Zero flux boundary condition for multicomponent flow

syble · June 7, 2015, 17:48

Hello,

My simulation is of a CO2-air variable composition mixture through a geometry. The geometry has an inlet (with a specific velocity profile and CO2 mass fraction) and an opening with a zero relative pressure. I would like to set a Neumann boundary condition (for CO2 mass fraction) on the opening boundary. Ie that the spatial gradient of CO2 mass fraction is zero on this opening boundary (since I do not know what the CO2 mass fraction at this boundary is but there should be no flux/rate of change of CO2 at the boundary).

However the only option in CFX pre is to specify the exact mass fraction.

Is there a way of having a Neumann boundary condition on this opening?

Thanks in advance for your help.

ghorrocks · June 8, 2015, 06:27

The Neumann boundary for incoming mass fraction does not seem to make physical sense. How can the gas entering the domain have a composition dependant on properties upstream? Isn't the composition dependant on what is downstream?

syble · June 8, 2015, 19:18

Thanks for your reply.

Yes I agree that when the opening acts as an inlet the flow coming into the domain will depend on the gas downstream, so that would require a specified mass fraction for the gas coming into the domain.

However when the opening acts as an outlet, I don't want to be specifying the mass fraction at the outlet as I do not know what the mass fraction of the gas will be as it leaves the domain. So I thought that setting a Neumann boundary condition would account for this. Let me know if this sounds incorrect/not physical.

ghorrocks · June 8, 2015, 19:43

CFX automatically switches between MF being determined by the upstream conditions on inflow to the domain, and the MF being determined by the MF of the fluid exiting the domain. So on outflow the MF is set by the convected fluid and no Neumann boundary need be set.

syble · June 8, 2015, 19:50

Okay that makes sense, thank you very much for clarifying this for me.

June 7, 2015, 17:48	Zero flux boundary condition for multicomponent flow	#1
syble New Member Join Date: Jun 2015 Posts: 10 Rep Power: 10	Hello, My simulation is of a CO2-air variable composition mixture through a geometry. The geometry has an inlet (with a specific velocity profile and CO2 mass fraction) and an opening with a zero relative pressure. I would like to set a Neumann boundary condition (for CO2 mass fraction) on the opening boundary. Ie that the spatial gradient of CO2 mass fraction is zero on this opening boundary (since I do not know what the CO2 mass fraction at this boundary is but there should be no flux/rate of change of CO2 at the boundary). However the only option in CFX pre is to specify the exact mass fraction. Is there a way of having a Neumann boundary condition on this opening? Thanks in advance for your help.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Total pressure and mass flow boundary condition at inlet	bscphil	OpenFOAM Pre-Processing	3	July 9, 2017 14:39
Volume flow rate boundary condition in OpenFOAM	mayank.dce2k7	OpenFOAM Running, Solving & CFD	13	August 11, 2014 20:16
Question about heat transfer coefficient setting for CFX	Anna Tian	CFX	1	June 16, 2013 06:28
CFX13 Post Periodic interface	EtaEta	CFX	7	December 8, 2011 17:15
New topic on same subject - Flow around race car	Tudor Miron	CFX	15	April 2, 2004 06:18

June 8, 2015, 06:27		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	The Neumann boundary for incoming mass fraction does not seem to make physical sense. How can the gas entering the domain have a composition dependant on properties upstream? Isn't the composition dependant on what is downstream?

June 8, 2015, 19:18		#3
syble New Member Join Date: Jun 2015 Posts: 10 Rep Power: 10	Thanks for your reply. Yes I agree that when the opening acts as an inlet the flow coming into the domain will depend on the gas downstream, so that would require a specified mass fraction for the gas coming into the domain. However when the opening acts as an outlet, I don't want to be specifying the mass fraction at the outlet as I do not know what the mass fraction of the gas will be as it leaves the domain. So I thought that setting a Neumann boundary condition would account for this. Let me know if this sounds incorrect/not physical.

June 8, 2015, 19:43		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	CFX automatically switches between MF being determined by the upstream conditions on inflow to the domain, and the MF being determined by the MF of the fluid exiting the domain. So on outflow the MF is set by the convected fluid and no Neumann boundary need be set.

June 8, 2015, 19:50		#5
syble New Member Join Date: Jun 2015 Posts: 10 Rep Power: 10	Okay that makes sense, thank you very much for clarifying this for me.