Flow regime

dvolkind · August 15, 2011, 06:08

Greetings!
I'd like to ask two questions:
1. How to identify flow regime?
2. Can CFD software be used for that?

Comments:
I would greatly appreciate if someone helped me understand the difference between laminar and turbulent flow regimes. I have already posted this question in the CFX Forum and I was told to check one of the turbulence books, such as "Turbulence modelling for CFD" by Wilcox. I've looked through it and found no explicit explanation about that.
A particular problem for my question could be von Karman's vortex street. If we consider any point downstream the bluff body, the velocity in it will be oscillating. But the flow is still considered laminar. So, vorticies and velocity fluctuation do not necessarily make the flow turbulent. How to identify the flow regime then? Maybe Fourier transformation has something to do with it?

Thanks in advance,
Dmitry

leflix · August 15, 2011, 15:43

Hi Dmitry,

Really the best I or we can do here for you is to advise you to find a good fluid mechanic book.
You will find a tone of them for free on internet. You can either start to have a look on
http://en.wikipedia.org/wiki/Turbulence

laminar flow does not mean absence of vortical structures. But if you don't have any vortices in the flow then you are surely laminar.
if the flow have periodic velocity fluctuations it is laminar. When these fluctuations become cahotic transition to turbulence occurs. There is a lot to say on turbulence characterization start to read good stuffs.

pete · August 15, 2011, 16:02

CFD-Wiki contains a lot of information about turbulence. See for example http://www.cfd-online.com/Wiki/Intro..._of_turbulence

dvolkind · August 16, 2011, 01:22

Thank you for your response!

Florent · August 16, 2011, 05:57

The spectra of the flow are a good indication of the regime. A turbulent spectra is characteristic.

dvolkind · August 16, 2011, 07:36

Thank you, Florent!

Could you please suggest any conventional criterion?

When reading scientific articles (mostly from IJHMT) sometimes I come across a phrase like "by performing CFD-analysis we have established the threshold Reynolds number (when transition occurs)". But nobody ever mentions how exactly. I'd like to know if there is a common procedure of doing this.

Thanks in advance,
Dmitry

leflix · August 16, 2011, 11:18

There is not one universal or as you say conventional criteria to determine if a flow is turbulent or not. It is very flow dependent. One criteria applied to a given flow won't fit to another one. For example turbulent intensity which could seem a logical criteria is a good example of what I mean. There is not a universal value of the turbulent intensity which could indicate for every type of flow, that the flow is turbulent.

As mentioned Florent checking the spectra is a good tool. Record the time history of a given variable, Apply an FFT on the signal and check how the peak frequency(ies) evolve with the parameter which governs the flow (Re, Ray, etc...). A broad range spectra is a good indicator that the flow became turbulent.

Threshold value is more a term appropriated to the development of an instability those appearance is generally very sudden. We speak generally of critical value of that governing parameter.
Transition from laminar to turbulence is smoother and there is not a clear and well determined value which decides wether the flow is turbulent or not....

Another point is that if your flow is 2D one generally do not speak of turbulence which is a pure 3D phenomena...(check the different material suggested here on turbulence).
In these 2D cases "chaotic behavior" is more appropriated.

dvolkind · August 16, 2011, 13:21

Thank you, leflix!

I hate to bother you, but I have one stupid question left.
"A broad range spectra is a good indicator that the flow became turbulent." - that's what I was told on the CFX forum. Is it possible to determine exactly whether the range is broad (for turbulent) or narrow (for laminar)?

leflix · August 16, 2011, 14:11

when the spectra is broad you obviously notice it, the problem is more to determine when it is sufficiently broad.
Because as I mentioned the change between laminar and turbulent regimes occurs over a transition period where it is difficult to say whether you have still a laminar regime or already a turbulent one.
You have to determine your own criterion to decide it. You can find several different criteria, but what is sure, none of them will give you a well defined value of your governing parameter.

August 15, 2011, 06:08	Flow regime	#1
dvolkind Member Dmitry Volkind Join Date: Jan 2010 Location: Ekaterinburg, Russia Posts: 64 Rep Power: 16	Greetings! I'd like to ask two questions: 1. How to identify flow regime? 2. Can CFD software be used for that? Comments: I would greatly appreciate if someone helped me understand the difference between laminar and turbulent flow regimes. I have already posted this question in the CFX Forum and I was told to check one of the turbulence books, such as "Turbulence modelling for CFD" by Wilcox. I've looked through it and found no explicit explanation about that. A particular problem for my question could be von Karman's vortex street. If we consider any point downstream the bluff body, the velocity in it will be oscillating. But the flow is still considered laminar. So, vorticies and velocity fluctuation do not necessarily make the flow turbulent. How to identify the flow regime then? Maybe Fourier transformation has something to do with it? Thanks in advance, Dmitry

