【转帖】another concentricity vs runout thread

yang686526

another concentricity vs runout thread
i realize there has been several of these threads but the more i read the more i get confused.  unfortunately there is no one where i work to go to with these questions.  for the longest time i was always under the assumption that "concentricty for the most part will be half your tir" assuming round features (center point to center point).  then i read this article
the article has little bearing on the concentricity vs. runout issue.  it is only exploring the difference between radial and diametrical tolerance zones.
the simple fact is that if you are measuring tir (i.e. deflection on mandrel) then you are not measuring concentricity, you are measuring runout.  in almost all cases, runout and diameter are sufficient definition for both form and function.
to truly measure concentricity, one must measure the full form of the datum and the feature to determine effective centers.  even then, this does not guarantee runout, as the runout would also depend on the circularity of each of the surfaces.
why was his concentricity twice the radial distance?
put a 0.04 dia. around the datum a centerpoint. note the 35.4 dia centerpoint falls outside of that diameter. use the 35.4 centerpoint to scibe a diameter around the centerpoint of datum a, it will measure dia0.046 which is outside of the dia0.04 requirement.
the concentricty has a diametrical tolerance zone called out.
i'm probably adding mud to the answers given above. another way of looking at concentricity, it is twice eccentricity. runout on an axis to a datum axis is controlling the form such as roundness and concentricity.
aaahhhh
don't all concentricty callouts have a diametrical tolerance zone called out?  should the end result be concentric within .023, not .046?
the end result would be .046. if the point is .023 from the center, then it is within a diameter of .046 and a diametric zone is how concentricity is specified and reported.
powerhound, gdtp t-0419
production supervisor
inventor 2008
mastercam x2
smartcam 11.1
ssg, u.s. army
taji, iraq oif ii
picture two ellipses w/ major axes 90掳 apart.  one could have concentricity = 0 and runout off the charts.
1.  would a cmm report .023 or .046 for concentricity in the example in the link?
2.  so the statement that i hear a lot that "in general tir is double concentricity" is false? (assuming roundness is perfect).
the report should report 0.046 for the example in the link.
on a perfect diameter, fully offset within the example zone, the full indicator movement would registar 0.04
another way to think of it, concentricity is identifying the center axis of the feature, where runout is more concerned with the surface of the feature.
thetick's example of two ellipses perfectly on center give you a concentricity measurement of 0. the runout would return the difference of the major and minor.
in general, asme concentricity seems best applied to spinning parts which require balance.
prior to the 1994 asme standard position rfs and concentricity was a distinction without much of a difference. the difference was that concentricity had to be applied to features that were coaxial with the datum features and with position rfs... in order to consider the tolerance zone round "cylindrical" the tolerance value had to be preceded by a diameter symbol. for either the estimated axis, center, or central tendency had to reside within the round "cylindrical" zone.
position rfs had greater latitude in that the toleranced features were not required to be coaxial with the datum features and the shape of the zone could be described other than round "cylindrical" either by the depiction of leader lines or the absence of a diameter symbol.
runout unlike position rfs or concentricity, does not consider a feature's "estimated axis, center, or central tendency" but rather its form and "estimated axis, center, or central tendency" simultaneously. as stated in the article you cited "if the feature is perfectly round" the summations are equivalent...if not they are not! form errors will always increase the measured value for runout! the prerequisites for specifying runout are like those for concentricity in that the toleranced feature must be coaxial to the datum feature(s).
in attempting to create a distinction for the evaluation of concentricity in the y14.5.1-1994 mathematical definition and subsequent y14.5m-1994 dimensioning and tolerancing the committees abandoned the "estimated axis, center, or central tendency" for concentricity and replaced it with a surrogate "midpoints of opposed surface elements". that definition, however well-meaning or intended, has rendered concentricity impotent in constraining features with asymmetrical balanced forms such as odd tooth gears, multi-lobbed cams, etc.
the 1982 y14.5 included a note of caution about specifying concentricity because of the difficulty in determining the center or central tendency of a feature... how is that any different from determining the axis of a feature????
bottom line if the feature is coaxial, perfectly round, and perfectly oriented to the datum features and but off location鈥?position rfs = concentricity = runout = total runout