Circle or cylinder

[[sc...]]

When choosing between the two for a secondary datum, I usually go with a cylinder. My reasoning is a cylinder has a center line axis, makes a better datum, verses a circle which is a point.

Which is the better choice for a secondary datum, circle or cylinder. Specifically when measuring position and profile? Thanks

Scott

[[Za...]]

If the depth of the cylinder is less than the diameter, I would use a circle. If its the same or larger, I would use a cylinder.

[[To...]]

Are you working to a standard, such as ASME Y14.5? You may want to consider a cylinder.

The Zeiss Cookbook suggests using a circle when the depth is less than the diameter but I don't that reference anywhere in Y14.5. However, on a thin sheet metal part, I get it.

This should wake up some of the sleeping GD&T bears out there.

[[De...]]

According to Y14.5, the location is the center of the cylinder once the cylinder is aligned (constrained) using the primary datum.

If the primary Datum is a plane that is perpendicular to the axis of the cylinder, then the plane is controlling two rotational degrees of freedom and the translational degree of freedom perpendicular to the axis of the cylinder.

The axis of the cylinder will control the two translational degrees of freedom the plane does not.

So the end result is determining a location in two directions like a 2d point, however where that point comes from becomes critical if the actual cylinder is not perfectly perpendicular to the primary datum.

If the cylinder is perpendicular to the primary datum, then a circle would give you the location you need, but your results will become more and more uncertain as the error between the cylinder and the primary datum increase

Generally cylinders that have a large ratio of diameter to length will work better as circles than cylinders because the amount of perpendicularity is minimized due to that same ratio.

Cylinders that are have a length that meets or exceed the diameter, are generally best checked as cylinders in order to account for perpendicularity to the main datum.

[[Ra...]]

Please sign in to view this quote.

Right or wrong, this is how we approach circles/cylinders.

I have always wondered though... when the cylinder is not near perpendicular to the primary plane, does Calypso resolve the measured cylinder to the axis of the largest theoretical inscribed perpendicular cylinder axis for secondary alignment?

Reading that again, I'm not sure my question is clear. Anyone else thought about this?

[[Aa...]]

Please sign in to view this quote.

If the ISO 5459 switch is on, then yes, that's how Calypso evaluates it. If you right-click the secondary cylinder in the DRF, you can see the fitting algorithm and constraints being used.