I have flow through a pipe with a reynolds number(calculated analytically) of around 800. So I simulate the flow using a laminar model. I get a residual value between 1E-3 and 1E-4.Also I find the average scale reynolds number in the out file above 3000. The CFX_mod document recommends a turbulence model if the average scale reynolds number is above 1000. so i use a k epsilon turbulence model and get better convergence 1E-5. So my question is how do I make a decision on using a laminar model or a turbulence model... is it based on 1) a reynolds number calculated analytically 2) the average scale reynolds number in the CFX out file(I find from previous posts that the average scale reynolds number is based on the average length sacle which is the cube root of mesh volume and may therefore be not always relevant) 3) convergence results

regards ranjith

Hi,

The choice of turbulence model depends on what physically happens in the pipe. If the pipe is laminar then use the laminar model, otherwise turbulent.

A pipe at Re=800 will normally be laminar. Therefore unless something exists which means a flow which would normally be laminar is in fact not (such as highly turbulent upstream conditions) then use the laminar model. Ignore the out file comment - some previous posts on this issue have discussed this issue in more detail.

Glenn Horrocks