Fluids

Fluids

Pressure across porous zone in FLUENT

    • gautelt
      Subscriber

      Dear community,


      I am simulating the supply airflow into a room through a 9m2 wall made out of ventilation textile, similar to that used in fabric ducts. A pressure chamber is located behind the wall, and a velocity inlet slot measuring 0.314 m2 is located in the chamber ceiling. In reality, the textile has lamellas in front, making the supply area only 4.5m2.


      From a manufacturer, I have been informed that the chosen fabric delivers 450 m3/h at 40Pa static pressure drop (for 4.5m2), and that I can assume the airflow to have a linear relationship with the pressure (e.g. 225 m3/h at 20 Pa).


      The textile is made as a porous zone, with the coefficients 1/alpha=4.75e11 and C2=0, made from a linear trendline of the pressure-air velocity plot and adjusted for a wall area of 9m2. When I simulate, I get air velocities and temperatures within the expected range, but the pressure inside the chamber is about 4500 Pa, which is way off the 40Pa needed for the airflow. The outlet is located in the back wall of the room and defined as outflow.


      Can anyone help me understand what is going on here? Thank you!


      Geometry


      Coefficient calculations

    • Karthik R
      Administrator

      Hello,


      How did you obtain the value of your resistance? Also, is your model fully converged?


      What are your boundary conditions?


      Thanks,


      Best,


      Karthik

    • gautelt
      Subscriber

      Thank you for helping, Karthik.

      The resistance is calculated from the trendline following the pressure curve, using the relationship C1 = 1440x, and that C1 = (mu/alpha)*delta_m. Due to linearity in pressure drop (according to the manufacturer), C2 is 0. Since the actual opening area is only half of the porous zone face area in Fluent, a conversion was made for the constant C1 (Fluent user guide eq. 7.2-11, link below), increasing C1 to 2880. It is then solced for 1/alpha as 1/alpha=2880/(mu*delta_m) = 4.75e11 m^(-2).


      The model is fully converged.  For BC, I have adiabatic walls, a total heat load in people an PC's of 480 W, operating temperature of 288K, inlet temperature 285K (want 3deg under temperature on inlet air), inlet velocity 0.4095 m/s.


      Fluent ug: https://www.afs.enea.it/project/neptunius/docs/fluent/html/ug/node233.htm

    • Rob
      Ansys Employee

      Use a pressure outlet, outflow is an obsolete boundary condition and shouldn't be used. 


      Are you using porous media or porous jump conditions?

    • gautelt
      Subscriber

      Thank you, I will try that.


      The proposed model is for a porous zone. Since the porous section is only 0.4 mm thick, I also tried to model as a porous jump where i simply used alpha=1/(4.75e11) m2 as face permeability. Pressure-Jump Coefficient and Thermal Contact Resistance as 0.


      I can't get the model to converge this way, and I am having difficulties understanding what the problem is also here. If you are able to help me out, that would be great!

    • gautelt
      Subscriber

      Attached you can see the velocity magnitude for both the porous media and porous jump, using a pressure outlet. The magnitude and distribution pattern for porous media is at least within the expected range, while the porous jump gives very high velocities. Do you understand why this happens? Have I calculated the input parameters wrong?


      Porous media velocity


      Porous jump velocity

Viewing 5 reply threads
  • You must be logged in to reply to this topic.