Surface Profile - Catastrophe?

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

  • Surface Profile - Catastrophe?

    To preface my question/comments.... I have looked though this forum for information on this topic and have come to the conclusion that the situation I am in may be worse then I had suspected.

    After working in an area of inspection in our company for some time I inherited it by default. The other programmer quit and I was left holding the bag.

    What is in the Bag:
    Die inspection

    Process: (which has been locked into place as an established process) Align to the top face/front edge(which is machined)/end(which is also machined) in a 3-2-1 structure. After which the impression is inspected using section or linear open scans that are dimensioned as a surface profile.

    Results: Are inconclusive at best. This has not bred much confidence in the whole inspection process. Form and location can look ok for most of the profile and some of the points magnified and distorted or completely distorted. Note - the Deviation and OT are reported. which are not a good representation of how it is out of tolerance. Form only also can show some major distortion or results. By this I mean the graphical analysis view shows some of the profile points as way out of tolerance bands. Since taking over this area I have changed some of the Form Only profiles from Least squares to Vector Least squares. This has improved how the profile is represented in the Analysis view a the results in general.

    The dies designs have been changing and the weakness of the to date inspection process is really showing. What is needed is the most accurate measurement of the impressions surface profiles and well as impression position on the Die.

    What I have tried is to measure the impressions as is normally done as well as reporting the position of points within each scan section. After reading other posts I am uncertain that this is a viable, or even repeatable, way to go about this. See Code Below.

    Bottom line: From what I have been reading, we have not been going about measuring and reporting the surfaces correctly. Simple sentence, but it carries the weight of upwards to 10,000 or more programs.

    Goal: Appropriately/accurately measure and report the surface profile of each section of the impression.(perhaps an overall profile as well) Report the position of the impression to the die "matched edge". Ideally: be able to create a report of the position of the impression to the other half of the die set.(This is a future goal and haven't really looked into this yet)


    I know this has been long, but I am in need of insight from some skilled programmers.

    Thank you in advance!!

