Turbulence modeling is a complex field. Trying to strike a balance between accuracy and computational efficiency has given rise to a relatively large number of different turbulence models. For more details see "RANS-based turbulence models". There is no definitive list to match turbulence models to applications.

In Caedium the default turbulence model is k-omega SST which is considered to be a relatively robust, accurate, and efficient model for a wide range of applications. Unless you are seeing results that call this into question I would suggest you stick with this model. Before considering a different turbulence model first remember that poor results are often due to:

• Poorly positioned boundaries, e.g., an outlet too close to strong velocity gradients
• Poor mesh elements (skewed)
• Insufficient mesh resolution in critical flow regions

Also note that using wall functions (as for the majority of turbulence models in Caedium), to avoid extremely fine near wall meshes, requires that the y+ values be in the range 30-300 for the best results.

Independent of the turbulence model to improve accuracy, sometimes at the expense of robustness, you can try:

• Substance->Solver->Schemes->Divergence = Linear Upwind
• Physics:Wall->Type:Default->nut = U Spalding Wall Function

If you know that the fluid flow is likely to be entirely laminar due to the Reynolds number (e.g., in a pipe flow with Re < 2300) then you should switch the turbulence model to Laminar.

If you'd like to try another widely used and reliable turbulence model then I'd suggest the Realizable k-epsilon.