- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report
Thanks, Wayne.
It would indeed be ideal to not use rails at all. This was my first try, but the loft would almost always fail with just the profiles as inputs. Providing a centerline fixed a couple of simpler cases, but a majority still couldn't compute the loft. Manually adding rails successfully created every loft I tried, but as I mentioned earlier it will get tedious for a large number of boundary points and profiles.
You are correct that a large portion of the sample image is composed of a constant radius arc. However this is a specified constraint to limit the radial boundary of the profile. For the general case the generated profiles will be completely irregular based on the assembly geometry being analyzed so I won't be able to divide the solid into sub-shapes.
I've attached some parts that demonstrate what's going on, however there is another problem:
LoftPoints_Broken.ipt is the direct output from my script. The part is created inside an assembly and edited in place to add the loft points from the cut profiles of solid bodies in the assembly environment. I can continue to edit the part in context but as soon as I try to save the part independently I get invalid segment reference errors like this, which totally breaks the part. The definitions of the planes appear to be the problem, but I don't know what causes this error or how to correct it so if you have any thoughts they would be much appreciated.
LoftPoints_Derived.ipt is just the sketches derived from the first part to get around the errors. If you look at this part you'll see that trying to loft all the profiles together at once with no rails/centerline will not compute.
Hopefully there is something that can be done for both issues. Thanks!