CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   characteristic Length (http://www.cfd-online.com/Forums/main/5057-characteristic-length.html)

CFD MAN August 19, 2002 21:43

characteristic Length
 
Hallo,

For example, if I want to calculate Renould number on a plane (contour) which is velocity*density*charactersticLength/viscosity

What value of Characteristic Length should be used?

When we want to determine which force is dominant we need to know dimensionless numbers and many of them are dependent on Characteristic Length. But this Characteristic Length is a random number (I know sometime we use D, L of the domain), then how we can make sure which force is dominant??

CFD Man

versi August 19, 2002 23:14

Re: characteristic Length
 
A characteristic lenghth is a qualitative feature. Sometimes is has some freedom, say R ~ D ~L, is qeuivalent, but some times it is very local, say BL momentum thickness , Taylor scale, y+, and so on. In Latter cases, oen always want to obtain some scaling laws. A quality numerical result should not depend on these selections unless one definition of L is on macroscopic, the other is on microscopic.

CFD MAN August 19, 2002 23:36

Re: characteristic Length
 
Thanks Versi If you calculate Re, based on domain Length and get a value of the order 1000, and if you calculate Re based on local mesh size and get 0.001

Now my question is, is this flow inertia dominant or viscous dominant?

Re=inertia force/viscous force ?

I have many confusions, I have just started with the easiest one :)

CFD MAN

puppy September 11, 2002 07:26

Re: characteristic Length
 
The truth is that Re in itself means nothing, so you cant say if its inertia or viscous dominant, or in other words, laminar or viscous. As far as I know, only for straight pipes one can say that, with D taken as characteristic lenght, Re>2500-5000 means turbulent. For other geometries, experimental results say more then just calculating Re on whatever base. Re is only usefull when comparing different test cases with the same geometry. Good luck!


All times are GMT -4. The time now is 11:55.