H2 Separation

Benfa · November 16, 2012, 15:06

Hi,

I am simulalting the transient separation of H2 in air in a room. The rooms boundaries are walls. The volume is initilaized with air while in the lower volume (bottom of the room) h2 with a specifi mass fraction is initialized. The fluid is defined as variable mixture (H2 and air). Air is used as constraint while H2 is calculated with the transport equation. I also activated kinematic diffusivity for H2. Boyuoncy is activated as well.

Problem is that I do not observe the expected diffusion and density difference driven flow of H2 to the top of the room. Normally the h2 concentration at the top should increase.

Anyone who simulated a similar phenomena or did I anything wrong with the mixture?

Thanks in advance

ghorrocks · November 17, 2012, 20:05

Have you done all the normal checks: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

evcelica · November 19, 2012, 19:56

Well diffusion is an extremely slow process, and the H2 won't actually separate from the air, they will always diffuse into each other. A problem may be that since you are doing this problem in a 1D manner, the buoyant force on the hydrogen can't really do much since the air is kind of "in the way" equally of it forming a buoyant plume and rising to the top. Think of this as being analogous to a pendulum that you have initialized as being upside down and being perfectly balanced. Of course the slightest disturbance would cause it to fall down (your hydrogen would go up) but if you initialize it in a perfect 1D manner, buoyancy ( or weight on the pendulum) would do nothing, and you would only see molecular diffusion taking place.

I would start the simulation from somewhere earlier in time. How did the bottom of the room start getting filled with the hydrogen?

November 16, 2012, 15:06	H2 Separation	#1
Benfa Member Benny Join Date: Apr 2012 Posts: 40 Rep Power: 14	Hi, I am simulalting the transient separation of H2 in air in a room. The rooms boundaries are walls. The volume is initilaized with air while in the lower volume (bottom of the room) h2 with a specifi mass fraction is initialized. The fluid is defined as variable mixture (H2 and air). Air is used as constraint while H2 is calculated with the transport equation. I also activated kinematic diffusivity for H2. Boyuoncy is activated as well. Problem is that I do not observe the expected diffusion and density difference driven flow of H2 to the top of the room. Normally the h2 concentration at the top should increase. Anyone who simulated a similar phenomena or did I anything wrong with the mixture? Thanks in advance

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Flow regime behind the separation	husker	Main CFD Forum	1	November 2, 2012 06:34
wall function and flow separation	O.D.Y.	FLUENT	1	October 3, 2011 17:27
Separation prediction over a 2d cylinder....	Kyle	FLUENT	2	December 9, 2005 05:17
Separation	An Modh Coinniolach	FLUENT	4	April 1, 2002 12:31
Euler + separation again	Oliver	Main CFD Forum	23	June 19, 2001 13:47

November 17, 2012, 20:05		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	Have you done all the normal checks: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

November 19, 2012, 19:56		#3
evcelica Senior Member Erik Join Date: Feb 2011 Location: Earth (Land portion) Posts: 1,188 Rep Power: 23	Well diffusion is an extremely slow process, and the H2 won't actually separate from the air, they will always diffuse into each other. A problem may be that since you are doing this problem in a 1D manner, the buoyant force on the hydrogen can't really do much since the air is kind of "in the way" equally of it forming a buoyant plume and rising to the top. Think of this as being analogous to a pendulum that you have initialized as being upside down and being perfectly balanced. Of course the slightest disturbance would cause it to fall down (your hydrogen would go up) but if you initialize it in a perfect 1D manner, buoyancy ( or weight on the pendulum) would do nothing, and you would only see molecular diffusion taking place. I would start the simulation from somewhere earlier in time. How did the bottom of the room start getting filled with the hydrogen?