High Resolution (CFX) vs 2nd Order Upwind (Fluent)

gravis · May 10, 2010, 08:11

Hi all,

I am wondering about the similarities between the advection schemes "High Resolution" and "Second Order Upwind" in CFX and Fluent, respectively. As I understand it, the principles behind the schemes are the same;

CFX: $\phi_{ip}=\phi_{up}+\beta\nabla\phi\bullet\bar{r}$
Fluent: $\phi_{f,SOU}=\phi+\nabla\phi\bullet\bar{r}$

In CFX, $\beta$ is limited to prevent under/overshoots. In Fluent, the gradient is limited to prevent under/overshoots.

In the CFX documentation, it is decribed how $\beta$ is calculated at each integration point/control volume surface while in Fluent it is not. Does this mean that in Fluent, basically a value corresponding to $\beta$ in CFX is calculated to limit the gradient, and that the Fluent scheme is also bounded? If it is so, is this value calculated similarly to how $\beta$ is calculated in CFX?

Thanks in advance.

Best regards,
Reine

gravis · May 10, 2010, 11:13

I realized my mistake: the details of the limiter in Fluent are actually described in the documentation. It seems that the limiter used in the High Resolution scheme corresponds to the "Standard Limiter" for Second Order Upwind in Fluent; both are based on principles by Barth and Jespersen.

gravis · May 11, 2010, 03:35

In CFX-Post, there are variables with names "-beta" that indicate the value of the limiter $\beta$ in each control volume.

Does anyone know if there are corresponding variables available in Fluent, and what their names are? I have not found any variables that I think match. In the Fluent documentation, the scalar that limits the gradient is denoted $\psi$ .

lzw2003 · March 24, 2011, 03:43

i just read the cfx theory , and find its high resolution schemes. as a comment question, i want to know how the wpwind node is found? just mathmatic method or other method?

May 10, 2010, 08:11	High Resolution (CFX) vs 2nd Order Upwind (Fluent)	#1
gravis Member Reine Granström Join Date: Apr 2009 Location: Sweden Posts: 43 Rep Power: 17	Hi all, I am wondering about the similarities between the advection schemes "High Resolution" and "Second Order Upwind" in CFX and Fluent, respectively. As I understand it, the principles behind the schemes are the same; CFX: $\phi_{ip}=\phi_{up}+\beta\nabla\phi\bullet\bar{r}$ Fluent: $\phi_{f,SOU}=\phi+\nabla\phi\bullet\bar{r}$ In CFX, $\beta$ is limited to prevent under/overshoots. In Fluent, the gradient is limited to prevent under/overshoots. In the CFX documentation, it is decribed how $\beta$ is calculated at each integration point/control volume surface while in Fluent it is not. Does this mean that in Fluent, basically a value corresponding to $\beta$ in CFX is calculated to limit the gradient, and that the Fluent scheme is also bounded? If it is so, is this value calculated similarly to how $\beta$ is calculated in CFX? Thanks in advance. Best regards, Reine

May 10, 2010, 11:13	Wrong...	#2
gravis Member Reine Granström Join Date: Apr 2009 Location: Sweden Posts: 43 Rep Power: 17	I realized my mistake: the details of the limiter in Fluent are actually described in the documentation. It seems that the limiter used in the High Resolution scheme corresponds to the "Standard Limiter" for Second Order Upwind in Fluent; both are based on principles by Barth and Jespersen.

May 11, 2010, 03:35	Additional question	#3
gravis Member Reine Granström Join Date: Apr 2009 Location: Sweden Posts: 43 Rep Power: 17	In CFX-Post, there are variables with names "-beta" that indicate the value of the limiter $\beta$ in each control volume. Does anyone know if there are corresponding variables available in Fluent, and what their names are? I have not found any variables that I think match. In the Fluent documentation, the scalar that limits the gradient is denoted $\psi$ .

March 24, 2011, 03:43	about upwind node	#4
lzw2003 New Member zhiwei liu Join Date: Mar 2009 Posts: 22 Rep Power: 17	i just read the cfx theory , and find its high resolution schemes. as a comment question, i want to know how the wpwind node is found? just mathmatic method or other method?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Gradient Discretization CFX vs Fluent	Scott Nordsen	FLUENT	1	December 3, 2009 19:50
Gradient Discretization CFX vs Fluent	Scott Nordsen	CFX	1	December 2, 2009 17:46
OpenFOAM - 2nd order of discretization?	makaveli_lcf	OpenFOAM Running, Solving & CFD	0	November 12, 2009 08:15
turbFoam vs CFX - 2D case high Re	ffucile	OpenFOAM Running, Solving & CFD	17	October 21, 2009 05:01
Reynolds Stress model in CFX vs Fluent	Tim	CFX	1	October 7, 2009 07:19