CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Symmetry Boundary Condition (http://www.cfd-online.com/Forums/main/10314-symmetry-boundary-condition.html)

Noname November 17, 2005 17:57

Symmetry Boundary Condition
 
when we model flow past a cylinder how do we mathematically express the symmetry boundary condition when we want to do calculations only for the half domain.


boling November 17, 2005 18:43

Re: Symmetry Boundary Condition
 
keep all the physical variables same except reverse the normal velocity component (normal to the symmetry plan)

Mani November 17, 2005 21:57

Re: Symmetry Boundary Condition *NM*
 

ramp November 18, 2005 14:27

Re: Symmetry Boundary Condition
 
The flow regimes for the flow across a circular cylinder are: (a) steady flow without separation (b) steady flow with symmetric vortices behind the cylinder, (c) laminar vortex shedding flow (d) 3-D unsteady flow, etc

Upto to the regime (b), the computations can be carried out using half domain because the flow is symmetric in the upper and lower half of the domain. In case of Newtonian flow, the occurance of laminar vortex shedding is approx @ Re=46 (few have reported 49) and so the half domain can be used for Re < 46 ......


vut February 18, 2014 03:54

symmetry boundary condition
 
Hi all,

I have the following geometry: 3 cylinders (2 concentric and the remaining excentric). So, the geometry has a symmetry plane.

Have I define each plane of each cylinder as symmetry?

Thanks in advance,

vut

lovecraft22 February 18, 2014 16:55

Not sure I understand your question as I did not understand what you are trying to simulate.

Anyways, as ramp already said, the fact that a cylinder has a symmetric shape doesn't mean that the resulting flow will also be symmetric. In most of the cases it will not actually.

vut February 19, 2014 05:43

Geometry of the injector
 
1 Attachment(s)
You can see in the attachment the geometry of the injector that I want to simulate.

In fact, I hope to be able to consider a half of the geometry. Thus, I define three surfaces as geometry planes. The simulation is converged. However, the mass flow rate I obtained is quite small comparing to that measured.

Let's assume that I define a pressure inlet and a pressure outlet.

My first question is:
do you have an idea to mesh this geometry?

And the second one is:
what are possibly the reason of low mass flow rate in this case?

Thanks in advance,

lovecraft22 February 19, 2014 14:11

I guess it is ok to use a symmetry plane in this case.

What do you mean with "how to mesh this geometry?"

Can you set the inlet condition with a specified massflow instead of pressure?

vut February 20, 2014 17:48

Mass flow rate not matched
 
Hi,

I mesh each cylinder sepately. Then I assembly them by tmerge. Everything seems OK.

The problem I have is the following:

I set the inlet as pressure inlet and the outlet as pressure outlet.

However, the mass flow rate given by Fluent does not match the experimental one!!!

Some body has a great idea to solve this stuff?

Thanks in advance,


All times are GMT -4. The time now is 19:02.