Why at copying of the sketch consisting from 2D curves will transform them in the new sketch to parabolas (and at reorganisation of the first sketch the interrelation and parametrization is naturally lost)?

Why at copying of the sketch consisting from 2D curves will transform them in the new sketch to parabolas (and at reorganisation of the first sketch the interrelation and parametrization is naturally lost)?

I consider is certain error SW.

first the sketch two splines

second the sketch the copied splines with 1ого the sketch, but they became parabolas. And further at change first the sketch - in second the sketch there will be errors of interrelation with first the sketch.

Shon, you've discovered an interesting defect.

To clarify, Shon creates a sketch with two 3-point splines with no tangent constraints. He then created a second sketch on top of the first one and used 'convert entities' to clone the splines. the resulting entities were parabolas. I did a little poking around and discovered that this only happens with 3 point splines. Equation driven splines, splines with 2 or 4 or more points and splines with tangent constraints will all produce spline entities when selected for convert-entities.

Shon, the work around is to add tangent lines to the ends of your splines. following that, convert entities will produce splines.

see attached for an example.

But you're right, that's probably unintended behavior and I doubt it's documented. You should call your VAR and report it as a bug.

John,

I wonder if it is not so much a bug as a glimpse behind the curtain. Perhaps the math behind a 3 point spline with no tangencies is, in fact, a parabola. Any math wizards care to comment?

Jerry Steiger

Jerry, from my reserach into the way B-splines work, they approxomate the graphs of undiferentiable functions using second-order curves joined and tangent at their endpoints, so an unconstrained 3-point spline is by the nature of that math a parabola.

However, I don't think this is behavior consistent with the rest of solidworks. If you create an unconstrained two point spline and do a convert-entities, you don't get a line segment, you get a two point spline. Personally, I would be happy if all sketch segments were splines that could be parameterized as analytic or forms as the user needed.

Imagine if you could configure an arc to be a straight line? How useful would that be for lens design?

John,

Cool, so I guess it really is a glimpse behind the curtain. Apparently a slip up by the folks writing code, who should have deprived us of the useful information by hiding it behind a spline symbol, like they do for the two point spline/straight line. Thanks!

I love your idea of being able to play with the definition of a curve.It would be great for many other purposes, like showing the deformed shape of sheet metal springs.

Jerry Steiger

I played with this file in SW 2013 and looks like the references are kept. The converted entities in the second sketch are changing from conic to spline if the curve in the first sketch increases in complexity.

I played with this file in SW 2013 and looks like the references are kept. The converted entities in the second sketch are changing from conic to spline if the curve in the first sketch increases in complexity.