The interface between hexagonal and tetrahedral cells offers other challenges, not just a change in accuracy but a change in stability.

RANS can handle this mostly since it's more diffusive, but DDES type simulations suffer lots of issues at these types of interfaces

Success includes making a product that is cost effective, safe, environmentally friendly, efficient, durable, and maintainable among other attributes.

Ivor Clifford.
Autocorrect out here just renaming researchers.

There actually seems to be quite a bit of nuclear use/application with OpenFoam. Some of the most interesting was a coupled matrix work by Ivory Clifford.

sourceforge.net/projects/ope...

This was later applied to conjugate heat/mass transfer systems and implicit pressure velocity coupling methods.

Finally, for simulations that do strive for accuracy or alignment with reality or to be simply useful in understanding a phenomena, apply some standards.

www.grc.nasa.gov/www/wind/val...

When a simulation uses lots of resources and is still wrong, there needs to be another reason to run it.

If that reason is an incremental improvement in a method or computer science or programming paradigm... Great.

Let's not automatically assume a large simulation is correct.

It looks amazing for sure, and I am amazed by the hugeness of the mesh and overall timescales.

However, I do question the validity. For an automotive case ( much lower Re number) one would use a smaller cell than this simulation to correctly capture the near wall physics in DDES turbulence model.

That's pretty awesome!

I'm attempting to start over with rust this time. We will see.

Enough to feel a little clever but not enough to quit my day job 😕

I didn't have this one on my bingo card. This is a bit worrying.

A more official new release from Cadillac

news.cadillac.com/newsroom.det...