Short Axis Cylinder

[[Da...]]

I have a part with a short axis cylinder that is a datum (1.750" diameter X 0.6" long). On the other side of the part 3.6" away is a cylinder of the same dimensions. The 2nd cylinder has a true position callout of .001" to the first cylinder, the first cylinder is the only datum in the FCF. I'm struggling to get repeatability. Would it be "legal" to constrain the vector of the datum cylinder since it is the only datum?
Thanks.
-Dave

[[De...]]

I don't have the right answer for you, but I suspect that the correct answer would be that it depends on what standard you are using for print definition. ASME and ISO handle cylinder evaluation differently.

[[Da...]]

I'm sure for ISO the answer is no, you must not constrain. Pretty sure the same goes in ASME.
Because constraining would yield different datum system than on a drawing.

[[Da...]]

The drawing is ASME 14.5 1994

[[Jo...]]

Is there a perpendicularity requirement of the cylinder Datum to anything else ?
Are there roundness or cylindricity requirements, other than F.O.S ? Diameter tolerance ?
Also the max inscribed evaluation to calculate cylinder vector is going to be a repeatability issue.
.001" position would be .0005" deviation from location, and vector. Are these used in conjunction as Datums ?

If those other elements are undefined, or large tolerance, you can make the case for a modifier, or maybe qualify a position with the caveat that you have constrained Datum vector, and report its perpendicularity(unconstrained)? GRR data ?

After re-reading your question, are the cylinders coaxial ? That's another discussion entirely

[[Ke...]]

I guess a 0.6" cylinder is considered "short" in some applications 🤣
I would report everything possible:
- Position as called out (with Plot)
- Position if constrained to another feature (with Plot) (marked Reference Only)
- (maybe) Position using LSQ evaluations (Marked Reference Only)
- Cylindricity of the Datum Cylinder
- If the cylinders are in-line and the same size/tolerance, I would report the size of both cylinders as a single feature... I do this if there is an application such as a bolt or some other component that passes through both features (Marked as Reference Only)

Then run a short repeatability study of the results, to try and show a different/better possible way evaluating the features (if there is one)

[[Jo...]]

Please sign in to view this quote.

I don't have a 1994 edition, but these methods would negate the projection error. In the case of A-B Datum, you could reference each Datum back to A-B . The considered axis tolerance zone would only be .6" long while the Datum axis is the 3.6" or the total span to provide stability of reference.

Capture2.JPGCapture1.JPG

[[Da...]]

Please sign in to view this quote.

-The datum is perp to main bore within .001", I get acceptable repeatability on this.
-No form requirement other than rule #1
-both cylinders are O.D. TP is being evaluated outer tangential
-only one cylinder is a datum
-total tolerance on dia is .001"
-the cylinders are straight across from each other, basic is zero (diametrical)

[[Da...]]

Please sign in to view this quote.

Running R&R now, including the total runout you referenced (preliminarily runout with the compound datum is repeating around .0001", TP is around .0003"). We are a "job shop" so we make to print. We're looking for alternate methods to present to our customer.

[[Ke...]]

Please sign in to view this quote.

Yeah, Co-datums would definitely add the stability; need the other results to show the need for it.