Sandia Flame D OpenFoam tutorial case - bad result?
 

Hi there!

I am trying to study combustion modelling within OpenFoam and am currently looking into the Sandia Flame D case - /opt/openfoam6/tutorials/combustion/reactingFoam/RAS/SandiaD_LTS

I copy the case into my own directory and execute provided Allrun script.
The results of the temperature field at the last time step are shown in the first picture (myD).

However (to me) - these results seem to be incorrect. If you google Sandia D flame tutorial in openFoam you get these screenshots that make more sense in terms of results.

For example from the following thesis(http://repozitorij.fsb.hr/9279/1/Del..._diplomski.pdf), page 79 (picture D2) - the results are very different yet according to the information there the conditions are identical to those found in the OpenFoam tutorial case.

Also there is a paper (https://link.springer.com/article/10...494-014-9561-5) looking into Sandia D flames and their OpenFoam simulation looks as follows (D3).

So my question is do I need to work with the case and modify to achieve these more sensible results or is it just a simple question of post-processing -as I assume the in-built openFoam tutorial on this case is correct.

Thank you!

Update on this -- I installed the most recent dev version of OpenFoam, which contains the updated Sandia Flame D tutorial and produces much better results.

Hi Brandani,

Would it be possible for you to share the updated results you mentioned? We are implementing the flamelet/progress variable approach in OpenFOAM, and obviously Flame D is one of the most used benchmark cases. I wonder what the results look like using the existing solvers in OpenFOAM.

Thanks!

So just to clarify:

1) What I was saying is that SandiaD flame in OpenFOAM version 6 seemed to produce inaccurate results (I believe there was no ignition with the EDC combustion model - 1st picture in my first post)

2) Then, what I found is that the OpenFOAM development version (https://openfoam.org/version/dev/) + OpenFoam v7 (https://openfoam.org/release/7/) must have had their in-built tutorial on Sandia Flame D updated because there was ignition and results seem to match those pictures (2nd and 3rd) I attached in my first post

3) So install v7 or the dev version of OpenFOAM and you will have a working base case. I did not do any modifications to the case in these versions.

Hope this helps!

This is the T field at last time step produced by the in-built case in the dev version.

Does anyone have an explanation for the results in v6? One gets actually the same "bad" result in v2106.

Hi dear Brandani
I would appreciate if you could answer my question.
Actually I'm using SandiaD_LTS test case, but the time unite which OpenFOAM is using does not make sense me at all. Is the unite of time in second? Because in the default setting, 'end time' is set to 1500, but this time for combustion (injection) is not a reasonable value ! In combustion the time is ususallly in millisecond scale.
Do you have probably any Idea about time units in Openfoam, specially by using this start time and end time and also Delta T.
In attachment you can also find the picture of the ControlDict.orig file.

Thanks alot in advance

Sandia Flame D OpenFoam tutorial case - bad result?
 

All quantities in OpenFOAM should be in SI system of unit.

End time 1500 here refers to the maximal number of time steps (intended as number of iterations).

The value to the time step (or time increment) in controlled by the PIMPLE dictionary inside system/fvSolution