|
[Sponsors] |
February 27, 2006, 17:26 |
local time stepping and pseudo time
|
#1 |
Guest
Posts: n/a
|
Hi,
I am trying to understand the basic methods of NUMECA but there are still some things not clear to me. I really hope somebody will help me. So, let's start: 1. NUMECA uses a four-stage explicit Runge-Kutta method to solve the equation: int(dU/dt dV) + sum(F_I*D_S) + sum(F_V*D_S) = int(Q dV) (F_I, F_V: inviscous and viscous fluxes; U: solution vector). What happens if the computed problem is "steady"? 2. On the numeical model page, the user has to specify a CFL-number, which is basically v*Delta_t/Delta_x. I have heard of methods that use a time stepping for steady computations to increase stability. How does it work with NUMECA? 3. Moreover to the general Navier-Stokes equation, a term with pseudo time is added. What does it mean? Is the pseudo time defined by the CFL number? Thanks-. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Local time Stepping | phi | Main CFD Forum | 0 | January 19, 2008 08:38 |
local time stepping | cfdbeginner | Main CFD Forum | 4 | March 3, 2006 05:26 |
Local time stepping | Mehdi Benna | Main CFD Forum | 1 | June 19, 2003 23:26 |
local time-stepping | Pierre-Yves Lesage | Main CFD Forum | 2 | February 19, 1999 18:59 |
local time-stepping | Pierre-Yves Lesage | Main CFD Forum | 3 | February 18, 1999 18:01 |