Trouble with offset srf after fillet

I attached a simple GH file to demonstrate the problem I have.
If you use the last two sliders you will notice, that the two non-planar surfaces (or the fillets) sometimes don’t get calculated properly.
Can someone help me to A) understand what’s going on here and B) how to fix this?

bend-help.gh (10.4 KB)

Try to use pufferfish plugin, In this case, has more options and more stable…


3 Likes

That plugin actually helps. Works fine with it, thanks a lot!

Anyway I though that my grasshopper defenition was pretty simple, so I’m still wondering what’s wrong there. The behaviour seems really strange. Is it a bug?

I’d like to pick this up again.

@DavidRutten regarding the “native” / “built in” GH component:
Why does it not calculate correctly? Why offer it when it isn’t working? Is it a bug?

@Pufferfish :sweat_smile: :
I attached the file again, this time done with pufferfish. There is still a problem. I can tell the component to create a “solid offset”. It results in many closed breps instead of a single one. In this example I can just use SolidUnion to make it one but with more complex geometry that doesn’t work anymore either. So my question is how can I get pufferfish to output a single brep?

Thanks!
File:
bend-help-2.gh (12.3 KB)

regarding Pufferfish:
though the surface offset component from pufferfish still can’t give me a single brep in some cases, I did manage to work around this. when setting up the different surfaces that the brep is composed of, I made sure that their normals all pint into the right direction. that way the SolidUnion component was able to join the breps to a single one after the pufferfish surface offset component. SolidUnion wasn’t able to do this when I had just flipped the surfaces before joining the together.
seem to me like carefully putting thegeometry together is always a good thing.
really a great plugin!

regarding the native grasshopper surface offset component:
I’m always a bit puzzled by the way the mcneel team treats bug reports. bugs that are probably painful to approach seem to get ignored completely. this kind of cherry picking reflects badly on the mcneel team, imo. the definition I uploaded in the start post looks simple to me and I assume teh behavior of the offset surface component qualifies as a bug - I still hope this will get addressed.

anyway, for now the work around does the trick for me.