[celestial]

Hello everyone

The heat flux is continuous at the interface between two slabs of different conductivity, the boundary condition being of course

k1*Grad T1 = k2*Grad T2

What amazes me is that everybody uses the segregated coupling approach to the heat diffusion solver: namely for these two slabs with their individual meshes, iterate that solver over both slabs until the desired level of convergence is achieved.

Nobody uses the block coupled version. Otherwise, there would be a solver and accompanying tutorial made available.

Beats me. Can't figure it out as I believe block coupling is the way to go. Imagine what a nightmare this iteration would be like if we were dealing with not two adjacent slabs but an arbitrary 3D complex where each simplex would have a different conductivity.

Have a nice weekend