3 Datum Points

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

  • 3 Datum Points

    'm Having a problem. I have a Part with 3 datum points for the Datum A. 2 of the points are 4mm above the one point. There is a perpendicular callout of bolts in the higher points to datum A, not sure how to do this. Not familiar with iterative alignments. Any help would be greatly appreciated.

  • #2
    If your points are square to the coordinate system ( ie 0,0,1) and all share the same vector, then you can construct an offset plane. Select your three points, and then for the two points, set the offset to 4mm and create.

    For help with iterative alignments, do a forum search. There are a multitude of topics discussing iterative alignments.
    "This is my word... and as such is beyond contestation."

    Comment


    • #3
      In this example I have 3 sphere's that are oriented to simulate the datum A target points. I measure the sphere's and offset the basics like this:

      Code:
      SPH1       =FEAT/CONTACT/SPHERE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
                  THEO/<-10.5863,5.75,6.2>,<0,0,1>,0.25
                  ACTL/<-10.5863,5.75,6.2>,<0,0,1>,0.25
                  TARG/<-10.5863,5.75,6.2>,<0,0,1>
                  START ANGLE 1=0,END ANG 1=360
                  START ANGLE 2=0,END ANG 2=90
                  ANGLE VEC=<1,0,0>
                  SHOW FEATURE PARAMETERS=NO
                  SHOW CONTACT PARAMETERS=YES
                    NUMHITS=7,NUMROWS=3
                    SAMPLE HITS=0
                    AVOIDANCE MOVE=NO,DISTANCE=0
                  SHOW HITS=NO
                  MOVE/CLEARPLANE
      SPH2       =FEAT/CONTACT/SPHERE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
                  THEO/<-1.894,2.1784,8.3535>,<0,0,1>,0.25
                  ACTL/<-1.894,2.1784,8.3535>,<0,0,1>,0.25
                  TARG/<-1.894,2.1784,8.3535>,<0,0,1>
                  START ANGLE 1=0,END ANG 1=360
                  START ANGLE 2=0,END ANG 2=90
                  ANGLE VEC=<1,0,0>
                  SHOW FEATURE PARAMETERS=NO
                  SHOW CONTACT PARAMETERS=YES
                    NUMHITS=7,NUMROWS=3
                    SAMPLE HITS=0
                    AVOIDANCE MOVE=NO,DISTANCE=0
                  SHOW HITS=NO
                  MOVE/CLEARPLANE
      SPH3       =FEAT/CONTACT/SPHERE/DEFAULT,CARTESIAN,OUT,LEAST_SQR
                  THEO/<-1.894,9.3217,8.3535>,<0,0,1>,0.25
                  ACTL/<-1.894,9.3217,8.3535>,<0,0,1>,0.25
                  TARG/<-1.894,9.3217,8.3535>,<0,0,1>
                  START ANGLE 1=0,END ANG 1=360
                  START ANGLE 2=0,END ANG 2=90
                  ANGLE VEC=<1,0,0>
                  SHOW FEATURE PARAMETERS=NO
                  SHOW CONTACT PARAMETERS=YES
                    NUMHITS=7,NUMROWS=3
                    SAMPLE HITS=0
                    AVOIDANCE MOVE=NO,DISTANCE=0
                  SHOW HITS=NO
      G_SPH1     =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                  NOM/XYZ,<SPH1.X.NOM,SPH1.Y.NOM,3.8381>,$
                  MEAS/XYZ,<SPH1.X,SPH1.Y,SPH1.Z - 2.3619>,$
                  NOM/IJK,<0,0,1>,$
                  MEAS/IJK,<0,0,1>
      G_SPH2     =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                  NOM/XYZ,<SPH2.X.NOM,SPH2.Y.NOM,3.8381>,$
                  MEAS/XYZ,<SPH2.X,SPH2.Y,SPH2.Z - 4.5154>,$
                  NOM/IJK,<0,0,1>,$
                  MEAS/IJK,<0,0,1>
      G_SPH3     =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                  NOM/XYZ,<SPH3.X.NOM,SPH3.Y.NOM,3.8381>,$
                  MEAS/XYZ,<SPH3.X,SPH3.Y,SPH3.Z - 4.5154>,$
                  NOM/IJK,<0,0,1>,$
                  MEAS/IJK,<0,0,1>
      PL_SPH_1   =FEAT/PLANE,CARTESIAN,TRIANGLE,NO
                  THEO/<-4.7914,5.75,3.8381>,<0,0,1>
                  ACTL/<-4.7914,5.75,3.8381>,<0,0,1>
                  CONSTR/PLANE,BF,G_SPH1,G_SPH2,G_SPH3,,
                  OUTLIER_REMOVAL/OFF,3
                  FILTER/OFF,WAVELENGTH=0
      Then I can just level and origin to the plane.

