Compressible flow solver in Fluent

jwillie2000 · May 21, 2012, 12:19

Hi All,

I have a case where the flow is compressible due a small gap in the geometry where the velocity shuts up, with even the possiblity of a shock wave. It is air and i am wondering whether using the segregated solver with coupled pressure-velocity can do in Fluent? And what about the boundary condition at the inlet? Currently, i am using a velocity inlet but i am not getting the mass balance. The mass at the inlet is exaggerated. Plus using the density solver does not seem to be getting the velocity profile i expect. I know CFX would do a better job but we do not have a cfx license. Just wanted to know your thoughts? U think changing to a pressure inlet boundary would do?

Thanks! Jimmy

Far · May 22, 2012, 01:54

Quote:

segregated solver with coupled pressure-velocity can do in Fluent

You can say this is the semi coupled solver as energy equation is solved in segregated manner.

Pressure based coupled solver is good at both incompressible and compressilble flows. So you should go for it.

jwillie2000 · May 22, 2012, 09:11

Hi Far, and thanks for your input. I am using the segregated solver with coupled velocity-pressure coupling. But the mass balance is not just working. The mass flow at the inlet has increased and that at the outlet is about the value i expect. For spatial discretization i am using the QUICK scheme. Any idea why this may be happpening? Theoretically, i expect the velocity after the duct to shoot up to about 670 m/s, which is quite high. I have a velocity inlet and a pressure outlet. Will be nice if you can comment. Thanks!

Jimmy

clgs.1903 · May 22, 2012, 11:35

Hi Jimmy,

The documentation states that velocity inlets should not be used for compressible flow, and to use mass flow inlets instead.

http://hpce.iitm.ac.in/website/Manua...ug/node222.htm

I was having a similar issue until I found this bit of information. Hope this helped!

jwillie2000 · May 25, 2012, 10:58

Hi Far,
Thanks for your input. I have it now with mass flow inlet and it is converging and the mass balance is kind of okay but only after running for long time. The question is with the energy equation. The residuals for temp keep fluctuating. They drop and then increase and yesterday for example i had a sudden jump and the solver crashed. I lower the under-relaxation factors and it is runnig again but has just starting going up again after being stable for a long time. The under-relaxation factor for the temp is now at 0.8. Have you had a similar problem?
Thanks! Jimmy

May 21, 2012, 12:19	Compressible flow solver in Fluent	#1
jwillie2000 Member James Willie Join Date: Mar 2009 Posts: 81 Rep Power: 17	Hi All, I have a case where the flow is compressible due a small gap in the geometry where the velocity shuts up, with even the possiblity of a shock wave. It is air and i am wondering whether using the segregated solver with coupled pressure-velocity can do in Fluent? And what about the boundary condition at the inlet? Currently, i am using a velocity inlet but i am not getting the mass balance. The mass at the inlet is exaggerated. Plus using the density solver does not seem to be getting the velocity profile i expect. I know CFX would do a better job but we do not have a cfx license. Just wanted to know your thoughts? U think changing to a pressure inlet boundary would do? Thanks! Jimmy

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
SteadyState solver for compressible flow	matteo_gautero	OpenFOAM Running, Solving & CFD	32	November 21, 2012 05:27
solver for subsonic compressible turbulent flow in OF 1.7	nileshjrane	OpenFOAM Running, Solving & CFD	20	February 13, 2012 06:54
compressible flow solver in OF1.6	mecbe2002	OpenFOAM	9	December 25, 2010 10:17
Natural Convection using Compressible Flow (chtMultiRegionFOAM)	msarkar	OpenFOAM	2	September 7, 2010 01:13
Can segregated solver be used for compressible flow?	Steven-GY	ANSYS	0	May 14, 2009 11:37

May 22, 2012, 09:11		#3
jwillie2000 Member James Willie Join Date: Mar 2009 Posts: 81 Rep Power: 17	Hi Far, and thanks for your input. I am using the segregated solver with coupled velocity-pressure coupling. But the mass balance is not just working. The mass flow at the inlet has increased and that at the outlet is about the value i expect. For spatial discretization i am using the QUICK scheme. Any idea why this may be happpening? Theoretically, i expect the velocity after the duct to shoot up to about 670 m/s, which is quite high. I have a velocity inlet and a pressure outlet. Will be nice if you can comment. Thanks! Jimmy

May 22, 2012, 11:35		#4
clgs.1903 New Member Join Date: May 2012 Posts: 15 Rep Power: 14	Hi Jimmy, The documentation states that velocity inlets should not be used for compressible flow, and to use mass flow inlets instead. http://hpce.iitm.ac.in/website/Manua...ug/node222.htm I was having a similar issue until I found this bit of information. Hope this helped!

May 25, 2012, 10:58		#5
jwillie2000 Member James Willie Join Date: Mar 2009 Posts: 81 Rep Power: 17	Hi Far, Thanks for your input. I have it now with mass flow inlet and it is converging and the mass balance is kind of okay but only after running for long time. The question is with the energy equation. The residuals for temp keep fluctuating. They drop and then increase and yesterday for example i had a sudden jump and the solver crashed. I lower the under-relaxation factors and it is runnig again but has just starting going up again after being stable for a long time. The under-relaxation factor for the temp is now at 0.8. Have you had a similar problem? Thanks! Jimmy