Repeatibility of measured angle and center of slot

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

  • Repeatibility of measured angle and center of slot

    The repeatibility of the angle I measured and the center of a slot is nowhere near what I hoped. Even the same part, I measured twice the angle values of the two measurement will be more than half degree. Is this normal? So is the center of a slot.

  • #2
    first off what had you hoped? if you are looking for 2nd or 3rd decimal place you may be chasing your tail. that being said a lot depends on the condition of your part, the accuracy of your machine and the precision of your program. Meaning PCDMIS will report angles all day long if the information plugged into it is good.
    Last edited by Ironhoe; 07-06-2010, 02:22 PM. Reason: speeling

    Comment


    • #3
      Originally posted by Heming View Post
      The repeatibility of the angle I measured and the center of a slot is nowhere near what I hoped. Even the same part, I measured twice the angle values of the two measurement will be more than half degree. Is this normal? So is the center of a slot.
      You mentioned 2 measurements.
      Were they both DCC?
      Were they both with exactly the same probed points?
      Were they both with the same part load?
      Lately, it occurs to me
      What a long, strange trip it's been.

      2017 R1 (Offline programming)

      Comment


      • #4
        Originally posted by Heming View Post
        The repeatibility of the angle I measured and the center of a slot is nowhere near what I hoped. Even the same part, I measured twice the angle values of the two measurement will be more than half degree. Is this normal? So is the center of a slot.
        Is the slot as cast or machined? If machined is it with one pass or each wall a separate pass? Are you using a program to measure this slot or using manual hits?

        If each wall is machined separately the center of each arc will not be the same and your call out for the slot will change with the changing of which arc you build your slot from. Play with your call out until you get the results you expect then run the same part several times. The variances in your output should be with in your machine capabilities.

        Comment


        • #5
          My part is CNC machined and should hold the dimension pretty well. So for CMM part, I am using a Shilfield Discover III. It's not a high precision CMM but should repeat itself. Therefore, the problem will be my measurement. I don't know how to increase the accuracy of my measurement. The angle is a flat area have 20 degree to X plane. It is really an straightforward measurement.

          Comment


          • #6
            All my final measurement is carried out by DCC. I only do the initial allignment manually.

            Comment


            • #7
              DCC measurement, angle does not repeat within 1/2 degree on same part?

              Sounds like alignment is not correct.

              B&S CHAMELEON/PCDMIS CAD++ V2011

              There are no bugs, only "UNDOCUMENTED ENHANCEMENTS!"

              sigpic

              Comment


              • #8
                Originally posted by Heming View Post
                All my final measurement is carried out by DCC. I only do the initial allignment manually.
                So, were the 2 different results the result of running the program complete (through the angle reporting) including the manual alignment?

                The manual alignment process can dramatically effect the DCC alignment. For instance, If you run the manual points VERY fast, or have grossly incorrect approach vectors, or hit on sharp edges, all of that will affect where the DCC points end up probing.

                If you only ran the manual alignment once, and then ran through the DCC twice (using CTRL+U) you should see similar results for both measurements.

                There are many reasons for non-repeatability.
                We can only make guesses (and we can keep guessing all day) if we do not have more info (like code) to review.
                Lately, it occurs to me
                What a long, strange trip it's been.

                2017 R1 (Offline programming)

                Comment


                • #9
                  I measured the top face (which is a thin ring of a cylinder) as my Z plane and the ID to get my X and Y zero and a flat along the X to rotate the part. I tried to make sure the face measured correctlly. I don't know where else I should look at for the allignment.

                  Comment


                  • #10
                    Originally posted by Heming View Post
                    All my final measurement is carried out by DCC. I only do the initial allignment manually.
                    Originally posted by Heming View Post
                    I measured the top face (which is a thin ring of a cylinder) as my Z plane and the ID to get my X and Y zero and a flat along the X to rotate the part. I tried to make sure the face measured correctlly. I don't know where else I should look at for the allignment.
                    Originally posted by John Riggins View Post
                    So, were the 2 different results the result of running the program complete (through the angle reporting) including the manual alignment?

                    The manual alignment process can dramatically effect the DCC alignment. For instance, If you run the manual points VERY fast, or have grossly incorrect approach vectors, or hit on sharp edges, all of that will affect where the DCC points end up probing.

                    If you only ran the manual alignment once, and then ran through the DCC twice (using CTRL+U) you should see similar results for both measurements.

                    There are many reasons for non-repeatability.
                    We can only make guesses (and we can keep guessing all day) if we do not have more info (like code) to review.
                    So you mention that you do a manual alignment and then dcc measurement, Did you follow up the manual alignment with one done in DCC mode?

