[omidiut]

Hi all
I am solving a turbulence flow around a bluff body, using DES model.
How can i calculate strouhal number for it?

Best regards

[ojha.mayank485]

Find the values of coefficient of lift for the body at different time values and do a FFT (fast fourier transform). Excel or even Matlab have built in functions to perform FFT. That will give you the frequency of vortex shedding (f).

For a cylinder St=f*D/U, for any bluff body I do not have any idea. You may look up literature.

Hope that helps.

-Mayank

[omidiut]

Thanks ojha.mayank485
can time step size effect the frequency?

[ghorrocks]

Yes. Time step size is one of many parameters you will need to check to get an accurate simulation http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

[ojha.mayank485]

Your timestep size should be smal enough to predict the flow properties accurately. For example if you monitoring the vortex shedding, then use time step atleast 1 to 2 % of vortex shedding period.

BTW what bluff body are you using ???

[pssk1988]

Originally Posted by ojha.mayank485

Your timestep size should be smal enough to predict the flow properties accurately. For example if you monitoring the vortex shedding, then use time step atleast 1 to 2 % of vortex shedding period.

BTW what bluff body are you using ???

I would like to add clarity. In general, the vortex shedding period is very large compared to the simulation timestep (considering time-explicit schemes).But here's the actual constraint: The timestep must be smaller than the vortex shedding period, mainly to avoid Aliasing effect. Read: https://en.wikipedia.org/wiki/Aliasing

To avoid aliasing, the sampling frequency must be greater than f/2, where f is the measured signal frequency. Therefore, maintaining atleast 3 measuring points per shedding period captures frequency information. However, more the data points better the shedding frequency captured.

[JoaoRoque]

Hi
I'm not able to calculate the Strouhal number using the FFT on excel.
I'm following the instructions on the following file, being the data, the CL values, but I obtain a shedding frequency of 849 Hz for a Re of 40 (d=1.5 mm, U=0,4m/s, air kinematic viscosity), which is absurd.

Can someone pls tell me what am I doing wrong?
Thanks in advance!

[ghorrocks]

How can we tell you what you are doing wrong when you don't show us your simulation or the excel file?

So ask your self these questions:
Is the simulation accurate? This FAQ can help: https://www.cfd-online.com/Wiki/Ansy..._inaccurate.3F

Are you calculating the FFT correctly? You should not use excel for any serious numerical work. You will go a lot further with a package designed for serious work, like matlab, python or something like that. But to check if your FFT is working then look at the most obvious frequency in your signal and work out the approximate frequency by hand and compare to FFT.

[JoaoRoque]

I can't show you the simulation files, because I have not been able to charge the PC which I ran it with. But here are the excel files and some information about the simulations:

Inlet is 5d away from the cylinder center, outlet is 25d, being d the cylinder diameter
top and bottom are 10d away from the cylinder center
73800 cells, I checked the mesh and it seemed good in terms of structure
i ran pimpleFoam in piso mode, with turbulence off. I ran pimpleFoam because of the fact that it has the ability to adjust the time step to guarantee the Courant number doesn't get higher than specified (1, in my case)

boudary types are patch on inlet, outlet, top and bottom, noSlip on the cylinder.

Excel files
https://drive.google.com/file/d/1g6y...ew?usp=sharing
https://drive.google.com/file/d/1b6a...ew?usp=sharing

[karachun]

Originally Posted by JoaoRoque

i ran pimpleFoam in piso mode, with turbulence off. I ran pimpleFoam

This is Ansys CFX forum, not OpenFoam forum. Please post OpenFoam questions here.
https://www.cfd-online.com/Forums/openfoam/