Learning Forum

[andrewl]

I'm injecting water particles (inert particle) from nose while inhalation. inlet and outlet were defined in Space Claim. 

In the Fluent, I set the injection velocity as well as the inhalation velocity. After running the simulation (it converged), nothing showed up on the scene. I set the outlet BC "trapped" to prevent overflow, but still not showing anything 

• Duration of injection is 2 second.
• Particle injection velocity 8 m /s
• Inhaltion velocity 0.5 m/s

Below is the model, arrow with an angle is the inlet for water particles, the other one is air. I'm using DPM model, and using surface injection (defined by the geometry)

Anyone know how to get the particle injected successfully?

[Rob]

That looks sensible. What is written in the Console panel relating to tracked particles?

[andrewl]

looks like its just post processing issues, I re-run the simulation again today and I can certainly see some really really large particles stuck in the airway. I think I need to fine tune the injected particles size also the particles scale value in "particle-track-1"

[andrewl]

the diameter of my injected particles was 0.001 mm , too small to be able to see.

[Rob]

Ah, yes. In the Particle Tracks panel you can set a display diameter - that tends to help. For info 1micron particles will typically follow the flow, miss the sides in the nose and be exhaled on the next breath. You may want to do some reading on drug & carrier particle sizes, or I do if the practice has moved on in the last few years. 

[andrewl]

No exhale simulated, I was just simulating inhalation and particle injection. Outlet boundary was set to "trap" so even pass the side of the nose it should be trapped and shown somewhere in the geometry.

Is "scale" in particle tracks panel same as what you said the display diameter?

[andrewl]

 

I found the diameter option in particle track panel, now my solution doesn’t converged. Error message in below, do you know what could be the reason? look like something to do with my temperature setting.

 

============================================================================ 

iter  continuity  x-velocity  y-velocity  z-velocity      energy           k       omega     time/iter

  1040  2.3827e-03  1.2309e-07  1.3347e-07  2.6021e-07 1.6748e+293  3.8479e-06  6.8937e-07  0:00:21   20

        Stabilizing temperature to enhance linear solver robustness.

        Stabilizing temperature using GMRES to enhance linear solver robustness.

 

Divergence detected in AMG solver: temperatureFilm time = 5.210000e-02 with timestep = 1.00e-04, (max_cfl: 6.705262e+00)

Film time = 5.220000e-02 with timestep = 1.00e-04, (max_cfl: 3.464841e+00)

Film time = 5.230000e-02 with timestep = 1.00e-04, (max_cfl: 3.045373e+00)

Film time = 5.240000e-02 with timestep = 1.00e-04, (max_cfl: 5.991012e+00)

Film time = 5.250000e-02 with timestep = 1.00e-04, (max_cfl: 6.446741e+00)

Film time = 5.260000e-02 with timestep = 1.00e-04, (max_cfl: 1.304315e+01)

Film time = 5.270000e-02 with timestep = 1.00e-04, (max_cfl: 1.205507e+01)

Film time = 5.280000e-02 with timestep = 1.00e-04, (max_cfl: 3.353779e+00)

Film time = 5.290000e-02 with timestep = 1.00e-04, (max_cfl: 3.320096e+00)

Film time = 5.300000e-02 with timestep = 1.00e-04, (max_cfl: 3.241072e+00)

 

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

 Artificial walls on 97 faces (0.4% area) of pressure-outlet 20 to prevent fluid from flowing into the boundary.

 

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Divergence detected in AMG solver: temperature

Error at host: floating point exception

 

===============Message from the Cortex Process================================

Compute processes interrupted. Processing can be resumed.

==============================================================================

Error at Node 1: floating point exception

Error: floating point exception

Error Object: #f

 

[Rob]

The wall film CFL is very high, why are you modelling a wall film? What settings are on the film as it's a function that needs more care than most realise. 

[andrewl]

Im not sure why its high, im modeling a wall film to simulate water particles deposited on the wall and form a liquid wall film. if the film greater than a thickness the liquid will transport to the adjacent surface. thats what i want to simulate.

generally the CFL number should be lower than 1, isn't it?

[andrewl]

that might be the reason causing divergence?

[Rob]

The film will move based on gravity and local flow, and it's likely the film is very thin. That can cause issues with momentum as another droplet hits the cell. Drop the film time step (or increase the number of film sub steps) to aim for a CFL around 0.05 and see how it behaves. The film calculation is generally fast relative to the flow so more film steps shouldn't slow things down too much. 

Do you need to be solving for energy? Just wondering how significant the temperature changes will be on the solution? 

[andrewl]

ok, I will try to lower the time step first.

Not necessary have to be solving for energy, particle spray in nose should not have much changes in temperature. I can even do the particle spray without interaction with contineous phase.

[Rob]

OK, so turning energy off may well help too. In Fluent we'd typically try and use a few models as possible as it's computationally more efficient. 

[andrewl]

got it , I have resolved the issues, particles are injected now. I will close this topic, thank you mate! appreciated.

[Rob]

You're welcome. Good luck with the work.