Hi!

My problem is this: I'm simulating fluid enclosed in a cyllinder with adiabatic walls. At the begining of transient temperature is the same everywhere. I have no mass flows in or out, but the average temperature is changing with time(rising). When I looked at temeperature field through time, temperature in cells next to wall was falling and their neighbours temperature was rising:

ex.

time JLEN-2 JLEN-1 JLEN

s K K K

0 383.0 383.0 383.0

0.01 383.0 383.019 382.984

...

5 391.5 395.0 356.8

I noticed, that this effect is grid dependant and not so much model dependant, beacause I tried different configurations.

Does anyone have a clue what is all about and how to avoid this.

Thank you for your help

Miroslav

Looks like either a convergence issue (make sure that enthalpy is fully converged to a very low residual value) or maybe some machine round-off errors.

If your enthalpies are actually what are being solved. Make sure that Tref is set to something just under your lowest expected temperature value. This will keep the magnitude of your enthalpies as low as possible. Also try double precision.

Jeff