Coefficient loop iteration

vortex69 · May 20, 2016, 14:54

In CFX solver control , there are two options namely "Max. Coeff. Loops" and "Min. Coeff. Loops". What is meant by COEFFICIENT LOOPS here ?

Far · May 20, 2016, 15:24

outer loops for your simulation. no of iterations for that time step

vortex69 · May 20, 2016, 15:31

thank you, but what do u mean by "outer loops for your simulation"

vortex69 · May 20, 2016, 15:32

Quote:

Originally Posted by vortex69

In CFX solver control , there are two options namely "Max. Coeff. Loops" and "Min. Coeff. Loops". What is meant by COEFFICIENT LOOPS here ?

thank you, but what do u mean by "outer loops for your simulation"?

ghorrocks · May 21, 2016, 07:10

In a transient simulation you typically do 3-5 iterations per time step. These iterations are called coefficient loops as the equations are linearised for each iteration, so the non-linear part of the Navier Stokes equation is introduced by updating the linear coefficients.

vortex69 · May 21, 2016, 10:59

Quote:

Originally Posted by ghorrocks

In a transient simulation you typically do 3-5 iterations per time step. These iterations are called coefficient loops as the equations are linearised for each iteration, so the non-linear part of the Navier Stokes equation is introduced by updating the linear coefficients.

thank you very much, could you please shortly explain what do u mean by "non-linear part of the Navier Stokes equation is introduced by updating the linear coefficients" ? ( I really need to know that and I would appreciate it if u help me )

ghorrocks · May 22, 2016, 07:25

The convection term in the Navier Stokes equations (https://en.wikipedia.org/wiki/Navier...okes_equations) is non linear as it has the velocity multiplied by the gradient of velocity. You can't solve this with a linear equation solver. So you linearise it by assuming the non-linear component is constant thereby making the equations linear and solveable by linear solvers. But to account for the non-linearity you then re-calculate the constant you use to model the non-linear component and repeat the process.

May 20, 2016, 14:54	Coefficient loop iteration	#1
vortex69 New Member Ali Join Date: Apr 2016 Location: Germany Posts: 26 Rep Power: 10	In CFX solver control , there are two options namely "Max. Coeff. Loops" and "Min. Coeff. Loops". What is meant by COEFFICIENT LOOPS here ?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
GPU - GeForce GTX Titan Black?	frode86	Hardware	11	August 8, 2016 23:36
Pressure distribution on a wall	darazsbence	CFX	17	October 6, 2015 11:38
[snappyHexMesh] determining displacement for added points	CFDnewbie147	OpenFOAM Meshing & Mesh Conversion	1	October 22, 2013 10:53
Creating, loop or iteration using expression edito	Jan	CFX	3	July 28, 2003 12:01
Heat exchanger problem	chiseung	FLUENT	16	October 20, 2001 05:36

May 20, 2016, 15:24		#2
Far Senior Member Sijal Join Date: Mar 2009 Location: Islamabad Posts: 4,558 Blog Entries: 6 Rep Power: 54	outer loops for your simulation. no of iterations for that time step

May 20, 2016, 15:31		#3
vortex69 New Member Ali Join Date: Apr 2016 Location: Germany Posts: 26 Rep Power: 10	thank you, but what do u mean by "outer loops for your simulation"

May 21, 2016, 07:10		#5
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	In a transient simulation you typically do 3-5 iterations per time step. These iterations are called coefficient loops as the equations are linearised for each iteration, so the non-linear part of the Navier Stokes equation is introduced by updating the linear coefficients.

May 22, 2016, 07:25		#7
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,872 Rep Power: 144	The convection term in the Navier Stokes equations (https://en.wikipedia.org/wiki/Navier...okes_equations) is non linear as it has the velocity multiplied by the gradient of velocity. You can't solve this with a linear equation solver. So you linearise it by assuming the non-linear component is constant thereby making the equations linear and solveable by linear solvers. But to account for the non-linearity you then re-calculate the constant you use to model the non-linear component and repeat the process.