|
[Sponsors] |
Flow Past a cylinder- Re-1000-PisoFoam Solver |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
October 24, 2015, 14:02 |
|
#21 |
New Member
vito
Join Date: Jun 2015
Posts: 22
Rep Power: 10 |
Davi,
Firstly, My suggestion is to use pimple Foam Solver when you are setting up LES, not pisoFoam ( they are essentially same when no. of outer corrector loops is 1). MESH As far Mesh refinement goes, the approach was very simple. I kept refining my mesh near wall of the cylinder till i got the Yplus value around 1, in my my case yplus was around 1.5, but lot of literature suggests it should be less than 1 to ofcourse capture the viscous effects in the boundary layer. But for your initial cases you can start of with a y plus a little greater than1. Also when you use Yplus LES utility in Openfoam it tends to give a slightly larger value in the initial few time steps but then tends go down much later in the simulation when the flow steadies out. Time step As you may already know, the beauty of pimple foam is such that you can specify a courant number and the solver will vary the time step to suit your courant number specification, so I started off 1e-04 for a reynolds of re=3900 and you should be fine with this. I used max courant number to be 0.5. My files are in a different computer and will not be able to post them until monday. But essentially the mesh is same as i have posted before but finer near the wall.Hope this helps till then. |
|
October 24, 2015, 14:07 |
|
#22 |
Member
Davi Barreira
Join Date: Apr 2014
Location: Fortaleza
Posts: 76
Rep Power: 11 |
It does help, thank. I'm also using pimple with outer corrector equal to 1. I've been using y+ around 1 already for some of the RANS models, so I guess the meshes Im using could work, the only thing is that I use a lot of local refinement and I've read that they might affect the solution, since the filtering will be different in the domain. What model for LES did you use? OneEquation, Smagorisnky...?
|
|
October 24, 2015, 14:09 |
|
#23 |
New Member
vito
Join Date: Jun 2015
Posts: 22
Rep Power: 10 |
I used the no. of outer loops 3
I used one-eqn and no damping or anything. Yeah I have not dived in to the depth of filters although I know that the domain size should be same in x, y and z directions. |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Fluent + flow past cylinder at Re=40 | m.vegad | Main CFD Forum | 25 | January 9, 2018 01:34 |
Flow past a cylinder. Help needed | Lurapa | Main CFD Forum | 1 | August 18, 2013 12:25 |
[ICEM] Flow past a 2D cylinder | arun7328 | ANSYS Meshing & Geometry | 0 | February 15, 2013 13:17 |
Drag coefficient of flow past cylinder vs time | pedroxramos | FLUENT | 0 | January 14, 2013 13:39 |
Flow past 2 smooth circular cylinder | slip | FLUENT | 0 | July 8, 2010 19:45 |