Comparison to Bespoke Conjugate Sim

Comparison to Bespoke Conjugate Simulations Having validated the performance of the CFD using the above simulation methodology, bespoke simulations were performed with the sole intent of validating the thermal resistance model. This would go some way to mitigating the additional conduction/flange effects commented upon above. As before, the simulations were steady-state in nature and utilised the realisable k-ε turbulence model with enhanced wall treatment. Solid conductivity was again specified to vary with solid temperature, and the fluid density was determined via the ideal gas law. Given the aforementioned comments on impingement zone Nusselt number, an increase in mesh density near the heat transfer surfaces was made with 15 prism layers towards the solid/fluid interfaces and a tetrahedral mesh away from the wall. The simulations utilised the repeating nature of the unit cell block via symmetry planes to minimise the domain size. An example of the domain (and mesh) used in the conjugate simulations along with some of the specified boundary conditions are provided in Figure 6 and Table 3. Note that, for geometries G1 and G2, the domain consisted of 8 unit cells, whilst for geometries G3 and G4, the domain was comprised of 4 unit cells due to the increased physical size of these geometries.