                    This will play an important part in how accurate your results will be as well as repeatability as John said.
                    Which one gets ridden today? MPH vs MPG..tough choice, both are FUN
                    sigpic

                    Starrett RGDC 4028-24 :alien:
                    Demon vintages 3.7, 4.1, 4.2, 4.3, 2009

                    Comment


                    • #11
                      STARTUP =ALIGNMENT/START,RECALL:,LIST=YES
                      ALIGNMENT/END
                      MODE/MANUAL
                      FLY/ON,0.1181
                      LOADPROBE/PH10-2MM
                      TIP/T1A0B0, SHANKIJK=0, 0, 1, ANGLE=0
                      TOPFACE =FEAT/PLANE,CARTESIAN,TRIANGLE
                      THEO/<0,0,0>,<0,0,1>
                      ACTL/<14.0164,11.8551,-12.6728>,<0.0010493,-0.0000084,0.9999994>
                      MEAS/PLANE,4
                      HIT/BASIC,NORMAL,<0,0,0>,<0,0,1>,<14.0615,12.3753,-12.6729>,USE THEO=YES
                      HIT/BASIC,NORMAL,<0,0,0>,<0,0,1>,<14.6063,11.8772,-12.6734>,USE THEO=YES
                      HIT/BASIC,NORMAL,<0,0,0>,<0,0,1>,<13.9235,11.247,-12.6727>,USE THEO=YES
                      HIT/BASIC,NORMAL,<0,0,0>,<0,0,1>,<13.4744,11.9207,-12.6723>,USE THEO=YES
                      ENDMEAS/
                      BORE =FEAT/CIRCLE,CARTESIAN,IN,LEAST_SQR
                      THEO/<14.0457,11.7851,-13.1793>,<0,0,1>,1.0033
                      ACTL/<14.0509,11.7879,-12.8061>,<0,0,1>,1.0052
                      MEAS/CIRCLE,4,ZPLUS
                      HIT/BASIC,NORMAL,<14.0345,11.2835,-13.181>,<0.0223701,0.9997498,0>,<13.5494,11.7587,-12.8069>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<14.5456,11.8262,-13.1799>,<-0.9966366,-0.0819481,0>,<14.0458,12.2907,-12.8068>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<14.0273,12.2865,-13.1787>,<0.0368758,-0.9993199,0>,<14.5517,11.7466,-12.8053>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<13.5445,11.8032,-13.1775>,<0.9993456,-0.0361704,0>,<14.0271,11.2857,-12.8052>,USE THEO=YES
                      ENDMEAS/
                      MOVE/POINT,NORMAL,<14.5809,11.1915,-6.2704>
                      TIP/T1A90B0, SHANKIJK=0, 1, 0, ANGLE=180
                      XAXIS =FEAT/LINE,CARTESIAN,UNBOUNDED
                      THEO/<13.8526,12.4385,-13.8067>,<1,0,0>
                      ACTL/<13.7755,12.4225,-13.5748>,<0.9999999,0,-0.0004115>
                      MEAS/LINE,3,YPLUS
                      HIT/BASIC,NORMAL,<13.8519,12.403,-13.7676>,<0.0173723,0.9998491,0>,<13.7755,12.4206,-13.5748>,USE THEO=YES
                      HIT/BASIC,NORMAL,<14.0613,12.3994,-13.7676>,<0.0173723,0.9998491,0>,<14.079,12.4227,-13.4968>,USE THEO=YES
                      HIT/BASIC,NORMAL,<14.2708,12.3958,-13.7676>,<0.0173723,0.9998491,0>,<14.2617,12.4241,-13.575>,USE THEO=YES
                      ENDMEAS/
                      ALIGN1 =ALIGNMENT/START,RECALL:STARTUP,LIST=YES
                      ALIGNMENT/LEVEL,ZPLUS,TOPFACE
                      ALIGNMENT/ROTATE,XPLUS,TO,XAXIS,ABOUT,ZPLUS
                      ALIGNMENT/TRANS,XAXIS,BORE
                      ALIGNMENT/TRANS,YAXIS,BORE
                      ALIGNMENT/TRANS,ZAXIS,TOPFACE
                      ALIGNMENT/END
                      TIP/T1A0B0, SHANKIJK=-0.0005, -0.0018, 1, ANGLE=-179.9318
                      MODE/DCC
                      MOVE/POINT,NORMAL,<0.0133,-0.4807,0.9071>
                      WORKPLANE/ZPLUS
                      PART NUMBER=15241 DATE=7/6/2010 TIME=3:24:41 PM PAGE#=2
