# Boundary Condition at outlet for supersonic flow

 Register Blogs Members List Search Today's Posts Mark Forums Read

 April 17, 2007, 10:33 Boundary Condition at outlet for supersonic flow #1 Benjamin Zachariah Guest   Posts: n/a Hi, I'm modelling an axisymmetric pipe flow from a compressed air canister at 8Bar (modelled as a pressure inlet) through a straight tube, into a constriction representing a valve, then again expanding into another tube, then into a C-D nozzle. As the flow will be supersonic, choked in the valve, then no boundary conditions should be imposed at the outlet of the nozzle, right? if this is the case, then what should i set the outlet boundary condition to be? the error i usually get is either floating point error, or some sort of divergence. any alterations of the URFs just delays the inevitable incorrect solution. i'm using the k-epsilon model, air is set as ideal gas, all the solution settings are second order, simple. thanking you all for your assistance

 April 18, 2007, 03:04 Re: Boundary Condition at outlet for supersonic fl #2 Rohan Guest   Posts: n/a Hi Benjamin, Generally for high speed flows its better to run the simulation at a lesser speed and then use this result as a intial guess to avoid divergence or floating point errors. Thanks Regards Rohan

 April 19, 2007, 06:07 Re: Boundary Condition at outlet for supersonic fl #3 Benjamin Zachariah Guest   Posts: n/a will try. thanks.

 April 19, 2007, 06:11 Re: Boundary Condition at outlet for supersonic fl #4 Benjamin Zachariah Guest   Posts: n/a a new error has just appeared that i've never seen before. it's Error: Specific_Heat_by_T_Integral: invalid lower bound, T0 <= 0 any ideas? thanks

 April 22, 2007, 23:47 Re: Boundary Condition at outlet for supersonic fl #5 Rohan Guest   Posts: n/a Hi, Looks like the T0 for the specific heat calculation is tending to zero. Iam not sure why this is happening. Please carefully go through your boundary conditions. Thanks Regards Rohan