Learning Forum

[xing]

Attached snapshot is a pressure contour in the domain for Open Channel flow simulation. Height of the domain is 1.5 m and the free surface is at 0.3 m. Origin is at the upstream end, at 0.3 m high where the free surface and the reference pressure (ANSYS default value: 101325 Pascal) is located. My question is: Should not the pressure value around the reference pressure location be close to the reference value that we set at that location? Comparing with this concept, our simulation does not have such pressure at the free surface. Could there be any other factor you suggest can be affecting the pressure? 

The image is the snapshot after calculations of about 2.6 seconds which is the steady-state flow condition.

[Amine Ben Hadj Ali]

Please insert the screenshot as we do not look into attachments (at least ANSYS Staff).

1/Add screenshot straight after initialization

2/Add more information about model and boundary conditions.

[xing]

More explanation about the model:
We are trying to simulate the real case of river water flowing through the Salmon Redd (Salmon digs shallow hole (Redd) in the water bed to lay their eggs). We are modeling two different domains, one with upper water-air domain using open channel flow model and another is, lower sediment domain where we will study the flow patterns and other flow behaviors. 

For this case of open channel flow in the upper domain we are tying to figure out the correct pressure at the reference pressure location. My question to you is, we have the free surface (water-air interface) at y = 0 m and since I haven't applied any external pressure, I believe the ANSYS default pressure value: 101325 Pascal should be there. If you see our snapshot, the pressure value is way different than this. We want to understand how are we getting this pressure value (around 500 Pascal) ? I am attaching Pressure contours for both, 1) right after initialization 2) after convergence.

Is the static pressure shown by FLUENT the absolute or the gauge pressure?

The Boundary Conditions applied are as in the image of pressure contour obtained right after initialization. The domain dimensions are as in the image.
We used, Transient Simulation; Gravity: -9.81 m/s; VOF-Open Channel Flow; K-omega (SST); Water and air have default values of viscosity and densities with Surface tension applied, 0.074 n/m. 

 

[Konstantin]

Fluent always uses gauge pressure which is absolute minus reference, so 101325 Pa is subtracted, and what you see at the reference point is gauge pressure. If you have a pressure boundary (pressure inlet or pressure outlet) then the reference pressure locations defaults to that boundary and your reference point location becomes irrelevant.

To observe expected hydrostatic pressure distribution, set reference pressure = 0 Pa. Make sure to correct boundary conditions and use absolute pressure values at ALL boundaries. Alternatively, you can add 101325 Pa to the calculated pressure field when post-processing the results.

Hope this helps.

[xing]

Thank you for your help!  We have additional questions below on the same problem. 

We are modeling a 2-D "Open Channel Flow" model (a river water flow over the waterbed and sediment underneath the bed).

Initially we started by modeling a separate (1) upper water+air domain and (2) porous sediment domain with certain permeability value assigned for water permission. Pressure profile exported from (1) at the bed was assigned as and Pressure Input (has both flux in and flux out) at the top boundary of Porous sediment domain(2). 

(1) Upper (Water+Air) domain      

(2) Sediment domain 

Now, we are trying to model a combined upper (air+water) and sediment (1+2) domain using a single simulation. With the similar conditions applied here, we haven't been able to get the solutions correctly. There is a slight fluctuation in the free surface at the upstream in the combined domain and so is for the pressure profile at the interface. I am attaching the Boundary Conditions and model information that we used. Could you please suggest any specific details we might have missed while modeling such open channel problem?

We used: Transient Simulation; Surface tension applied, 0.074 n/m; VOF (open channel); K-omega (SST) model, sand (density 2500 kg/m3) for the porous medium in sediment region with some permeability.  

---

Also, how to specify different velocity magnitudes for water and air at inlet in open channel flow simulations? The velocity of the air seem to follow the velocity applied to the water. Please have a look at the velocity vectors for water and air below (right at the entrance).

 

[ydai]

 

 

 

 

[ydai]