Hallo,

I'm working on problem of natural convection.(a tank with water inside,heated trom bottom) If I use bussinesq model of density everything is ok. The problem begins when I use polynominal ,piecewise-linear, or piecewise-polynominal model. The the static pressure field should grow with -Z direction , but in the middle of tank I have divergence. The problem exists even with high Ra.numbers. For every send message thanks a lot.

Michal

I don't have much experience in natural convection modelling, but what is your pressure differencing scheme? I believe there is an option that is better for problems with substantial body-forces than the default.