- **Main CFD Forum**
(*http://www.cfd-online.com/Forums/main/*)

- - **does the time step depend on constant source term?**
(*http://www.cfd-online.com/Forums/main/87344-does-time-step-depend-constant-source-term.html*)

does the time step depend on constant source term?In 3D Euler FVM steady flow solver, the time step for local time stepping acceleration is scaled as
(1) cell_volume /( lambda1+lambda2+lambda3) where lambda1,2,3 are spectral radii of the convective matrix. I do understand how formula (1) is derived. My question is, if there's a (constant) source term on the R.H.S. of the Euler equation, i.e., Original: du/dt + Flux(u)=0 Now: du/dt + Flux(u)=Constant then, will this constant on the R.H.S. change the time step? Can I still use formula (1) or should I formulate another time step to take into account the effect of the R.H.S term? Thanks. |

Hi
We know that CFL = (|U|*del_t)/delx, from which we get del_t = cfl*delx/IU| which can be written for 3D as delt = cfl * cell_volume / (|V.S|+c*|S|) . where S is the surface area vector. If the constant you were referring to is the artificial dissipation terms then the time step doesnot depend on those terms. If I am wrong please correct me. Thank you |

All times are GMT -4. The time now is 04:57. |