CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Strouhal number ? (

Tony Chen February 2, 1999 23:27

Strouhal number ?
Will somebody help me that what is the most simple way to calculate the vortex shedding frequency or the Strouhal number for flow past a vertical circular cylinder (3D) ? Is this value as same as the frequency of the fluctuating lift force ? Thanks in advance.

laliong February 3, 1999 05:46

Re: Strouhal number ?
For 3D cases, the model of vortex shedding may be oblique or parallel. Which is in your problem. Basicaly, I think you'd better to identify which part of the wake is involved to calculate the frequency.

I think the most simple way to estimate the frequecy is to check the velocity field with a specified initial velocity field succesively.

Tony Chen February 3, 1999 08:42

Re: Strouhal number ?
Thanks for your help. The Re=10,000 in my calculation, and the experimental shedding frequency St=0.206. To check the velocity field as you mentioned, are there some particularly representative positions for sampling ? Those points should be located in the vertical central plane, but how far from the origin ? Thanks again.

prasad Patnaik February 3, 1999 15:03

Re: Strouhal number ?
If the representative postions being sampled are with in the wake region, it should give a reasonably good measure of the temporal velocity field and there by the 'vortex shedding frequency'. May be, you can choose, distances of 0.1D, 1D, 2D, 5D and 10D, 20D in the downstream region along the central plane, for the purpose of monitoring the velocities. The time period of vortex shedding deduced from all of them should turnout to be 'almost' the same (To our pleasant surprise!). Even the steady periodic lift coefficient fluctuations can be used to cross check the same.

Debatin February 5, 1999 12:58

Re: Strouhal number ?
I think the simplest way to detect the frequency of the vortex shedding is to count the vortex shedding on its shedding location area (as the physics told us). This means: record the pressure signal - value on one side of your cylinder over a few periodes and then make a fourier transformation and your problem is solved.

Adrin Gharakhani February 5, 1999 19:35

Re: Strouhal number ?
"vortex shedding" frequency can best be accounted for by looking at the "vorticity" field. After all, you are trying to measure how often "vorticity" is "shed" off the cylinder surface!

The velocity field and/or streamlines, generally speaking, will give you nice pictures but won't tell you much about the flow physics. On the other hand, once you get a hang of vorticity (i.e., once it becomes as simple an idea as velocity) it reveals tremendous information regarding 3-D effects, separation, recirculation, etc.

In fact, if you get the vorticity contours (at some 2-D cut) you will be able to visually determine the vortex shedding frequency as a first cut. You will most probably be within 10 percent of your FFT analysis of the signals!

The vorticity field will show you the locations of primary, secondary, etc vortices that are shed off the cylinder (if your CFD method can actually capture them) and by visually checking their locations, you can decide where you should measure the signals for your analysis.

As was mentioned earlier, the exact location of your "probe" will not be as important, but since you have a 3D flow I would be concerned more about 3D effects. In 3D, vorticity is a vector, so which component should you check for comparison with the experiment? Usually there will be a dominant direction. The best answer, I think, is: check how and where the measurements were made and try to match that.

Adrin Gharakhani

Tony Chen February 5, 1999 23:16

Re: Strouhal number ?
Since the lift force is the dynamic effects of shed vortices, how is the fluctuation of lift coefficient related to the vortex alternating process ? are they share the same frequency in the sense of spatial integral or temporal average ?



Debatin February 6, 1999 04:42

Re: Strouhal number ?
In every signal you record in your field , local values as well as integral scales, your vortex shedding is embedded. The only difference is the contamination from other periodic effects as cross-waves vortex pairing - splitting and so on so that your powerspectra shows supharmonics of your vortex shedding frequency. Relatet to your question: yes the liftcoefficient exhibits among others the vortex shedding frequency. The strongest signal you will get is by using pressure information as drugcoefficient or pressure probes on the wall, thats our experience also in experiments.

John Lim March 3, 1999 22:03

Valve flow study with FLUENT
Hi. I am a mechanical enginering undergraduate senior at Western Michigan University currently undertaking a CFD-related Senior Design Project. My project partner and I am currently running FLUENT analyses on the flow characteristics through a Vee-ball valve. The fluid through the valve and piping system is water. For certain valve openings, we were able to obtain the velocity and pressure contours of the overall outflow. However, we are interested in finding out the velocity, pressure, and related properties of a single point in the downstream flow (behind the valve) in an unsteady flow analysis. We would appreciate it if you can furnish us with step-by-step information with regard to the above mentioned. We are currently running FIDAP7. We hope to be able to write the velocity and pressure data of that particular single point to a file. We also like to plot the velocity and pressure of that point against the time step.

The objective of this project is also to relate the Strouhal frequency and the Reynolds number of the flow. Is there a possibility to obtain frequency-related information of the flow, apart from the turbulence intensity? Due to the vortices and cavitation, the flow is generating vibration and pressure pulses which are the cause of the pipe ruptures in some paper and pulp plants.

We welcome suggestions and guidance from you. Thank you for your time and consideration.

Regards, John Lim; Department of Mechanical and Aeronautical Engineering, Western Michigan University, Kalamazoo, MI

Eric Grald March 5, 1999 14:20

Re: Valve flow study with FLUENT
Dear John,

You can generate the plots you desire (velocity or pressure versus time at a single point) using the HISTORY command in FIPOST. The steps to use this command are completely described in the FIPOST User's Manual. The HISTORY command will also write the data (say, pressure vs. time step) to the ident.FIOUT file.

You asked about obtaining other frequency-related information about the flow. You can also compute the fast Fourier transform of any quantity and plot the power spectrum versus frequency by using the FFT command in FIPOST. Again, please consult the Users Manual for details.


Eric Grald, Fluent Inc.

pedroxramos February 1, 2013 14:05

Hello there!

I don t find FIPOST in the internet... can you send m a link, please? It is possible export data from the fluent to Fipost? I want to do a plot of velocity vs time and a contour of velocity in time... Regards.

All times are GMT -4. The time now is 08:03.