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
benchmark: flow over a circular cylinder	goodegg	Main CFD Forum	12	January 22, 2013 12:47
Flow meter Design	CD adapco Group Marketing	Siemens	3	June 21, 2011 09:33
reversed flow at velocity inlet / mass flow inlet	ib	FLUENT	1	March 26, 2007 14:11
How to change from mass flow to volume flow rate	stanley	FLUENT	1	February 2, 2007 07:44
transform navier-stokes eq. to euler-eq.	pxyz	Main CFD Forum	37	July 7, 2006 09:42

August 15, 2011, 15:43		#2
leflix Senior Member Join Date: Aug 2011 Posts: 272 Rep Power: 16	Hi Dmitry, Really the best I or we can do here for you is to advise you to find a good fluid mechanic book. You will find a tone of them for free on internet. You can either start to have a look on http://en.wikipedia.org/wiki/Turbulence laminar flow does not mean absence of vortical structures. But if you don't have any vortices in the flow then you are surely laminar. if the flow have periodic velocity fluctuations it is laminar. When these fluctuations become cahotic transition to turbulence occurs. There is a lot to say on turbulence characterization start to read good stuffs.

August 15, 2011, 16:02		#3
pete Administrator Peter Jones Join Date: Jan 2009 Posts: 682 Rep Power: 10	CFD-Wiki contains a lot of information about turbulence. See for example http://www.cfd-online.com/Wiki/Intro..._of_turbulence

August 16, 2011, 01:22		#4
dvolkind Member Dmitry Volkind Join Date: Jan 2010 Location: Ekaterinburg, Russia Posts: 64 Rep Power: 16	Thank you for your response!

August 16, 2011, 05:57		#5
Florent New Member Florent Duchaine Join Date: Jan 2011 Location: Toulouse, France Posts: 25 Rep Power: 15	The spectra of the flow are a good indication of the regime. A turbulent spectra is characteristic.

August 16, 2011, 07:36		#6
dvolkind Member Dmitry Volkind Join Date: Jan 2010 Location: Ekaterinburg, Russia Posts: 64 Rep Power: 16	Thank you, Florent! Could you please suggest any conventional criterion? When reading scientific articles (mostly from IJHMT) sometimes I come across a phrase like "by performing CFD-analysis we have established the threshold Reynolds number (when transition occurs)". But nobody ever mentions how exactly. I'd like to know if there is a common procedure of doing this. Thanks in advance, Dmitry

August 16, 2011, 11:18		#7
leflix Senior Member Join Date: Aug 2011 Posts: 272 Rep Power: 16	There is not one universal or as you say conventional criteria to determine if a flow is turbulent or not. It is very flow dependent. One criteria applied to a given flow won't fit to another one. For example turbulent intensity which could seem a logical criteria is a good example of what I mean. There is not a universal value of the turbulent intensity which could indicate for every type of flow, that the flow is turbulent. As mentioned Florent checking the spectra is a good tool. Record the time history of a given variable, Apply an FFT on the signal and check how the peak frequency(ies) evolve with the parameter which governs the flow (Re, Ray, etc...). A broad range spectra is a good indicator that the flow became turbulent. Threshold value is more a term appropriated to the development of an instability those appearance is generally very sudden. We speak generally of critical value of that governing parameter. Transition from laminar to turbulence is smoother and there is not a clear and well determined value which decides wether the flow is turbulent or not.... Another point is that if your flow is 2D one generally do not speak of turbulence which is a pure 3D phenomena...(check the different material suggested here on turbulence). In these 2D cases "chaotic behavior" is more appropriated.

August 16, 2011, 13:21		#8
dvolkind Member Dmitry Volkind Join Date: Jan 2010 Location: Ekaterinburg, Russia Posts: 64 Rep Power: 16	Thank you, leflix! I hate to bother you, but I have one stupid question left. "A broad range spectra is a good indicator that the flow became turbulent." - that's what I was told on the CFX forum. Is it possible to determine exactly whether the range is broad (for turbulent) or narrow (for laminar)?

August 16, 2011, 14:11		#9
leflix Senior Member Join Date: Aug 2011 Posts: 272 Rep Power: 16	when the spectra is broad you obviously notice it, the problem is more to determine when it is sufficiently broad. Because as I mentioned the change between laminar and turbulent regimes occurs over a transition period where it is difficult to say whether you have still a laminar regime or already a turbulent one. You have to determine your own criterion to decide it. You can find several different criteria, but what is sure, none of them will give you a well defined value of your governing parameter.