post #1 of 1
Thread Starter 
This empirical study aims to confirm (or refute) the notion that PSU exhaust has a velocity low enough that it rises up not too far away from the case.

-Corsair HX1000 (140mm fan)
-SunbeamTech thermal sensors
-Zalman ZM-MFC-2

Testing conditions and environment:

-PSU has ample space in front of its exhaust grill (>10' unobstructed)
-Ambient temp 26.5C
-PSU wattage during the test = 299W (see Pic 2)
-PSU direct temperature 33.1C (Sensor 1)

Methodology: (see Pic 1)
-Measuring Sensor 2 sits at a distance (d) from face of PSU exhaust grill for 1 minutes and the reading is taken

Result: (see graph below)
d('') 1 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6
h 2'' (C) 27.0 27.0 27.0 27.1 27.2 27.2 27.4 27.3 27.2 27.1 27.1
h ½''(C) 27.5 27.7 27.9 27.8 27.5 27.3 27.3 27.3 27.2 27.2 27.2

Data Analysis:
1. Hot air exhausted from a PSU has a low velocity to escape far away from the grill and the PC
2. Temperature drops fast when the hot exhaust is mixed with ambient air in the open environment
3. It begins to rise up right after being exhausted, with the majority being 1'' to 5'' from the grill

1. With a PSU installed in the bottom of a case and as hot air exhausted from graphics card(2) would also exhibit a similar behaviour, these hot exhausted air will rise in the rear of the case.

Practical suggestions: (see pic 3)
1. Maintain 6'' (at least) from the rear of the case to the wall behind
2. Better yet, provide active ventilation to the outside rear of the case instead of relying on passive natural convection
3. Avoid using the rear top fan of case as intake as it will suck in these hot air, not only due to the rise factor but also due to its (larger) suction force

Future studies: (when I have lots of time to kill)
1. Establish 2D isothermal lines and perhaps even 3D
2. Do a study on a graphics card

Edited by windfire - 4/13/11 at 4:29am