Accessing the components of the resolved Reynolds stress tensor in a LES

flotus1 · January 11, 2013, 05:20

In a LES in Fluent, I can evaluate some of the Reynolds stresses via the "data sampling for time statistics" option.

My problem is that only the shear stresses (uv, uw, vw) are created. I am missing the normal components (uu, vv, ww) of the Reynolds stress tensor.

Am I missing something or do I really have to use a UDF to evaluate the normal components?

flotus1 · January 17, 2013, 06:42

Any ideas?

sbaffini · January 17, 2013, 12:05

You should really look better... does RMS mean anything to you?

flotus1 · January 18, 2013, 04:57

Yes it does.
But I thought that a root mean square for a velocity component with a non-zero mean value is not the same as the reynols stress, where only the fluctuating part is considered. In fact the RMS of a constant velocity in laminar flow is the velocity itself, so it cannot be the a reynolds stress.

sbaffini · January 18, 2013, 05:16

Your doubt is understandable, indeed the Fluent manual exactly report this definition for those quantities. However, if you compare them with your own statistics (in the most simple case by sampling the velocity in a point) you will see that, as a matter of fact, they are RMS of fluctuations. Also, if you look at the RMS of your streamwise (main) velocity component, it clearly is in term of fluctuations (with respect to the mean). Consider that this is also true for the non RMS statistics.

flotus1 · January 18, 2013, 05:29

Since I didnt find a definition in the fluent manual, i assumed the "usual" definitions for the statistics.
Thanks for the clarification.

flotus1 · January 18, 2013, 10:04

Sorry to bother you again, but where exactly did you find the information?

I searched again through the fluent manual but couldnt find any hint about the calculation of RMS statistics.

sbaffini · January 18, 2013, 13:42

It depends from the version of the manual. For the latest Fluent versions, check the users guide, field function definition, alphabetical listing or using the solver, performing time-dependent calculations, post-processing...

However, it just says that they are root mean squares of the variables, no fluctuation (with respect to the mean) is ever mentioned.

flotus1 · January 21, 2013, 04:34

This is where my confusion came from.
Anyway, thanks a lot for sharing your insight.

January 11, 2013, 05:20	Accessing the components of the resolved Reynolds stress tensor in a LES	#1
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,428 Rep Power: 49	In a LES in Fluent, I can evaluate some of the Reynolds stresses via the "data sampling for time statistics" option. My problem is that only the shear stresses (uv, uw, vw) are created. I am missing the normal components (uu, vv, ww) of the Reynolds stress tensor. Am I missing something or do I really have to use a UDF to evaluate the normal components? jzc likes this.

January 18, 2013, 05:16		#5
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,195 Blog Entries: 29 Rep Power: 39	Your doubt is understandable, indeed the Fluent manual exactly report this definition for those quantities. However, if you compare them with your own statistics (in the most simple case by sampling the velocity in a point) you will see that, as a matter of fact, they are RMS of fluctuations. Also, if you look at the RMS of your streamwise (main) velocity component, it clearly is in term of fluctuations (with respect to the mean). Consider that this is also true for the non RMS statistics. flotus1 and jzc like this.

January 18, 2013, 13:42		#8
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,195 Blog Entries: 29 Rep Power: 39	It depends from the version of the manual. For the latest Fluent versions, check the users guide, field function definition, alphabetical listing or using the solver, performing time-dependent calculations, post-processing... However, it just says that they are root mean squares of the variables, no fluctuation (with respect to the mean) is ever mentioned. jzc likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
How to calculate Reynolds Stress using LES data?	Jongdae Kim	Main CFD Forum	7	February 13, 2020 04:09
Reynolds Stress Models	Jade M	Main CFD Forum	0	April 21, 2010 17:38
reynolds stress components in CEL	Roberto	CFX	0	December 3, 2008 04:29
get the value of six Reynolds Stress components?	July	CFX	2	June 24, 2008 10:19
How to get Reynolds stress of LES in FLUENT ?	lcw	FLUENT	2	June 5, 2006 00:39

January 17, 2013, 06:42		#2
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,428 Rep Power: 49	Any ideas?

January 17, 2013, 12:05		#3
sbaffini Senior Member Paolo Lampitella Join Date: Mar 2009 Location: Italy Posts: 2,195 Blog Entries: 29 Rep Power: 39	You should really look better... does RMS mean anything to you?

January 18, 2013, 04:57		#4
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,428 Rep Power: 49	Yes it does. But I thought that a root mean square for a velocity component with a non-zero mean value is not the same as the reynols stress, where only the fluctuating part is considered. In fact the RMS of a constant velocity in laminar flow is the velocity itself, so it cannot be the a reynolds stress.

January 18, 2013, 05:29		#6
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,428 Rep Power: 49	Since I didnt find a definition in the fluent manual, i assumed the "usual" definitions for the statistics. Thanks for the clarification.

January 18, 2013, 10:04		#7
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,428 Rep Power: 49	Sorry to bother you again, but where exactly did you find the information? I searched again through the fluent manual but couldnt find any hint about the calculation of RMS statistics.

January 21, 2013, 04:34		#9
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,428 Rep Power: 49	This is where my confusion came from. Anyway, thanks a lot for sharing your insight.