    Regards,

    David

  • #2
    Code:
    REE        =FEAT/SCAN,SECTIONSCAN,NUMBER OF HITS=59
                ,SHOW HITS=NO,SHOWALLPARAMS=NO
                MEAS/SCAN
                BASICSCAN/LINE,NUMBER OF HITS=59,SHOW HITS=NO,SHOWALLPARAMS=NO
                ENDSCAN
                ENDMEAS/
    DIM R_EE= PROFILE OF SURFACE OF SET REE    FORMANDLOCATION  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  ARROWDENSITY=100  OUTPUT=BOTH
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    M       0.0000     0.0000     0.0010     0.0010     0.0000     0.0000 ----#----
                ANALYSISVIEW/R_EE,,
                MOVE/CLEARPLANE
    RSS        =FEAT/SCAN,SECTIONSCAN,NUMBER OF HITS=49
                ,SHOW HITS=NO,SHOWALLPARAMS=NO
                MEAS/SCAN
                BASICSCAN/LINE,NUMBER OF HITS=49,SHOW HITS=NO,SHOWALLPARAMS=NO
                ENDSCAN
                ENDMEAS/
    DIM R_SS= PROFILE OF SURFACE OF SET RSS    FORMANDLOCATION  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  ARROWDENSITY=100  OUTPUT=BOTH
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    M       0.0000     0.0000     0.0010     0.0010     0.0000     0.0000 ----#----
                MOVE/CLEARPLANE
    RDD        =FEAT/SCAN,SECTIONSCAN,NUMBER OF HITS=48
                ,SHOW HITS=NO,SHOWALLPARAMS=NO
                MEAS/SCAN
                BASICSCAN/LINE,NUMBER OF HITS=48,SHOW HITS=NO,SHOWALLPARAMS=NO
                ENDSCAN
                ENDMEAS/
    DIM R_DD= PROFILE OF SURFACE OF SET RDD    FORMANDLOCATION  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  ARROWDENSITY=100  OUTPUT=BOTH
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    M       0.0000     0.0000     0.0010     0.0010     0.0000     0.0000 ----#----
                ANALYSISVIEW/R_SS,R_DD,,
                MOVE/CLEARPLANE
    RCC        =FEAT/SCAN,SECTIONSCAN,NUMBER OF HITS=53
                ,SHOW HITS=NO,SHOWALLPARAMS=NO
                MEAS/SCAN
                BASICSCAN/LINE,NUMBER OF HITS=53,SHOW HITS=NO,SHOWALLPARAMS=NO
                ENDSCAN
                ENDMEAS/
    DIM R_CC= PROFILE OF SURFACE OF SET RCC    FORMANDLOCATION  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  ARROWDENSITY=100  OUTPUT=BOTH
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    M       0.0000     0.0000     0.0010     0.0010     0.0000     0.0000 ----#----
                MOVE/CLEARPLANE
    RBB        =FEAT/SCAN,SECTIONSCAN,NUMBER OF HITS=56
                ,SHOW HITS=NO,SHOWALLPARAMS=NO
                MEAS/SCAN
                BASICSCAN/LINE,NUMBER OF HITS=56,SHOW HITS=NO,SHOWALLPARAMS=NO
                ENDSCAN
                ENDMEAS/
    DIM R_BB= PROFILE OF SURFACE OF SET RBB    FORMANDLOCATION  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  ARROWDENSITY=100  OUTPUT=BOTH
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    M       0.0000     0.0000     0.0010     0.0010     0.0000     0.0000 ----#----
                ANALYSISVIEW/R_CC,R_BB,,
                MOVE/CLEARPLANE
    ROAL       =FEAT/SCAN,SECTIONSCAN,NUMBER OF HITS=300
                ,SHOW HITS=NO,SHOWALLPARAMS=NO
                MEAS/SCAN
                BASICSCAN/LINE,NUMBER OF HITS=300,SHOW HITS=NO,SHOWALLPARAMS=NO
                ENDSCAN
                ENDMEAS/
                FORMFEED/
    DIM R_OAL= PROFILE OF SURFACE OF SET ROAL    FORMANDLOCATION  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=25.00  ARROWDENSITY=100  OUTPUT=BOTH
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    M       0.0000     0.0000     0.0010     0.0010     0.0000     0.0000 ----#----
                ANALYSISVIEW/R_OAL,,
                FORMFEED/
                RECALL/ALIGNMENT,INTERNAL,Dcc
                ASSIGN/EEMINPTZ=MININDEX(REE.HIT[1..REE.NUMHITS].Z)
    REEZ       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,0,REE.HIT[EEMINPTZ].Z>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM REE_Z= LOCATION OF POINT REEZ  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Z      -1.3240    -1.3240     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION REE_Z
                ASSIGN/EEMINPTY=MAXINDEX(REE.HIT[1..REE.NUMHITS].Y)
    REEY       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,REE.HIT[EEMINPTY].Y,0>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM REE_Y= LOCATION OF POINT REEY  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Y       5.1277     5.1277     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION REE_Y
                ASSIGN/SSMINPTZ=MAXINDEX(RSS.HIT[1..RSS.NUMHITS].Z)
    RSSZ       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,0,RSS.HIT[SSMINPTZ].Z>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RSS_Z= LOCATION OF POINT RSSZ  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Z      -0.8650    -0.8650     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION RSS_Z
                ASSIGN/SSMINPTY=MAXINDEX(RSS.HIT[1..RSS.NUMHITS].Y)
    RSSY       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,RSS.HIT[SSMINPTY].Y,0>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RSS_Y= LOCATION OF POINT RSSY  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Y       5.1354     5.1354     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION RSS_Y
                ASSIGN/DDMINPTZ=MAXINDEX(RDD.HIT[1..RDD.NUMHITS].Z)
    RDDZ       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,0,RDD.HIT[DDMINPTZ].Z>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RDD_Z= LOCATION OF POINT RDDZ  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Z      -0.5456    -0.5456     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION RDD_Z
                ASSIGN/DDMINPTY=MAXINDEX(RDD.HIT[1..RDD.NUMHITS].Y)
    RDDY       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,RDD.HIT[DDMINPTY].Y,0>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RDD_Y= LOCATION OF POINT RDDY  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Y       5.1281     5.1281     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION RDD_Y
                ASSIGN/CCMINPTZ=MAXINDEX(RCC.HIT[1..RCC.NUMHITS].Z)
    RCCZ       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,0,RCC.HIT[CCMINPTZ].Z>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RCC_Z= LOCATION OF POINT RCCZ  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Z      -0.2358    -0.2358     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION RCC_Z
                ASSIGN/CCMINPTY=MAXINDEX(RCC.HIT[1..RCC.NUMHITS].Y)
    RCCY       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,RCC.HIT[CCMINPTY].Y,0>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RCC_Y= LOCATION OF POINT RCCY  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Y       5.2797     5.2799     0.0010     0.0010     0.0002     0.0000 -----#---
    END OF DIMENSION RCC_Y
                ASSIGN/BBMINPTZ=MAXINDEX(RBB.HIT[1..RBB.NUMHITS].Z)
    RBBZ       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,0,RBB.HIT[BBMINPTZ].Z>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RBB_Z= LOCATION OF POINT RBBZ  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Z      -0.0233    -0.0233     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION RBB_Z
                ASSIGN/BBMINPTY=MAXINDEX(RBB.HIT[1..RBB.NUMHITS].Y)
    RBBY       =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,RBB.HIT[BBMINPTY].Y,0>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM RBB_Y= LOCATION OF POINT RBBY  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Y       5.5161     5.5177     0.0010     0.0010     0.0016     0.0006 -------->
    END OF DIMENSION RBB_Y
                ASSIGN/ROALMINPTZ=MININDEX(ROAL.HIT[1..ROAL.NUMHITS].Z)
    ROALZ      =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<0,0,ROAL.HIT[ROALMINPTZ].Z>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM ROAL_Z= LOCATION OF POINT ROALZ  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    Z      -1.3241    -1.3241     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION ROAL_Z
    ROALX      =GENERIC/POINT,DEPENDENT,CARTESIAN,$
                NOM/XYZ,<0,0,0>,$
                MEAS/XYZ,<ROAL.HIT[ROALMINPTZ].X,0,0>,$
                NOM/IJK,<0,0,1>,$
                MEAS/IJK,<0,0,1>
    DIM ROAL_X= LOCATION OF POINT ROALX  UNITS=IN ,$
    GRAPH=OFF  TEXT=OFF  MULT=10.00  OUTPUT=BOTH  HALF ANGLE=NO
    AX    NOMINAL       MEAS       +TOL       -TOL        DEV     OUTTOL
    X      -3.0010    -3.0010     0.0010     0.0010     0.0000     0.0000 ----#----
    END OF DIMENSION ROAL_X
                RECALL/ALIGNMENT,INTERNAL,Main

