CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   very strange flow behavoir...pls help (http://www.cfd-online.com/Forums/cfx/21080-very-strange-flow-behavoir-pls-help.html)

caty March 14, 2005 20:43

very strange flow behavoir...pls help
 
Dear CFX users,

I have a problem with the CFX. It seems like he doesn't want to yield the results I want.

The model is a following:

I have to simulate the fire in an open environment(100*100). Wind is assumed to blow from west to east. The inlet wind velocity is 3 m/s. The east part is considered to be the opening. The upper part is treated as an opening and the bottom part is considered to be slip wall. The other two walls (front and back wall) are considered to be slip walls. Buoyancy model is activated.

I contact the support as I can't see what i am doing wrong but the results doesn/t look better with their settings.

The problems:

1. I set the outlet pressure to static averaged pressure = 0

I changed the upper boundary plane to opening as suggested by CFX support. The condition was static pressure =0, flow perpendicular to the boundary. Is anyway this a correct enough condition as I have a wind blowing from inlet to the outletů.so no perpendicular flow components should exists at an appreciable distance of the fire. From the beginning the solver announced about the reversed flow on the outlet and the visual analyze of the solution shows a strange flow behavior. The flow is leaving the domain only on the lower part of the inlet. In the upper part a flow vertical (perpendicular on the wind direction can be seen).

2. The other possible option, static pressure and no direction seems to not work as always when I set it I have an error seeing that the solver need to be closed by windows.

2. The variable outlet pressure conditions have been also tried. This was intended due to the high model dimensions (100 m) and a pressure correction (relative pressure) have been applied dp= ro*g*h as it can be a factor with influence. Using this equation also did not solve the problem. Also it was noticed that in the velocity inlet area is a different relative pressure calculated by the solver.(cfx support says: "i doubt that you need this condition")

It was intended to apply also a velocity condition at the outlet (forcing to have the same velocity at the inlet and outlet) but big instabilities are noticed.

So, what I am doing wrong and which are the corect boudary conditions?

Any suggestion is deeply appreciated,

caty


Glenn Horrocks March 15, 2005 17:16

Re: very strange flow behavoir...pls help
 
Hi,

I think you mentioned that your top boundary is an opening/outlet. I would make it a slip wall. That is a more reliable top boundary I find.

Glenn Horrocks


All times are GMT -4. The time now is 08:45.