CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   reverse flow problem (http://www.cfd-online.com/Forums/fluent/28843-reverse-flow-problem.html)

cisong August 20, 2001 22:36

reverse flow problem
 
there is a reverse flow when it is iterating , can you give me some suggestion what should i do , i have reduced the under relaxation value but it seem doesn't work.

bete August 21, 2001 01:41

Re: reverse flow problem
 
I think you to need learn alot about fluid mechanics, in certain cases of courses there are reverse flow, depend on geometry that you generated. Arek goblok & congok gak oleh mlebu forum iki, congok, gatel, taek koen .......

Armin Gips August 21, 2001 01:57

Re: reverse flow problem
 
Reverse flow only appear during the early convergence stages of your calculation, or later if something with your geometry is wrong

Yours Armin

John C. Chien August 21, 2001 02:24

Re: reverse flow problem
 
(1). It is hard to know what is going to happen when using iterative method with some arbitrary initial flow field guess. (2). No one will be able to know how to prevent certain flow behavior to happen. (3). On the other hand, if you start with a well defined initial flow field and run a transient flow calculation, then each time step solution will be acceptable because it is following the physics of the flow. (4). So, the suggestion is: start with a well-defined initial flow field distribution first, and run transient calculation. (5). It is probably all right to run the inerative steady-state calculation with a well-defined initial flow field. But then, you don't know whether you will run into flow recirculation at the exit (or inlet). (6). This issue has been discussed before. So, the first thing to do is to move the inlet and the exit location further away from the point of interest in your calculation. This is one simple way to avoid the reverse flow at the exit when the solution converges.

Raza Mirza August 22, 2001 20:01

Re: reverse flow problem
 
Is reverse at the exit ALWAYS bad? I have cases where i see some reversed flow even after 4 orders drop in residuals and after the flow from outlets have stabilized (i have multiple pressure outlets). It does not seem to bother me though as the faces which have reversed flow have very little (negligible) flux. Hence I feel that the effect of boudary conditions will not be significant. Am I wrong in making this assumption?

John C. Chien August 22, 2001 20:50

Re: reverse flow problem
 
(1). If you are designing an engine inlet, and the calculated engine inlet face (the exit of the duct)has reverse flow, then you could stall the engine in actual flight condition.

D.Tandra August 23, 2001 11:53

To BETE, WATCH YOUR LANGUAGE !!!!!!!
 
Bete:

This forum was created for CFD engineer to share our problem, help each other, and exchange information.

I think this forum certainly welcome a beginner in CFD, but definitely prohibit "UNEDUCATED MAN" like you to join the forum.

I can not belive that you can use such a RUDE INDONESIAN languange to reply cisong. You are such a coward ! Why didn't you say it in english.

Let me translate it for you: "Arek goblok & congok gak oleh mlebu forum iki, congok, gatel, taek koen ....... "

"YOU STUPID IDIOT PERSON CAN NOT JOIN THIS FORUM, MORON, BOZO, YOU ARE LIKE A SHIT ........"

-----------------------------------------------------

Raza Mirza August 24, 2001 23:33

Re: reverse flow problem
 
John,

Fortunately I am not designing an engine inlet, rather an automotive HVAV air distribution system!

david September 7, 2001 22:02

Re: reverse flow problem
 
Mr. Cisong,

I think there are two cases which may appear reversal flow. Most encountered one is due to not physically defined initial condition or non-proper defined computational domain size. The other cases are due to physically in your flow process there indeed has reversal flow. for example, if you want to investigate Richardson Flow, one of simple case is flow osciallting in a circular channel, there indeed may appear reversal flow.

Suggest you check whether physically your flow process should have reversal flow.

Good Luck!

David


All times are GMT -4. The time now is 18:21.