I have a bore that is Datum B. It has a perpendicularity callout to Datum A, which is a flat surface. So, I have to measure it as a cylinder. I have another bore that has a circular runout callout to AB. So do I have to measure B again as a circle? Because I would have to assign that as another datum to build the feature control frame. But if I just use B as a cylinder, then wouldn't it be measuring total runout and give me a different result than circular runout?
In that case, I have another similar part. It wants the circular runout of Datum A (flat surface) to Datum B (inner bore). Do I still measure Datum B as a cylinder? Because it always reports way out of tolerance if I measure the circular runout of A to B. But if I measure B as a circle, then the circular runout of A to B looks good. If I measure total runout of A to B as a cylinder, it's always exactly the same as if I measure circular runout of A to B. In this case, there are no other callouts on the print that would require me to measure B as a cylinder.
basically think of it as if you were to measure it with an indicator. you would put something cylindrical in to datum b and rotate around that while measuring the runout of A. But if your runout is drawn to a cylinder then you need to measure it as such. a circle does not have a center axis. Runout is the high to low point deviation of the surface elements relative to a datum Axis.
Here is what the ASME Standard says about Datum Features for Runout Tolerances (9.3.1):
"The datum axis for a runout tolerance may be established by a cylindrical datum feature of sufficient length, two or more cylindrical datum features having sufficient axial separation, or a cylindrical datum feature and a face at right angles to it. Features used as datum features for establishing axes should be functional, such as mounting features that establish an axis of rotation."
So, using a 2D circle as a datum reference would not be allowed.
Circular Runout of a plane to a datum cylinder is unusual but allowed. It's basically checking a bunch of concentric rings on the face, one ring at a time. Total runout checks the whole face at the same time within a tolerance zone of 2 parallel planes.
I think the idea is to have Circular Runout refine Total Runout.
The intent of the control is to limit "in and out" movement of the face as the part rotates. As a test, you could dimension perpendicularity of the plane to the cylinder and see how those results compare to your runout dimensions. I'm not certain of the math differences but it would be interesting to compare.
Total runout and circular runout are being applied to the plane. not the datum B cylinder. circular runout only applies to one circular element at a time. Total runout applies to the entire surface combined. So if you only have one row of hits on datum A then the two values will be identical.
Thanks. Before my current job, I worked at a place that did fabricated parts from a brake press with sheet metal and had never seen runout before. I'm struggling to understand it.
