[student666]

Hi all,

I've been always thinking that openfoam version 4.1 (and previous) that comes by cfdDirect is suitable for almost all physics, in a general way of speaking.
Anyway today, a colleague of mine advised me different.
He says that segregated solvers for cht and especially for combustion, are not giving you any chance to have a stable solution.
So trivial question:
Is it true?
He also said that by using Fluent, he had never been able to get a good solution, so he switched to CFX.

So....What are the main difference between coupled and segregated?

And so:

What is the most suitable openfoam fork to be used to solve cht and combustion cases? Or better, what I should consider in using one or another fork?

[p.b]

See here:
Segregated vs Coupled Solver
Segregated or Coupled flow?

Both segregated and coupled approaches have advantages and disadvantages.

Coupled solvers are available in foam-extend (block-matrices):
https://openfoamwiki.net/index.php/I...oam-extend-4.0
https://sourceforge.net/p/foam-exten...lvers/coupled/
https://sourceforge.net/p/foam-exten...rials/coupled/

But you probably need some time to make yourself familiar with the idea behind it, e.g.:
http://www.tfd.chalmers.se/~hani/kur...ersion-2.0.pdf
http://www.personal.psu.edu/dab143/O...ord_slides.pdf
http://www.sciencedirect.com/science...45794916306046
https://www.researchgate.net/profile...nce-Frames.pdf

[akidess]

You can tell your colleague Fluent also has a coupled solver, it's just not the default.

[student666]

Thank you guys for the info and for the reference papers. Regards

Sent from my ASUS_X008D using CFD Online Forum mobile app

[arjun]

Originally Posted by student666

Hi all,

I've been always thinking that openfoam version 4.1 (and previous) that comes by cfdDirect is suitable for almost all physics, in a general way of speaking.
Anyway today, a colleague of mine advised me different.
He says that segregated solvers for cht and especially for combustion, are not giving you any chance to have a stable solution.
So trivial question:
Is it true?
He also said that by using Fluent, he had never been able to get a good solution, so he switched to CFX.

So....What are the main difference between coupled and segregated?

And so:

What is the most suitable openfoam fork to be used to solve cht and combustion cases? Or better, what I should consider in using one or another fork?

I have been coding finite rate chemistry and testing the solver against various benchmarks.
So far i would say that your friend is right as far as fluent goes. I have had gotten strange results too from it even for relatively simple cases sometimes.

However it is not correct that you can not get good or stable results from segregated solver. It depends on segregated solver used. The classic segregated approach do indeed has stability issues , but segregated solver could be rewritten thinking that now energy is also a player in the game. (the standard segregated algorithm is not designed thinking non isothermal case).

This is why I provide another segregated solver for combustion cases and so far I am happy.