      Comment


      • #4
        I tried that, just not sure of my results. The part is like a ballooned letter A, with top portion flat then 2 ears raised 4mm with a bolt thru each ear. My lack of confidence is I originally measured a plane on flat surface, offset that 4mm and measured a small plane around each ear then intersected the 3 planes, which I used as my datum for perpendicular. My struggle is I'm getting a lot better numbers that way. This is a prototype part and I sent 6 samples to my customer to evaluate (all samples were .007 out of tolerance) and they said they found 1 that was at the max.....I asked how they were measuring these and all I got back is using 3 points with a offset. I appreciate the feedback just want to see if any other ideas or suggestions. Thanks

        Comment


        • #5
          Here is what I tried, is this correct?
          PNT1 =FEAT/POINT,CARTESIAN
          THEO/<0.297,12.601,0>,<0,0,1>
          ACTL/<0.298,12.6,-0.011>,<0,0,1>
          MEAS/POINT,1,WORKPLANE
          HIT/BASIC,NORMAL,<0.297,12.601,0>,<0,0,1>,<0.298,12.6,-0.011>,USE THEO=YES
          ENDMEAS/
          MOVE/POINT,NORMAL,<-25.319,74.944,13.309>
          PNT2 =FEAT/POINT,CARTESIAN
          THEO/<-25.353,74.939,3.555>,<0,0,1>
          ACTL/<-25.351,74.933,3.574>,<0,0,1>
          MEAS/POINT,1,WORKPLANE
          HIT/BASIC,NORMAL,<-25.353,74.939,3.555>,<0,0,1>,<-25.351,74.933,3.574>,USE THEO=YES
          ENDMEAS/
          MOVE/POINT,NORMAL,<-25.305,84.704,8.106>
          MOVE/POINT,NORMAL,<87.273,80.785,7.660>
          PNT3 =FEAT/POINT,CARTESIAN
          THEO/<87.261,80.784,2.971>,<0,0,1>
          ACTL/<87.263,80.781,2.675>,<0,0,1>
          MEAS/POINT,1,WORKPLANE
          HIT/BASIC,NORMAL,<87.261,80.784,2.971>,<0,0,1>,<87.263 ,80.781,2.675>,USE THEO=YES
          ENDMEAS/
          MOVE/POINT,NORMAL,<87.643,80.854,100.596>
          PLN2 =FEAT/PLANE,CARTESIAN,TRIANGLE,NO
          THEO/<20.765,55.798,4.824>,<0.0111865,-0.1162072,0.993162>
          ACTL/<20.774,55.797,4.728>,<0.0139237,-0.1155679,0.993202>
          CONSTR/PLANE,OFFSET
          ID = PNT1,PNT2,PNT3,,
          OFFSET = 0,4,4

