Regarding problem in Setting up the supersonic jet impingement case in rhoCentralFoam
 

Hello everyone,
I am trying to set up a case on the problem of supersonic jet Impingement. I am using rhoCentralFoam solver and kOmega SST model for this case study. My geometry consists of an inlet to the nozzle and after it exits the nozzle, it has four outlets ( 2 at the sides and 2 at the top on the left and right side of the nozzle exit) and a bottom plate on which the jet should impinge. I am using a total pressure BC at the inlet and wave transmissive BC at outlets and zero Gradient at the bottom wall for Pressure.
Actually at first I am trying to validate by trying to obtain the Cp curve as in this paper https://www.researchgate.net/publica...Impinging_Jets by using all the same parameters
The main problem is that after my jet exits the nozzle, rather than coming down and impinging on the ground majority of it gets dispersed sideways, and only a small portion of the jet is hitting the ground. What might be the reason for it?
I have also tried both zero gradients and fixed value BC for the outlets, but the results are unsatisfactory.
So please, if anyone could help and tell me where am I going wrong.
As a result, I am attaching the case I set up and the pressure contour obtained after the simulation through this link https://drive.google.com/drive/folde...Cb?usp=sharing.

Hi,


I think the problem is the following:


The top and left patch next to the outlet of the nozzle are set to waveTransmissive. But they share a boundary with the nozzle, which is
a wall since you have defined it as noSlip and also you have defined those patches left and right of the nozzle as noSlip, if I see that correctly. Those boundary conditions cant coexsist. My solution would be the following:


1. Decide whether your boundary conditions left and right of the patch should be a physical wall or a free enviroment.


2. If its a wall: Use noSlip for U, zeroGradient for p and T


3. If its a free enviroment: You should rebuild your mesh. Let the nozzle stick out a little bit out of the left and right patches, so that there are no big gradients at the boundary of wall and waveTransmissive BC.

Hi
It is environment everywhere, i.e., atmospheric conditions except at the walls. Here in my geometry, the only boundary which acts as walls are the nozzle walls, and the bottom plate rest all are outlets into the atmosphere.
I don't know if I got you correctly, but you asked me to put the nozzle inside the domain and move the top right and left patches beside the nozzle to somewhat upside. So I tried by placing them beside the nozzle inlet and remaining all case setup with BC's where same as I had shown earlier but still I got the same thing as you can see in this pressure contour and velocity contour here https://drive.google.com/drive/folde...56?usp=sharing the wave is getting separated and only a small part of it is impinging on the bottom wall.

For me it looks like what you would expect. The flow will be seperated since the bottom is a wall and the jet impigning the wall.


But your domain might be too small. Increase the domain size to the left and right and see if your boundary condition might be a source of error.

Btw you should use waveTransmissive for U too.

Should I also use waveTranmissive for T also?

I think there is no need to do that according to how the BC works. But actually I have forgotten that. The advection velocity and pressure need to calculate in a manner, that the wave reflection gets reduced. I think accoustic waves are still reflected.


Long story short: I would set zeroGraident for T and waveTransmissive for p and U