CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   -   turbulent velocity profile model ? (http://www.cfd-online.com/Forums/cd-adapco/55139-turbulent-velocity-profile-model.html)

 cfdq February 10, 2006 04:33

turbulent velocity profile model ?

I model a turbulent pipe flow which Re=5000 geometry:(vc3d 0 54.5 20 0 360 72 0 3862 500 using csys 2) at the inlet boundary W=-0.689025(for adjusting Re=5000) and turbulent intensity=0.05 turbulent length=0.001 Another setting is outlet boundary. also I give the geometry scalar factor=0.001(pipe radius =54.5 mm) so final result is as symbol "3" as following web site:

http://www.wretch.cc/album/show.php?...1139464550&p=1

but what I want is trublent velocity profile as following web site

http://www.wretch.cc/album/show.php?...1139464458&p=0

can anyone help me?

 Yacine February 13, 2006 06:23

Re: turbulent velocity profile model ?

use a longuer pipe in the streamwise direction or periodic boundary conditions and you will eventually end up with the profile you want after the solution is converged!!!!

 cfdq February 13, 2006 10:49

Re: turbulent velocity profile model ?

where to use periodic condition? In the theta direction?

 cfdq February 13, 2006 12:04

Re: turbulent velocity profile model ?

do you locate the wall boundary at the R(r)=Radius. or just treat it as defaults setting and needn't to set the wall boundary at the R(r)=Radius. thank you.

 Alex February 13, 2006 17:26

Re: turbulent velocity profile model ?

Not sure what you mean by that last post.

Periodic boundary = Cyclic in STAR-CD language. You need a cylinder with cyclic boundaries on each end, so what goes out one end, comes in the other (or a longer pipe). Also, keep reducing your convergence tolerance. The changes on a iteration by iteration basis will be very small, so the default tolerance may be too large.

 cfdq February 13, 2006 21:43

Re: turbulent velocity profile model ?

can it use cyclic condiction in the z direciton in csys 2 (cylinder coordinate).using cyclic condition must be periodic repeating phenomenon.but pipe flow velocity is changing along the pipe(z direction) .does it work ?

 cfdq February 13, 2006 22:04

Re: turbulent velocity profile model ?

you mean orginal Max Residual Tolerance=0.001 .change it to 0.0001 (smaller).?

 Anton Lyaskin February 14, 2006 06:08

Re: turbulent velocity profile model ?

Actually it should be done in a bit different way - see Methodology, p. 5-8. What you need is a sector mesh (with cyclic boundaries in theta direction), only one cell long along the axis. At "inlet" you should apply uniform pressure and at outlet you should apply mass flow rate.

 asmaa ahmed March 6, 2006 15:20

Re: turbulent velocity profile model ? *NM*

 All times are GMT -4. The time now is 17:07.