Numerical schemes and artificial dissipation

MarcelloAsciolla · May 25, 2018, 15:24

Is there a way to see what is the exact scheme for these options when I choose the numerical scheme in "equation class settings"?
-central difference
-high resolution
-upwind
-specified blend factor

Maybe it is in the documentation, but I did not found them.
For istance I have an idea for some of them (I know the classical formulation for the "upwind" and the "central difference" ones, but I do not know if there are modified version), but what is the expression used for the "high resolution" or the "blend factor"?

Is there a way to use a homemade scheme in CFX?

Which scheme can assure the minimum numerical dissipation (I mean the artificial one)?

Opaque · May 25, 2018, 17:15

Documentation section:

CFX Solver Theory Guide/Discretization and Solution Theory/Numerical Discretization/Discretization of the Governing Equations/Advection Term

There is no entry for a User Defined treatment of the advection term.

Is there a particular advection scheme you would like to try?

It would be great to know what such scheme advantage is?

MarcelloAsciolla · May 25, 2018, 22:54

Thank you, I will try to give a look there.

There is not a particular numerical scheme that I wanted to use, I wanted to know if there is this opportunity.

I need a CFD simulation with the requirement of little dissipation because it is for aeroacoustic purpose.

ghorrocks · May 26, 2018, 04:51

Then have a look at the central difference scheme. CFX also has a variant of central difference scheme for LES modelling - read the documentation for details.

MarcelloAsciolla · May 26, 2018, 06:05

Yes, I was thinking about a central difference one. I do not know if there is anything better than it, for this reason I also asked for a suggestion about different schemes... And I am glad to hear that there is a different version for LES, so I must think that they already thought about that.

ghorrocks · May 26, 2018, 07:00

Low dissipation is essential for LES, so you have a few low or no dissipation options for LES models.

May 25, 2018, 15:24	Numerical schemes and artificial dissipation	#1
MarcelloAsciolla Member Marcello Asciolla Join Date: Apr 2018 Location: Italy Posts: 43 Rep Power: 8	Is there a way to see what is the exact scheme for these options when I choose the numerical scheme in "equation class settings"? -central difference -high resolution -upwind -specified blend factor Maybe it is in the documentation, but I did not found them. For istance I have an idea for some of them (I know the classical formulation for the "upwind" and the "central difference" ones, but I do not know if there are modified version), but what is the expression used for the "high resolution" or the "blend factor"? Is there a way to use a homemade scheme in CFX? Which scheme can assure the minimum numerical dissipation (I mean the artificial one)?

May 26, 2018, 04:51		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Then have a look at the central difference scheme. CFX also has a variant of central difference scheme for LES modelling - read the documentation for details. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

May 26, 2018, 07:00		#6
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Low dissipation is essential for LES, so you have a few low or no dissipation options for LES models. __________________ Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Melt front artificial dissipation with VoF	steph79	FLUENT	0	February 22, 2016 08:13
Artificial dissipation for NS equations, flow in a pipe	RameshK	Main CFD Forum	8	November 13, 2011 11:29
numerical scheme without artificial diffusion artifact	mpeti	OpenFOAM Running, Solving & CFD	2	October 26, 2011 12:36
Two questions on Numerical schemes for NS equations	RameshK	Main CFD Forum	0	December 21, 2010 08:46
Standard for checking and testing numerical schemes?	X. Ye	Main CFD Forum	7	August 31, 1999 18:05

May 25, 2018, 17:15		#2
Opaque Senior Member Join Date: Jun 2009 Posts: 1,880 Rep Power: 33	Documentation section: CFX Solver Theory Guide/Discretization and Solution Theory/Numerical Discretization/Discretization of the Governing Equations/Advection Term There is no entry for a User Defined treatment of the advection term. Is there a particular advection scheme you would like to try? It would be great to know what such scheme advantage is?

May 25, 2018, 22:54		#3
MarcelloAsciolla Member Marcello Asciolla Join Date: Apr 2018 Location: Italy Posts: 43 Rep Power: 8	Thank you, I will try to give a look there. There is not a particular numerical scheme that I wanted to use, I wanted to know if there is this opportunity. I need a CFD simulation with the requirement of little dissipation because it is for aeroacoustic purpose.

May 26, 2018, 06:05		#5
MarcelloAsciolla Member Marcello Asciolla Join Date: Apr 2018 Location: Italy Posts: 43 Rep Power: 8	Yes, I was thinking about a central difference one. I do not know if there is anything better than it, for this reason I also asked for a suggestion about different schemes... And I am glad to hear that there is a different version for LES, so I must think that they already thought about that.