[Jan995]

Hello there,

I have written finite difference predictor-corrector code for incompressible 2D flow over an obstacle, as well as supersonic flow in a nozzle from MacCormack scheme.
However, I have huge problems with modifying this code to work for low Reynolds number 2D flow between two plates with heat transfer on the walls. I want to produce strict laminar flow, but my code is not converging. In next section, I also want to add an obstacle.

Can anyone help me with modifications needed for the algorithm from Anderson`s CFD book on MacCormack scheme, that need to be made so that laminar solution will converge?

If there is any better finite difference (preferably explicit) algorithm that is easy to implement in Matlab/Python, please let me know.

Thank you for all suggestions.

[AshwaniAssam]

Hi,


MacCormack scheme is essentially meant for simulating compressible flows. The essential problem with scheme designed for compressible flow is the presence of large dissipation. You have to search for a generalized Maccormack scheme which is also applicable for low speed calculation. However, you can use SLAU convective scheme. It is a convective scheme proposed by Kitamura and Shima.



A probable recent work on generalized MacCormack Schme is


"Development of the Generalized MacCormack scheme and its extension to low Mach number flows: Generalized MacCormack Scheme for Low Mach Numbers"

• 10.1002/fld.4377