Question about grid scaling...

Iaroslav · May 14, 2009, 17:00

Hi everyone,

I have a question about grid scaling in Fluent 6.3.16. At the moment I am working on modeling a dimensionless domain (axi-symmetric) for free jet flow (supersonic, compressible ideal gas). My intention is to have all my dimensions in terms of diameter lengths (ie, my domain in 41D long, 27D high with the inlet being 0.5D high which is located on the bottom left boundary, with the rest of the boundary being a wall).

Initially I believed that is I though of D being a meter in fluent, I would get a correct solution. However, when I found a new solution doing the each same thing, except scaling my grid by a factor of 0.01 (cm), I got a different result (jet core length is reduced by a factor of 3). Is there something I am missing?
(Note: I have used pressure inlet for inlet, pressure far field for the top boundary, pressure outlet for the right boundary, axis for the bottom boundary, and wall for the left boundary that is above the inlet).

Any help would be appreciated.

Thanks,
Iaroslav

Daniel Tanner · May 14, 2009, 17:24

You can use whatever dimensions you want, but they must be consistent!!
The length scale units must be consistent with the pressure scale units.

For instance, in dimensional space if length is in meters then velocity must be meters per unit time! Similarly if length is x diameters long then velocity must be diameters per unit time.

Thus one variable cannot be changed independently of the others! Remember pressure is proportional to 1/Area => 1/length^2. This must be changed to compensate for a change in length units.

Iaroslav · May 14, 2009, 17:45

Thanks Daniel,

What you said makes a lot of sense. However, is there an option in Fluent to make my boundary values relative to some other length (i.e in terms of diameter lengths) or would I have to calculate these manually?

Thank you,
Iaroslav

Daniel Tanner · May 15, 2009, 05:03

Irrespective of any automatic options to do this in fluent (I am not aware of any) I would always advise that you work it out. It gives you some insight into the problem and avoids any ambiguity in setting the boundary conditions.
Good Luck!

May 14, 2009, 17:00	Question about grid scaling...	#1
Iaroslav New Member Iaroslav Join Date: May 2009 Posts: 16 Rep Power: 17	Hi everyone, I have a question about grid scaling in Fluent 6.3.16. At the moment I am working on modeling a dimensionless domain (axi-symmetric) for free jet flow (supersonic, compressible ideal gas). My intention is to have all my dimensions in terms of diameter lengths (ie, my domain in 41D long, 27D high with the inlet being 0.5D high which is located on the bottom left boundary, with the rest of the boundary being a wall). Initially I believed that is I though of D being a meter in fluent, I would get a correct solution. However, when I found a new solution doing the each same thing, except scaling my grid by a factor of 0.01 (cm), I got a different result (jet core length is reduced by a factor of 3). Is there something I am missing? (Note: I have used pressure inlet for inlet, pressure far field for the top boundary, pressure outlet for the right boundary, axis for the bottom boundary, and wall for the left boundary that is above the inlet). Any help would be appreciated. Thanks, Iaroslav

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Global scaling - grid independency testing	Jenny	CFX	6	December 10, 2013 04:52
MapFields to New Grid For Extreme Grid Deformations due to Body Motion	albcem	OpenFOAM	0	May 5, 2009 15:17
Obtain Geometry Data from a Grid File in CFX 10.0	ARJUN	CFX	2	August 17, 2006 08:20
Moving Body Grid Question	Matt	Main CFD Forum	5	September 13, 2004 01:45
Numerical methods for discontinuous grid interfaces?	Hansong Hang	Main CFD Forum	12	September 16, 1998 23:26

May 14, 2009, 17:24		#2
Daniel Tanner Member Daniel Tanner Join Date: Apr 2009 Posts: 54 Rep Power: 18	You can use whatever dimensions you want, but they must be consistent!! The length scale units must be consistent with the pressure scale units. For instance, in dimensional space if length is in meters then velocity must be meters per unit time! Similarly if length is x diameters long then velocity must be diameters per unit time. Thus one variable cannot be changed independently of the others! Remember pressure is proportional to 1/Area => 1/length^2. This must be changed to compensate for a change in length units.

May 14, 2009, 17:45		#3
Iaroslav New Member Iaroslav Join Date: May 2009 Posts: 16 Rep Power: 17	Thanks Daniel, What you said makes a lot of sense. However, is there an option in Fluent to make my boundary values relative to some other length (i.e in terms of diameter lengths) or would I have to calculate these manually? Thank you, Iaroslav

May 15, 2009, 05:03		#4
Daniel Tanner Member Daniel Tanner Join Date: Apr 2009 Posts: 54 Rep Power: 18	Irrespective of any automatic options to do this in fluent (I am not aware of any) I would always advise that you work it out. It gives you some insight into the problem and avoids any ambiguity in setting the boundary conditions. Good Luck!