how to justify if a flow is turbulent in the ocean modeling

rob3rt 0ng · June 23, 2011, 04:41

Hi All,

My apology for asking such a fundamental question, but I've been struggling to understand on what basis should we say that the flow is turbulent (apart from the experimental results)? For ideal cases such as duct or pipe, then we normally use Reynolds number to justify the flow characterisation, however, for open channel I don't think this is quite the case. How about using Froude number as opposed to Reynolds number?

Thanks very much for your time and attention.

Kind Regards,
Robert

Bernhard · June 23, 2011, 04:48

You should not focus to much on a Reynolds number. Even for pipe flow it is possible to have laminar flow for Re=10 000 in special occasions. The Froude number cannot be a measure, since the molecular viscosity plays an important role in the dissipation and break up of eddies at small scales.

rob3rt 0ng · June 23, 2011, 05:08

Hi Bernhard,

Thanks for your reply. So to go back to the question, how do I know if a flow is turbulent prior to choose solvers, etc?

Kind Regards,
Robert

Bernhard · June 23, 2011, 05:34

In general you know what your simulating, so you know what to expect. This is the only option to tell it on beforehand, but I think it is really essential to understand what you are modelling when you start your simulations. And how can you judge the results if you don't really understand the case?

Of course you can look at your results, and check in what regime the turbulent viscosity is, but that doesn't make sense for me.

rob3rt 0ng · June 23, 2011, 05:48

Thanks Bernhard.

I judge my results based on comparison with the experimental observations. The way I do it now is similar to your second suggestion (I know this isn't the right way - which is the reason why I post this discussion on the board) which is to cross reference the turbulent viscosity ratio based on the Reynolds number and match the results with the experimental data.

Thanks and Regards,
Robert

vatavuk · June 24, 2011, 12:22

Hi Robert,
You can use the results for channel flows. Re=V B/nu
were V=medium velocity, B=channel depth, nu=kinematic viscosity. Transition from laminar to turbulent flow is expected for appproximately Re=1300.
Paulo

r08n · October 27, 2015, 14:02

I have a similar question (I decided to ask here instead of creating a new thread; sorry if I'm stealing this thread, but it seems relevant if you omit the "ocean modelling" part).

So, is it possible to somehow check quantitatively if the flow is turbulent? In my case, there is a pipe with a wire grid/mesh inside. It is possible to calculate Re from a known flow velocity and some characteristic lengths of the structure. However, how to assess the turbulence (if any) more precisely and to know at least approximately the velocity at which the flow begins to turbulize? Most solvers are based on the assumption of the flow nature, so one should select the appropriate solver, but how to select the correct one if I don't know in advance the flow regime (except for the value of Re and maybe some other parameters)?

There were already some questions (threads) about modelling the laminar/turbulent transition, but 1) they were rather old and probably things have changed in newer OF versions, 2) I think this question is simpler, because we don't have to model the transition itself.

June 23, 2011, 04:41	how to justify if a flow is turbulent in the ocean modeling	#1
rob3rt 0ng Member Robert Ong Join Date: Aug 2010 Posts: 86 Rep Power: 15	Hi All, My apology for asking such a fundamental question, but I've been struggling to understand on what basis should we say that the flow is turbulent (apart from the experimental results)? For ideal cases such as duct or pipe, then we normally use Reynolds number to justify the flow characterisation, however, for open channel I don't think this is quite the case. How about using Froude number as opposed to Reynolds number? Thanks very much for your time and attention. Kind Regards, Robert

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Turbulent flow + VOF	CK	FLOW-3D	3	January 13, 2009 07:30
Natural convection - Inlet boundary condition	max91	CFX	1	July 29, 2008 20:28
Laminar flow or Turbulent flow	mech	FLUENT	0	January 27, 2007 18:51
Reynolds and Turbulent Flow	Frederic Dubinski	Main CFD Forum	2	October 20, 2004 13:57
Hydrostatic pressure in 2-phase flow modeling (long)	DS & HB	Main CFD Forum	0	January 8, 2000 15:00

June 23, 2011, 04:48		#2
Bernhard Senior Member Bernhard Join Date: Sep 2009 Location: Delft Posts: 790 Rep Power: 21	You should not focus to much on a Reynolds number. Even for pipe flow it is possible to have laminar flow for Re=10 000 in special occasions. The Froude number cannot be a measure, since the molecular viscosity plays an important role in the dissipation and break up of eddies at small scales.

June 23, 2011, 05:08		#3
rob3rt 0ng Member Robert Ong Join Date: Aug 2010 Posts: 86 Rep Power: 15	Hi Bernhard, Thanks for your reply. So to go back to the question, how do I know if a flow is turbulent prior to choose solvers, etc? Kind Regards, Robert

June 23, 2011, 05:34		#4
Bernhard Senior Member Bernhard Join Date: Sep 2009 Location: Delft Posts: 790 Rep Power: 21	In general you know what your simulating, so you know what to expect. This is the only option to tell it on beforehand, but I think it is really essential to understand what you are modelling when you start your simulations. And how can you judge the results if you don't really understand the case? Of course you can look at your results, and check in what regime the turbulent viscosity is, but that doesn't make sense for me.

June 23, 2011, 05:48		#5
rob3rt 0ng Member Robert Ong Join Date: Aug 2010 Posts: 86 Rep Power: 15	Thanks Bernhard. I judge my results based on comparison with the experimental observations. The way I do it now is similar to your second suggestion (I know this isn't the right way - which is the reason why I post this discussion on the board) which is to cross reference the turbulent viscosity ratio based on the Reynolds number and match the results with the experimental data. Thanks and Regards, Robert

June 24, 2011, 12:22		#6
vatavuk Senior Member Paulo Vatavuk Join Date: Mar 2009 Location: Campinas, Brasil Posts: 196 Rep Power: 17	Hi Robert, You can use the results for channel flows. Re=V B/nu were V=medium velocity, B=channel depth, nu=kinematic viscosity. Transition from laminar to turbulent flow is expected for appproximately Re=1300. Paulo

October 27, 2015, 14:02		#7
r08n Member Robertas N. Join Date: Mar 2009 Location: Kaunas, Lithuania Posts: 53 Rep Power: 17	I have a similar question (I decided to ask here instead of creating a new thread; sorry if I'm stealing this thread, but it seems relevant if you omit the "ocean modelling" part). So, is it possible to somehow check quantitatively if the flow is turbulent? In my case, there is a pipe with a wire grid/mesh inside. It is possible to calculate Re from a known flow velocity and some characteristic lengths of the structure. However, how to assess the turbulence (if any) more precisely and to know at least approximately the velocity at which the flow begins to turbulize? Most solvers are based on the assumption of the flow nature, so one should select the appropriate solver, but how to select the correct one if I don't know in advance the flow regime (except for the value of Re and maybe some other parameters)? There were already some questions (threads) about modelling the laminar/turbulent transition, but 1) they were rather old and probably things have changed in newer OF versions, 2) I think this question is simpler, because we don't have to model the transition itself.