Least Square and Inscribed Circles

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

  • Least Square and Inscribed Circles

    Hey everyone,

    I've got a bushing hole diameter in question. Is there a way to output the same circle as Least Square and Inscribed Circles without measuring it twice?

    Thanks!
    Homb.
    B&S 7-10-7::3.7 MR3
    DEA BRAVO Dual Arm::3.5 MR2
    sigpicHomburgler
    DEA Bravo Dual Arm | 3.5MR2
    B&S 7107 | 3.7MR3

  • #2
    NO.....sorry


    Why yes you can.....sorry
    Last edited by bw_bob; 08-17-2007, 05:37 PM.
    sigpic.....Its called golf because all the other 4 letter words were taken

    Comment


    • #3
      Let's see... How much error would I introduce if I used the "To Points" program on my circle and then construted it as the 2 different types of circles? I have a tolerance band of 0.2mm.

      Homb.
      sigpicHomburgler
      DEA Bravo Dual Arm | 3.5MR2
      B&S 7107 | 3.7MR3

      Comment


      • #4
        Originally posted by bw_bob View Post
        NO.....sorry
        Liar
        Links to my utilities for PCDMIS

        Comment


        • #5
          This should not introduce any error provided you update your measured values to match your theos prior to using TO POINTS. You could always create a vector pnt at say 0,0,1 and paste w/ a pattern. I take it you want to constuct two diff circles using diff algorythyms(sp?) using the same pnts?
          sigpic 1.61803399Δ
          ASSIMILATE INTERGRATE SYNTHESIZE

          Comment


          • #6
            Read thread from 8/10/2007
            sigpic
            3.7 MR4

            Comment


            • #7
              http://pcdmisforum.com/showthread.php?t=5778

              You mean this one....
              sigpic.....Its called golf because all the other 4 letter words were taken

              Comment


              • #8
                Originally posted by homburgler View Post
                Hey everyone,

                I've got a bushing hole diameter in question. Is there a way to output the same circle as Least Square and Inscribed Circles without measuring it twice?

                Thanks!
                Homb.
                B&S 7-10-7::3.7 MR3
                DEA BRAVO Dual Arm::3.5 MR2

                wouldnt this work? Measure Hole with max Inscribed, then construct diameter from the holes hits and do a least squared
                Code:
                CIR1       =AUTO/CIRCLE,SHOWALLPARAMS = YES,SHOWHITS = NO
                            THEO/-5,0.25,-0.5,1,0,0,25
                            ACTL/-5,0.25,-0.5,1,0,0,25
                            TARG/-5,0.25,-0.5,1,0,0
                            THEO_THICKNESS = 0,RECT,IN,CIRCULAR,MAX_INSC,ONERROR = NO,$
                            AUTO MOVE = NO,DISTANCE = 0.15,RMEAS = None,None,None,$
                            READ POS = NO,FIND HOLE = NO,REMEASURE = NO,$
                            NUMHITS = 25,INIT = 0,PERM = 0,SPACER = 0,PITCH = 0,$
                            START ANG = 0,END ANG = 360,DEPTH = 0,$
                            ANGLE VEC = 0,0,-1
                CIR2       =FEAT/CIRCLE,RECT,IN,LEAST_SQR
                            THEO/-5,0.25,-0.5,0,0,1,12.4409
                            ACTL/-5,0.25,-0.5,0,0,1,12.4409
                            CONSTR/CIRCLE,BFRE,CIR1.HIT[1..25],,,,
                            OUTLIER_REMOVAL/OFF,3
                            FILTER/OFF,UPR=0
                Last edited by cmmguy; 08-17-2007, 05:17 PM.
                Links to my utilities for PCDMIS

                Comment


                • #9
                  I stand corrected....
                  sigpic.....Its called golf because all the other 4 letter words were taken

