CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   TNT Explosion Simulation (http://www.cfd-online.com/Forums/openfoam/114231-tnt-explosion-simulation.html)

 CharlieTan84 March 7, 2013 09:00

TNT Explosion Simulation

Hello People,
I am recently conducting research on Blast simulations. I use the commercial code "Ansys-Autodyn" but I would like to use OpenFoam if it's possible.
I think Explosion Simulations are based on "Euler Equations", so therefore I think I can run an explosion simulation on Openfoam.

Any kind of help will be appreciated my friends.
Thank you very much and have a great day!!
Charlie

 CharlieTan84 March 11, 2013 16:42

 elvis March 13, 2013 09:34

Hi,

http://www.fsb.unizg.hr/?OpenFOAM_Summer_School_2012
shows a picture from a underwater explosion (the EGP in the filename of the pic, points to Professor Paterson), but I do not know they have done that.

But I guess that Prof Jasak´s Blockcoupled Matrix Solver (OpenFOAM 1.6-extend) was used for that solver. http://www.personal.psu.edu/dab143/O...ord_slides.pdf

sorry that I only posted hot air.

 elephantwalker March 13, 2013 21:05

explosions, the easy way

Explosions are a whole realm of modeling in themself separate from CFD.

It is true that you need to start with a compressible density based method. I would use one of the applications that includes the shock wave in it for hints.

so sonicFoam or sonicDyMFoam.

What's important are the inital conditions. The stick of TNT should be assumed to be at a certain pressure and density (say 1 kg, 1" diameter, 10" long, and 1000 ATM) with the temperature being whatever PV=NRT predicts. The initial TNT reaction is not modeled, you just model the response of the air to this impulse.

Use an ideal gas law (the temperature of TNT is supercritical, so it is a perfect gas). The form of the energy equation should be be in density, not pressure, as the Euler equation is also in density.

Use a moving mesh, because as the shockwave is produced, the moving mesh will follow the shock wave.

That's the easy way to do it.

Its kind of like modeling the dam break model with a liquid, only you are using a gas against another gas with a really high pressure/temperature. Things are going to move faster.

:)....

-Lawrence

 CharlieTan84 April 19, 2013 08:45

Hello elvis and elephantwalker,

Sorry for answering so late. Thank you very much for your help. I will try your methods and post my experience on the forum, so that other people can benefit from it.

Cya!

 newmancfd May 7, 2013 03:55

Hello CharlieTan84,

It's nice to see ur question here, recently I'm working on the same topic. But I'm still doing the theory preparation. In about 2 weeks I will begin the simulation. So I wish u can share ur experience with us. Perhapes we may share the same problem, and we may solve it together. XD

 CharlieTan84 May 8, 2013 05:07

Hello Newmancfd!
I am so happy to hear from you!
Nowadays I am very busy with other stuff, but I would like to start with openfoam as soon as possible!
What is your exact research topic? What kind of theory preparation are you doing? May you recommend me some books/materials?
And yes let's solve this problem together!

 newmancfd May 13, 2013 03:28

Hello!
My topic is on the air explosion in urban area. Now I'm reading <Fluid Mechanics> by Landau and <Computational Gasdynamics> by Laney, which are both very general. I suppose the theoretic preparation is important, although I'm stucked at comprehending some abstract conceptions at this moment. But I'll solve them all after a few days I think.
Hope my message will do u some help.

 hesamking June 24, 2014 19:41

Thank you so much.
I am doing blast with CFX. Would you suggest using two domains (one smaller as the detonation and the other one being big contatining the structure) and could you suggest how to implement this?
Thanks,

Hesam

Quote:
 Originally Posted by elephantwalker (Post 413821) Explosions are a whole realm of modeling in themself separate from CFD. It is true that you need to start with a compressible density based method. I would use one of the applications that includes the shock wave in it for hints. so sonicFoam or sonicDyMFoam. What's important are the inital conditions. The stick of TNT should be assumed to be at a certain pressure and density (say 1 kg, 1" diameter, 10" long, and 1000 ATM) with the temperature being whatever PV=NRT predicts. The initial TNT reaction is not modeled, you just model the response of the air to this impulse. Use an ideal gas law (the temperature of TNT is supercritical, so it is a perfect gas). The form of the energy equation should be be in density, not pressure, as the Euler equation is also in density. Use a moving mesh, because as the shockwave is produced, the moving mesh will follow the shock wave. That's the easy way to do it. Its kind of like modeling the dam break model with a liquid, only you are using a gas against another gas with a really high pressure/temperature. Things are going to move faster. :).... -Lawrence

 All times are GMT -4. The time now is 02:03.