Fluent UDF for Non linear wall temperature gives constant wall T after initialisation

anuarun · April 20, 2017, 09:13

Hello all,
I have a temperature profile file as my wall BC. My profile file is working fine (as evident from T plots after convergence). I have written a UDF to vary vapour pressure as a function of temperature. I have hooked it to the vapour pressure option while activating mass transfer from liquid to vapour>cavitation. It gets interpreted neatly and solution converges also.Now, the solution obtained is not showing the expected improvement (rather change) when compared to a case where Pv=const.
My UDF looks like this.

#include "udf.h"
/*for Vapour pressure*/
DEFINE_PROPERTY(Vapour_pressure,c,t)
{
real T = C_T(c,t);
real p_v;
p_v = (0.00329*exp(0.19039*T) - 0.05762)*1000000.0;
return p_v;
}

How will I know if this is working fine? Do I need to write the UDF specifically in terms of SUB_THREAD, SUPER_THREAD etc. I assumed that since I am calling UDF from vapour pressure options, it will take appropriately. Is that right? Has anybody done something similar? Any help is greatly appreciated. I am stuck here for more than a month.

anuarun · May 16, 2017, 11:17

Hello all,
I found that defining temperature BC as a profile file is better than including a UDF especially when I have to include(by interpreting) other UDFs(say for properties in terms of temperature). Now my profile file is working fine. I have written a UDF to vary vapour pressure as a function of temperature. I have hooked it to the vapour pressure option while activating mass transfer from liquid to vapour>cavitation. It gets interpreted neatly and solution converges also.Now, the solution obtained is not showing the expected improvement (rather change) when compared to a case where Pv=const.
My UDF looks like this.

#include "udf.h"
/*for Vapour pressure*/
DEFINE_PROPERTY(Vapour_pressure,c,t)
{
real T = C_T(c,t);
real p_v;
p_v = (0.00329*exp(0.19039*T) - 0.05762)*1000000.0;
return p_v;
}

How will I know if this is working fine? Do I need to write the UDF specifically in terms of SUB_THREAD, SUPER_THREAD etc. I assumed that since I am calling UDF from vapour pressure options, it will take appropriately. Is that right? Has anybody done something similar? Any help is greatly appreciated. I am stuck here for more than a month.

anuarun · May 16, 2017, 11:25

I am really sorry for the bad editing. My first post was regarding a Temperature UDF for wall temperature not working fine. I have solved it by including a profile file instead of a UDF. That post got deleted while editing.

April 20, 2017, 09:13	UDF for vapour pressure	#1
anuarun New Member Anuja Vijayan Join Date: Mar 2017 Location: Thiruvananthapuram Posts: 23 Rep Power: 9	Hello all, I have a temperature profile file as my wall BC. My profile file is working fine (as evident from T plots after convergence). I have written a UDF to vary vapour pressure as a function of temperature. I have hooked it to the vapour pressure option while activating mass transfer from liquid to vapour>cavitation. It gets interpreted neatly and solution converges also.Now, the solution obtained is not showing the expected improvement (rather change) when compared to a case where Pv=const. My UDF looks like this. #include "udf.h" /for Vapour pressure/ DEFINE_PROPERTY(Vapour_pressure,c,t) { real T = C_T(c,t); real p_v; p_v = (0.00329exp(0.19039T) - 0.05762)1000000.0; return p_v; } How will I know if this is working fine? Do I need to write the UDF specifically in terms of SUB_THREAD, SUPER_THREAD etc. I assumed that since I am calling UDF from vapour pressure options, it will take appropriately. Is that right? Has anybody done something similar? Any help is greatly appreciated. I am stuck here for more than a month. Last edited by anuarun; May 16, 2017 at 11:22. Reason: Encountered a new problem*

May 16, 2017, 11:17	Vapour Pressure UDF	#2
anuarun New Member Anuja Vijayan Join Date: Mar 2017 Location: Thiruvananthapuram Posts: 23 Rep Power: 9	Hello all, I found that defining temperature BC as a profile file is better than including a UDF especially when I have to include(by interpreting) other UDFs(say for properties in terms of temperature). Now my profile file is working fine. I have written a UDF to vary vapour pressure as a function of temperature. I have hooked it to the vapour pressure option while activating mass transfer from liquid to vapour>cavitation. It gets interpreted neatly and solution converges also.Now, the solution obtained is not showing the expected improvement (rather change) when compared to a case where Pv=const. My UDF looks like this. #include "udf.h" /for Vapour pressure/ DEFINE_PROPERTY(Vapour_pressure,c,t) { real T = C_T(c,t); real p_v; p_v = (0.00329exp(0.19039T) - 0.05762)*1000000.0; return p_v; } How will I know if this is working fine? Do I need to write the UDF specifically in terms of SUB_THREAD, SUPER_THREAD etc. I assumed that since I am calling UDF from vapour pressure options, it will take appropriately. Is that right? Has anybody done something similar? Any help is greatly appreciated. I am stuck here for more than a month.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Divergence in AMG solver!	marina	FLUENT	20	August 1, 2020 12:30
UDF for linear temperature variation	nonstopthinker	Fluent UDF and Scheme Programming	2	September 4, 2016 01:36
specified shear at wall - temperature gradient - UDF - access violation error	senD	Fluent UDF and Scheme Programming	9	September 18, 2014 08:29
UDF of linear distribution of temperature in a 2D tank	arailos	Fluent UDF and Scheme Programming	1	August 19, 2010 15:14
udf variable for wall temperature	Kiran	FLUENT	0	July 31, 2008 09:31

May 16, 2017, 11:25		#3
anuarun New Member Anuja Vijayan Join Date: Mar 2017 Location: Thiruvananthapuram Posts: 23 Rep Power: 9	I am really sorry for the bad editing. My first post was regarding a Temperature UDF for wall temperature not working fine. I have solved it by including a profile file instead of a UDF. That post got deleted while editing.