CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Announcements from Other Sources (http://www.cfd-online.com/Forums/openfoam-news-announcements-other/)
-   -   LTS based Lagrangian particle solver and test cases (http://www.cfd-online.com/Forums/openfoam-news-announcements-other/103505-lts-based-lagrangian-particle-solver-test-cases.html)

ulli June 19, 2012 12:25

LTS based Lagrangian particle solver and test cases
 
Dear FOAMers,

a Lagrangian particle solver for the simulation of evaporative cooling of flue gas in a quenching device is hereby provided to the community.

The solver uses the Local Time Stepping (LTS) acceleration technique.

Further more a series of test cases is provided along with a short instructions manual in PDF format. The focus of this publication lies on the comparison of the LTS approach with the PISO/PIMPLE based solution method.

Solver, test cases and documentation can be found here:
http://www.dhcae-tools.com/contributions.htm

Feel free to use this thread for remarks, suggestions and questions.

Martin Becker and Ulrich Heck

Chrisi1984 August 1, 2012 05:07

visualization of the Lagrangian particles
 
Hello Martin, hello Ullrich,

first of all thank you very much for sharing that solver.

It seems to work very well.

But is it possible to visualize the lagrangian particles in this solver?

I did not get it working until now.

Thanks in advance!

Kind regards

Christian

MartinB August 1, 2012 05:17

Hi Christian,

it is not really possible in the cases that we provided. The lagrangian particles evaporate completely within the particle transport iterations, so there is nothing left to be visualized.

As we pointed out in the PDF document provided with the test cases and as you can see at these slides presented at the 7th OpenFOAM Workshop (http://www.openfoamworkshop.org/2012.../BeckerMartin/) it is necessary to append another simulation with another solver (for example reactingParcelFoam) to get the particles for visualization purposes.

Martin

Chrisi1984 August 1, 2012 09:10

Hi Martin,

thank you very much for the information.

Where should I start working on the solver that not all particles evaporate immediately?

Kind regards,

Christian

ulli August 3, 2012 02:37

Hi Christian,

there is no need to modify the solver. If you reduce the temperature of the hot gas in the test case, not all of particles should evaporate.

Best regards

Ulrich

Chrisi1984 November 8, 2012 03:11

injecting a liquid mixture
 
Hi,

its me again.

Your solver works fine for injecting water.

Now I would like to inject a liquid mixture of water and urea.

Then the water should evaporate as first fraction from the droplets. Later the urea concentration in the droplets should increase and the evaporation of the second fraction urea should evaporate.

I think therefore I have to switch the composition model from "singleMixtureFraction" to "singlePhaseMixture".

The only problem is that "singlePhaseMixture" is not available in your solver.

Can you please give me a hint, how I can make that composition model available and working in your solver?

Thanks in advance!

MartinB November 8, 2012 08:53

Hi Christian,

you can use the "singlePhaseMixture" in the solver this way:

In the dhcaeLTSThermoParcelSolver.C switch the cloud classes:
Code:

//#include "basicReactingMultiphaseCloud.H"  // <--- remove
#include "basicReactingCloud.H"  // <--- add

And in createClouds.H change:
Code:

//basicReactingMultiphaseCloud parcels  // <--- remove
basicReactingCloud parcels  // <--- add

Now the solver uses the singlePhaseMixture model selected in the reactingCloud1Properties file.

Good luck

Martin

Chrisi1984 November 10, 2012 10:17

Thank you Martin!

I can now use the singlePhaseMixture approach!

But it is a pity that the two components of the mixture still do not evaporate one after each other.

Both mixture components start evaporating to the same time. Although in real life the lower boiling component should evaporate first before the higher boiling component should evaporate.

Do you know how the injected mixture can be really handled as a mixture consisting of two different liquids with differnet boiling points?


Kind regards

Christian

MartinB November 12, 2012 12:53

Hi Christian,

you can try to add a second cloud by doubling the cloud definition, the source terms etc in the solver sources. Then you can handle two different fluids with there individual Tvap and Tbp etc.

Martin

Chrisi1984 February 4, 2013 15:38

Hi Martin,

by using two clouds I think I can dose two different fluids, but not a mixture of both. I am right?

I have now a new idea. Therefore I need the chemical reactions. How can I reintroduce that feature into your solver?

Thanks in advance.

Kind regards

Christian

MartinB February 5, 2013 09:49

Hi Christian,

you should compare the source code files of LTSReactingParcelFoam and dhcaeLTSThermoParcelSolver with each other line by line. For example in hsEqn.H the term "+ combustion->Sh()" must be added and so on. The make/options file must be adjusted, too.

Martin

JuanRodriguez February 24, 2015 13:18

Hello Ullrich,

Perhaps I arrived here late, but could you please share again your solver? (The links in the DHCAE Tools page are dead)

Thank you.

elvis February 26, 2015 09:05

Quote:

Originally Posted by JuanRodriguez (Post 533248)
Hello Ullrich,

Perhaps I arrived here late, but could you please share again your solver? (The links in the DHCAE Tools page are dead)

Thank you.

I can confirm those links http://www.dhcae-tools.com/Contributions.html are not working

you do not get the presentation
http://www.openfoamworkshop.org/2012.../BeckerMartin/
but you will get the presentation
http://sourceforge.net/projects/open...SlidesOFW7.pdf

ulli February 26, 2015 13:15

Dear Juan, dear Elvis

sorry for this. Now it should work again.

Best regards

Ulli

P.S. : I had to update the link: The solver is now here

http://www.dhcae-tools.com/Media.html


All times are GMT -4. The time now is 14:31.