how to define a material of gas

kiwishall · May 7, 2013, 11:28

Hello!　I'd like to build a simulation of expander with natural gas, so is it better to define the material as a mixture in CFX? or defining it as a pure material of which the properties data are obtained in the NIST refprop(LNG). What's more, can the density and the specific heat both be set as the function of T and P, is there any restrictions in CFX. While I set the density as the function of T and P, CFX always occurs error. But if I set it only as the function of T, it works even with a wrong result. Of course, for a gas, the density should be determined by T and P. So what can I do? Please help me. Thahks!

ghorrocks · May 7, 2013, 20:33

For compressible gases CFX has a number of approaches. The simplest is ideal gas and you should use this initially when you set the model up and get the basic stuff working. Then you can consider the different equations of state, real gas or mixture models which are available. But do all the setup on a simple ideal gas before going to a complex material properties model.

kiwishall · May 8, 2013, 01:17

Thank you for your advice! I have used the ideal gas model and gotten a result. But to get a better result, I must consider it as a real gas and then I met the above problems. So could you help me ?

ghorrocks · May 8, 2013, 01:45

There is advice in the CFX documentation on obtaining convergence. The key things are to use a small time step (at least initially), double precision if you need it and start with as good an initial condition as you can get. Good mesh quality is a big help too.

kiwishall · May 8, 2013, 02:03

I think you misunderstood what I mean. I have gotten a good result with the ideal gas supplied in the CFX. The mesh quality is good. But now I want to simulate with the data obtained by the NIST REFPROP. So how to transform the data into the CFX, and how to determine the density and the specific heat with these data?

May 7, 2013, 11:28	how to define a material of gas	#1
kiwishall New Member Lee Join Date: Oct 2012 Posts: 26 Rep Power: 14	Hello!　I'd like to build a simulation of expander with natural gas, so is it better to define the material as a mixture in CFX? or defining it as a pure material of which the properties data are obtained in the NIST refprop(LNG). What's more, can the density and the specific heat both be set as the function of T and P, is there any restrictions in CFX. While I set the density as the function of T and P, CFX always occurs error. But if I set it only as the function of T, it works even with a wrong result. Of course, for a gas, the density should be determined by T and P. So what can I do? Please help me. Thahks!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Continuing User Defined Real Gas Model issues	aeroman	FLUENT	6	April 8, 2016 04:34
Free surface boudary conditions with SOLA-VOF	Fan	Main CFD Forum	10	September 9, 2006 13:24
UDF FOR UNSTEADY TIME STEP	mayur	FLUENT	3	August 9, 2006 11:19
How to define the mixture of air and other gas?	dxmgws	FLUENT	5	March 21, 2003 17:03
Material define	Q Chen	FLUENT	5	February 20, 2002 22:02

May 7, 2013, 20:33		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	For compressible gases CFX has a number of approaches. The simplest is ideal gas and you should use this initially when you set the model up and get the basic stuff working. Then you can consider the different equations of state, real gas or mixture models which are available. But do all the setup on a simple ideal gas before going to a complex material properties model.

May 8, 2013, 01:17		#3
kiwishall New Member Lee Join Date: Oct 2012 Posts: 26 Rep Power: 14	Thank you for your advice! I have used the ideal gas model and gotten a result. But to get a better result, I must consider it as a real gas and then I met the above problems. So could you help me ?

May 8, 2013, 01:45		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	There is advice in the CFX documentation on obtaining convergence. The key things are to use a small time step (at least initially), double precision if you need it and start with as good an initial condition as you can get. Good mesh quality is a big help too.

May 8, 2013, 02:03		#5
kiwishall New Member Lee Join Date: Oct 2012 Posts: 26 Rep Power: 14	I think you misunderstood what I mean. I have gotten a good result with the ideal gas supplied in the CFX. The mesh quality is good. But now I want to simulate with the data obtained by the NIST REFPROP. So how to transform the data into the CFX, and how to determine the density and the specific heat with these data?