Intake And Exhaust Question

[Cloudy]

I'm going to be building a portible gaming PC using a Corsair 380T case, but a have a question about intake and exhaust fans. The case will have a 200mm intake fan at the front, a 120mm exhaust fan at the back, a multi-fan GPU on the left side blowing air in, and will have a 240mm radiator on the right side for the CPU liquid cooler.

I assume my best bet would be to make the fans for the radiator intake so cool air from outside the case blows through the radiator. Thought that was better than making them exhaust fans blowing hot air from inside the case through the radiator. My only slight concern will be there will be a ton of posative air pressure and only a single 120mm fan for exhaust. What would you do if you used this case with a 240mm liquid cooler?

http://imgur.com/gallery/QRpl0Ee

[Omega]

Very nice LAN case.

Looking at the case I think it doesn't matter that much, The entire case is just mesh panels which provide more then enough breathing room and the font/back fans will provide enough airflow to get all the hot air out already.

If you put any fans on the left where the GPU is I would let them pull air in, since the GPU will be doing that also. And you are free to install a RAD in the left as long as it pulls air out of the case.

So yeah. front in, left in, right out, back out.

It's a very open case, I wouldn't worry too much about the airflow aslong as some air is being pushed in to it, the heat will find it's own way out.

[Cloudy]

I have asked this questions on multiple forums and I have been receiving tons of different answers. I'm probably going to take the time to try different layouts and benchmark CPU and GPU temps under load for each one. Either way I appreciate the feedback, interesting seeing what everyone else would do themselves.

[Omega]

Some people are very serious about this stuff. They want to optimize the crap out of the airflow, and sure, you can. But it's not worth the effort.

My way of doing it is most likely not the best, the mobo and GPU might run a few C hotter. But 55c vs 57c isn't even worth talking about, it's miniscule and makes no noticable difference except in the temp monitoring software.


"Airflow elitists"

[Monk]

Have the radiator exhaust, else you are just filling the case with more hot air.

Technically, negative pressure (more exhaust than intake) offers slightly better temps, but, it'll pull dust in through every gap, but as omega says, the difference really is minimal.

[someevildude]

Monk is correct. I've learned a great deal on cars and also racing in my youth. If you really want to eliminate any question here it is: Positive pressures have a way of building heat on it's own. That's caused by added friction against surfaces. Circular airflow keeps things more stabilized, but also has a vacuum down the center of the the pipe in effect. Bigger fans will also help with this since they cover more surface area, they won't have to push as hard to move all that air at once. It's also more quiet.

The radiator is not a bad idea, but an easy solution is to flip those fans around. It's not important for what side of the radiator it's on, just have it exhaust the airflow.

[Cloudy]

引用自 someevildude：

The radiator is not a bad idea, but an easy solution is to flip those fans around. It's not important for what side of the radiator it's on, just have it exhaust the airflow.

I will eat my words if I end up being wrong but for better CPU temps I disagree... I predict that the radiator fans set to exhaust will improve GPU temps but increase CPU.

Since I've had so many mixed answers I've come to the conclusion that I'm just going to benchmark both methods and see which one benefits me more. (I'm well aware they are going to be very similar and I won't see real world differences, but at this point to curiosity is killing me and I'll just say it's "for science".)

[TehSpoopyKitteh]

General rule is this:

Intake:
Front, Sides, Bottom

Exhaust:
Rear, top.


Airflow is usually pulled through the by open side of the fan. Back of fan should always face the direction of the airflow.

[hawkeye]

I doubt that intake v outtake is that important in cooling. It is more important to get a non-laminar airflow across the coolers, which is why they have fans. Laminar flow of air across a cooler results in inefficient heat transfer.

The objective is to achieve a turbulent flow across a large area to maximize heat transfer. The large area is the total area of the fins on the cooler.

There are other factors like the profile of the fins - flat or corrugated. At high air velocities corrugated will give better performance in general.

Dusty fins will reduce performance. As will a cooler blowing in the wrong direction against the air flow.

The purpose of the case fans is to push cold air into the case and remove warm air from it. If the incoming air mixes with the case air it could reduce efficiency. So the intake should channel air towards the cpu and gpu coolers.

At the boundary between the metal surface of the cooler and the air, the rate of cooling is proportional to the difference in fin and air temperature. So having cooler air coming in blowing on the coolers should help.

Outtake fans are going to suck out case air - the exhaust of the cpu and gpu coolers mixed with case air - which will be warm. Less effective I suspect.

So for maximum cooling I think a well directed incoming airflow onto the coolers would potentially give better results. You could actually fabricate a channel or nozzle out of a suitable material and tape it onto the coolers. Much like the way air is sucked into the engines on fighter aircraft.

I just leave the side off my case. Works better than anything else. The cooler fans blow the air out of the case, no need for intake or outtake fans.

[Monk]

Think however you wish, plenty of tests show negative air pressure is technically the best approach, unless you can perfectly balance intake and exhaust, but, that's near impossible given fan speed variance along with mesh differences and a hell of alot of other variables.

Also, given that the gpu performance is affected by temp while the cpu doesn't really give a damn as long as its not throttling (on new cards with automatic boost clocks), exhausting from the rad makes more sense.

[Cloudy]

引用自 Monk：

Think however you wish, plenty of tests show negative air pressure is technically the best approach, unless you can perfectly balance intake and exhaust, but, that's near impossible given fan speed variance along with mesh differences and a hell of alot of other variables.

Also, given that the gpu performance is affected by temp while the cpu doesn't really give a damn as long as its not throttling (on new cards with automatic boost clocks), exhausting from the rad makes more sense.

I will be overclocking the CPU, so it will likely make more sense to make intake. Besides, the slight GPU temp increase probably won't be enough to it preform much slower (if at all).

[Monk]

Even if you are overclocking, hell, there is a reason most custom loops exhaust, besides, if your cpu gets close enough to throttle with a few degrees, then it's probably too hot, where as the temp difference on a modern gpu will drop clock speeds.

Raising the temperature in your case will also raise the temp of the cpu technically, plus, a radiator is highly restrictive, by the time the fans push air through it, it's lost the majority of its pressure so is then not really helping cool anything, not that it would be as it'll be hot already.

Mind you, this is on the presumption you are using a well designed case, which most aren't (especially with glass panels).