search

mutationpp / Mutationpp

The MUlticomponent Thermodynamic And Transport library for IONized gases in C++
GNU Lesser General Public License v3.0
101 stars 58 forks source link

Two-Temperature model using NASA-9 #65

Open grgbellasvki opened 5 years ago

grgbellasvki commented 5 years ago

Currently, the only way to do multi-temperature models is with RRHO model. Often for increased accuracy one should use instead the NASA-9 polynomials. The aim of this issue is to extend the library to work with multi-temperature models and NASA polynomial databases.

This is an extension necessary for the coupling of Mutation++ with US3D.

Briefly, for a two temperature model the total enthalpy for heavy is computed as: hi(T,Tv) = 5/2RT + (hi(Tv)- 5/2RiTv) (To be checked)

jbscoggi commented 5 years ago

Hey @grgbellasvki this would be great to have added. Of course, you could generalize it as a way to extend any single-temperature thermo to a two-temperature thermo. For example, you could even use a RRHO taking only a single temperature which would be useful for testing because you should retrieve the same answer as for two-temperature RRHO. Just a thought...

lkampoli commented 5 years ago

Hello, in this regards, as I understand, there is no "default" way to switch to a 2T-model, for example Park 2T for air 5, right? But, in the past several simulations were done with COOlFluiD with 2T.

Of course here the hint is that, I'm gradually moving to implement from 1T -> 2T -> STS. So, I'm interested in the topic.

Best regards, Lorenzo

lkampoli commented 5 years ago

Hello, I understand that I wrote it wrong! Of course it is possible to use a 2T model, just I would like to add our formulation.

Regards, Lorenzo

lkampoli commented 5 years ago

Hello, may I ask a typical set of database (mixture, mechanism, transfer) file set to run air 5 2T Park model with Mutation, please?

grgbellasvki commented 5 years ago

The mixture and mechanism files for single and multi-temperature models are identical. What changes is that you should choose the correct StateModel, "ChemNonEq1T" for single temperature, "ChemNonEqTTv" for Park's two temperature model. Note that if you choose "NASA" polynomials for thermo_db the state model "ChemNonEqTTv" should be avoided; "RRHO" is the recommended option. We want to create a new thermo_db model which is suitable for multi-temperature models using the NASA polynomials for the total energy. The idea is similar to the one presented in Candler and MacCormack 1991 paper: "Computation of Weakly Ionized Hypersonic Flows in Thermochemical Nonequilibrium", Equation (5).

Considering energy transfer terms, again the StateModel is responsible for choosing the appropriate ones. Please give a look to the ChemNonEqTTv class for what is currently implemented.

Hope I answered your question.

air5_example.zip

lkampoli commented 5 years ago

Hello, yes thanks. Just I was in doubts because I get some problems when I run the simulation. Since I still get error, may I ask you whether some of you has a working testcase with air5 2T to run in COOLFluiD, please?

jbscoggi commented 5 years ago

Hey @lkampoli, just taking a look at this conversation and wanted to make sure you realized that the mixture file for air5 was recently renamed to be air_5 in order to be consistent with other mixtures. Maybe that is the source of the error?

lkampoli commented 5 years ago

Hello, yes I noticed. It is not that, but thanks for the warning! Could also be something similar somewhere else ...

jbscoggi commented 5 years ago

You can always use checkmix to make sure the mixture you want is at least loaded correctly.