[NTNHAN]

Dear Team REEF3D and all members,

Respect thanks to your software.

I am trying to simulate the local scour around the cubic solid that has an open hole, for the small domain around the cubic solid, the scour shape is nice, but when I extend the domain around the cubic object (just extend the domain and change the suitable Discharge volume and keep another thing), there are no scour around cubic solid.

My problem (question): Why when I extend the domain, erosion doesn't happen? and How can I solve it?

The cubic dimension is 5.1x5.1x5.1 cm and the opening holes have dimensions is 3.9x3.9x3.9 cm located at the center of each surface of the cubic. The position of the cubic is on the top of the sediment domain. The velocity is 0.22m/s. I am using REEF3D 23.01 version.

I have attached two cases (small_domain and extend domain) for each case conclude control.txt, ctrl.txt and solid.stl (ASCII) below.

Link: https://drive.google.com/drive/folde...usp=share_link

Here are the control.txt and ctrl.txt for small_domain

control.txt

C 11 1
C 12 3
C 13 3
C 14 2
C 15 21
C 16 3

B 1 0.003
B 10 0 0.11 0 0.11 -0.051 0.102
S 1 1
S 7 0.03 0.03 0

M 10 16
M 20 2

ctrl.txt

B 10 1
B 50 0.0001
B 60 1
B 61 2

D 10 4
D 20 2
D 30 1

F 30 3
F 40 3
F 50 2
F 60 0.10

I 12 1

M 10 16

P 10 1
p 27 1
P 30 1
p 180 1

N 10 14
N 11 11
N 40 3
N 41 60
N 47 0.3

S 10 1
S 11 1
S 13 5
S 14 0.3
S 15 0
S 16 4
S 20 0.00035
S 22 2650
S 23 0.3
S 24 0.4
S 25 30
S 30 0.047
S 41 2
S 42 1
S 44 3
S 45 15
S 50 3
S 57 0
S 71 0.01
S 72 0.14
S 80 2
S 90 4
S 91 1
S 93 2

T 10 2
T 36 2

W 10 0.002664
W 22 -9.81


For the extend_domain I just changed the dimension of the domain.

B 10 -0.06 0.171 -0.03 0.141 -0.051 0.153 (Rectangular domain)

F 60 0.15 (water level)

W 10 0.005643 (Discharge volume)


Best regards

Nhan

[Felix_Sp]

Hey

first of all:

Is there a reason you basically have no air phase and F60 is nearly identical with your domain height? I do not know how stable such an application is.

In regard to your question. My guess is: You have symmetry boundary conditions in y-direction and therefore your domain is basically mirrored on that position. You could think of it like infinite numbers of hollow blocks in a row. Therefore, the cross-section for your flow is way smaller for the small domain than it is for your large domain. Therefore, even though you scale your discharge, the flow velocities are smaller for the extended domain than for the small domain (also compare the figure I appended). A lower flow velocity results in a lower shear stress, and therefore the scour is not present in the extended domain.

Cheers

[NTNHAN]

Dear Felix_Sp

Respect thank you for your response. I think this is good information for me.

According to your information, to improve my model I will change something:

1. Increase the air phase domain.
2. Change the boundary condition for two sides from "symmetric" to become "wall"

I also want to consult you on two issues.
1. Mesh size for the solid object,
a. How many mesh elements (at least) for a solid object to get high accuracy?
b. Does it always create meshes that coincide with the edges of the object?

2. With my ctrl.txt file, what are the parameters I need to change to get good results?

All the best

Nhan

[Felix_Sp]

Hey

To 1. Yeah, I guess that would not hurt. I actually do not know how well REEF3D performs as a one-phase (two-phase I guess with sediment) model.

To 2. If you make the boundary a wall, you will have the same problem, as the cross-section is again bounded by walls (in the end it's just the continuity equation )

Seeing whether your mesh is good is I guess your job to find out (maybe have a look at grid convergence studies: https://www.grc.nasa.gov/www/wind/va.../spatconv.html)

In regard to your question whether meshes are always coinciding with the boundary edges, maybe have a look at this thread: Mesh not adapted to the solid (Boundary layers)
Short answer: No, as REEF3D is not using a body fitted mesh, but the immersed boundary method.

And I obviously cannot say how you can get good results. I guess if it is physical, it is good My experience with sediment transport modelling is yet limited. Maybe somebody else has a better answer. Sorry

Cheers and good luck

[NTNHAN]

Dear Felix_Sp

Thank you once again.

Sometimes, my problem is very big for me, but with you or other people, it is easy. I will try my best to improve my model.

Looking forward to some comments from other guys.

Best regards.

Nhan