runout -vs- concentricity

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

  • runout -vs- concentricity

    Could anyone give me an easy answer as to the difference between runout and concentricity? Ex. my origin is a circle, I checked another circle. Then, I checked the runout from that dia. to the origin. I got like .0058. Then, I checked the concentricity from that diameter to the origin and got .0035. Any explanation? The print does call for single runout to the origin. Why the big difference? If using runout is the correct way, did I do it right? PLN, (Dat_A), CIR1(Dat_B).....XY origin to Dat_B. Z origin to Dat_A. Then checked the circle at hand. Clicked runout...selected that circle....plugged in my tolerance of .002

    I am using INCHES....FYI

    Thanks!

    Josh
    ~Josh~ Giles Rescue "Ut ceteri vivant"

  • #2
    Runout > includes form error of the feature being dimensioned (think indicator)
    Concentricity > center to center misloacation of co-axial features (form is not part of the math).

    Craig
    <internet bumper sticker goes here>

    Comment


    • #3
      Concentricity is an axis to axis measurement. Runout is a surface to surface measurement. You could have a rectangular feature with perfect concentricity but horrible runout.


      HTH

      Dennis

      Comment


      • #4
        So....does this look right....The runout of Cir6 to the origin is .0058?

        DAT_A2 =AUTO/CIRCLE,SHOWALLPARAMS = NO,SHOWHITS = NO
        THEO/0,0,0,0,0,1,8.999,0
        ACTL/-0.001,0.0014,0,0,0,1,8.999,0
        TARG/0,0,0,0,0,1
        DCC_ALN2 =ALIGNMENT/START,RECALL:ALN2, LIST= YES
        ALIGNMENT/TRANS,XAXIS,DAT_A2
        ALIGNMENT/TRANS,YAXIS,DAT_A2
        ALIGNMENT/END
        DIM RNOUT3= RUNOUT OF CIRCLE CIR6 TO THE ORIGIN UNITS=IN ,$
        GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=BOTH
        AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
        M 0.0058 0.0000 0.0020 0.0000 0.0058 0.0038 -->
        DIM CONCEN1= CONCENTRICITY FROM CIRCLE CIR6 TO THE ORIGIN UNITS=IN ,$
        GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=BOTH
        AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
        M 0.0035 0.0000 0.0020 0.0000 0.0035 0.0015 -->
        ~Josh~ Giles Rescue "Ut ceteri vivant"

        Comment


        • #5
          Originally posted by craiger_ny View Post
          Runout > includes form error of the feature being dimensioned (think indicator)
          Concentricity > center to center misloacation of co-axial features (form is not part of the math).

          Craig

          Is this why my runout is showing .004 and a v-block/indicator is showing .00005?

          Has anybody got any documents that supports that this is happening.

          D@mn operators are trying to say the cmm is wrong,,, dirty button pushers.....
          I talk dirty to my cmm. Justn

          Comment


          • #6
            Originally posted by justncredible View Post
            Is this why my runout is showing .004 and a v-block/indicator is showing .00005?

            Has anybody got any documents that supports that this is happening.

            D@mn operators are trying to say the cmm is wrong,,, dirty button pushers.....
            The cmm is probably wrong. I check runout with an indicator. To me its the best & most accurate way. My .02
            sigpic.....Its called golf because all the other 4 letter words were taken

            Comment


            • #7
              ji are you dimensioning two circles? I too do not like the results I get for runout on the CMM but there may be things to consider in light of you getting 0.004 vs 0.00005. First would be the feature vectors. If you are doing 2 2D features such as circle to circle then you absolutely have to be in an alignment scenario that gives them a common vector (when they are measured). If you are leveled to a plane we'll call A and measure circle 1 you need to be in the same level scenario when you measure circle 2 or their vectors will not match and they can show error.

              Knowing that you filter a lot of your outliers I am confident you are getting better form than most so form measuring errors are not much of a susspicion here but should be considered. I am not much of a fan for roundness with a CMM and that carries over to runout.

              Any chance of throwing up some code for the features in question and the dimension.

              My first answer is runout with PCDMIS sucks but I think you should get better than 0.004 when you are getting 0.00005 on the bench.

