This shows you the differences between two versions of the page.
Both sides previous revision Previous revision | |||
spin:esc201_hs2019 [2019/12/09 14:52] stadel [Assignments] |
spin:esc201_hs2019 [2019/12/09 14:53] (current) stadel [Assignments] |
||
---|---|---|---|
Line 101: | Line 101: | ||
10. Compare Finite Difference Upwind and Corner Transport Upwind (finite volume) in 2-D using a Gaussian on a 2-D periodic mesh. due ** 8.12.2019 ** | 10. Compare Finite Difference Upwind and Corner Transport Upwind (finite volume) in 2-D using a Gaussian on a 2-D periodic mesh. due ** 8.12.2019 ** | ||
- | 11. Last exercise: 2-D Sedov Taylor Blast Wave. Define a 2-D **periodic** grid of variables (rho, rho_u, rho_v, E). Set P = e = 1e-5, rho_u = rho_v = 0, and rho = 1.0 everywhere (Note: gamma = 2). Set one cell (either in the corner, or center of the grid) to have e = 1. Adapt the timestep delta_t at each step to satisfy the Courant condition (given by the maximum of D_max across the grid). The timestep should be very small at first and increase with time as the shock wave expands. | + | 11. Last exercise: 2-D Sedov Taylor Blast Wave. Define a 2-D **periodic** grid of variables (rho, rho_u, rho_v, E). Set P = e = 1e-5, rho_u = rho_v = 0, and rho = 1.0 everywhere (Note: gamma = 2). Set one cell (either in the corner, or center of the grid) to have e = 1. Adapt the timestep delta_t at each step to satisfy the Courant condition (given by the maximum of D_max across the grid). The timestep should be very small at first and increase with time as the shock wave expands. You should use the corner transport upwind method with the predictor-corrector scheme outlined in the lecture. However, you can make a test using the 2-D basic LAX scheme. |