Min opening . . .

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

  • Min opening . . .

    OK, I explained this a couple ways in the past. Didn't seem to make clear what I need. I'm back on the project and once again trying to resolve this issue.
    I'm doing this all off line on a CAD model. Not an actual part. So I can compare results to print, etc.

    So . . . Let's say you have a tennis ball sitting in the lower left corner, and a basketball in the upper right corner.

    I need to know the shortest distance between the farthest point on the tennis ball in X (farthest point to the right), to the closest point to the basketball surface, surface to surface, not center to center. PLUS . . . they aren't really round. Nearly round. (Actual part, 2 curved surfaces. See attached sketch)

    So . . . I did work on this. I was able to do a couple things.

    1). I wrote a little code . . . finds the highest point (point in "X" that's farthest to the right.) of 20 measured points on a curve. THIS is my origin. Highest point on my trailing edge in X direction!

    Problem 1 . . . I need the Y & Z value of that same point. Yeah, I got the "value" of the highest X point. But I need to create a generic point using that X valus, AND the YZ value for that same point. (See sketch)

    I scroll up. Find the point that had the greatest X value. Let's say that point was at . . .

    X .1250, Y .2511, Z 1.5000

    I need to "match up" the values. My code "found" the .1250 to be the greatest value of the points in X, but I need to associate that now with the other two values.

    So when I make the generic point, I can use the value of the variable for X, and also the real value for Y & Z for the point that was highest at the X+.. (Make sense?)

    Basically, hit points along the surface, and find the XYZ value of THE 1 point farthest in X+.

    Problem 2 . . . Then I need to get the shortest distance between that point, and the curved surface in the upper right corner.

    I scanned, and converted to points, along that surface, and know I covered where that shortest point is. (I made a series of lines, and 2d/3d measurements between the high point on leading edge to the points I measured on curve 2 to prove it out.)

    Now I need to make it programmatic.

    To prove it out, I found my high points in my variable statements and confirmed them looking at the actual points measured. Then set my origin at the high point.

    But I need to do it programmatically.

    Any help would be much appreciated.

    Wayne
    Attached Files

  • #2
    Well, it seems to me...

    lets put the tennis ball on our left and the basketball on our right. You want the distance from the rightmost point on the tennis ball to the leftmost point on the basketball.

    If I understood correctly, what I would do is measure them each as a sphere. Then, create a 3d line from center to center. Now, create a pierce point where the line pierces the tennisball's right surface and where it pierces the basketball's left surface. Now, just dimension from one point to the other (center to center).

    Maybe I'm not getting it. Also, that method doesn't account for out of round conditions.
    Last edited by #2#; 07-02-2007, 12:15 PM.

    Comment


    • #3
      When you say 'shortest point' it isn't exactly clear if you mean 'direct' or of you mean in a particular Axis. I say this because you chose the faor right point on the 'tennis ball' as the start point. This would indicate that you want the closest in a particular axis and not necessarily a direct, closest point.
      Bill Jarrells
      A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

      Comment


      • #4
        Min opening . . .

        Here lies the problem. Yeah, it's the farthest point on the "tennis ball" in X. That's my new origin. But on the "basketball", it's not the farthest point to anything. I need to know the smallest distance from the new origin to the "basketball". Look at the attached drawing.

        So, from my new origin, which I can't set until I get part one of this resolved (pulling the X & Z values of the same point it finds as Y high point) . . . I need to know the distance to the points measured on the "basketball", and then determin the shortest of the distances between new or9i9gin and point on the "ball".

        In more technical terms . . . I need to measure the shortest distance from the high point on the trailing edge to the closest point at the same height, on the opposing blade. The DO, or Min opening of the vane.

        Comment


        • #5
          Could you not do variables to find your min point and then a 3d distance to that point. What else am i missing
          sigpic.....Its called golf because all the other 4 letter words were taken

          Comment


          • #6
            You need to post a drawing of this requirement and what the alignments are. You drawing is not clear.
            Links to my utilities for PCDMIS

            Comment


            • #7
              BW, yer close to the issue . . . cmmguy, you must not understand my question. The drawing is SO clear. 2 curves. One on left, one on right.I need the distance from the farthest point in X+ on the lower curve, to the CLOSEST point on the opposing curve surface.

              Look at it this way . . . I'm sticking a gage pin thru the vane opening to measure the smallest opening size. The pin will ultimately contact the trailing edge, and "some point" on the wall of the opposing vane. What would that pin size be?

              Comment


              • #8
                Originally posted by thenettrainer View Post
                Here lies the problem. Yeah, it's the farthest point on the "tennis ball" in X. That's my new origin. But on the "basketball", it's not the farthest point to anything. I need to know the smallest distance from the new origin to the "basketball". Look at the attached drawing.

                So, from my new origin, which I can't set until I get part one of this resolved (pulling the X & Z values of the same point it finds as Y high point) . . . I need to know the distance to the points measured on the "basketball", and then determin the shortest of the distances between new or9i9gin and point on the "ball".

