hello

i'm using FLUENT to find Cd values for 3D rectangular body (parallelopiped)....i have got some results from fluent...i kept the frontal area same and increased the Length of the cylinder(length parallel to flow)....

frontal area : 3x3

length : 0.5m, 1, 2,3, 6,12, 18, 24

Cd values are first decreasing as the length increases and after 3m length, the values are increasing....i did for a range of Re....my fluid is water....

and i'm not able to give a logical explanation for the increase in Cd values as the length is increased..... please help me in this regard...if u know any journal,please give me the title.... any kind of help is highly appreciated... Thanks a lot in advance...

regards

karthik.S

It's a little difficult to say offhand, especially without knowing what the Re range is (giving that would help). Here are a few things to look at: What is the boundary layer on the body doing as length increases, and what is happening to the pressure behind (and in front) of your body as the length increases? The force on the body should be made up of shear forces + pressure forces, so that is how I would start.

jason

hi jason

i'm modelling for turbulent flow...Re from 10^5 and above.....let me check the pressure behind the body... thanks for ur help and others if u have any idea..pls do post ur replies... thanks regards..

karthik

The dependence of the Cd on the length may indicate three-dimensional effects. Take a look at the solution to see if there is any variation in spanwise direction. There should be... What boundary conditions are you using on the side?

Another important question: What do you exactly mean by Cd? Are you running steady state, or are you running unsteady and taking the time-averaged Cd? You should do the latter, because at such high Reynolds number the flow is surely unsteady, and the steady state equations will not give you the correct drag.Not only is the flow unsteady, but you should also expect significant spanwise variations at high Re. Depending on the cylinder length, you may get different modes of vortex shedding, which may result in different time-averaged drag coefficients.