                      FTF =FEAT/PLANE,CARTESIAN,TRIANGLE
                      THEO/<0,0.0025,0>,<0,0,1>
                      ACTL/<0.0001,0.0073,0>,<-0.0010132,0.000503,0.9999994>
                      MEAS/PLANE,4
                      HIT/VECTOR,<0,0.579,0>,<0,0,1>,<0,0.5789,-0.0003>
                      HIT/VECTOR,<0.56,0,0>,<0,0,1>,<0.5597,0.0001,0.0006>
                      HIT/VECTOR,<0,-0.55,0>,<0,0,1>,<0.0004,-0.5499,0.0002>
                      HIT/VECTOR,<-0.5599,0,0>,<0,0,1>,<-0.5596,-0.0001,-0.0006>
                      ENDMEAS/
                      MOVE/POINT,NORMAL,<-0.0692,-0.01,0.1384>
                      FID =FEAT/CYLINDER,CARTESIAN,IN,LEAST_SQR
                      THEO/<0.0001,0,-0.16>,<0,0,1>,1.006,0.12
                      ACTL/<-0.0046,-0.0002,-0.1592>,<-0.0023584,0.0001221,0.9999972>,1.0053,0.1223
                      MEAS/CYLINDER,8
                      WORKPLANE/YMINUS
                      HIT/BASIC,NORMAL,<0.503,-0.0087,-0.22>,<-0.9998491,0.0173723,0>,<0.4981,-0.0078,-0.2191>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<0.0088,0.5029,-0.22>,<-0.0173723,-0.9998491,0>,<0.0083,0.5023,-0.22>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<-0.5029,0.0087,-0.22>,<0.9998491,-0.0173723,0>,<-0.507,0.0089,-0.2201>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<-0.0087,-0.5029,-0.22>,<0.0173723,0.9998491,0>,<-0.0063,-0.5028,-0.2184>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<-0.0087,-0.5029,-0.1>,<0.0173723,0.9998491,0>,<-0.0066,-0.5028,-0.0995>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<-0.5029,0.0087,-0.1>,<0.9998491,-0.0173723,0>,<-0.5074,0.0076,-0.0992>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<0.0088,0.5029,-0.1>,<-0.0173723,-0.9998491,0>,<0.0079,0.5023,-0.0988>,USE THEO=YES
                      MOVE/CIRCULAR
                      HIT/BASIC,NORMAL,<0.503,-0.0087,-0.1>,<-0.9998491,0.0173723,0>,<0.4978,-0.0073,-0.0986>,USE THEO=YES
                      ENDMEAS/
                      MOVE/POINT,NORMAL,<0.1456,0.2775,0.2507>
                      MOVE/POINT,NORMAL,<-0.4477,6.1657,-0.8176>
                      TIP/T1A90B0, SHANKIJK=0, 1, 0.0048, ANGLE=179.8143
                      MOVE/POINT,NORMAL,<-0.1503,0.9,-0.68>
                      WORKPLANE/YPLUS
                      LINEX =FEAT/LINE,CARTESIAN,UNBOUNDED
                      THEO/<-0.15,0.6391,-0.7409>,<1,0,0>
                      ACTL/<-0.15,0.6389,-0.7402>,<0.9999999,0,-0.000354>
                      MEAS/LINE,3,WORKPLANE
                      HIT/VECTOR,<-0.15,0.6,-0.78>,<0,1,0>,<-0.15,0.599,-0.78>
                      HIT/VECTOR,<0,0.6,-0.78>,<0,1,0>,<0,0.5998,-0.78>
                      HIT/VECTOR,<0.15,0.6,-0.78>,<0,1,0>,<0.15,0.6007,-0.78>
                      ENDMEAS/
                      MOVE/POINT,NORMAL,<0,1.7,1.1579>
                      TIP/T1A0B0, SHANKIJK=-0.0032, -0.0048, 1, ANGLE=0.0003
                      MOVE/POINT,NORMAL,<0.8684,0.2541,0.4819>
                      WORKPLANE/ZPLUS
                      FINAL =ALIGNMENT/START,RECALL:ALIGN1,LIST=YES
                      ALIGNMENT/LEVEL,ZPLUS,FTF
                      ALIGNMENT/ROTATE,XPLUS,TO,LINEX,ABOUT,ZPLUS
                      ALIGNMENT/TRANS,XAXIS,FID
                      ALIGNMENT/TRANS,YAXIS,FID
                      ALIGNMENT/TRANS,ZAXIS,FTF
                      ALIGNMENT/END
                      MOVE/POINT,NORMAL,<0.8539,0.6394,-0.4477>
                      MOVE/POINT,NORMAL,<0.8726,0.0251,-0.52>

