(Very) Short Radius Bends

The minimum radius bend CAESAR will accept is equal to 1/2 the OD of the pipe.

When viewed out of plane, the intrados is formed by a right angle of two straight lines.

It appears the intent of our client is to install ducting accordingly, which falls within d/D ratios permissible by code.

I get an uncomfortable feeling in that CAESAR beam analysis might not be enough, but to escalate it to FEA, there has to be a better definition of geometry where there isn't a convergence to 0 at the outside surface of the intrados.

Has anyone run into this before?

So mulling through the Minimum thickness calcs and SIF calcs, it would seem that while the minimum thickness for the intrados converges to 0, but the thickness for the extrados converges to a tangible number.

Similarly, flexibility factor and SIFs converge to real numbers.

Therefore, it would seem that there should be no other hiccup.