CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM (https://www.cfd-online.com/Forums/openfoam/)
-   -   To ALL Combustion Foamers please join (https://www.cfd-online.com/Forums/openfoam/76573-all-combustion-foamers-please-join.html)

hk318i May 29, 2010 10:10

To ALL Combustion Foamers please join
 
Dear Foamers,

I am a new OpenFOAM user working on non-premixed combustion flames (diffusion flames). I was using Fluent :mad: before.
Currently, I am searching for all available combustion models in OpenFOAM and I noticed the following;
  1. The models descriptions in OpenFOAM are not enough to figure out which model it is.
  2. The current combustion models in OpenFOAM 1.6.x are missing many important models, however there are many universities and research groups already implemented theses models. Unfortunately they are not sharing their codes.
For the the first point, I think by more searching and reading the codes itself man can figure out the details of the solver which is one of the main OpenFOAM advantages.
The problem is the second point because it makes open source codes lose it's essence. That will make each new user or group decides to use OpenFOAM to build its own solvers and models which already done by others. I hope that these codes are shared somewhere and it was my fault to find it. Can anyone show me where can I find the combustion models codes especially for diffusion flames? and which models are never implemented in OpenFOAM before? Where are the active combustion groups? or should we make a one?!

chandramurthy May 30, 2010 11:15

Wait for few days, in the upcoming 5th OpenFOAM workshop considerable advancement in combustion modelling is expeted.

http://web.student.chalmers.se/group...stractOFW5.pdf

hk318i May 30, 2010 11:22

Thank you :)

alberto May 30, 2010 13:05

Quote:

Can anyone show me where can I find the combustion models codes especially for diffusion flames? and which models are never implemented in OpenFOAM before? Where are the active combustion groups? or should we make a one?!
I'm interested in the combustion group too for a project with a student. Is there one somewhere? :-)

However I don't see a major problem in re-implementing the model. Sometime it is easier than understanding others code, if not written very cleanly and documented. And in these years I have learnt that "announcements" of new features and solvers by members of the community do not mean these features will be released by the authors, and, sometime, they don't mean they are actually ready. In other words, if you need a model, start coding, don't wait :D

hk318i May 30, 2010 13:23

You are right. I should start coding what I need. What is your advice for a new foamer? :)

What do you think, should we start a new combustion group? I did not find anyone also. :(

l_r_mcglashan May 30, 2010 17:13

What exactly are you trying to model? If your chemistry is complex then you'll have to use flamelet methods. My group has a flamelet equation solver available on sourceforge which is written in C++, and I have used that to solve lagrangian style flamelets for a laminar diffusion flame successfully. Takes a long time to solve though... You're probably best off just writing it yourself to be honest, it didn't take that long.

Regarding point 2), have you actually contacted other research groups? In my experience they are often very willing to share code, you just have to ask nicely :D

alberto May 30, 2010 18:11

Quote:

Originally Posted by hk318i (Post 260907)
You are right. I should start coding what I need. What is your advice for a new foamer? :)

What do you think, should we start a new combustion group? I did not find anyone also. :(

It would be useful.

My knowledge of combustion is purely theoretical and quite limited though, since I have never actually worked on it (my interest is elsewhere, aka multiphase flow), so I doubt my contribution could be of much use :-)

hk318i May 31, 2010 04:32

Quote:

Originally Posted by l_r_mcglashan (Post 260913)
What exactly are you trying to model? If your chemistry is complex then you'll have to use flamelet methods. My group has a flamelet equation solver available on sourceforge which is written in C++, and I have used that to solve lagrangian style flamelets for a laminar diffusion flame successfully. Takes a long time to solve though... You're probably best off just writing it yourself to be honest, it didn't take that long.

Regarding point 2), have you actually contacted other research groups? In my experience they are often very willing to share code, you just have to ask nicely :D

I am trying to model turbulent non-premixed flames of hydrogen-hydrocarbons fuels.
I am new OpenFOAM user and I do not have experience yet about the working groups in this area :(.
What is the name of your group project in sourceforge?

l_r_mcglashan May 31, 2010 04:56

http://sourceforge.net/projects/mopssuite/ is the site, you can svn clone the repository and the flamelet solver is called 'camflow'. It can use CVODE or RADAU5 which come with the repository. There are some examples you can run it on and also you can generate some partially completed documentation using doxygen.

l_r_mcglashan May 31, 2010 05:02

Also, if you want an overview of turbulent combustion, I like this book:

http://www.amazon.co.uk/INTRODUCTION.../dp/1860947794

Maybe I'm slightly biased because I went on the course, but it covers almost everything and you can take it as a starting point.

hk318i May 31, 2010 05:03

Quote:

Originally Posted by l_r_mcglashan (Post 260966)
http://sourceforge.net/projects/mopssuite/ is the site, you can svn clone the repository and the flamelet solver is called 'camflow'. It can use CVODE or RADAU5 which come with the repository. There are some examples you can run it on and also you can generate some partially completed documentation using doxygen.


Thank you :)

hk318i June 8, 2010 14:47

Quote:

Originally Posted by l_r_mcglashan (Post 260913)
What exactly are you trying to model? If your chemistry is complex then you'll have to use flamelet methods. My group has a flamelet equation solver available on sourceforge which is written in C++, and I have used that to solve lagrangian style flamelets for a laminar diffusion flame successfully. Takes a long time to solve though... You're probably best off just writing it yourself to be honest, it didn't take that long.

Regarding point 2), have you actually contacted other research groups? In my experience they are often very willing to share code, you just have to ask nicely :D

I downloaded it, but I did not find any .C files. :confused:

l_r_mcglashan June 8, 2010 16:06

Are you sure? What can you see? There should be trunk folders within each software directory, and they'll have include/ and source/ folders in them.

hk318i June 8, 2010 16:24

Quote:

Originally Posted by l_r_mcglashan (Post 262149)
Are you sure? What can you see? There should be trunk folders within each software directory, and they'll have include/ and source/ folders in them.


I found the following folders/
auxilliary
sweep2
xmlf90
gpl.html

l_r_mcglashan June 8, 2010 16:49

Ah, this code is under development, so you have to get it from the svn repository:

svn co https://mopssuite.svn.sourceforge.net/svnroot/mopssuite mopssuite

You can view the svn repository here:

http://mopssuite.svn.sourceforge.net/viewvc/mopssuite/

hk318i June 8, 2010 17:01

Quote:

Originally Posted by l_r_mcglashan (Post 262153)
Ah, this code is under development, so you have to get it from the svn repository:

svn co https://mopssuite.svn.sourceforge.net/svnroot/mopssuite mopssuite

You can view the svn repository here:

http://mopssuite.svn.sourceforge.net/viewvc/mopssuite/

Thank you
I downloaded it

hk318i June 9, 2010 08:30

Dear All,

I need your advice. I am wondering, is it possible for a new OpenFOAM user to implement the flamelet model for turbulent non-premixed flames of hydrocarbons mixture fuels like (ch4/h2) alone (as a master student)? How long will it take? How complex is it? concerning that, I will try to follow the available codes as mush as possible, It will be my full time work and I need to have some output by next September.

What do you think? :confused:


All times are GMT -4. The time now is 00:47.