|
[Sponsors] |
![]() |
![]() |
#1 |
Senior Member
Join Date: Jan 2013
Posts: 134
Rep Power: 12 ![]() |
HI everyone!
I'm doing flow around cylinder at Re=100, and I chose laminar as the RASModel in RASproperties file. Dose it mean the simulation is running in Direct numerical simulation? |
|
![]() |
![]() |
![]() |
![]() |
#2 |
Senior Member
Lieven
Join Date: Dec 2011
Location: Leuven, Belgium
Posts: 299
Rep Power: 21 ![]() |
That is indeed what it means, pretty cool he ;-).
But note that if the flow is turbulent, a very fine mesh is required to get an accurate result. This is in fact already the case for you (you can check the discussion on the flow regimes at http://en.wikipedia.org/wiki/Reynolds_number ). So to get accurate results for the Re = 100, you should: * run in 3D * do a transient simulation * perform a grid sensitivity study (to find the required number of cells) If any of the above is not achievable, you should keep Re < 40 in which case the flow should be laminar so that a 2D, steady state, crude mesh simulation should be acceptable (if it is an infinite cylinder). Note however that even in this case turbulent structures might appear due to numerical inaccuracies. Cheers, L |
|
![]() |
![]() |
![]() |
Thread Tools | Search this Thread |
Display Modes | |
|
|
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
CFX Treatment of Laminar and Turbulent Flows | Jade M | CFX | 18 | September 15, 2022 08:08 |
DNS always returns to laminar initial solution | kat | Main CFD Forum | 10 | September 9, 2007 10:20 |
The difference between Laminar model and DNS at lo | fpz | FLUENT | 3 | March 14, 2005 07:33 |
Laminar solver? LES? or DNS? | Ray | Main CFD Forum | 5 | March 19, 2003 11:19 |
Laminar model the same as DNS? | Kerrin | Main CFD Forum | 6 | February 24, 2003 16:12 |