CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Channel Flow Development Length (

SamR May 8, 2010 11:16

Channel Flow Development Length

I am writing a code in Matlab that computes the velocity profile between 2 stationary parallel plates. Set up: steady, incompressible, laminar, 2D flow. Reynolds number of 100, based on a velocity inlet of 0.02m/s and channel height 0.1m. Density and viscosity corresponding to that of air. I'm using the SIMPLE algorithm on a backward staggered grid, with upwind differencing.

However I am finding that my development length is much too short. Based on the Reynolds number, and relationship L/D=0.5+0.05*Re, L/D should be around 5 - whereas I am seeing L/D of closer to 1. The profile of the fully developed region is in close agreement of the analytical solution, it just occurs earlier than it should do.

Any ideas on what might be causing this?


JSoares March 18, 2011 15:08


First let see if understand your problem:
Air proprieties fluent: ρ=1.225 [kg/m3] and μ=1.7894e-5
Distance between plates = 0.1 [m] as 2h=0.1m
Dh (Hydraulic diameter) = 4h = 0.2 m
So v= 0.0073037 [m/s]
You want de Le:
Le/Dh ≈ 0.06*Re
So Le/Dh = 6 m
Residuals: 1e-6
Iterations: 2000 (for example) – solutions should converge before 2000 iterations
Import Solution:
1) Create a line x=0 and y between -0.05 and 0.05 m
2) File-export-solution data…
a. Fyle Type: ASCII
b. Surfaces: chose the line created
c. Quantities: Chose x-velocity
d. Save it
3) Open excel
4) File-open-choose the file saved
5) Don’t forget to set separation comma to read the results


All times are GMT -4. The time now is 13:56.