# Simple Box - Gravity with Pressure Outlet - Unrealistic Reverse Flow

 Register Blogs Members List Search Today's Posts Mark Forums Read

July 22, 2020, 12:36
Simple Box - Gravity with Pressure Outlet - Unrealistic Reverse Flow
#1
New Member

Bs
Join Date: Apr 2020
Posts: 2
Rep Power: 0
Hi all

I intend to simulate a tidal turbine in a tidal current. First, I am trying to test the application of gravity/hydrostatic pressure gradients to a simple box of sea water of the same dimensions as my domain so that I can get this right before adding the turbine.

Inlet velocity 0.1m/s, the nearest wall to the right is inlet (flow in +y direction)
Seawater properties: density 1024kg/m3, visocisty 0.00111
the domain is 48x48x80m
Gravity applied with operating density 0
A reference pressure at the origin represents that which would be experienced at 24m depth below sea level (about 340,000Pa)
pressure outlet with 0 gauge pressure
Remaining boundaries are walls
using k-w SST
Coupled method
Coarse mesh for the sake of this experiment

As you can see I am getting a bizarre reverse flow circulating flow pattern, as well an incorrect and inconsistent pressure values along the height of the box. I think the issue is with the pressure outlet: by applying a fixed pressure at outlet and applying gravity to the domain so that fluid with a pressure gradient arrives at the outlet, the velocity at outlet has to vary along the outlet to change pressure to a constant value?
I have selected average pressure specification, I am not sure how else to allow fluent to maintain the
The values of pressure at top and bottom (Z direction) are not as expected and not consistent along the flow path - the upper boundary should see 1atm pressure as if it is sealevel. I believe the pressure contour shows relative/gauge pressure, so max and min should be about +-2.4e5Pa

I have been able to achieve a more realistic result (consistent and correct pressure gradient, max and min pressure expected, relatively smooth flow from inlet to outlet) by applying mass flow inlet and outlet. This is fine for a confined channel but is not applicable to my physical scenario and as such shouldnt be used.

As I say I think this is an issue of the options selected for the pressure outlet, but I am not sure. Any advice is great!
Attached Images
 Capture2.jpg (123.7 KB, 19 views) Capture.PNG (44.2 KB, 9 views)

July 23, 2020, 09:02
#2
New Member

Bs
Join Date: Apr 2020
Posts: 2
Rep Power: 0
For reference, this is what I would like to achieve (attached)

This is by applying gravity along with equal mass flow inlet and outlet
Attached Images
 Capture3.PNG (174.6 KB, 17 views)

 December 1, 2021, 17:31 #3 New Member   Chris Join Date: Mar 2021 Posts: 23 Rep Power: 4 I know this is an old post but I have recently encountered a similar problem. I haven't completely solved this issue but what I did do is specify the outlet pressure profile with a UDF to be (rho*g*h) the hydrostatic pressure. There is still a small discontinuity in the pressure gradients of the outlet and interior domain. This significantly reduced the reverse flow but the pressure gradients do not perfectly align hence there is still reverse flow. I'm still working on a solution. I cannot use outflow in my case because I have incompressible flow. Were you able to find a solution?