# Viscosity UDFs

 User Name Remember Me Password
 Register Blogs Members List Search Today's Posts Mark Forums Read

 LinkBack Thread Tools Display Modes
 April 21, 2009, 12:22 Viscosity UDFs #1 New Member   Richard Join Date: Apr 2009 Posts: 3 Rep Power: 9 Dear all, I am trying to get some udfs for variable viscosity working in Fluent 6.3, although I'm having a great deal of problems in doing so... I want to interpret the following code: #include "udf.h" //Works Casson DEFINE_PROPERTY(cell_viscosity,c,t) { double mu_lam double tiny = 0.0000000001; double tauy = 0.01083; double muinf = 0.0031; double m = 200; double strain; double first; strain = pow(C_DUDX(c, t),2)+ pow(C_DVDY(c, t),2)+ pow(C_DWDZ(c, t),2)+ 2*pow(C_DUDY(c, t)+ C_DVDX(c, t),2)+ 2*pow(C_DUDZ(c, t)+ C_DWDX(c, t),2)+ 2*pow(C_DVDZ(c, t)+ C_DWDY(c, t),2); strain = pow(strain, 0.5); first = 1-exp(-pow(m*strain,0.5)); mu_lam = pow(muinf,0.5) + pow(tauy/(strain + tiny),0.5)*first; mu_lam = pow(mu_lam, 2); if(cell==100){printf("viscosity = %f",mu_lam);} return mu_lam; } However, when I read my case file, and try to interpret this, I get the following error: Error: c:\fluent.inc\fluent6.3.26/src/dpm.h: line 1192: parse error. I would greatly appreciate any help I could get on this. Many thanks in advance. Richard

 April 22, 2009, 12:13 #2 New Member   Richard Join Date: Apr 2009 Posts: 3 Rep Power: 9 Can anyone please help me with this? My final year project depends on this, and I don't know much about programming... Thank you.

 April 23, 2009, 06:36 #3 New Member   Richard Join Date: Apr 2009 Posts: 3 Rep Power: 9 Actually forget it, I figured it out in the end: I had my header file in the same folder as my code, so it didn't work!

 May 11, 2009, 02:30 #4 New Member   Jane Join Date: Mar 2009 Posts: 18 Rep Power: 9 Hi, Supernova strain = pow(C_DUDX(c, t),2)+ pow(C_DVDY(c, t),2)+ pow(C_DWDZ(c, t),2)+ 2*pow(C_DUDY(c, t)+ C_DVDX(c, t),2)+ 2*pow(C_DUDZ(c, t)+ C_DWDX(c, t),2)+ 2*pow(C_DVDZ(c, t)+ C_DWDY(c, t),2); may i know how do you difine the C_DUDX (c,t) and the rest in your UDF??

 May 11, 2009, 04:37 #5 Senior Member   Max Join Date: Mar 2009 Posts: 133 Rep Power: 9 hello Jane, C_DUDX(c,t) is a predefined macro which returns the derivative of x-velocity with respect to x-coordinate. You do not have to define it by yourself. However there is also a macro C_STRAIN_RATE_MAG(c,t) to access the strain rate for each cell itself, thus there is no reason to compute this value by hand. cheers

 May 11, 2009, 05:14 #6 New Member   Jane Join Date: Mar 2009 Posts: 18 Rep Power: 9 Hi, coglione Thank you for your reply. for example if i wan to code the equation as show in figure. below is the equation that i code for UDF. shear_rate = sqrt (SQR(C_DUDX(c,t))+ SQR(C_DVDY(c,t) + SQR(C_DWDZ(c,t)); may i know the equation that i code into UDF is correct or not?

May 11, 2009, 20:46
#7
New Member

Jane
Join Date: Mar 2009
Posts: 18
Rep Power: 9
Quote:
 Originally Posted by Jane Hi, coglione Thank you for your reply. for example if i wan to code the equation as show in figure. below is the equation that i code for UDF. shear_rate = sqrt (SQR(C_DUDX(c,t))+ SQR(C_DVDY(c,t) + SQR(C_DWDZ(c,t)); may i know the equation that i code into UDF is correct or not?

Help me please...
i'm just a UDF beginner...please tell me

 May 12, 2009, 02:57 #8 Senior Member   Max Join Date: Mar 2009 Posts: 133 Rep Power: 9 Hello Jane, the correct equation for strain rate is given by supernova in the first message of this thread. Use his coding or simply the macro C_STRAIN_RATE_MAG(c,t). It will return excactly the same and is much more efficient in terms of cpu-time. cheers

May 12, 2009, 03:46
#9
New Member

Jane
Join Date: Mar 2009
Posts: 18
Rep Power: 9
Quote:
 Originally Posted by coglione Hello Jane, the correct equation for strain rate is given by supernova in the first message of this thread. Use his coding or simply the macro C_STRAIN_RATE_MAG(c,t). It will return excactly the same and is much more efficient in terms of cpu-time. cheers
Hi coglione,

thank you for your reply

i've tried macro C_STRAIN_RATE_MAG(c,t) in the UDF, the problem occurs when i simulate using this predefined strain rate. the continuity graph increase until fluent show errors message.

i suspect my shear rate equation caused the increase of continuity, but i still cannot find the solution.

can you give me some idea?

