Measuring a radius with a limited arc

[[Da...]]

I've considered going to tech support with this, but before I do, I thought I'd see if anyone here knew of a work around in Calypso for measuring a radius with an arc of less than 90 degrees. I've worked with PCDMIS fairly extensively and I know they have a work around, since the standard method of measuring the radius of a circle when the radius arc is less than 90 degrees does not work with their software. It's a multi-step process in PCDMIS... measure the circle with it set to "Fixed Radius", create an alignment setting that circle to zero in all three axis, then take additional points in the radius and dimension them as polar radius.
Does anyone know if Calypso has any sort of work around for this sort of feature? I know I can't be the only person that has found it necessary to measure radii with a limited arc, so if someone with experience in this sort of thing would like to offer the benefit of that experience, it would be greatly appreciated...

[[Ro...]]

Use a 5 point circle, touch points instead of scanning.
I use this method for most instances where i have a very small section of a 3d feature such as a cylinder or cone or sphere.
Its actually very repeatable and correlates well to secondary gage checks
Also 2 touch point lines works well for very short lines (less than 1mm long) on chamfers for angle and distance intersections,

[[Ri...]]

You need an alignment for the circle. If it is a corner, use two points on the surface before and after to set the XY. Then when you measure the circle, measure it to that alignment. When you report the radius, change the evaluation parameters to lock the XYZ coordinates. If you are reporting linear distances, lock the radius.

Advanced settings: I've also used Curves or Free-Form surfaces to create the alignment to measure the arc to.

Like Roberto said though, just take a few single data points instead of scanning.

[[Da...]]

Please sign in to view this quote.

These are actually radii on the outside profile of a cylindrical part, so there's no "corner" I can align to. Would locking the XYZ still work if I'm still in the part alignment? Or would it only work if it's a more localized alignment?
I've already tried switching the feature to where it's taking five individual points instead of a scan... the results did improve, but they're still checking out of tolerance. We've verified that the radii in question are in fact good by other methods, so I know it's not the part that's the problem.

[[Ri...]]

Please sign in to view this quote.

Curve is the best tool then to give you an alignment.

You can try just locking the XYZ as is, and see what kind of results you get, but the issue will be is that if the radius is indeed far away from where it should be the results will be far off.

There is also a radius measurement Characteristic that I've had some luck with. Size - Standards - Radius measurement. You can test that out - I believe it will be close to what you were doing with PCDMIS.

[[Ja...]]

The math behind PCDMIS and Calypso is the same. I saw a demonstration made on PCDMIS that gave me the exact same results when I did it with Calypso, or any CAD program. In your feature, you can go into evaluation and constrain location to get an accurate radius or constrain radius to get an accurate location. The problem is if you constrain location it is locking XYZ relative to the alignment of the feature. If that is the base alignment that is far away from your radius, then you may need to create a better alignment just for that and change the alignment of your feature to that. I have created a 3D best fit alignment that worked really well. It's under resources- utilities. That may be why PCDMIS measured twice.

[[Ro...]]

Features / Special Geometries / Radius Point

[[Ke...]]

Depending on how limited the radius arc is... If the arc is 60-90°, I have had good correlation when scanning and using the L1 evaluation setting (If closer to 60°, I start the scan a little early, so that I'm not masking too much of the radius). My results correlated very well to handheld radius gages, as well as MarSurf/TalySurf measurements.
A supplier used the touch point method (using 6 points), and got results that were within spec, even though they were clearly under spec using handheld gages, and other evaluation methods. These were plastic injection molded parts and the features were being affected by shrinkage and they were also on the mold parting line, so the form was not great... I tried the touch point method myself, but could not get similar results to theirs, even when using a local alignment...so, if you are considering using the touch point method make sure that the features have a good form.

[[Ri...]]

Create your feature.
Edit to restrain the Radius value (Always nominal)
Copy the feature and recall the original feature but uncheck "Re-calculate Nominal geometry".
Edit the new feature and select X,Y and Z as restrained allowing Radius to be active.
This will give you your true radius value.
You can then create a third Theoretical feature and assign X to X, Y to Y and Z to Z from the first feature.
Recall the diameter (radius) from the second feature.
The third feature will have X, Y, Z and Diameter (Radius) values for use.
I've used this method successfully for over a decade.

[[Ma...]]

Please sign in to view this quote.

How you can recall feature to be able to uncheck that?
I thought that these restrictions are done like this:

• measure feature with R locked
• copy previously measured feature, recall measured points with no recalculation and use measured XYZ ( A1,A2, ... ) from original and restrict XYZ ( angle ) to get radius.

Your guide tells me to restrict XYZ same as nominal at original feature.

[[Is...]]

There is a place in hell reserved for the people who invented the measurement of segmented geometric elements like the radius.

[[Ph...]]

Please sign in to view this quote.

🤣 Could not have said it better myself!!

[[Ri...]]

Please sign in to view this quote.

I do it all the time, works great.

[[Ma...]]

Please sign in to view this quote.

Can you share pictures with steps? I think we don't understand each other.

[[Is...]]

Please sign in to view this quote.

I normally don’t accept these specifications on designs; and this point of view has damaged several times my reputation.
Is unbelievable the level of uncertainty that we have with these measurements...