                      20DEGLEFT =FEAT/PLANE,CARTESIAN,TRIANGLE
                      THEO/<-0.515,0.1858,-2.825>,<-0.9396976,0.3420064,0.0000105>
                      ACTL/<-0.5132,0.193,-2.8245>,<-0.9413301,0.3374872,-0.0002677>
                      MEAS/PLANE,4
                      HIT/VECTOR,<-0.44,0.3919,-3.25>,<-0.9396976,0.3420064,0.0000105>,<-0.4416,0.3924,-3.2495>
                      HIT/VECTOR,<-0.48,0.282,-2.4>,<-0.9396976,0.3420064,0.0000105>,<-0.4813,0.2823,-2.3998>
                      HIT/VECTOR,<-0.53,0.1446,-2.4>,<-0.9396976,0.3420064,0.0000105>,<-0.5306,0.1449,-2.3994>
                      HIT/VECTOR,<-0.6,-0.0477,-3.25>,<-0.9396976,0.3420064,0.0000105>,<-0.5994,-0.0477,-3.2494>
                      ENDMEAS/
                      SLOTRITE =FEAT/ROUND_SLOT,CARTESIAN,IN
                      THEO/<0.5295,0,-2.5575>,<1,0,0>,<0,0,1>,0.125,0.705
                      ACTL/<0.5296,0.0048,-2.5571>,<1,0,0>,<0,-0.001087,0.9999994>,0.125,0.7042
                      MEAS/SLOT,6,XPLUS
                      HIT/VECTOR,<0.5295,-0.0625,-2.8475>,<0,1,0>,<0.5301,-0.0576,-2.8458>
                      HIT/VECTOR,<0.5295,0,-2.91>,<0,0,1>,<0.5297,-0.0025,-2.9087>
                      HIT/VECTOR,<0.5295,0.0625,-2.8475>,<0,-1,0>,<0.5296,0.0679,-2.8477>
                      HIT/VECTOR,<0.5295,0.0625,-2.2675>,<0,-1,0>,<0.5294,0.0668,-2.2687>
                      HIT/VECTOR,<0.5295,0,-2.205>,<0,0,-1>,<0.5293,-0.0014,-2.2052>
                      HIT/VECTOR,<0.5295,-0.0625,-2.2675>,<0,1,0>,<0.5294,-0.0578,-2.266>
                      ENDMEAS/
                      MOVE/POINT,NORMAL,<0.7506,0.0211,-2.3149>
                      MOVE/POINT,NORMAL,<-0.8765,8.7161,-1.8903>

                      The above is the aliignment section (both manual and DCC) and the measuring angled plane and the slot part. Please point out where the problem comes from.

                      Comment


                      • #12
                        How are you dimensioning the angle?
                        To an axis, or to another feature?
                        Lately, it occurs to me
                        What a long, strange trip it's been.

                        2017 R1 (Offline programming)

                        Comment


                        • #13
                          Mode/dcc
                          move/point,normal,<0.0133,-0.4807,0.9071>
                          workplane/zplus
                          part number=15241 date=7/6/2010 time=3:24:41 pm page#=2
                          ftf =feat/plane,cartesian,triangle
                          theo/<0,0.0025,0>,<0,0,1>
                          actl/<0.0001,0.0073,0>,<-0.0010132,0.000503,0.9999994>
                          meas/plane,4
                          hit/vector,<0,0.579,0>,<0,0,1>,<0,0.5789,-0.0003>
                          hit/vector,<0.56,0,0>,<0,0,1>,<0.5597,0.0001,0.0006>
                          hit/vector,<0,-0.55,0>,<0,0,1>,<0.0004,-0.5499,0.0002>
                          hit/vector,<-0.5599,0,0>,<0,0,1>,<-0.5596,-0.0001,-0.0006>
                          endmeas/
                          you said your part is a thin walled tube.....your DCC hit points on the top plane are not evenly placed about the top face, you have 4 hits at Y.579,X.56,Y-.55, and X-.5599, a variation of .030. Make them all the same value. It's possible that one of the hits is not "centered" on the edge and is tilting your DCC alignment off in a weird direction

                          so start there
                          Which one gets ridden today? MPH vs MPG..tough choice, both are FUN
                          sigpic

                          Starrett RGDC 4028-24 :alien:
                          Demon vintages 3.7, 4.1, 4.2, 4.3, 2009

                          Comment


                          • #14
                            in your measure cylinder feature, you have a yminus workplane.

                            HTH

                            B&S CHAMELEON/PCDMIS CAD++ V2011

                            There are no bugs, only "UNDOCUMENTED ENHANCEMENTS!"

                            sigpic

                            Comment


                            • #15
                              Thank you all very much. The Yminus is absolutely wrong there. I dimension the angle to Y-axis not to another feature of the part. I'll try to prrof read my whole program (it's a long one 14 pages) and run again to see any change.

                              Comment

                              Related Topics

                              Collapse

                              Working...
                              X