[kri321shna]

Hi,

I am using udf.in and I selected some outputs in udfs.

I am getting below error message while executing. How to overcome it.

"Error, could not find CONVERGE UDF LIBRARY, libconverge_udf.so"

which files i need to complile using make command.

can somebody help me out.
Thanks

[MFGT]

Depends on what you want to output with your UDF.

I use the UDF for monitor points with extra variables.

[MFGT]

So, for my monitor points i include these 5 objects in the makefile.win file:

OBJS =  user_dynamic_routines.obj\
        user_dt.obj\
        user_inputs.obj\
        user_points.obj\
        user_points_data.obj\

then run the make_udf.bat and copy the converge_udf.dll into C:\Program Files\CONVERGE

[kri321shna]

Thank you so much MFGT for your response.

I cannot able to see these *.obj and converge udf manual in my windows or linux installation folders.

I am looking for wall htc values. Please find attached udf.in (changed file extn due to uploading error)

Please suggest me, how to go further to solve this error.

Thank you again.

Krishna

[jhuang]

Hi Krishna,

Do you use Linux or Windows? The following steps are for linux:

1. Please download the two zip files of UDF_Dev_Packages from our download site. They must be of the same version of the solver you are using.

2. Unzip the two zip files and copy all the files into a folder udf_lib.

3. Copy the udf_lib folder into your case directory (where you put all the .in and .dat files for the case)

4. Under udf_lib folder, run make command. With the default udfs, it should run successfully and generate this library file libconverge_udf.so.

5. I see that you want to use user_wall_value and user_heat_transfer, you can make modifications to those two files based on your need. Then compile the library file again.

6. Run the simulation in your case directory (which contains the udf_lib subfolder). If CONVERGE still complains it cannot find the libconverge_udf.so, add ./udf_lib to your environment variable LD_LIBRARY_PATH

Hope this helps. Thanks!

[MFGT]

If its with Linux, i am out.
We are on Windows here

[jhuang]

Hi MFGT, thanks for your answers on Windows system!

[kri321shna]

hi I got all these files.

Based on my attached udf.in ,

please tell me the exact procedure in linux.

Thank you so much.

Krishna

[jhuang]

Hi Krishna,

As first step, you don't need to do any modification to the files you downloaded. Just do the compiling, make sure you got the libconverge_udf.so generated correctly.

Run the simulation with the udf.in you uploaded and udf_flag in inputs.in set to 1. If the simulation runs fine, then go to the next step. What's happening here is, the two udf files user_heat_transfer.c and user_wallvalue.c are called and taking control of calculating wall heat transfer flux and wall values. In user_heat_transfer.c, you can see that heat_model 0 is used, therefore no matter what you set in your turbulence.in, model 0 will be used.

Now you can modify the two files user_heat_transfer.c and user_wallvalue.c based on your need. Please refer to UDF manual for explanations of variables, data structures. If you have any questions, ask here or send a request to support@convergecfd.com. Thanks!

[kri321shna]

hi Jhuang,

after step 4 (when i am executing make command), i am getting this below error

make: *** No targets specified and no makefile found. Stop.

how can i overcome this

[kri321shna]

one more doubt

if i develop a .so file in linux, will it work in windows execution??

Thank you

[kri321shna]

I got the solution.

Thank you