loft - bending tables

You might want to have a look at this link.

This is not a fault of SolidWorks. There is not a CAD program in existence that can accurately unfold a loft by means of simple neutral-axis equation. The equation is only valid for simple bends. You may get a result, but it will not be accurate.

Unfolding non-gaussian geometry is a task fit for FEA-based specialty software. Many stamping toolmakers have such software.

Please define Gaussian geometry for us. A quick Google search didn't turn up anything helpful.

Basically, the difference between a Developable and Non-Developable (topology which can not be flattened in a straight line, requires stretching to flatten) surface. Generally used as a color range to illustrate surface deviations. So,.. if the color range over the surface is the same or transitions gradually from blue to green, it can be developed, easily flattened. If the color range across the surface changes dramatically or chaotically, it's non-developable, most likely needs to be deformed or stretched to flatten.

Gaussian geometry basically is any surface which can be unfolded without distortion (conversely, any surface that can be made from a flat piece of paper without kinks). As far as sheet metal and CAD goes, planes, cylinders, cones, and untapered extrusions are Gaussian. Almost everything else is not.

I got into this some years ago when I was tasked to sculpt surfaces that could have labels applied without wrinkling. Unigraphics actually has a special surface analysis tool to check Gaussian curvature.

I also figured out how to do this in SW. It is a lot of work though. The better you want the approximation, the more work it is.

Interesting reading. I should have thought to look there or on wikipedia.

So should we request SW add a Gaussian curvature plot to it's surface shading capabilities.

Looking at the definition of Gaussian curvature one might quickly determine whether a surface is developable and how to unbend it by using TOOLS/SKETCH TOOLS/SURFACE CURVES. Turn on curvature combs for two intersecting face curves. If one of the face curves doesn't show a comb then there is the possibility that the Gaussian curvature at that intersection point is 0 or close to it. However, the curvature combs don't seem to be very good at representing principal curvature values so results should be taken with a grain of salt. Two perpendicular planes and intersection curves might also be useful.