                In more technical terms . . . I need to measure the shortest distance from the high point on the trailing edge to the closest point at the same height, on the opposing blade. The DO, or Min opening of the vane.
                Sounds like a GE Part - LOL

                I understand what you are up against. I had to do this for Turbine Blades with another CMM Software.

                1) You should have been able to capture the X, Y, and Z of the highest X Point as those were available and depending on how you extracted the highest Point you should have captured it. Try indexing to it with array. If all else fails brute force will work. Sort for highest and capture X, Y, and Z as you go, replacing them as you need to.
                2) From there can't you just grab Radial Points sweeping the Nominal Closest Point and points before and after the nominakl closest point and capture the Shortest Radii Point?
                Bill Jarrells
                A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

                Comment


                • #9
                  Originally posted by thenettrainer View Post
                  BW, yer close to the issue . . . cmmguy, you must not understand my question. The drawing is SO clear. 2 curves. One on left, one on right.I need the distance from the farthest point in X+ on the lower curve, to the CLOSEST point on the opposing curve surface.

                  Look at it this way . . . I'm sticking a gage pin thru the vane opening to measure the smallest opening size. The pin will ultimately contact the trailing edge, and "some point" on the wall of the opposing vane. What would that pin size be?
                  Well it wouldnt be the first time...

                  The furthest point in X, when rotated up to find the closest point in 3D would not be the distance that would allow a pint to pass. As you rotate, that original point would give you a larger distance than what would pass a pin.
                  Links to my utilities for PCDMIS

                  Comment


                  • #10
                    seems like i would construct a line between the two spheres make that my x-axis>use an array to capture x values>use variables to sort>3d distance

                    Comment


                    • #11
                      Min opening

                      Look at the new attached sketch . . .
                      Attached Files

                      Comment


                      • #12
                        Originally posted by thenettrainer View Post
                        Look at the new attached sketch . . .
                        Did this all the time for GE Blades.
                        Please see earlier post. You should be able to capture the X,Y, and Z of the Origin Point with Variables.
                        From there it is a simple sweep.
                        Bill Jarrells
                        A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

                        Comment


                        • #13
                          Originally posted by thenettrainer View Post
                          Look at the new attached sketch . . .
                          You said smallest opening for a pin to pass and that is where you lose me...
                          Links to my utilities for PCDMIS

                          Comment


                          • #14
                            He is looking for the DO measurement of a blade.
                            Bill Jarrells
                            A lie can travel half way around the world while the truth is putting on its shoes. - Mark Twain

                            Comment


                            • #15
                              Doh! Ya think! LOL

                              First of all, thanks for your help thru this. I really appreciate it.

                              Yeah, calculating the DO. You don't even want to know how their old programmer was doing it! UGGHHHH!

                              Anywho . . . I'm trying to do it programmatically. Instead of visually scanning for high points, etc.

                              Wingman . . . THAT was my original question. HOW? As I said, I was able to use a variable routine to capture the highest "Y" point from a group of points.

                              But I can't figure out how to get the actual "X & Z" from that SAME Y point.

                              As in . . .

                              I measure 20 points. The variable routine lets me know my largest Y+ is . . .

                              SY1M18 =AUTO/VECTOR POINT,SHOWALLPARAMS = YES
                              THEO/0.5736,0.4623,-4.381,-0.2159392,0.9756726,-0.0378584
                              ACTL/0.5736,0.4623,-4.381,-0.2159392,0.9756726,-0.0378584
                              TARG/0.5736,0.4623,-4.381,-0.2159392,0.9756726,-0.0378584
                              THEO_THICKNESS = 0,RECT,SNAP = YES,$
                              AUTO MOVE = NO,DISTANCE = 0

                              So I run the program. My variable value for highest Y is +0.4623.

                              Super! Wonderful. That's great! But . . .

                              I have no idea what the X & Z value is for the point measured as highest in Y!!!! In this example, I know that SY1M18 contain s the highest Y value.

                              SO . . . What I need to do is . . . make SY1M18 my generic point. Or a point with the same value as SY1M18.



                              How do I get a generic point made with the Hi Y point (I can do this part!) AND the X & Z value for SY1M18?

                              I can assign that variable, and can plug in the Y value . . .

                              HP_Y1 =GENERIC/POINT,DEPENDENT,RECT,$
                              NOM/XYZ,0,SY1M,0,$
                              MEAS/XYZ,0,SY1M_M,-4.381,$
                              NOM/IJK,0,0,1,$
                              MEAS/IJK,0,0,1

                              But how do I associate this Y value with the X & Z of that same point.

                              AS in . . . If the Y high pt is for example, the 8th point in the group (or array), how do I pull the X & Z value out of the same array to match the Y value, so I have THE high point values.

                              Then I can create a gen point from those XYZ values.

                              Summary: I measured 20 some points. Found point SY1M18 has highest Y value.

                              So I want to use the XY & Z values of SY1M18 as a gen point.

                              THEN . . . I want to "compare" that gen point to 20 some poinhts measured on opposing vane, and fine shortest distance.

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X