[raul_xusernamex_]

Problems in the outlet-Scour Pipeline- vertical mesh..

Problems in the outlet-Scour Pipeline- vertical mesh.

Hello, very good evening, some time ago I made a query related to the subject of scour in pipes under oscillatory flow due to waves.
According to my classes, the main thing is to adequately represent the hydrodynamics for my channel in the numerical channel. For that, it is necessary to determine how many vertical layers I need and obtain an adequate element size by means of the water level, which is known.
Reviewing the manual I realized that there is a command called B 2 in the control.txt file in which it allows me to define the element size in the vertical but when I tell REEF3D to generate the vertical decritization it rejects the execution and I cannot obtain what I need.

Maybe I'm mistaking the way I'm entering the values ​​in the command. Because many of the tutorials that are available there are none that use the B2 command.
I wanted to know how I should use this command in my current configuration. Below I show the configuration of my control.txt files.

Control.txt

C 11 6
C 12 3
C 13 3
C 14 8
C 15 21
C 16 3

B 1 0.01
B 2 0.01 0.01 0.01

B 10 -4.4 4.4 0.0 0.01 -0.3 1.0
B 130 1
S 32 0.0 0.1 0.05

M 20 2

M 10 4

crt.txt

B 10 1
B 50 0.0001
B 90 1
B 92 4
B 93 0.1 1.43
B 98 3
B 99 4

D 10 4
D 20 2
D 30 1

F 30 3
F 35 4
F 40 3
F 42 1.0
F 60 0.40

I 12 1

N 10 18
N 11 11
N 40 3
N 41 60.0
N 47 0.1

M 10 4

S 10 1
S 11 1
S 12 0
S 13 1.0
S 20 0.00058
S 21 3
S 22 2700
S 24 0.4
S 25 45
S 32 4
S 30 0.11
S 47 0.5
S 48 0.5
S 57 0
S 71 -3.0
S 72 3.0
S 77 -3.0 4.1
S 80 2
S 90 1


P 10 1
P 30 0.1
P 40 1
P 42 1.0

T 10 2

W 22 -9.81

P 52 0.005
P 53 1
P 55 0.40
P 122 1
P 126 1

P 51 -1.0 0.005
P 51 0.0 0.005
P 51 1.0 0.005
P 51 2.0 0.005

[kamath]

Hi Raul,

There is one major misunderstanding that might solve your problem.

B 1 in control.txt = uniform grid size for the whole domain

B 2 in control.txt = Number of CELLS in each direction.

B 2 is generally used along with grid stretching functions explained here:
Use of grid stretching functions in DiveMESH

B 2 values will overwrite B 1.

With your current input, the mesh will not be generated correctly as "number of grids" is supposed to be an integer and you provide 0.01

[raul_xusernamex_]

Originally Posted by kamath

Hi Raul,

There is one major misunderstanding that might solve your problem.

B 1 in control.txt = uniform grid size for the whole domain

B 2 in control.txt = Number of CELLS in each direction.

B 2 is generally used along with grid stretching functions explained here:
Use of grid stretching functions in DiveMESH

B 2 values will overwrite B 1.

With your current input, the mesh will not be generated correctly as "number of grids" is supposed to be an integer and you provide 0.01

Ohh thank you very much Lord Kamath !
I read well now and you are right.
Another query related to this topic.
My teacher tells me that I should pay attention to how many vertical layers I need to reproduce the hydrodynamics. What does he mean by that?
As I understand it, a vertical mesh should be obtained to describe the velocity profile in the water column and I think there is a command where the dimensions of each element are set.
So since I know the water head (channel depth) and then divide it by the number of vertical layers I need, I should get a suitable cell size.
So what command(s) should be involved in this case?

Thank you very much.

[kamath]

The concept of number of layers applies to shallow water equations. And there if the shallow water equations are being solved using the explicit non-hydrostatic pressure solution.

In the REEF3D framework, SFLOW uses the shallow water equations, but with an implicit non-hydrostatic pressure solution. Therefore, the question of layers does not arise.

In addition, from your ctrl.txt, you are using the Navier-Stokes solver, REEF3D::CFD, therefore the question of layers does not arise here either.

What is most important for free surface flows and free surface flows with two-phase modelling (eg. air/water) is the grid resolution near the free surface.
On that note, we have noticed in REEF3D that due to the level set method and high-order discretisation (fifth-order WENO), the grid resolution required to represent the wave length correctly (ca. 80 cells/ wavelength) is then sufficient to represent the wave heights accurately.