[CFD27]

Hello all,
I want to simulate the diffusion flow of a species through a wall to the other side of the wall. For this purpose, I have already determined the cells adjacent to the wall using a UDF and thus built in source terms from both sides. To make the sources now dependent on the concentration of the species, I want to determine the gradient of the species. For this I found in pressure-based solver after entering the command "(rpsetvar 'species/save-gradients? #t)" into the TUI. As a first step, I tried to determine the gradient and save it in a UDM so that I can use it later for computation. When compiling the UDF the following error occurs: error C2440: '=': 'real *' cannot be converted to 'real. When modifying the UDF to save the mass fraction, no error occurs. Can anyone give me a hint where my error is?

This is my Code for Saving Gradient:

#include "udf.h"
#include "mem.h"
DEFINE_EXECUTE_AT_END(Grad_calc)
{
	Domain* d;
	Thread* t;
	cell_t c;
	d = Get_Domain(1);
	thread_loop_c(t, d)
	{
		begin_c_loop(c, t)
		{
			C_UDMI(c, t, 1) = C_YI_G(c, t, 0);
		}
		end_c_loop(c,t)
	}
}

Thanks in advance.
Many Greetings
Joe

[AlexanderZ]

gradient is a vector so you have 3 scalar values for 3D problem
to be

C_UDMI(c, t, 1) = C_YI_G(c, t, 0)[0];
C_UDMI(c, t, 2) = C_YI_G(c, t, 0)[1];
C_UDMI(c, t, 3) = C_YI_G(c, t, 0)[2];

[CFD27]

Dear Alexander,
Thank you for your reply. I have already been able to fix the problem myself. Your hint that the gradient is a vector was part of it. I converted the gradient to a scalar using the Magnitude macro. After this error was fixed, a memory error occurred, which I fixed by having the solver access the gradient only after checking to see if it was in memory. In addition, I have replaced Execute at End macro with a Define Adjust macro. Thanks for your help.
Many greetings
Joe