    Comment


    • #3
      I think that we need a little more information, what is the actual Profile callout? IS the surfaces you trying to measure are flat or curved?
      B&S One
      PC-DMIS CAD v2014

      Romer Infinity

      Comment


      • #4
        The tolerance has been more to a process then a print call out. ±0.0125 of the smallest tolerance on the print. I suppose this was done to allow a percentage of the final tolerance as much as not having to create a separate print or callout for this inspection. It went from all programs as form and location to some being form only. The surfaces are curved(medical & aero).

        Comment


        • #5
          What type of probing system are you using?
          Do you need to filter the scans to remove noise?

          Comment


          • #6
            Renishaw PH10/SP25/SM25-2. There Is some rough surfaces due to benching, but that is an occasional occurrence.

            Comment


            • #7
              So from what I can gather, the routine's extracted Min/Max point indices of the scans are picking up outliers or data that you believe to be erroneous, due to benching or rough surface?
              Is your question of uncertainty related to the location points, the profile measurements, or both?

              If it's both- you likely have alignment issues, or like you are saying a random min/max index of the whole scan might not be appropriate/viable way to extract the values intended.

              What I mean by this, is if you are trying to extract max Z of a specific spot on the part, and are using
              Code:
              ASSIGN/CCMINPTZ=MAXINDEX(RCC.HIT[[B]1..RCC.NUMHITS[/B]].Z)
              , you are assessing the entire scan, instead of picking up only the 5 or 6 possible points that are truly going to tell you the max z point. If one of the points along the whole scan is in a blend radius and the probe jumps, your assignment picks up the blend radius error point, instead of the real max z of the part. if you simply narrow your assignment to just those specific points that the max Z can fall within, it should mitigate picking up erroneous data.
              Code:
              ASSIGN/CCMINPTZ=MAXINDEX(RCC.HIT[[B]15..22[/B]].Z)
              If it's one or the other, let's focus your strategy there.

              another concern is the point density. if your scans are only 50 or 60 points, you might be missing the true tangent or max value of a curved part. You might want to increase scan point density.

              Comment


              • #8
                louisd - very good points!

                Makes me want to go back to the beginning and ask... What is the best and most appropriate way to measure and a surface profile while giving the location of either that measured profile or the position of the entire impression of a die?

                Damb - any way I write this it comes out to bloody vague.

                Comment

                Related Topics

                Collapse

                Working...
                X