          Comment


          • #6
            Originally posted by Cobe73 View Post
            Here is what I tried, is this correct?
            PNT1 =FEAT/POINT,CARTESIAN
            THEO/<0.297,12.601,0>,<0,0,1>
            ACTL/<0.298,12.6,-0.011>,<0,0,1>
            MEAS/POINT,1,WORKPLANE
            HIT/BASIC,NORMAL,<0.297,12.601,0>,<0,0,1>,<0.298,12.6,-0.011>,USE THEO=YES
            ENDMEAS/
            MOVE/POINT,NORMAL,<-25.319,74.944,13.309>
            PNT2 =FEAT/POINT,CARTESIAN
            THEO/<-25.353,74.939,3.555>,<0,0,1>
            ACTL/<-25.351,74.933,3.574>,<0,0,1>
            MEAS/POINT,1,WORKPLANE
            HIT/BASIC,NORMAL,<-25.353,74.939,3.555>,<0,0,1>,<-25.351,74.933,3.574>,USE THEO=YES
            ENDMEAS/
            MOVE/POINT,NORMAL,<-25.305,84.704,8.106>
            MOVE/POINT,NORMAL,<87.273,80.785,7.660>
            PNT3 =FEAT/POINT,CARTESIAN
            THEO/<87.261,80.784,2.971>,<0,0,1>
            ACTL/<87.263,80.781,2.675>,<0,0,1>
            MEAS/POINT,1,WORKPLANE
            HIT/BASIC,NORMAL,<87.261,80.784,2.971>,<0,0,1>,<87.263 ,80.781,2.675>,USE THEO=YES
            ENDMEAS/
            MOVE/POINT,NORMAL,<87.643,80.854,100.596>
            PLN2 =FEAT/PLANE,CARTESIAN,TRIANGLE,NO
            THEO/<20.765,55.798,4.824>,<0.0111865,-0.1162072,0.993162>
            ACTL/<20.774,55.797,4.728>,<0.0139237,-0.1155679,0.993202>
            CONSTR/PLANE,OFFSET
            ID = PNT1,PNT2,PNT3,,
            OFFSET = 0,4,4
            Why aren't the nominal values for these points 4.000?
            "This is my word... and as such is beyond contestation."

            Comment


            • #7
              When I construct a plane out of the 3 it doesn't give me a option, just allows me to type in the offset for each point

              Comment


              • #8
                This is what I have for the drawing. Any input after looking at this would be a big help.
                IMG_0248.jpg

                Comment


                • #9
                  Originally posted by Cobe73 View Post
                  This is what I have for the drawing. Any input after looking at this would be a big help.
                  [ATTACH=CONFIG]10669[/ATTACH]
                  What Vinni said is correct - but the vector of the constructed plane doesn't look good, and as vinni pointed out the Z values of the points are not 4.0 theoretically.

                  Where are you getting 4mm from?
                  Automettech - Automated Metrology Technology

                  Comment


                  • #10
                    I picked up a plane around large hole, measured hole, and a line to get my manual alignment. Then measure a point at each datum reference, then constructed a plane and offset the 2 higher points 4mm from the large hole surface that was my z-0 from the manual alignment. If anyone can run me thru how they would set it up step by step would really appreciate it, I'm missing something here.

                    Comment


                    • #11
                      Your measured points should have a theoretical value of 4.000 if your basic is 4.000. Seeing what your values are from those measured points, tells me that your part is no good, or your alignment is no good.

                      Are you working with CAD?

                      Why aren't you using Auto Features?

                      Do you know what THEO is, and why it is important?
                      "This is my word... and as such is beyond contestation."

                      Comment


                      • #12
                        No I don't have a cad model, I understand the theos, but the part measures 3.5 mm to the offsets and varies. On the drawing you get 4mm +/-.5 on the offsets. From what you see on the print how would you set this up? I dont know what to try now. This is the 1st time I have used pcdmis, always ran a manual cmm and had different software.

