Won't let me do a radius and radius intersection

[[Ca...]]

I'm trying to find the intersection of two radius doing offline programming but Calypso said no result

 

Edited November 12

[[Je...]]

Hi.  I've seen this type of situation occur.  Usually, it means that the actual result calculated that there was in fact no intersection between the two calculated circles.

Instead, I usually evaluate via curves or by intersecting a line from the center of one of the two circles to the other at the nominal intersection.

There are a few ways to handle it.

[[Ca...]]

Jeff, the problem is Calypso cannot make a nominal intersection and I don't have curve. How would you tackle it?

[[Ky...]]

You can't intersect the circles because technically they don't intersect, at least nominally. They are tangent to each other, which the Calypso intersection constructions doesn't count.

 

Best option, especially without curve, would be to measure the circles at the same height, use recall in a 3D line to make a line connecting their centers and then intersect that line with either circle. You WILL get slightly different values depending on which circle you select, but the difference does give you an idea of how good the part is there.

[[Ja...]]

Could you also measure those radii as partial cylinders instead of circles? You should be able to get an intersection between those.

You may also need to add some constraints depending on the size of those radii for accuracy and repeatability.

Edited November 12

[[Je...]]

Kyle explained it well.  I'd recommend following his suggested strategy.

[[Je...]]

Have a look here, where the same problem is solved:

 

[[Ky...]]

Please sign in to view this quote.

No, I don't think that would work. First, I don't think cylinder/cylinder intersection works very well. Even if it did, it is still tangent, not intersecting.

[[Ja...]]

Please sign in to view this quote.

Understood.

I have had some luck using cylinders in the past when measuring an actual part, even if it isn't the best practice when the model has them as tangents I am able to get very close to what I can also verify on vision systems or comparators. We don't have Curve so I have had to get inventive. I really like the idea linked above about creating 3D Lines and intersecting them with the circles, keeping that in my back pocket for the next time I run across a similar issue to the original poster on this thread.

[[Ma...]]

Only reliable way is posted method with two circles -> 3d line from their centers -> intersection of circle and line -> symmetry of those intersections