There is a slight difference between my version of the SEM and Nicolas'. This, I think, comes from a different definition of the eddy length scales used:

$ \delta_{min} = \Delta x $ $ \delta_{min} = min{\Delta x;\Delta y;\Delta z} $

This difference lead us to bigger structure to the wall (where the cut-off here quoted is active) which leads to this different $ C_f $.

There are differences in the computational costs too: NJ-SEM is a bit faster to compute (compared to the RP-SEM) because these bigger structure are better captured by the grid. Anyway the new DF-SEM and DF-SEMr are both quicker than the two aforementioned methods!


What now?

1 - I want to close the channel flow simulations - Some final studies (like a eventual k-budgets) can be executed

2 - I am in contact with Prof. Davidson to define the 2DHump test case, defined within the ATAAC project.

All of these studies can be useful to define the final article (deadline 30th April) and, furthermore, I am in contact with D. Monfort because EDF would like to implement the SEM and the DF-SEM possibly into CS 2.1 (released in July 2011) ...



A problem rose to my mind: the imposed average velocity profile are different! The $ \kappa-\omega SST $ model overestimate the channel massflow. DNS data instead "underestimate" the mass flow, compared to a periodic LES!!!!

Current Tags:
create new tag
, view all tags
Topic attachments
I Attachment Action Size Date Who Comment
pngpng Cf_solo.png manage 7.1 K 2011-02-08 - 08:49 RuggeroPoletto  
pngpng pressureequationiteration.png manage 13.5 K 2011-02-08 - 08:59 RuggeroPoletto  
pngpng velprofileloglog.png manage 5.1 K 2011-02-08 - 12:15 RuggeroPoletto  
Topic revision: r2 - 2011-02-08 - 12:21:38 - RuggeroPoletto
Main Web
21 Jul 2018


Manchester CfdTm

Ongoing Projects


Previous Projects


Useful Links:

User Directory
Photo Wall
Upcoming Events
Add Event

Computational Fluid Dynamics and Turbulence Mechanics
@ the University of Manchester
Copyright © by the contributing authors. Unless noted otherwise, all material on this web site is the property of the contributing authors.