CPU

Experts or not, this is all about Fluid dynamics. let me try to explain this, the blow is more concentrated then the draw, since the draw can be pulled from a larger area, then pushed forward. the weakest part of the fan is the pull of it, the strongest the push, so the push is what you would use. try this simple test, put a card over a fan in the suck direction, notice how there is nearly no force? now try the opposite, you will notice much more force. that is why you use the blow to cool. as for case fans, it is about getting air-though the case, you would use the small case fan as intake, this is because it is smaller, your exhaust needs to be bigger tho since you have a much lower force draw.

[QUOTE=CommonTater;643602]

CarlTR6 said:

rmw020 said:

I agree that the CPU needs cool air and I personally think the the fan vlowing toward the heatsink and CPU is the most efficient. However, even if the fan is exhausting - blowing away from the heat sink, cooler air is still being drawn across the heatsink and CPU. I just don't think doing so is as efficient.

It's about the way air moves... If you stand behind a fan you don't actually feel the air motion, because 90% of it occurs very close to the fan in a very disorganized way as local air pours in to fill the vaccuum behind the blades.

This same air becomes the air stream you feel from a goodly distance in front of a fan. You feel it, because it's organized... it penetrates and travels.

This is what they call the Bernouli effect... moving air gathers more moving air along with it. The problem is that Bernouli's effect only works with pressure, not decompression of air (or fluids). The blowing on the palm example is one simple demonstration another is to bundle a garbage bag and blow it up like a baloon... it's going to take a lot of huffing and puffing to blow it up. But if you hold the end open and simply blow into the bag, you can inflate it with one good blast of breath. Because the air you blow into the bag gathers other air with it and causes a momentary pressure burst.

This relates to fans in that the front of the fan (which normally faces the heatsink, blowing air into it) creates pressure inside the fins, forcing air deeply into the blades getting far more heat conduction, whereas reversing the fan counts on a vaccuum to cause an air inrush which is mostly going to happen immediately behind the fan, with little or no penetration into the fins themselves. It will simply take the path of least resistance to fill the small pocket of reduced pressure right near the blades.

An odd confirmation of this is that a fan will spin faster when drawing air off a heatsink than when blowing air in... because the Bernouli effect causes pressure which causes resistance against the fan blades... and it's that pressure that drives cool air across the entire fin surface of the heatsink.

WHEW... does that help?

While I think that I agree with most of what you have said, your definition of the Bernouli effect leaves something to be desired. This is the effect of either a gas or liquid that is forced through a restricted aperture, causing it to speed up. When it speeds up, the pressure is reduced, This is what occurs in a carburetor, where the throat creates a restriction, which causes the air to speed up and lose pressure, which in turn causes fuel to be pulled from the jets by the lower air pressure.

Actually, it doesn't require an aperture as such, because the same effect occurs over the wing of an airplane, where the upper surface of the wing is curved in a fashion that requires to air to travel at a greater speed than the air below, creating a low pressure above the wing, producing lift and keeping the aircraft aloft.

Product FRED said:

So basically, exhaust.

I don't know what you mean by "exhaust"....

The fan on the cpu cooler should blow air into the fins of the heatsink. As I explained with the simple demonstration in my first message, this is at least 10 times more effective than counting on a vaccuum to draw air out of the fins.