Recall Feature Points from Curve

[[Sh...]]

Anyone ever use Recall Feature Points - Curve result points?  I stumbled across this yesterday and used it to recall a small rad from a torus and it seemed to be pretty accurate, I just don't know if it's advisable to use or not.

[[Cl...]]

You are recalling curve points into a circle? If this is the case, then be advised that

if the circle varies in size part to part, then the location of the recalled points can change.

This will affect the repeatability of the constructed circle.

[[Cl...]]

Not 100% sure but maybe creating a best-fit alignment from the curve might correct that.

Edited February 21, 2025

[[Ma...]]

Normally on older versions you are recalling points from curve even with masked ones - this option will allow you to pass measured points after filters - which can or cannot be helpfull depending on usage.

[[Je...]]

Please sign in to view this quote.

Shawn,

I use this option often depending on the application.  It's an important distinction between using a recalled curve feature primarily as a point container for actuals or recalling the curve-filtered points per the curve's nominals.

In theory, this can make the strategy more feasible per system resources because it reduces the quantity of recalled points.  It depends on the application.  Another advantage is that it imports a feature whose task of interpolation is already accomplished.  An example would be for evaluating profile of a freeform surface made up of multiple curves.

I can explain more specific use cases if helpful, but that's the overview.  Both options are important.  It's definitely not the case that one option is always correct and the other is "bad."

PS, the Zeiss moderators will probably relegate this thread to the purgatory known as the curve subtopic, so hopefully you're able to get your answer before that occurs.

Edited February 21, 2025

[[Cl...]]

I once askes a Zeiss applications engineer if I should tick the "Re-calculate Nominal Geometry" box or leave it un-checked. He didn't know. I don't know either?

[[Je...]]

[[Ma...]]

Please sign in to view this quote.

If you are recalling from curve, then this will happen:

• You have no actuals on curve -> nominals will be recalculated from nominal points of curve
• You have actuals on curve ( measured ) -> nominals will be recalculated from actuals

I am using that recalculation only from curve and without actuals for making basic features from model. It's not recalculating nominals every run ( i think ).

[[Sh...]]

Please sign in to view this quote.

 

[[Je...]]

Please sign in to view this quote.

Good question, Shawn.  I usually like a higher density of points to define arcs.   The situations where I use curve filtering are when the underlying geometry is a set of curves such as a spline and not a true, singular circle.  This usually occurs when I'm evaluating profile on a set of adjoining features such as an opening.

Edited February 25, 2025