Transport Equation issue in CFX

esicia · February 18, 2014, 23:51

Dear All,

I am modeling the air flow pattern inside a room for a multi-component air (dry air, H2O and CO2). like a typical HVAC system, there are supply air, two occupants as heat, water vapor and CO2 and also the exhausts. The problem I am facing at the moment is about the CO2 concentration at the outlets. I set a initial values for the whole space at first and at the end of the converged simulation, the CO2 and H2O just spreads close to the occupants and nothings reaches the oulet. The CO2 and H2O mass fraction at outlet does not change over 2000 iteration and it is always the same as the initial condition. The residual is around 10^-4 for CO2 and H2O.

Do you have any advice on this? let me know if you need more inf.
Changing from steady to unstedy is too troublesome here, because of extensive computation needs.

Thanks for the help
Regards,
Esmail

ghorrocks · February 19, 2014, 06:37

How are you modelling the flow? How are you modelling the CO2 and H2O? An image of your geometry and your CCL would help.

esicia · February 27, 2014, 23:33

Dear Glenn, thanks for your response.

the indoor air has been modeled with a multicomponent material. dry air considered as the constraint while H2O and CO2 are modeled as transport equations.

I proceeded with the simulation and went further iterations (from 2000 to 4000) and now the H2O and CO2 spreaded more in the space. It seems i need to go further than this to get reasonable results. Now i decided to change to high resolution turbulence numerics, in the hope to get steady state and logical value at exhaust air CO2 level (a user point) with less than 4000 iteration.

I have no idea, if there is any setting in solve control (like advection scheme or tubulence numerics, ...) of CFX which can speed up the transport equation solution or engagement per iteration of the main flow equations. it would be great if i can have your thoughts on this.

Regards,
Esmail

ghorrocks · February 27, 2014, 23:37

A larger physical time step will converge faster - as long as it remains numerically stable.

February 18, 2014, 23:51	Transport Equation issue in CFX	#1
esicia New Member Esmail Join Date: May 2009 Posts: 11 Rep Power: 17	Dear All, I am modeling the air flow pattern inside a room for a multi-component air (dry air, H2O and CO2). like a typical HVAC system, there are supply air, two occupants as heat, water vapor and CO2 and also the exhausts. The problem I am facing at the moment is about the CO2 concentration at the outlets. I set a initial values for the whole space at first and at the end of the converged simulation, the CO2 and H2O just spreads close to the occupants and nothings reaches the oulet. The CO2 and H2O mass fraction at outlet does not change over 2000 iteration and it is always the same as the initial condition. The residual is around 10^-4 for CO2 and H2O. Do you have any advice on this? let me know if you need more inf. Changing from steady to unstedy is too troublesome here, because of extensive computation needs. Thanks for the help Regards, Esmail

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
ANSYS CFX 14 on UBUNTU 12.04 64bit: PARALLEL ISSUE	david.pasquale	CFX	7	July 15, 2024 14:43
UDF Transport equation for transition momentum thickness Reynolds number	Wantami	FLUENT	0	April 9, 2011 03:50
Estimate convection and diffusion terms in transport equation	Lance	CFX	0	April 4, 2011 12:16
CFX Additional Variable Transport Equation	Scott Nordsen	CFX	3	January 30, 2010 06:36
Additional Variable - Diffusive Transport Equation	RadheShyam Das	CFX	3	October 18, 2007 13:35

February 19, 2014, 06:37		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	How are you modelling the flow? How are you modelling the CO2 and H2O? An image of your geometry and your CCL would help.

February 27, 2014, 23:33	hi	#3
esicia New Member Esmail Join Date: May 2009 Posts: 11 Rep Power: 17	Dear Glenn, thanks for your response. the indoor air has been modeled with a multicomponent material. dry air considered as the constraint while H2O and CO2 are modeled as transport equations. I proceeded with the simulation and went further iterations (from 2000 to 4000) and now the H2O and CO2 spreaded more in the space. It seems i need to go further than this to get reasonable results. Now i decided to change to high resolution turbulence numerics, in the hope to get steady state and logical value at exhaust air CO2 level (a user point) with less than 4000 iteration. I have no idea, if there is any setting in solve control (like advection scheme or tubulence numerics, ...) of CFX which can speed up the transport equation solution or engagement per iteration of the main flow equations. it would be great if i can have your thoughts on this. Regards, Esmail

February 27, 2014, 23:37		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	A larger physical time step will converge faster - as long as it remains numerically stable.