|
[Sponsors] |
![]() |
![]() |
#1 |
New Member
Join Date: Jan 2022
Posts: 2
Rep Power: 0 ![]() |
Hello,
I am pretty new to CFX. I am simulating water passing by some symmetric geometry through a restriction in 3D. Asymmetric simulation can not be used here for further step As shown in the picture, the turbulences and the flow shape after the restriction is not symmetric. I don't get to understand if my results are normal. Is it due to my turbulence model (SST)? Does CFX consider the non-linearities? Or, are there other points I am missing out? I have tried both Hexa and tetra meshes but both end up with (more or less) the same results. Either with the model - tried both with half of it or the full 3D model. Regarding the convergence criteria, I choose the flow either tend to go right or left. Would an unsteady simulation solve it? I am pretty lost. If these nonlinear results are normal can someone explain to me why? Appreciate your time! VR |
|
![]() |
![]() |
![]() |
![]() |
#2 |
Super Moderator
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,816
Rep Power: 144 ![]() ![]() ![]() ![]() |
When fluids flow over a symmetric geometry they do not always result in symmetric flows. That is part of the beauty of fluid mechanics. A famous example is the von Karman vortex street, where flow over a cylinder generates a periodic vortex pattern. (https://en.wikipedia.org/wiki/K%C3%A..._vortex_street)
So the effect you are seeing is probably real. The jet will flip to one side, and possible flap about. This means you should do a transient simulation to see which it is.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum. |
|
![]() |
![]() |
![]() |
![]() |
#3 |
Senior Member
Erik
Join Date: Feb 2011
Location: Earth (Land portion)
Posts: 1,173
Rep Power: 23 ![]() |
What you are seeing is very well known, and called the Coanda effect.
It is caused by the Bernoulli principal (higher velocity = lower pressure). Pressure is lower on the wall where the jet goes since it is traveling faster than the rest of the fluid. This low pressure wall pulls the jet to that side and holds it there. If your pipe were oval, the jet would probably lock to one side, but in a circular pipe, it may wander around the circumference. |
|
![]() |
![]() |
![]() |
![]() |
#4 |
Senior Member
Gert-Jan
Join Date: Oct 2012
Location: Europe
Posts: 1,885
Rep Power: 27 ![]() |
What question do you want to answer? Do you want to know the flow rate as function of pressure drop? Then your results will be reasonably fine.
But it you want to undertand the flow downstream the nozzle, then this is certainly not a problem to solve using symmetry. Your jet will travel around in all directions. You will need a full transient 3D model to solve this properly. In this case, you should look in literature for help. Depending on the expansion ratio, the jet in the chamber downstream will behave very differently. You can find this in e.g. Guo, Langrish & Fletcher - Numerical simulation of unsteady turbulent flow in axisymmetric sudden expansions, 574(123), Transactions of the ASME, 2001 |
|
![]() |
![]() |
![]() |
Tags |
asymmetric, cfx, restriction, symmetric, water |
|
|
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Duplicating/pattern results for same geometry but multiple separate instances | gopisandy | FLUENT | 0 | June 17, 2019 11:41 |
how cfd results can be trusted in a very complex geometry | garrison | Main CFD Forum | 1 | November 8, 2014 04:13 |
Comparing Air numerical results with water experimental results | cristian2009es | Main CFD Forum | 1 | July 2, 2014 18:11 |
Problem Importing Geometry ProE to CFX | fatb0y | CFX | 3 | January 14, 2012 19:42 |
[DesignModeler] Error while Loading a Geometry | JD944 | ANSYS Meshing & Geometry | 0 | October 25, 2011 14:51 |