 May 12, 2009, 07:44 #10 Senior Member   Max Join Date: Mar 2009 Posts: 133 Rep Power: 9 Hello Jane, non-newtonian fluids are always prone for numerical problems due to the highly non-linear nature of the momentum equation involved. I usually start the simulation with a moderate shear dependency of the viscosity (or even constant one) and switch to the actual rheological model when the approximate solution has converged. This provides a realistic and quite smooth strain field and may help convergence. If instability is still observed lower your relaxation, use first order discretization and if nothing helps at all switch to transient simulation using a small time-step. Hope this helps

May 12, 2009, 08:00
#11
New Member

Jane
Join Date: Mar 2009
Posts: 18
Rep Power: 9
Quote:
 Originally Posted by coglione Hello Jane, non-newtonian fluids are always prone for numerical problems due to the highly non-linear nature of the momentum equation involved. I usually start the simulation with a moderate shear dependency of the viscosity (or even constant one) and switch to the actual rheological model when the approximate solution has converged. This provides a realistic and quite smooth strain field and may help convergence. If instability is still observed lower your relaxation, use first order discretization and if nothing helps at all switch to transient simulation using a small time-step. Hope this helps
Thank you for your advice, finally i get the stable continuity.
Thank you very much

I found my UDF unable to print the numbers to console.

Below is my "message" in UDF.

/*Message("UDF: time=%f,shear_rate=%f");*/

 May 13, 2009, 11:06 #12 Member   Daniel Tanner Join Date: Apr 2009 Posts: 54 Rep Power: 9 Leave out the outer quotation marks. "Message("UDF: time=%f,shear_rate=%f");" should be of the form Message("UDF: time=%f,shear_rate=%f", X, Y); where X and Y are the time and shear_rate variables, i.e., you have not told the MESSAGE macro where to find the time and shear rate variables.

May 13, 2009, 20:49
#13
New Member

Jane
Join Date: Mar 2009
Posts: 18
Rep Power: 9
Quote:
 Originally Posted by Daniel Tanner Leave out the outer quotation marks. "Message("UDF: time=%f,shear_rate=%f");" should be of the form Message("UDF: time=%f,shear_rate=%f", X, Y); where X and Y are the time and shear_rate variables, i.e., you have not told the MESSAGE macro where to find the time and shear rate variables.
Thank you very much

March 4, 2011, 11:53
Help
#14
New Member

Join Date: Mar 2011
Posts: 1
Rep Power: 0
Quote:
 Originally Posted by coglione hello Jane, C_DUDX(c,t) is a predefined macro which returns the derivative of x-velocity with respect to x-coordinate. You do not have to define it by yourself. However there is also a macro C_STRAIN_RATE_MAG(c,t) to access the strain rate for each cell itself, thus there is no reason to compute this value by hand. cheers
Hi Coglione,

Your input here has been very useful in my project.
I would like to ask a few questions:

C_STRAIN_RATE_MAG(c,t) deos it give a dimensionless value? or just the strain rate?

This is me code for my model:
#include"udf.h"
DEFINE_PROPERTY(c_effective_viscosity, cell, thread)
{
double e_viscos;
double m = 200;
double a_viscos = 0.0031;
double y_stress = 0.01082;
double strain;
double a;
double b;
double c;
strain = C_STRAIN_RATE_MAG(cell,thread);
a = y_stress/strain;
b = m*strain;
c = 1-exp(-pow(b,0.5));
e_viscos = pow(a_viscos,0.5)+(pow(a,0.5)*c);
e_viscos = pow(e_viscos,2);
return e_viscos;
}

However, when I try to run it, Fluent does not start iterating and shows the following error message.
Error: Floating point error: invalid number
Error Object: ()

Can you please help?

Regards,
Maple

 April 10, 2011, 11:03 #15 Member   xianghong wu Join Date: Mar 2009 Posts: 57 Rep Power: 9 Hello, I am writing a nonNewtonian viscosity model of my own, in the myViscosityModel.c, I wrote shearrate=max(minshearrate, sqrt(2.0*(symm(fvc::grad(u)))&&(symm(fvc::grad(u)) )/3.0); when I compile it with "wmake", error is given as : arguments in max()function has different dimensions [0 0 -1 0 0 0 0] and [0 0 0 0 0 0 0], then I tried to delete one argument, as: shearrate= sqrt(2.0*(symm(fvc::grad(u)))&&(symm(fvc::grad(u)) )/3.0; wmake it, still one error exist, but I didn't found out the error, then I changed it back, as original: shearrate=max(minshearrate, sqrt(2.0*(symm(fvc::grad(u)))&&(symm(fvc::grad(u)) )/3.0); then I wmake it, still one error ,but this time the output is much more than the first time, I can not find where the error is, I am confused, the same code, leads to different output, what is the problem? I checked the dimensions of the two arguments of max(), there are both 1/second, why it thinks the second argument 's dimension is [0 0 0 0 0 0 0]? anybody has any idea? thank you. and sorry for disturbing. wendy

 April 11, 2011, 03:08 #16 Senior Member   Max Join Date: Mar 2009 Posts: 133 Rep Power: 9 Are you sure this is the right forum for this question? It sounds pretty much like OpenFoam which has its own userforum here on cfd-online. cheers

April 12, 2011, 09:08
#17
Member

xianghong wu
Join Date: Mar 2009
Posts: 57
Rep Power: 9
Quote:
 Originally Posted by coglione Are you sure this is the right forum for this question? It sounds pretty much like OpenFoam which has its own userforum here on cfd-online. cheers
yeah, Thank you for reminding.

 November 19, 2014, 03:11 strain rate and scale of strain rate #18 Member   Qureshi M Z I Join Date: Sep 2013 Posts: 74 Rep Power: 5 hi, anybody know the difference between "strain rate" and "scale of strain rate". please share your knowledge. regards

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are On Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post tom FLUENT 0 February 13, 2009 11:27 Madhukar Rapaka FLUENT 0 June 26, 2006 03:17 Mecobio Main CFD Forum 0 November 7, 2005 13:55 ap FLUENT 8 April 19, 2003 08:00 David Yang FLUENT 3 June 3, 2002 06:13

All times are GMT -4. The time now is 01:09.

 Contact Us - CFD Online - Top