Total Axial Runout, two Datums?

[[Mi...]]

I have a callout that I'm not quite sure I have right. Excuse the MS Paint drawing.

A is a plane
B is a cylinder perpendicular to A

Total Axial runout and Parallelism applied to another plane parallel to -A-

Runouts only allow you to populate one datum. So in this case, the toleranced surface is measured in Base Alignment, which was made using -A-, I then just used Datum B in the Total Axial Runout characteristic.

Questioning that because the parallelism sometimes has a higher result than the Total runout.

Runout.png

[[Ja...]]

Here is a previous post about this:

viewtopic.php?f=13&t=2383&hilit=ronout

Looks like the solution is to check the feature back to the B and constrain B to A in the characteristic.

[[Mi...]]

IMO -B- is the only one that matters for runout.. Since runout includes not basic dimension controlling position along the datum, establishing -A- doesn't have any effect on the outcome.

Parallelism may be higher, but that's because -A- and -B- aren't necessarily parallel either, and the difference of the two readings may incorporate that error.

[[Ph...]]

Since the parallelism control has a tighter tolerance, it will supersede the runout control. If each characteristic (as shown) is setup properly within Calypso per the stated DRF, they will produce the same result. Since you appear to being using an older version, you’ll need to follow the steps discussed in the thread Jaime shared. Alternatively, you can use the angularity characteristic to accomplish the same thing.

It’s my understanding that the newer versions of Calypso will now allow for multiple datum references within Runout characteristics, so you may look into updating if you're not into the workaround methods.

[[Mi...]]

Please sign in to view this quote.

Thanks, constraining to A was the easiest solution. I should have thought of that myself.

Please sign in to view this quote.

Out of curiosity I opened up Calypso 2020 on my offline seat and you are correct, the ability to select multiple datums is there. Unfortunately the machine I have to use only has 2018 installed - and this won't be enough to convince anybody we need to pay for the upgrade.

For comparison sake, in 2020 I did both methods and got the same result for both with subsequent evaluation.

Thanks all!

[[Da...]]

Please sign in to view this quote.

Please ask the designer if what he really meant is |total runout|0.06|B|A|
I can't see it otherwise.

[[Ja...]]

Please sign in to view this quote.

I agree. I think runout of that surface relative to A as primary is like parallelism and adding the datum B as secondary doesn't change anything. But, there is an example of this in ASME Y14.5 2009. According to that, the part should be mounted to A and rotated about B.

[[Ri...]]

Please sign in to view this quote.

Yeah, but this only matters for hard gaging. Doesn't apply to CMMs.

Drafters just put stuff on a drawing, and it's up to you to figure out what they want.

[[Mi...]]

Please sign in to view this quote.

Even that, I think, isn't strictly correct, though. It should really be Surface Profile or Position. And in either of those cases, either datum order could be correct, depending on what the designer wants.

But I agree, ask the designer.

[[Da...]]

Please sign in to view this quote.

I failed with secondary datum.
So, A as secondary in |total runout|0.06|B|A| also isn't making sense here with axial direction.

Maybe more like with datum B only -> |total runout|0.06|B|

[[Ma...]]

Total Axial Runout to |A|B| = Parallelism to |A|
Total Axial Runout to |B|A| = Perpendicularity to |B|

Runout to |A| would not be valid without |B| to specify the center of rotation