div schemes error using Gauss linear
 

Hi,
I had already asked a question in a previous post: https://www.cfd-online.com/Forums/op...mperature.html


I could restrain my problem of wrong enthalpy and wrong temperature to the convection part of the energy equation in EEqn.h fvm:div(phi,h) in other words .

When using the second order Gauss linear my enthalpy is dropping and the temperature falls under 293K, which is not physical. However when using Gauss upwind the enthalpy remains to it initial state, which is correct and the minimal temperature is still at 293K.
Remark phi is very small in my case and I doubt that the second order linear scheme causes dispersion error and the dissipative upwind smooth the error out.



Is this possible?