perimeter scan help

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

  • perimeter scan help

    I just started a new job and they have a tesastar scanning probe, but it's not giving me any variance between the nominal and actual when a do a perimeter scan. i have the scan set to find-nominal and defined and it scans fine but no usable data comes back. if i change defined to normal it pecks at it very slowly but at least that seems to give me some real data, but life is too short for that. I'm using pC- Dmis 2015

  • #2
    Would you post the code of the scan after execution showing the parameters and the hits?
    153010 Global Advantage w/ LSPX1H_T Analog Probe
    7107 Global Classic TP20
    2019R1 SP1 CAD++

    Comment


    • #3
      Tesastar is not scanning, it is a TTP touch-trigger probe, functionally equivalent to Renishaw TP20. It collects a single data point when it collides with the part, via a kinematic/conductive circuit. the probe moves into contact with the part and in the process a switch in the stylus modules opens an electrical circuit, thus taking a single point...

      Scanning probes can collect hundreds of points per inch while dragging the probe along a surface. they are a whole different monster.

      Comment


      • NinjaBadger
        NinjaBadger commented
        Editing a comment
        Tesastar is actually the probe head and can support scanning probes.

        I would surmise by the fact he mentions it scanning and then changing to normal it peck scans that he does infact have a scanning probe on there.

        My guess would be the find nominals tolerance not being big enough (should be probe radius plus a bit more than the deviation you'd expect to find in the part)

      • louisd
        louisd commented
        Editing a comment
        https://store.hexagonmetrology.us/te...-trigger-probe
        what's this, then?

    • #4
      SCN6 =FEAT/SCAN,PERIMETER,NUMBER OF HITS=37
      ,SHOW HITS=NO,SHOWALLPARAMS=YES
      EXEC MODE=DEFINED, NOMS MODE=FIND NOMS,CLEARPLANE=YES,SINGLE POINT=NO,THICKNESS=0
      FINDNOMS=0.6,SELECTEDONLY=YES,USEBESTFIT=NO,PROBEC OMP=YES,AVOIDANCE MOVE=NO,DISTANCE=0,CAD Compensation=NO
      LINE,INCR=0.1,CAD TOL=0.02,OFFSET=0.04,OFFSET TOL=0.02
      HITTYPE=VECTOR
      INITVEC=0.0151786,0,-0.9998848
      DIRVEC=0.9854221,-0.0140346,0.1695475
      POINT1=1.7826,0.9474,-0.4994
      POINT2=5.1752,0.9173,0.7547
      MEAS/SCAN
      BASICSCAN/PERIMETER,NUMBER OF HITS=37,SHOW HITS=NO,SHOWALLPARAMS=YES
      <1.7826,0.96,-0.4994>,<5.1079,0.96,0.6998>,DirVec=0.999703,-0.0000282,0.0243689
      InitVec=0.0151786,0,-0.9998848,THICKNESS=0
      FILTER/DISTANCE,0.1
      EXEC MODE=DEFINED
      BOUNDARY/
      HITTYPE/VECTOR
      NOMS MODE=FINDNOMS,0.6
      ENDSCAN
      ENDMEAS/

      Comment


      • #5
        As NinjaBadger says, it could be the FINDNOMS tolerance. You have a FINDNOMS tolerance of 0.6 which is probably fine for a program using inches but would be way too small for one using millimetres. As a general rule of thumb, you should set it to a minimum of the tip radius plus whatever amount of deviation you are expecting. So for a Ø4mm tip I would normally use a FINDNOMS tolerance of around 3mm.

        Another thing to try would be to switch the EXECMODE to RELEARN rather than DEFINED. With DEFINED scans, they will throw away any hits taken outside the measuring range of the probe you are using. It is also possible to scan "fresh air" - because it is a pre-defined path, the CMM will drive the probe along that path, even if it is not in contact with the part but it will obviously not return any points if this happens.
        Neil Challinor
        PC-DMIS Product Owner

        T: +44 870 446 2667 (Hexagon UK office)
        E: [email protected]

        Comment


        • #6
          I've played with the findnoms tolerance, but sill no results and in the exec controls for the perimeter scan it only give me two options defined and normal, I've been using the relearn when i do an open scan and that is working fine

          Comment


          • neil.challinor
            neil.challinor commented
            Editing a comment
            What version of PC-Dmis are you using and what type of probe is it?

            I seem to remember there were some issues with MASTER & NOMINALS mode in some old versions of PC-Dmis but I'm not aware of any issue with FINDNOMS.

          • NinjaBadger
            NinjaBadger commented
            Editing a comment
            It's while since I've done a perimeter scan but is it possible to define the scan without having select the surfaces first?

            You have SELECTEDONLY=YES, but if there are no surfaces selected this might be your problem.

        • #7
          Just grabbing at straws, but do you have the correct "OPTPROBE" statement before the scan? Also, under scan construction I always use NULLFILTER instead of line...
          Last edited by Hi-Tech Redneck; 11-27-2019, 11:46 AM.
          Sheffield Endeavor3 9.20.8, Tesastar-SM, Leitz LSP-X1s & LSP-X1M, PCDMIS 2011 MR1

          Comment

          Related Topics

          Collapse

          Working...
          X