                  Comment


                  • #10
                    Originally posted by cmmguy View Post
                    wouldnt this work? Measure Hole with max Inscribed, then construct diameter from the holes hits and do a least squared
                    Code:
                    CIR1       =AUTO/CIRCLE,SHOWALLPARAMS = YES,SHOWHITS = NO
                                THEO/-5,0.25,-0.5,1,0,0,25
                                ACTL/-5,0.25,-0.5,1,0,0,25
                                TARG/-5,0.25,-0.5,1,0,0
                                THEO_THICKNESS = 0,RECT,IN,CIRCULAR,MAX_INSC,ONERROR = NO,$
                                AUTO MOVE = NO,DISTANCE = 0.15,RMEAS = None,None,None,$
                                READ POS = NO,FIND HOLE = NO,REMEASURE = NO,$
                                NUMHITS = 25,INIT = 0,PERM = 0,SPACER = 0,PITCH = 0,$
                                START ANG = 0,END ANG = 360,DEPTH = 0,$
                                ANGLE VEC = 0,0,-1
                    CIR2       =FEAT/CIRCLE,RECT,IN,LEAST_SQR
                                THEO/-5,0.25,-0.5,0,0,1,12.4409
                                ACTL/-5,0.25,-0.5,0,0,1,12.4409
                                CONSTR/CIRCLE,BFRE,CIR1.HIT[1..25],,,,
                                OUTLIER_REMOVAL/OFF,3
                                FILTER/OFF,UPR=0

                    Sure will!

                    TK
                    sigpicHave a homebrew

                    Comment


                    • #11
                      Originally posted by cmmguy View Post
                      wouldnt this work? Measure Hole with max Inscribed, then construct diameter from the holes hits and do a least squared
                      Code:
                      CIR1       =AUTO/CIRCLE,SHOWALLPARAMS = YES,SHOWHITS = NO
                                  THEO/-5,0.25,-0.5,1,0,0,25
                                  ACTL/-5,0.25,-0.5,1,0,0,25
                                  TARG/-5,0.25,-0.5,1,0,0
                                  THEO_THICKNESS = 0,RECT,IN,CIRCULAR,MAX_INSC,ONERROR = NO,$
                                  AUTO MOVE = NO,DISTANCE = 0.15,RMEAS = None,None,None,$
                                  READ POS = NO,FIND HOLE = NO,REMEASURE = NO,$
                                  NUMHITS = 25,INIT = 0,PERM = 0,SPACER = 0,PITCH = 0,$
                                  START ANG = 0,END ANG = 360,DEPTH = 0,$
                                  ANGLE VEC = 0,0,-1
                      CIR2       =FEAT/CIRCLE,RECT,IN,LEAST_SQR
                                  THEO/-5,0.25,-0.5,0,0,1,12.4409
                                  ACTL/-5,0.25,-0.5,0,0,1,12.4409
                                  CONSTR/CIRCLE,BFRE,CIR1.HIT[1..25],,,,
                                  OUTLIER_REMOVAL/OFF,3
                                  FILTER/OFF,UPR=0
                      It's been a couple of years, but I have to ask because I have the exact same problem here in 3.5. It's not working this way. I can't combine a number of hits like CIRCLE1.HIT[1..12], I always have to use CIRCLE1.HIT[1] up to CIRCLE1.HIT[12]. That is a lot of typing, and now I have to combine a great number of hits (I want to combine four sphere sectors to one sphere), until PC-DMIS says "line too long".

                      On our other machine with 3.7 I think the solution described above will work, but not with 3.5. Any ideas?
                      DEA Mistral (PC-DMIS 3.5)
                      Zeiss Contura G2 (Calypso)
                      Zeiss Accura (Calypso)
                      Zeiss MC 550 (PC-DMIS 3.7)
                      Mitutoyo BH 706 (MCOSMOS Geopak)

                      Comment


                      • #12
                        I don't know if this will work

                        Create a hole using: CIRCLE1.HIT[1] up to CIRCLE1.HIT[12].

                        Then edit that and change it to: CIRCLE1.HIT[1..12].

                        I think that I used that method to make it work.

                        Good luck.

                        B&S CHAMELEON/PCDMIS CAD++ V2011

                        There are no bugs, only "UNDOCUMENTED ENHANCEMENTS!"

                        sigpic

                        Comment


                        • #13
                          Unfortunately, this doesn't work.

                          I managed to combine hits to a sphere, though, by not measuring a number of spheres but take all hits as points, then construct a sphere with the points taken. Problem solved for now. :-)
                          DEA Mistral (PC-DMIS 3.5)
                          Zeiss Contura G2 (Calypso)
                          Zeiss Accura (Calypso)
                          Zeiss MC 550 (PC-DMIS 3.7)
                          Mitutoyo BH 706 (MCOSMOS Geopak)

                          Comment

                          Related Topics

                          Collapse

                          Working...
                          X