
[Sponsors] 
May 3, 2010, 23:39 
Can Flow3D solve the JouleThomson effect?

#1 
New Member
wuyu
Join Date: Mar 2010
Posts: 24
Rep Power: 9 
I am doing some projects about VOF. So I want to buy a Flow3D license. But my supervisor askd me if Flow3D can solve JouleThomson effect. I think he is thinking to use it for another project under him, which is about gas leaking. I told him Flow3D is good at VOF, but not so powerful to deal with other processes. My quesiton is if Flow3D can solve JouleThomson effect? If can, is there any example I can use to show my professor?
from wikipedia about JT effect. Joule Thomson effect describes the temperature change of a gas or liquid when it is forced through a valve or porous plug while kept insulated so that no heat is exchanged with the environment 

May 4, 2010, 10:35 
FLOW3D & Joule Thomson Effect

#2 
Senior Member
Jeff Burnham
Join Date: Apr 2010
Posts: 204
Rep Power: 9 
The answer is yes: FLOW3D can model compressible gases (simultaneously with incompressible fluids, if necessary), and has full heat transfer calculations available. I've modeled steam through an orifice to show the Joule Thomson (throttling) effect: as you can see the jet temperature is much lower than the upstream temperature. The steam jet is at an angle because it's 'swinging' back and forth (oscillating). Two images are attached. Let me know if you have any other questions.
[IMG]file:///C:/Users/Jeff/AppData/Local/Temp/mozscreenshot.png[/IMG] 

May 4, 2010, 12:13 

#3  
New Member
wuyu
Join Date: Mar 2010
Posts: 24
Rep Power: 9 
Thanks a lot, Jburnham. And your example is quite helpful.
Quote:


May 4, 2010, 22:34 

#4 
New Member
wuyu
Join Date: Mar 2010
Posts: 24
Rep Power: 9 
JBurnham, sorry to bother you. My supervisor, asked me how to define joulethomason coefficient in flow 3D. Oh my god, I even do not have flow 3D. So i came here again to ask, is there a direct way to define jt coefficient, or there is a indirect way to define it.


May 5, 2010, 11:00 
Joule Thomson Coefficient

#5 
Senior Member
Jeff Burnham
Join Date: Apr 2010
Posts: 204
Rep Power: 9 
Seasoul  the Joule Thomson coefficient is usually defined as
Kjt = V/Cp (aT1), where V is the gas volume, Cp is the constantpressure heat capacity, a is the coefficient of thermal expansion, and T is the absolute temperature. I assume you're looking at gas or vapor for this problem, not liquid. In FLOW3D, activating the fullycompressible 2fluid model 'turns on' the heat transfer and density evaluation physics packages. Then, for Fluid #2 (the fully compressible fluid), you can define the constantvolume heat capacity Cv. Temperature T and volume V are dynamically computed. The coefficient of thermal expansion is an available input in FLOW3D, but it controls density evaluation as a function of temperature only. For compressible fluids, only the specific gas constant R needs to be defined (specific to the fluid  not the universal gas constant). Then the constantpressure heat capacity will be dynamically computed as a function of Cv and R. For example, in the steam problem I showed, Cv (cv2) is about 9128 ft^2/s^2/R, R (rf2) is 2760 ft^2/s^2/R, and compressibility coefficient a is not specified because the steam is compressible and the density is determined from the equation of state (rho = p/RT). FLOW3D can use other unit systems as well, this problem just happened to be in ftslugRankine units. A heat transfer coefficient for internal heat transfer is also specified for the fluid, but that's not explicitly related to the JouleThomson effect. So, in short, you don't actually need the JT coefficient, but you can calculate it from R if necessary, and you don't need Cp, but again, you can calculate from Cv (which you do need): cp  cv = R. The specific gas constant is an intrinsic property of the fluid you're working with. FLOW3D assumes that the ideal gas law applies for evaluating the relation between T, p, and rho. If the problem involves a nonideal gas (Cv varies significantly with pressure and/or temperature), then you might be looking at a codecustomization. However, I'll point out that the ideal gas assumption is OK for most problems and most gases and gas mixtures, and as shown in the example I sent earlier, the pressure drop does correlate with a temperature change (small departures from the ideal gas rules that occur due to molecular internal work won't be modeled). Hope that helps. 

May 11, 2010, 04:48 

#6 
New Member
wuyu
Join Date: Mar 2010
Posts: 24
Rep Power: 9 
Thanks.
After I get the software, I will try this case as soon as possible. 

May 11, 2010, 13:45 

#7 
Senior Member
Jeff Burnham
Join Date: Apr 2010
Posts: 204
Rep Power: 9 
Sounds good. Once you get it installed, let me know if you'd like the test case that generated the images I sent earlier: I can provide the case with some commentary on the setup to help you get started.


May 12, 2010, 00:50 

#8 
New Member
wuyu
Join Date: Mar 2010
Posts: 24
Rep Power: 9 
JBurnham,
Can you send me the case file (or the input text file) now? My supervisor has started the purchasing, and I want to start learning from now on. I am learning flow 3d by reading the early tech notes in the flow3d official website. But the cases in the quite early days seem to be obsolete, I think now flow3d is different and more powerful than at that time. Thanks in advance. 

December 21, 2011, 01:01 

#9 
New Member
leechun
Join Date: May 2011
Posts: 1
Rep Power: 0 
how do i simulate the joule thomson effect? i input Cv and R for the fluid properties (fluid 2) but couldnt get the same results as JBurnham


Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Abt: Added mass effect in compressible flow.  jinwon park  Main CFD Forum  0  August 2, 2007 14:43 
inviscid  Sylvain  FLUENT  6  October 30, 2005 14:58 
Inviscid Drag at subsonic, subcritical Mach #  Axel Rohde  Main CFD Forum  1  November 19, 2001 13:19 
any better way to solve 3D stokes flow?  Yangang Bao  Main CFD Forum  0  September 29, 1999 10:26 
Question on 3D potential flow  Adrin Gharakhani  Main CFD Forum  13  June 21, 1999 05:18 