Free surface analysis boundary conditions between liquid surfaces

[Anonymous]

Hey!

I have a question regarding free flow simulations. I am trying to model the water flow through a water dam. For this I have drawn the dam in Inventor and also added parts for water in an assembly and imported them into Autodesk CFD.

From what I have understood there should be parts for both the initial height of the water (the bottom block) and for the height the water can reach to while for example passing through the dam. Both these "parts" are seen below in the picture, they are both chosen to have water as material. The dam is chosen concrete, and is suppressed.

My question is about the boundary conditions(please see second pic). Especially the one between the two blocks. If I choose P=0 in between the two, I get crazy high levels of water. If I choose nothing the analysis stops due to errors. I believe the answer will be obvious to many but I have never done free surface before so would highly appreciate input on this matter.. What is the right way to go with boundary conditions in this case? 

 

[marwan_azzam]

Hello Edonder,

Without seeing the CFD model here I suggest the following:

1- Do not apply any internal boundary conditions (i.e. at the interface between the fluid bodies).

2- I would do a search on YouTube for videos on Free Surface in Autodesk CFD.  There are at least a couple on there that you would find very helpful.

If after doing the above you are still having problems with the model, please post the support share file (*.cfz) on the forum and we can look it over.

Regards,

Marwan

[Anonymous]

Hello Marwan,

Thanks for your reply. I had watched the videos butwas still a bit confused about this as I thought having no BC between two walls represented a wall. But I did a simpler simulation now and got results. Although they seem to be very unrealistic. I am attaching the files for it here. It is the case "Design - Copy (3)". 

It is a model where I want to simulate a straight canal with water flowing from one side to the other. There is another inlet from one side where water flows in at a faster rate. I want to then see if this flow coming in from the side would cause a water rise on the side across, where there is supposedly an establishment.

I autosized the mesh, decreased the size of the mesh overall and made it also finer around the inlets and around the surface of the water. I made 5 wall layers with a growth rate 1,20 (something I had seen in other free surface videos, do tell me please if it was an unnecessary step here). What happens is at iteration 1200 (t=4) there occurs a back-wave from the boundary where p=0, and the water just returns and piles up on the inlet side. I am obviously doing something wrong but can't really figure out what? 

Thanks for your assistance.

Deniz

[Anonymous]

The attachments don't seem to follow, sending them here again!

[Anonymous]

I believe you will need the support file (.cfz) but I can't seem to attach it? Only the cfdst file seems to attach, does that work?

[Anonymous]

Hello again,

I unfortunately can't get this to work after following the steps and also following various advices from others. The water just ends up following unrealistic paths and in certain areas I get velocities up to 400m/s after 9000 iterations. Is it the mesh? Solver? Boundary conditions? Geometry?

Attaching still doesn't work so I added a copy on my drive. Here is the link;

https://drive.google.com/file/d/1PPPGYvQJDjNSzm8Z5Rh6sKg2WiEgzQD1/view?usp=sharing

Hope you can help!

//Deniz