Shell failed: another example

I would like to run one of the two flat sides schelling rear (0.8), but the command will not correctly emptied (I tested with different tolerances).
You could control it.
shelling failed.3dm (123.3 KB)

Hi Davide - thanks - do you have an example of this done successfully?


I have tried everything, but the shilling does not work well. As you see, it is a simple volume that is applied a chamfer edge. Emptying does not occur to any set tolerance (I tried them all, from 0.5 to 0.0001)

Hi Pascal,
Reading Clement’s other post today made me try to use OffsetSrf on this on because Shell doesn’t have options of having rounded or sharp corners. I saw that OffsetSrf didn’t have problems in the areas where Shell was having problems but the offset wasn’t cap’able straight away. Extracting one face and then running OffsetSrf and then running Shell on the solid and then combining the two results made for a pretty straight-forward way of getting there. Perhaps Shell should round corners?
shelling failed-wd.3dm (150.5 KB)

Another example in which the shilling fails. Empty the rear flat face of a unit (the critical part is the one where there is the upper fitting)
Give it a look.
shelling failed 2.3dm (1.6 MB)

I think it’s a problem of intersections, the shilling should work, it is not a very complicated example.

Hi Wim - thanks- I’ll check with the developer, I cannot now remember of roundy corners was considered for shelling and rejected as being not what users are after, or what…


Yeah, in the example I refer to, the user didn’t want the roundy corners. The trick with the BoxEdit scaling of individual surfaces does not to be as straight-forward in this example…

I ask if this type of example was solved. Both the shelling and the offset surface can not close the surfaces well (Rhino 6).
It’s been almost three years since I posted the example… maybe Rhino 7 can do well, I do not know…

Thank you

when the resulting edges do not hit each other geometrically you will have a hard time getting anything automated to work. why are you so eager to get something like this resolved? have you got millions of such nut cases to crack?

Is a valid geometry, a closed polysurface, formed by 27 edges manifold, I do not understand where is the problem?

do it manually, you might find out. i have seen far more expensive software failing here.

If we follow your advice, we buy Moi, which costs 1/5 of Rhino, and we do everything manually, from fillet to offset, etc. Why spend more?
The commands must evolve over time, they have to upgrade, otherwise we would still use Rhino 1.0!

then why did you ever use anything else than Rhino 1.0 if nothing evolved? You should´ve stopped right at version 2 then.

"Rhinoceros 0.0 - (null) "

hmmmm. 1/5 of nothing must be very cheap indeed…


In fact, I used Rhino for pleasure, while working with a BIM (ArchiCad) software. My speech wanted to be understood in this sense: from Rhino, which costs more than double that of Moi, for example, we expect a lot, is a matter of “expectations”, considering the development time (from operating system, not so much from CAD ).
Do not take every word as a controversy, please.

get a job