Issue regarding Two Temperature Gas Model

[aStRo-Jim]

Hello,
I am trying to do a simulation to validate the following experiment:

https://www.sciencedirect.com/science/article/pii/S1000936122002217

which consists of reacting hypersonic gas flow over a half-cylinder. It uses Gupta's 5 species model and the Two Temperature Gas Model. To do this I modified the '2D/sphere-lehr/' example simulation in Eilmer 5. The results, however, do not match with the ones from the paper:

The vibrational temperature is especially dissimilar and this is leading me to believe that I am not setting up the thermochemical model correctly. Also, my simulation does not fully converge and gets stuck around relative_residual = 1e-6, where it oscillates indefinetely between 3 values. I have attached the energy exchange file, the reaction file, the gas model file, the job.lua file and the diagnostics:

I have also seen that the stop_on_relative_residual does not work sometimes, as I have set it at 1e-6 and the simulation does not stop when that value oscillates below the threshold (it reaches 8e-7 when it oscillates).

chem.txt
energy_exchange.txt
gas.txt
job.txt

[rjgollan-on-github]

I don't think iterative convergence is the issue. It seems deep enough to me to make an assessment about shock layer thickness and structure.

Instead, I think the physical modelling needs some attention. Please check Knudsen number and make a judgement as to whether a Navier-Stokes model is appropriate at these conditions.

[Mr-Tun]

Hi @aStRo-Jim ,

The reference paper you are working on said in Section 3 Thermochemical nonequilibrium models as "More details about
these constants appearing in the above Arrhenius expressions are given in Zhao’s dissertation", I donnot know if you have already checked or access to this reference, I just had a quick look of Zhao's PhD dissertation, your chem.txt file is not completely same as the one he used. Also, the thermal and viscous property of gas are also given in his dissertation appendix.

To validate the numerical simulation maybe you need to set your rate and property as same as the appendix of his reference dissertation.

I attached the screenshot of his appendix.

Best Regards,
Yufan