Learning Forum

[Nida]

Dear Sir,

Appended is my UDF for flow profile. I am providing it to Mass Flow Inlet. As it is starting from negative sinusoidal.

It should move flow out in the start and then flow should come it in sinusoidal pattern.

But there is NO difference even I put here negative sign or not.

F_PROFILE(f, thread, position) = (-0.000096)*sin(2 * PI * 0.001 * t);

I want to get flow out from Fluid Domain in start of simulation. I am initializing the fluid domain with zero velocity and 300K (by default Temperature)

 

#include "udf.h"

#define PI 3.141592654

DEFINE_PROFILE(first_velocity_in_out, thread, position)

{

face_t f;

real t;

begin_f_loop(f, thread)

t = RP_Get_Real("flow-time");

{

F_PROFILE(f, thread, position) = (-0.000096)*sin(2 * PI * 0.001 * t);

}

end_f_loop(f, thread)

}

[Rob]

Have you checked the mass flux on the boundary, and run the model for a few time steps?

[Amine Ben Hadj Ali]

Can you share a screenshot of the boundary? What are giving as direction?

[Nida]

Model is simple. This is a cylinder where flow is coming in and out from bottom.

I am making bottom as mass flow inlet and providing that UDF to it. Flow is coming in and out in sinusoidal manners. 

But the  problem is that in the start flow should go out then come in(According to sin function with negative sign that I am providing).

But there is no difference for negative or positive sign to this Sin function.  First flow go in then go out.

This is the setting for direction of flow "Normal to Boundary"

[Nida]

Mass Flow rate at 10th time step is 6.0278908e-06 (positive)

I also shared the model  and Mass flow Inlet settings in this post. 

[Amine Ben Hadj Ali]

The most important thing is the direction. You need to use the direction vector to tell Fluent that flow will cone out. Check the manual as it describes it.

[Nida]

Sir, I tried with both methods "Direction Vector" and " Normal to Boundary".

I am not saying that flow is not coming out. It coming out but not in start of simulation.

My question is that there should be a difference between these two profiles.

F_PROFILE(f, thread, position) = (-0.000096)*sin(2 * PI * 0.001 * t);

F_PROFILE(f, thread, position) = (0.000096)*sin(2 * PI * 0.001 * t);

One is starting with positive profile and other is starting with negative profile.

But why I am getting same output for both of these???

For (-0.000096) flow should come out in start of running simulation if I am initialization my domain.

 

[Rob]

How are you initialising the flow? 

[Nida]

Sir, now it is hybrid initialization.

Previously I also used zero absolute velocity and 300K Temperature for initialization.

[Rob]

So, in that case why would the velocity be negative at t=0?  Think what initialisation does then consider how the values are calculated in the UDF. 

[Amine Ben Hadj Ali]

At t=0 both Profiles return Zero. Still thinking better to make the direction vector by profile.