Referencing a datum to another with different tolerances

[Bill65]

Hello, let me just tell you guys what a great forum you have here. I was some sort of lurker for a long time ( sorry for that) and now I have a problem of my own.

I work at a machine shop and last week we receive a part which I don't have any idea where it goes but it has something weird ( or maybe weird to me). The CMM guy left and I have little experience controlling parts, so that's why I come to you for help.

The part has 3 datums "A"curved surface, "B" a 3 hole pattern and "C" a semi-circle. The thing is datum C and B reference each other but with different tolerances. Is this possible? I mean isn't it redundant? Because when I control the part with a CMM the semicircle with respect to the pattern passes but the pattern with respect to the semicircle don't.

Looking at the drawing A controls the Y position, B the Z position and C the X position. I know that when you are measuring C you want to know the real position in X and Y. I think that's why it references A also. I just want an explanation to clear my mind on this. I can't scan the original drawing but I am attaching a very similar one that I made.

Thanks in advance

[Kelly_Bramble]

Technically, one can define a part with these datums...

When measuring one simply establishes the datums as specified... So, the R .250 +/- .005 is measured after establishing a datum zero from primary datum feature A and secondary datum feature B.

The three holes Dia. .125 +/- .001 is measured from a datum reference established from datum C feature and so on. Each measurement is a separate setup.

I agree that the specifications are suspect from a design point of view. Also, I suspect that the units are inches as opposed to mm as indicated by the leading zeros usage.

[Bill65]

Thanks for your answer and sorry for not telling you that, yes the drawing is in inches.
But what about the positions tolerances they are different by a lot. Isn't the same B with respect to C than C with respect to B? I mean, the part will be ok when B with respect to C and C with respect to AB will be below 0.001 not 0.01. I know that there is also A wich controls Y position for the circle but position in X direction is the center of the semi circle to the Datum B.

I think the best will be defining C with respect to AB and B with respect to A.

Thanks in advance.

[Kelly_Bramble]

Yes, the tolerances are different however without knowing function one cannot speculate on intent.

Be aware that when a datum is established for measurement the true geometric counter part is the actual measurement reference. The geometric counterpart technically only contacts the most-apt to mate surface areas or points of the datum feature. Therefore, regardless of how accurate the feature is tolerance specified the origin of measurement is the datum geometric counter part.

[Bill65]

Thanks for all Kelly, I have a better insight of this issue