how to turn on the low-Mach approximation?

Questions and remarks about code_saturne usage
Forum rules
Please read the forum usage recommendations before posting.
Post Reply
daniele
Posts: 148
Joined: Wed Feb 01, 2017 11:42 am

how to turn on the low-Mach approximation?

Post by daniele »

Hello,

I am a little bit confused about the correct way of setting Code_Saturne to use a Boussinesq approach or a low-Mach approach.
I use v6.0.0. In the GUI, one has the possibility of selecting Incompressible or Compressible flow. I guess that in both cases the Incompressible option has tu be chosen... Then, for the Boussinesq approach, the g*Drho term is added through the cs_user_source_terms subroutine. I am right?
But what about the low-Mach formulation? How can I turn it on?
I am running simulations with multi-species: density varies AS A FUNCTION OF THE SCALAR MASS FRACTION, ENERGY EQUATION IS NOT REQUIRED AND NOT SOLVED (i.e. density is not a function of temperature).

Thank you very much.
Best regards,
Daniele
Erwan Le Coupanec
Posts: 45
Joined: Sun Sep 08, 2013 8:50 pm

Re: how to turn on the low-Mach approximation?

Post by Erwan Le Coupanec »

Hello Daniele,

Through the GUI, you're correct you have to select the standard (incompressible) model.
Then in 6.0, you've got the possibility to enable a Boussinesq approach by setting in cs_user_parameters.c, idilat=0 (dans la structure stokes_model, voir exemple dans doc Doxygen); then you fill in the thermal expansion field (field name is "thermal_expansion") in cs_user_physical_properties.c.

The Low-Mach model can be enabled with idilat=3, though I'm not sure of what you exactly want.
This model adds up an energy equation with a source term taking into account the time variation of the thermodynamic pressure (an integral thermodynamic pressure for this model).

What is the relative variation of rho expected in your case ?

Regards,
Erwan
daniele
Posts: 148
Joined: Wed Feb 01, 2017 11:42 am

Re: how to turn on the low-Mach approximation?

Post by daniele »

Hello Erwan,

Thank you for your help.
Actually I am simulating the cold leg mixing experimental test, where density difference depends on the species (scalar) mass fraction. Density difference is around 200 kg/m3, with the average density (between the two species) equal to around 1000 kg/m3. If I remember well you used a Boussinesq approach..? In this case I should simply have to add a source term equal to (rho-rho_ref)*g in the momentum equation parallel to gravity.

By the way, idilat=0 exists in v6, but not is v5. What is the value of idilat to be chosen in v5 in order to retain the Boussinesq approach (1 or 2)?

Regarding the Low-Mach approach, I thought it simply meant the the density was kept inside the d/dt and div terms. I misunderstood the meaning. If it requires the energy equation, it is definitely not what I need.

Thank you. Best regards,
Daniele
Post Reply