CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   convergence problem in a long pipe (https://www.cfd-online.com/Forums/cfx/72410-convergence-problem-long-pipe.html)

 feizaghaee February 4, 2010 18:04

convergence problem in a long pipe

1 Attachment(s)
hi freinds
i'm trying to simulate a long pipe (180 m length) with 1.5 m diameter that has a 90 degree curve in one third of length from begining.inlet boundry conditions are velocity of .8 m/s and static pressure. outlet boundry condition is velocity of .8 m/s normal to the boundry. after about 2000, residules don't change and they approximately became a smooth line parallel to each other. i use hexa mesh that each cell along the pipe has 1 meter length. i use multiphase option. does anyone know what can i do to obtain convergence.

 ckleanth February 4, 2010 18:57

do you need to model the problem using cfd? you didnt say what you want to model anyway.

 feizaghaee February 4, 2010 19:37

i want to model the water and air flow.

 feizaghaee February 5, 2010 11:49

i work with CFX 11.0.

 ckleanth February 5, 2010 12:22

I will reiterate; you did not say what you want to model, no information on the physical or how you modeled it. if you want an answer please post an apropriate question.

 ghorrocks February 7, 2010 17:20

For general comments on convergence see here:
http://www.cfd-online.com/Wiki/Ansys...gence_criteria

 AliTr February 8, 2010 18:45

1m mesh length is large if you are modeling multiphase Air/water, you haven't told the other mesh dimensions but relatively they should be large as well. you need smaller mesh size for better resolved water surface, in most of the Air/Water models, this is the reason for flat line convergence.
you can also try smaller physical timescale while reducing the mesh size.

 feizaghaee February 16, 2010 09:05

i solved it by changing the time scale. tanks a lot everyone

 All times are GMT -4. The time now is 23:26.