CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Reynold's number and incompressible flow (

Neil October 23, 2006 09:58

Reynold's number and incompressible flow
Dear all,

I am a beginner at modelling incompressible flow conditions. I have read some papers which model incompressible internal flow and they mention that they are restricted to low Reynold's number for stability reasons. Does solving high Re number flows present a particular difficulty when solving the incompressible flow equations? How are stability problems resolved? by using a finer mesh?



Mani October 23, 2006 16:25

Re: Reynold's number and incompressible flow
I guess the answer to your question depends on clarification of some points you brush over:

a) Who are "they", in your sentence "they are restricted to low Reynold's number"? The flows, the algorithms, ... (if I read your sentence literally, the papers mention that the papers are restricted... which would be very bad behavior for a paper)

b) Related to question a), what type of stability are we talking about: Flow stability, numerical stability...

c) What Reynolds number are we talking about, i.e. how is it defined. Is it the cell Reynolds number (Peclet number)?

A clear question has the best chance to prompt a clear answer. Lucky for you, this forum is not an oracle which gives you only one shot. :) Try again.

Neil October 24, 2006 02:56

Re: Reynold's number and incompressible flow
Sorry for not making my question clearer

a) It is the algorithm that is restricted to low Reynold's numbers

b) numerical instability

c) Reynold's number is defined based on the diameter of the flow channel

Thanks for your help


Mani October 24, 2006 08:11

Re: Reynold's number and incompressible flow
What type of internal flow is it? Pipes, pumps... How low a Reynolds number is required? The only thing I can imagine right now is that you have an algorithm for laminar flow. If that's the case, the numerical instability observed at higher Reynolds numbers is actually related to the physical instability of the laminar flow. If the algorithm does handle turbulent flows, there may be some limit associated with the turbulence model. Other than that, a Reynolds number restriction would seem odd to me.

To answer your question about grid resolution, yes, there is a dependence on the Reynolds number. The higher the Reynolds number, the thinner the boundary layers, the higher the requirement on the grid resolution. This is primarily an accuracy requirement. If coarse grids also cause a stability problem will depend on the algorithm.

All times are GMT -4. The time now is 17:03.