Hi all,

I'm working on a project where I need to consider wind speed and direction over and through a barn.  The barn is rectangular, and I have a large box that is offset from the barn on all faces by 10 feet or so, with the environmental pressure B.C. applied to the inside faces of the large box so air can enter and leave the computational domain.  I need to add wind to this simulation in a way that will allow me to change the direction easily, without having to reselect faces.  I tried to simply select a solid body that was the same size as the computational domain for the inlet volume flow B.C., but it doesn't behave properly (the flow is always coming up from the bottom of the solid body, even if I specify Vx, Vy, and Vz independently).

Does anyone know how I can achieve this?  The initial conditions allow me to specify Vx, Vy, and Vz in the way I want to use them, but they are only valid until the solver begins its calculations.  Can I set up global boundary conditions in a similar way without using geometry to define a face?  The reason I don't want to use faces is that if the angle changes from perpendicular to the barn (along X), I need to change the values for the end faces, and also change them from environmental pressure to inlet volume flow with some X/Y velocity component values.

I know I could do something with an external flow project, and the ambient conditions would do what I want, but I need the computational domain to rebuild to the model geometry in the same way as the internal flow projects - i.e. the computational domain should be coincident to the solid body boundaries.  I have a large solid space defined with in-ground tanks that are beneath the barn, and I have fans ventilating those spaces at the same time as the wind moving through the barn above.  I really need a way to include ambient conditions in an internal study.

Thanks,

Dan