CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   coefficient of drag(Cd) (

haris December 31, 2007 02:40

coefficient of drag(Cd)
dear friends,

i am doing my project in fluent(laminar flow over a circular cylinder ).i have a problem with the plot- coefficient of drag verses iteration. instead of getting an increasing value of drag coefficient with increasing no. of iteraiotn, i am getting a decreasing value of drag coefficient with increasing no. of iteraiton....can u please help me out...

thanks in advanced....and wish u all happy new year.

Sham December 31, 2007 08:41

Re: coefficient of drag(Cd)
Could you pls explain it clearer? Are u using unsteady solver?

haris January 1, 2008 06:40

Re: coefficient of drag(Cd)
hello, i am using segregated solver.its a 2D problem. Formulation- implicit. Velocity Formulation-absolute. Gradient option - cell based. Pressure velocity coupling- simple Discritization Method for pressure- standard.

for momentum-first order upwind. Boundary condition-1)cylinder as wall

2)inlet- velocityinlet

3)outlet-pressure outlet

4)right and left wall as symmetric.

haris January 1, 2008 23:40

Re: coefficient of drag(Cd)
yes i am using an unsteady solver.

Sham January 2, 2008 02:19

Re: coefficient of drag(Cd)
What do you mean by increasing value of drag coefficient with increasing no. of iteration?

If you use unsteady solver, you should be looking at Cd vs flow time not iteration (only for steady solver).

Usually once the solution converge, you Cd value will assume a regular fluctuating pattern. Time step size is important here.


haris January 2, 2008 06:07

Re: coefficient of drag(Cd)
i have done both steady and unsteady case of unsteady , cd decreases with flow time and attains a constant value very near to zero.The nature of curve is same for steady case but as you said cd decreases with no. of iteration.i was doing for Reynold No.100 and the time step used was 0.02

Sham January 2, 2008 06:28

Re: coefficient of drag(Cd)
Time step size looks fine. If i get your sentence right when you say cd decreases with flow time and attains a constant value very near to zero for unsteady solver, that should not be the case in unsteady solver. That will only happen in steady solver.

Do you get vortex shedding?

If no, it looks that you need to stimulate some perturbation in the system to create disturbance. This will allow instability and vortex will shed. Then I would expect you get a fluctuating Cd graph. However, the average value should not be zero at Re 100.


haris January 3, 2008 05:21

Re: coefficient of drag(Cd)
hello sham,

i got a good vortex shedding without using the help of disturbance.but i got a decresing value of cd with increasing flow time and the pattern of curve was "L" shape

haris January 11, 2008 00:25

Re: coefficient of drag(Cd)
hello sham,

i reduced the under relaxation factor for body force to zero and done the iterations(unsteady). i got an increasing cd value with flow time.i think ia am going the right way..

All times are GMT -4. The time now is 10:37.