Learning Forum

[andrew.m.hayes]

hi -

I am trying to model air flow starting inside a computer case and exiting through various means.  I created an enclosure around the case and did a volume extract using the walls of the enclosure.  That seemed to work, but I am not sure if that was correct. I am also not sure if I am supposed to define the walls of the enclosure as 'symmetry'.  My largest hurdle is creating an air input.  Currently, I have a line that enters the top and exits in the middle of the case.  i can do a volume extract on that line and define the Airin and Airout, but when I do a volume mesh they get lost since the air line is inside the enclosure.  Is there a way to define a point source for an air exit?

[Rob]

Do you need the "outside" or would domain boundaries on the casing be more suitable? Are you using a forced flow ie fan at one end of the server rack with a few holes or a box with holes and a load of fans like a more common desktop? Are you using Fluent?

[SRP]

Hi,

A line has no area, so the mesher will ignore it in a 3D volume mesh. You need a surface for a boundary condition. That’s why your “Airin/Airout” get lost.

[andrew.m.hayes]

I am using an air input inside the computer case, so I 'think' I need the domain outside the computer (the enclosure) case to model the air flow out of the case

When I do a volume extract of the air line I do have two surfaces that I have defined as "in" and "out", but when I do the volume mesh the fluid zones get merged, since both the in and out are inside the enclosure

 

[Rob]

OK, so air is flowing around the outside of the case. Normally we'd ignore this but if there's a risk of pulling warm air back in it may be useful. I'm not sure what the line is? I'd expect holes/fans etc to draw air back into the box and would set these as "fan" or "interior" or "porous-jump" labels at the geometry stage. That should also avoid Fluent Meshing losing surfaces and cell zones.