Vortex shedding frequency from cylinder

UOWmecheng · July 12, 2016, 10:12

Hey everyone, I'm trying to simulate VIV of a rigid cylinder in CFX. After running the file and exporting data I found that the lift coefficient fluctuates at a frequency of 1.39 Hz, however the shedding frequency as estimated by the Strouhal number is 0.97 Hz. In post-CFD I found that y+ max is ~6.4 and the max courant number is ~0.25. Does anyone have a suggestion to get the shedding frequency closer to that predicted by the Strouhal number? Any help is greatly appreciated!

markomilosevic · July 12, 2016, 18:14

Hi Matt,

what Re are you dealing with, and what numerical solution did you use

UOWmecheng · July 12, 2016, 21:39

Hi Marko,

The Re is 10 000 and I am using DES

Other specs: fluid is water with density = 1000 kg/m^-3 and dynamic viscosity = 1.16E-3, cylinder diameter = 0.05m, first layer mesh height = 1E-4m, velocity is steady in x direction with magnitude 0.232 m/s

urosgrivc · July 13, 2016, 02:30

I wonder if SAS SST with y+~1 would predict the frequency even better:
How did you model the elastic part of your simulation (rigid body springs?)
For VIV mass of the cilinder and springs will efect the solution frequency also, not just the fluid that is surounding your cilinder.
Cilinder must have the wright amount of inertia as it will be prone to acelleration and decellerations.
Are you doing a 2D or 3D?
Is your domain big enough?...

It is a nice case you have got there, I want to be informed of what you will come up with and what you will be able to achieve.

This was also done with cfx I think it is quite nice and hope you will get there.
It is just a fixed cilinder but the frequency looks to be spot on.
https://www.youtube.com/watch?v=mNQ9NRdrKCM

It would be nice to get a lot of data of this reserch than compare the results, my future plans;
https://www.youtube.com/watch?v=9FRTj6_1J2k

markomilosevic · July 13, 2016, 21:06

ok, not expert but trying to solve similar situations... I would observe frequency with realisable k e with first layer height between 30-300y+, and coefficients drag/lift with sst and first layer 1y+, and 10 cv in sub/buffer layer, and check correlation with literature, and than move to des and les... as things are going with me les is solution, inform me what results you got with des, how big is your domain and mesh quality

urosgrivc · July 14, 2016, 03:15

Are you doing:
A = a VIV simulation with oscilating cilinder
or
B = just wortex sheding with stationary cilinder?

markomilosevic · July 14, 2016, 13:33

scoping thru vortex shedding phenomenon, to see range of interest viv for chimneys, finishing with viv for one case , but did a range of re... as it is air, I am neglecting mass of air vibrating with cylinder, so the case is different, not water... would compare what I can get from vortex shedding and structural dynamics, and later viv simulation

ghorrocks · July 14, 2016, 20:05

Before you dive into the complexities of turbulence modelling - have you checked the basics? See FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F

If you have not checked that your time step, mesh and convergence tolerance are correct then you are wasting your time looking at the turbulence model.

July 12, 2016, 10:12	Vortex shedding frequency from cylinder	#1
UOWmecheng New Member Matt Join Date: Jul 2016 Posts: 2 Rep Power: 0	Hey everyone, I'm trying to simulate VIV of a rigid cylinder in CFX. After running the file and exporting data I found that the lift coefficient fluctuates at a frequency of 1.39 Hz, however the shedding frequency as estimated by the Strouhal number is 0.97 Hz. In post-CFD I found that y+ max is ~6.4 and the max courant number is ~0.25. Does anyone have a suggestion to get the shedding frequency closer to that predicted by the Strouhal number? Any help is greatly appreciated!

July 12, 2016, 18:14		#2
markomilosevic New Member marko milosevic Join Date: Jun 2016 Posts: 4 Rep Power: 10	Hi Matt, what Re are you dealing with, and what numerical solution did you use UOWmecheng likes this.

July 13, 2016, 02:30		#4
urosgrivc Senior Member urosgrivc Join Date: Dec 2015 Location: Slovenija Posts: 365 Rep Power: 11	I wonder if SAS SST with y+~1 would predict the frequency even better: How did you model the elastic part of your simulation (rigid body springs?) For VIV mass of the cilinder and springs will efect the solution frequency also, not just the fluid that is surounding your cilinder. Cilinder must have the wright amount of inertia as it will be prone to acelleration and decellerations. Are you doing a 2D or 3D? Is your domain big enough?... It is a nice case you have got there, I want to be informed of what you will come up with and what you will be able to achieve. This was also done with cfx I think it is quite nice and hope you will get there. It is just a fixed cilinder but the frequency looks to be spot on. https://www.youtube.com/watch?v=mNQ9NRdrKCM It would be nice to get a lot of data of this reserch than compare the results, my future plans; https://www.youtube.com/watch?v=9FRTj6_1J2k Last edited by urosgrivc; July 13, 2016 at 07:21.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
vortex shedding frequency	naveen	OpenFOAM	23	May 17, 2016 09:12
Vortex Shedding in a 2D cylinder Using buoyantBoussinesqPimpleFoam	veerag96	OpenFOAM	0	April 16, 2016 23:22
Cylinder vortex shedding (3D)	Apocolapse	STAR-CCM+	2	April 3, 2014 21:33
frequency of vortex shedding	HaKu	Main CFD Forum	3	March 18, 2011 01:00
3D cylinder and Vortex Street shedding issues	josik_1982	FLUENT	2	July 17, 2010 11:19

July 12, 2016, 21:39		#3
UOWmecheng New Member Matt Join Date: Jul 2016 Posts: 2 Rep Power: 0	Hi Marko, The Re is 10 000 and I am using DES Other specs: fluid is water with density = 1000 kg/m^-3 and dynamic viscosity = 1.16E-3, cylinder diameter = 0.05m, first layer mesh height = 1E-4m, velocity is steady in x direction with magnitude 0.232 m/s

July 13, 2016, 21:06		#5
markomilosevic New Member marko milosevic Join Date: Jun 2016 Posts: 4 Rep Power: 10	ok, not expert but trying to solve similar situations... I would observe frequency with realisable k e with first layer height between 30-300y+, and coefficients drag/lift with sst and first layer 1y+, and 10 cv in sub/buffer layer, and check correlation with literature, and than move to des and les... as things are going with me les is solution, inform me what results you got with des, how big is your domain and mesh quality

July 14, 2016, 03:15		#6
urosgrivc Senior Member urosgrivc Join Date: Dec 2015 Location: Slovenija Posts: 365 Rep Power: 11	Are you doing: A = a VIV simulation with oscilating cilinder or B = just wortex sheding with stationary cilinder?

July 14, 2016, 13:33		#7
markomilosevic New Member marko milosevic Join Date: Jun 2016 Posts: 4 Rep Power: 10	scoping thru vortex shedding phenomenon, to see range of interest viv for chimneys, finishing with viv for one case , but did a range of re... as it is air, I am neglecting mass of air vibrating with cylinder, so the case is different, not water... would compare what I can get from vortex shedding and structural dynamics, and later viv simulation

July 14, 2016, 20:05		#8
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,852 Rep Power: 144	Before you dive into the complexities of turbulence modelling - have you checked the basics? See FAQ: http://www.cfd-online.com/Wiki/Ansys..._inaccurate.3F If you have not checked that your time step, mesh and convergence tolerance are correct then you are wasting your time looking at the turbulence model.