Questions about inlet/outlets and pressure
Hello. I have two questions:
1) What is the effective difference between the ‘default’ setting for an inlet/outlet compared with having an inlet/outlet set to the ‘atmosphere’? I have three inlet/outlets in my simulation and with them on the ‘default’ setting, I kept having falsely excessive local velocities. Once I realized this and changed them to the ‘atmosphere’ setting (which better reflects the reality my simulation is supposed to represent), the simulation converged, without any persistent unrealistic velocities.
2) My simulation uses air as the substance. The ambient pressure is sea-level atmospheric pressure. As noted above, the simulation ran fine with the pressure of the substance air being set to ‘0’. (I thought I had set it to 2216 pounds per square foot, but had not.) I then saw that the air pressure for the three inlet/outlets, when set to ‘atmosphere’, was close to this value, set by Caedium at 2205.77 pounds per square foot. When I went back and set both the substance air pressure and the atmosphere pressure to 2216 pounds per square foot, I again got extremely high (unrealistic) velocities. My question is: Should I set the substance air pressure to zero again? If so, why? It seems like I should have both the substance pressure of the simulation and the atmosphere pressure set to the same pressure (sea-level atmospheric pressure in my case). The inlets/outlets are intended only to be borders of the simulation through which air may enter or leave, depending on what is happening with the simulation, at relatively low velocities (few feet/sec).
Thanks in advance for any help.
Feedback
Questions? Ideas? Problems?
Inlet-Outlet Type = Atmosphere
The Inlet-Outlet->Type = Default allows flow to exit and enter the flow volume just as the Type = Atmosphere, except Default imposes the specified inlet velocity when it switches to an inlet. Whereas Atmosphere allows velocity to vary according to the local conditions and adjusts the local pressure to match the specified total pressure.
Assuming you are simulating incompressible flow, then the pressure is relative (gauge pressure) not absolute. For numerical accuracy (and solver robustness) it is critical that you specify a reference pressure of zero. The Inlet-Outlet->Type = Atmosphere requires total pressure, which is equivalent to:
p + 1/2 * rho * v * v
where dynamic pressure = 1/2 * rho * v * v
Typically an Inlet-Outlet->Type = Default works well when the local conditions are likely to be close, e.g., on the far field boundary of an external aerodynamic simulation. If you have other inlet and outlet combinations in your model then this far field boundary will likely not be as robust.
If you have a definite inlet or outlet then you should not use the Inlet-Outlet condition.
This makes a lot of sense.
This makes a lot of sense. Thank you.