Hi

I am simulating the flow of air through an orifice meter in a circular pipe for my 4th year CFD project.

Diameter of pipe (D) = 0.1m, Re = 1.1exp05 P = 2bar @ 20 degrees C (all at inlet). Turbulence intensity at inlet is 5% and a requirement is that the total number of nodes should be no more that 10000. Turbulence length scale at inlet is 0.5D. The diameter of the orifice is 0.4m.

Does anyone know what I should take for the usptream and downstream lengths for my model and why?

I have read Fluid mechanics by F.M. white and I think this says that a length of 50D is required for fully developed flow at the orifice. Obviously this would produce a massive model and take about 5 years to solve on my poor old uni pc.

For this type of problem, I made successful calculations using fully-developed inlet conditions 10 diameters upstream of the plate, and an outlet plane located about 15 to 20 diameters downstream of the plate. This type of CFD problem has been studied in detail by Asbjorn Erdal (PhD thesis) at STATOIL, and earlier by Michael Reader-Harris at NEL, among several others.