              Craig
              <internet bumper sticker goes here>

              Comment


              • #8
                Originally posted by jkb_mcg View Post
                So....does this look right....The runout of Cir6 to the origin is .0058?

                DAT_A2 =AUTO/CIRCLE,SHOWALLPARAMS = NO,SHOWHITS = NO
                THEO/0,0,0,0,0,1,8.999,0
                ACTL/-0.001,0.0014,0,0,0,1,8.999,0
                TARG/0,0,0,0,0,1
                DCC_ALN2 =ALIGNMENT/START,RECALL:ALN2, LIST= YES
                ALIGNMENT/TRANS,XAXIS,DAT_A2
                ALIGNMENT/TRANS,YAXIS,DAT_A2
                ALIGNMENT/END
                DIM RNOUT3= RUNOUT OF CIRCLE CIR6 TO THE ORIGIN UNITS=IN ,$
                GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=BOTH
                AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
                M 0.0058 0.0000 0.0020 0.0000 0.0058 0.0038 -->
                DIM CONCEN1= CONCENTRICITY FROM CIRCLE CIR6 TO THE ORIGIN UNITS=IN ,$
                GRAPH=OFF TEXT=OFF MULT=10.00 OUTPUT=BOTH
                AX MEAS NOMINAL +TOL -TOL DEV OUTTOL
                M 0.0035 0.0000 0.0020 0.0000 0.0035 0.0015 -->
                Lots of things to look at here JKB, the location of CIR6 to the ORIGIN would help. Also, roundness of the two diameters. Those together can affect runout. Looking at each individual point would tell us a lot. I'm not saying do that, but it would give more information about the part.
                When in doubt, post code. A second set of eyes might see something you missed.
                sigpic

                Comment


                • #9
                  SCN-B- =BASICSCAN/CYLINDER,SHOWHITS=NO,SHOWALLPARAMS=YES
                  0,0,0.125,CutVec=0,0,1,OUT
                  InitVec=1,0,0,DIAM=0.507,ANG=1440,PITCH=0.1,DEPTH= 0,THICKNESS=0,PROBECOMP=YES,AUTO MOVE=NO,DISTANCE=0
                  FILTER/DISTANCE,0.1
                  EXEC MODE=FEATURE,USEHSSDAT=YES,USEDELAYPNTS=NO
                  BOUNDARY/
                  HITTYPE/VECTOR
                  NOMS MODE=MASTER
                  ENDSCAN
                  MOVE/CLEARPLANE
                  SCN064 =BASICSCAN/CIRCLE,SHOWHITS=NO,SHOWALLPARAMS=YES
                  0,0,-0.3,CutVec=0,0,1,OUT
                  InitVec=1,0,0,DIAM=2.15395,ANG=360,DEPTH=0,THICKNE SS=0,PROBECOMP=YES,AUTO MOVE=NO,DISTANCE=0
                  FILTER/DISTANCE,0.1
                  EXEC MODE=FEATURE,USEHSSDAT=YES,USEDELAYPNTS=NO
                  BOUNDARY/
                  HITTYPE/VECTOR
                  NOMS MODE=MASTER
                  ENDSCAN
                  MOVE/CLEARPLANE
                  CYL-B- =FEAT/CYLINDER,RECT,OUT,LEAST_SQR
                  THEO/0,0,0.32233,0,0,1,0.507,0.39466
                  ACTL/-0.00009,0.00003,0.32538,0.0008414,-0.0005587,0.9999995,0.50672,0.39823
                  CONSTR/CYLINDER,BF,SCN-B-,,
                  MOVE/POINT,NORMAL,2.90782,6.74994,8.84181

                  Due to the wide open tols, I did not filter the scan cyl to create the cyl-b-.
                  I did increase the scan on the cyl, from .4 pitch to .1 and the rotation is at 1440, which means I make 4 complete revs around the axis of the cyl, each .1 apart at the end, like a screw. The 2nd is just one scan since the tols .00195...
                  I talk dirty to my cmm. Justn

                  Comment


                  • #10
                    JKB,
                    If your drawing is speaking ASME:
                    Runout is a composite tolerance. Runout is a SURFACE control (to a datum axis). It will control surfaces constructed around a datum axis, and those constructed at right angles to a datum axis. It will control form, orientation and location of features. Using "total runout" to a surface around a datum axis will control circularity, straightness, coaxiality, angularity, taper, and profile of that surface. Pretty powerful huh?

