Weak Enforcement of Heat Flux BC

[GomerOfDoom]

Hello all,

I'm trying to wrap my head around the heat flux wall boundary condition. My question arises because, in running a simulation of flow through a serpentine diffuser, despite specifying a zero heat flux wall bc, I am seeing significant heat fluxes in the surface output (see attached picture).

I would have expected the code to drive the temperature gradient much closer to zero, but maybe I don't understand quite how strongly (or weakly) the heat flux BC is being enforced?

Can anyone shed light on how close to zero I should expect the heat flux to be, and why I might be seeing such large values in my results?

EDIT: It was pointed out in another discussion that the heat flux output in the paraview files is calculated from the gradient between the wall vertex and the first off-wall vertex, and that this gradient value is not really "at the wall," but rather slightly off the wall. I think it's still worth discussing whether these values are reasonable and/or if there is a better way to output the heat flux values to the surface output file.

Thanks,
Paul