[z_mehrdad]

Hi Everybody!!

I'm trying to simulate the shock waves caused by the explosion of gas (such as propane) in a tunnel using ANSYS fluent 14.0.

Firstly, I assumed an imaginary tunnel with dimention of 0.5*0.5*10m and tried the approach of patching a high pressure in a small region in the middle of the tunnel.

The solver was set to "coupled-Pressure Based-Transient"
The "Real Gas Peng-Robinson" model was used for calculating the Air density.
The two ends of the tunnel were set as "Pressure Outlet" boundary condition with 0 Gauge pressure.
The "standard" scheme was used for discretization of Pressure Eq. Also, "2nd Order Upwind" scheme for density, momentum and energy Eqs.
Whole the domain was initialized with the default values of FLUENT. But a pressure of 10atm was patched in a small region in the middle of the tunnel.

By running the case, after only one iteration the error "Divergence detected in AMG solver: Pressure Coupled" is shown in the consul.

By changing the Air density to "Constant" the calculations continues with a rapid decrease in "Continuity" to about the order of 1e-40!! But even at the first time step (0.01 Sec.) the pressure in whole domain is calculated about the order of 1e-19 Pa!!

By decreasing the time step from 0.01 Sec. to 1e-6 Sec. identical results were achieved.


Does any body have any idea??!!

[sfotovati]

Use Pressure Far-Field instead of Pressure Outlet.