[subsemitonium]

Hello,

I'm trying to validate my calculations with a simple example from VDI Heat Atlas, please find the attached case. I want to make sure that I can produce reasonable results before going any further. Unfortunately I get about 30% error - and all measures to reduce it failed. This is why I'm asking for help now.

Theory: The case is a 1D plate in laminar flow. Theory from VDI Heat Atlas (2013) says:
Nusselt_x = 0,332*sqrt(Re_x)*Pr^(1/3)
with the local Reynolds Number Re_x = w * x /nu with x being the distance over the plate.
In my example, the fluid is water with Prandtl = 7, beta = 2e-4, nu = 1e-6, mu = 1e-3, rho = 1000, cp = 4185

When I run my case, htc is about 30% too high. The solver is buoyantBoussinesqSimpleFoam. Turbulent Prandtl Number is without influence (laminar flow), turbulence models are turned off. I changed many parameters, the only influence I could found is the laminar Prandtl number. To my best knowledge, the laminar Prandtl number in transportProperties is the same "normal" Prandtl number as in thermophysicalProperties - which is a fluid property. Reducing it to a value of 3.5 corrects my results but seems very wrong. On the other hand, I am wondering, why Prandtl number is double-defined.

Any help is very welcome! Thank you.

Edit: Sorry, case file was too large - deleted mesh files. Please execute blockMesh ...

[subsemitonium]

Nobody interested?

I checked the wallHeatFlux utility by calculating snGrad(T) and using thermal conductivity of water as well as T_wall and T_inlet. I could approximately verify the values by the utility .. which are still about 30% too high compared to the VDI equations. Since I've checked everything else (even tried other solvers for the pressure field) I absolutely don't know ...

Any good ideas?