                        Comment


                        • #13
                          I think your plane should be constructed with a MINUS 4 offset for the 2 planes at 4 from the first. You're telling the demon how far the points must move to be on the same level as the one at zero. In this case they need to move -4 from where they are. Try it, it'll work

                          TK

                          PNT1 =FEAT/POINT,RECT
                          THEO/0,0,0,0,0,1
                          ACTL/0,0,0,0,0,1
                          CONSTR/POINT,OFFSET,,0,0,0
                          PNT2 =FEAT/POINT,RECT
                          THEO/10,10,4,0,0,1
                          ACTL/10,10,4,0,0,1
                          CONSTR/POINT,OFFSET,,10,10,4
                          PNT3 =FEAT/POINT,RECT
                          THEO/10,-10,4,0,0,1
                          ACTL/10,-10,4,0,0,1
                          CONSTR/POINT,OFFSET,,10,-10,4
                          PLN1 =FEAT/PLANE,RECT,TRIANGLE
                          THEO/4.828,0,4.598,-0.6896552,0,0.7241379
                          ACTL/4.828,0,4.598,-0.6896552,0,0.7241379
                          CONSTR/PLANE,OFFSET
                          ID = PNT1,PNT2,PNT3,,
                          OFFSET = 0,4,4
                          PLN2 =FEAT/PLANE,RECT,TRIANGLE
                          THEO/6.667,0,0,0,0,1
                          ACTL/6.667,0,0,0,0,1
                          CONSTR/PLANE,OFFSET
                          ID = PNT1,PNT2,PNT3,,
                          OFFSET = 0,-4,-4

                          How are the code tags done again? Enter the code above in a new program and look at the vectors of the constructed planes, look at it from -Y in the graphics window
                          Last edited by tking; 07-16-2015, 12:42 PM.
                          sigpicHave a homebrew

                          Comment


                          • #14
                            Originally posted by tking View Post
                            I think your plane should be constructed with a MINUS 4 offset for the 2 planes at 4 from the first. You're telling the demon how far the points must move to be on the same level as the one at zero. In this case they need to move -4 from where they are. Try it, it'll work

                            TK

                            Code:
                            PNT1       =FEAT/POINT,RECT
                                        THEO/0,0,0,0,0,1
                                        ACTL/0,0,0,0,0,1
                                        CONSTR/POINT,OFFSET,,0,0,0
                            PNT2       =FEAT/POINT,RECT
                                        THEO/10,10,4,0,0,1
                                        ACTL/10,10,4,0,0,1
                                        CONSTR/POINT,OFFSET,,10,10,4
                            PNT3       =FEAT/POINT,RECT
                                        THEO/10,-10,4,0,0,1
                                        ACTL/10,-10,4,0,0,1
                                        CONSTR/POINT,OFFSET,,10,-10,4
                            PLN1       =FEAT/PLANE,RECT,TRIANGLE
                                        THEO/4.828,0,4.598,-0.6896552,0,0.7241379
                                        ACTL/4.828,0,4.598,-0.6896552,0,0.7241379
                                        CONSTR/PLANE,OFFSET
                                        ID = PNT1,PNT2,PNT3,,
                                        OFFSET = 0,4,4
                            PLN2       =FEAT/PLANE,RECT,TRIANGLE
                                        THEO/6.667,0,0,0,0,1
                                        ACTL/6.667,0,0,0,0,1
                                        CONSTR/PLANE,OFFSET
                                        ID = PNT1,PNT2,PNT3,,
                                        OFFSET = 0,-4,-4
                            How are the code tags done again? Enter the code above in a new program and look at the vectors of the constructed planes, look at it from -Y in the graphics window
                            [ code ]*code sample*[ /code ] No spaces in the brackets.

                            Nice catch with the 4 / -4 thing.
                            "This is my word... and as such is beyond contestation."

                            Comment


                            • #15
                              Originally posted by VinniUSMC View Post
                              [ code ]*code sample*[ /code ] No spaces in the brackets.

                              Nice catch with the 4 / -4 thing.
                              Maybe just maybe I'll be able to remember the code tags someday. Naw, I'm just getting older everyday, ain't gonna happen I'll depend on the younger ones to remind me.

                              Thanks,

                              TK
                              sigpicHave a homebrew

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X