PCDMIS has a very good profile algorithm but you have to both understand profile and the way the software attacks it. First, a basic look at profile. The software will want to know a couple of things about what you're trying to do. If you are looking a surface or contour that is defined with basic dimensions relative to specified datum axes or targets, then you're probably looking a case of profile using both form and location. Those basics are critical as if you don't have CAD you're going have to define the theorhetical basics of the manually measured features so that the software knows where to establish it's tolerance envelope. That's also so you can establish the tolerance yourself when you access the menu. Say you have a profile callout of .005 and it's shown on the print to bilateral about the theorhetical edge then you will put +/- .0025 in as your tolerance. If the print implies a unilateral tolerance in one direction or other to the edge you would of course enter something like +.000/-.005 or whatever based on the callout.
        Form only is based on a callout where say you have a plane or maybe 3 small feet that form the basis of a plane surface that the print wishes to express as a coplanar condition only and cares not where those measured points lie from a specific target or datum.

        When you use profile in PCDMIS, using it the correct way is important. Say you're doing the later example of form only. Let's say we're doing 3 small feet that define a plane and there's a callout of profile .005, no basic dimensions shown. What you're going to do is to take say, for lack of a better number, 20 individual points on all three of those feet. That is, each hit on those surfaces will be it's own point, you know, hit and then enter. Not part of a bunch of points and then hitting enter to define a plane. So what you have is now 20 individual points that if you're using CAD, have a common target location in the direction or axis you're measuring. If you're using a manual machine with no CAD, you'll have to go back and define the "perfect" theorhetical targets in the axis you're measuring so the software knows where those points are supposed to be located relative to your alignment. Now, what you're going to do is construct a feature set based on those 20 points. You'll see it offered in your feature construction menu. The result will be defined as a scan. This array of points can now be evaluated on form alone. Go into the profile menu, click form only, (you'll get a response that there can be no minus tolerance when using form only) and evaluate your profile based on the feature set. PCDMIS is going to look at those individual points and try to place them within a .005 wide envelope encompaasing all of them if possible. That's what form only is all about. It doesn't care about the location of those points to another datum or axis. You can have 20 points with a profile of form of .0005 but if they may be out of parallel by .05 if you use and create a plane based on planar location from a feature or datum target. I know this may read like stereo instructions but trust me, I run into quite a few engineers that don't know heads or tales about the usage or non usage of profile when they stick those nice little symbols on the print. If you have some specific questions, drop a line. We're all here!

        B
        Physics dictates to man why his world acts the way it does....Chemistry tells him why it smells the way it does.

        Comment


        • #5
          Thanks, but I need this done as Form and Location as it controls the Z height of a plane. I get what you're saying though. I was just miffed as to why when the plane was made off from the CAD in autofeatures it wouldn't dimension properly, but when contructed from points, it came in good.
          CMM Programmer
          Jackson Michigan
          Mistral 7.7.5
          4.3MR2

          Comment


          • #6
            Try running program offline---just against the model
            if your program generates any error in profile report you have an issue in alignment or how points were generated in scan--just a thought that has worked for me in the past

            Comment

            Related Topics

            Collapse

            Working...
            X