Learning Forum

[srath9]

Hello,

I am trying to simuate a simple case with GRID_MOTION UDF. The main issue I am facing is mass imbalance (not conserved). 

2D Geometry: x (0,150 mm) and y (0,20 mm) 

Case: 

1. Single phase: Air (ideal gas)
2. Dynamic mesh: 
   • Right wall - stationary
   • Left wall - Deforming
   • Top wall - Deforming
   • Bottom wall - UDF
3. Time step ~ Have even tried 1e-5s

Predicted motion of part of bottom wall using UDF:

Actual motion (as predicted):

At t=0;

At some time point (motion defined by UDF):

At t =0.5s;

 

 

UDF used:

#include "udf.h"

#include "dynamesh_tools.h"

#define amplitude 0.01

// Cycles is defined for periodic motion

#define cycles 30.0    

 

// Function to define the motion based on the time t

DEFINE_GRID_MOTION(trap, domain, dt, time, dtime)

{

    Thread *tf = DT_THREAD(dt);

    face_t f;

    Node *v;

    real x, y,z;

    real new_z;

    real periodicity = (60/cycles)/2;

 

    int n;

    real t = CURRENT_TIME; // Get the current time

 

    SET_DEFORMING_THREAD_FLAG(THREAD_T0(tf));

 

    begin_f_loop(f, tf)

    {

        f_node_loop(f, tf, n)

        {

            v = F_NODE(f, tf, n);

 

            if (NODE_X(v) > 0.050 && NODE_X(v) < 0.090 && NODE_POS_NEED_UPDATE(v))

            {

                NODE_POS_UPDATED(v);

                x = NODE_X(v);

                y = NODE_Y(v);

z = NODE_Z(v);

 

                // Update the coordinates

 

                NODE_Y(v) = (amplitude/M_PI) * (asin(sin(M_PI*t/periodicity)) + acos(cos(M_PI*t/periodicity))); 

NODE_X(v) = x;

            }

        }

    }

    end_f_loop(f, tf);

}

Mass imbalance:

 

Please share any info that might help. Thank you!!

[Federico]

Hello, 

the way your interior mesh is deforming is compromising mesh quality. I would recommend that you remesh this model using tetrahedral cells and use Remeshing in Fluent. This will help preserve mesh quality and should reduce any mass imbalance from highly skewed cells or faces.