[Obad]

Hi guys,

I'm currently writing a code to solve the laminar boundary layer equations over a flat plate for an arbitrary pressure distribution. To validate my code I compared my results with the solution of the Falkner-Skan equation for incompressible boundary layers with pressure gradient.

I compared the velocity profiles at different axial positions on the flat plate. In the front the numerical results seem reasonable but further in the back the results diverge. I attached a graph that shows both the numerical and the Falkner-Skan results, these results are with a favourable pressure gradient.

My code works fine for flow without pressure gradient and for flows with a moderate pressure gradient the results are good. But as I increase the pressure gradient (more favourable) the numerical results start to form a "bump" in the middle.

I use a rectangular grid and at the upper boundary I specify the u-velocity distribution as it is given by the pressure gradient and set the gradient of the v-velocity to zero.

Does anyone have a clue what could cause this problem?

Cheers!