                    Concentricity:
                    Is a condition where the median points of ALL diametrically opposed elements are congruent with the axis (or center point) of a datum feature. You'd be inspecting a point cloud - so to speak.

                    Pcdmis is going to treat concentricity and coaxiality the same way. This is wrong. Your concentricity results will be based the same way a position formula is. For runout, (and I would have to explore this) I believe pcdmis would be taking your larger and smaller inspection points - and reporting the difference between them. That is why you would be having the discrepency.
                    I would exercise extreme caution when inspecting runout on a cmm.

                    A side note:
                    ISO and ASME differ on concentricity definitions. our "coaxiality" (asme) is there concentricity (ISO). This just 1 of the many differences betwen them.

                    Kev
                    RFS Means Really Fussy Stuff

                    When all you have is a hammer - everything looks like a nail....
                    sigpic

                    Comment


                    • #11
                      OK ji I guess at this point I'd see if PCDMIS is giving you form or position error. Any chance you can do a 2D distance from feature to feature as well as a form dimension of each. Perhaps you'd have to align to one and dimension location. You'll need to narrow down where this is getting frogged up it is either form or location. like I said I do not like the way PCDMIS does runout so I don't use it but 0.004 sounds like too much of a discrepancy unless you are metric but I am assuming that 0.004 is inches. I do notice that in one instance you are doing a cylinder and the other a circle. What may be happening is that the cylinder has it's own vector obviously and the circle has a vector normal to your leveling feature. With different vectors they will report mislocation, this is my first suspicion. You could try leveling to the cylinder then dimensioning to see what you get also. What ever the case you'll need to break it down into individual elements to get to the bottom of it. Give location of one feature to the other a shot as well as form and possibly try leveling to the cylinder and see what you get. I am inclined to believe it is the vectors between the 3D feature and the 2D feature that is aggravating this.
                      <internet bumper sticker goes here>

                      Comment


                      • #12
                        If I'm not mistaking I read somewhere that concentricity is 2 times the distance. For what reason I don't know. Can someone check on this.
                        sigpic

                        Comment


                        • #13
                          Originally posted by AuRules View Post
                          If I'm not mistaking I read somewhere that concentricity is 2 times the distance. For what reason I don't know. Can someone check on this.
                          In a nut shell you are correct. Even though we are talking concentricity it is easiest to explain it in terms of indicating so lets assume there are no form or orientation errors just for the sake of example. Suppose diameter A is mislocated 0.001 from diameter B. Putting B in a workholding devise or V-Block (a no-no for true runout but usually acceptable for a hasty indication) as A is rotated it will indicate 0.001 in a positive direction on one side and then as it rotates around it will indicate 0.001 in a negative direction giving a total variation of 0.002 (0.001 minus -0.001 = 0.002). You have essentially measured the actual size of the target zone that A lands in. You could also picture it like TP 0.001 mislocation constructs a 0.002 tolerance zone.

                          These are examples to make it clear in your head not actually ASME definitions. I'm just throwing it out this way so it can be pictured. If I am even making sense.
                          <internet bumper sticker goes here>

                          Comment


                          • #14
                            [QUOTE=craiger_ny;46974]In a nut shell you are correct. Even though we are talking concentricity it is easiest to explain it in terms of indicating so lets assume there are no form or orientation errors just for the sake of example. Suppose diameter A is mislocated 0.001 from diameter B. Putting B in a workholding devise or V-Block (a no-no for true runout but usually acceptable for a hasty indication) as A is rotated it will indicate 0.001 in a positive direction on one side and then as it rotates around it will indicate 0.001 in a negative direction giving a total variation of 0.002 (0.001 minus -0.001 = 0.002). You have essentially measured the actual size of the target zone that A lands in. You could also picture it like TP 0.001 mislocation constructs a 0.002 tolerance zone.

                            These are examples to make it clear in your head not actually ASME definitions. I'm just throwing it out this way so it can be pictured. If I am even making sense.[/QUOTE]


                            Very good explanation Craig....My dumba$$ even understood it.
                            sigpic.....Its called golf because all the other 4 letter words were taken

                            Comment

                            Working...
                            X