【转帖】hi folks figure 6-49 of the y14.5

yang686526

hi folks figure 6-49 of the y14.5
hi folks
figure 6-49 of the y14.5 standard, how does one establish the two datums c-d and measures the c-d for runout?
that one isn't too bad.
one would place datum d in a divider head or chuck but not too far in. the other end would end up wobbling.
now place a digital indicator on a stand at the top of datum c and zero off. rotate the part 180 degrees and take a reading. tap the part carefully about 1/2 the reading difference and then tighten the chuck.
one would keep doing this until datum c is running true to datum d. once you have done that, one may take the lateral circular runout on the faces shown.
the centre line was created through both datum d and also datum c.
dave d.
thanks dave
is it possible that a cmm can measure this in the free state without locking on to the datums?
circular runout is not conducive to cmms but anything can be done "half baked".
all product measurement is performed in free state but one would have to set up on the datums and shown.
dave d.
the 鈥渉alf baked鈥?way to do it on a cmm would be to create an axis between both journals by circularly probing near the center of each journal and aligning the centroids or just by using both circularly probed data sets in a single cylinder solve routine. if the journals are different sizes and you used the cylinder solve routine the form will be erroneous but the axis alignment will be ok. then just use the runout solve routine at various cross sections of the journals. the more points that you take the more comprehensive your estimation will be. a continuous scan at any given cross section will deliver results similar to an instrument scan. the runout solve routine will return the difference between the maximum radius and minimum radius values from the resolved axis at any given cross section.
all inspections are 鈥渉alf baked鈥?estimations of the true condition! with dave鈥檚 method you cannot access the portion of the journal that is in the chuck, the jaws of the chuck itself are assumed to be running true to the bearing journals of the chuck (there is error), and only a limited number of the infinite number of the possible cross section traces will be performed.
paul