How to best cool computer?

chico1st · Post by **chico1st** » Fri Jul 02, 2010 8:57 pm

Hi there, I am wondering how to best cool my computer quietly.
I have a P182, ATI 5770 graphics card, and ATI X3 440 CPU in the upper chamber of the P182.

I like fairly high performance computing and I was looking at the P180 Whole nine yards.

It seems to me that having good CPU heatsink with 2 fans on it being baffled out as the only source of case exhaust is the best bet but im scared that the air will be too hot going through my CPU, especially while running my system full tilt.

I suppose my question is, do you think exhausting the upper chamber through 2 fans on a CPU heatsink is best. Would having a fan in the front of the chamber (where the hard drive bay is) be good for bringing in air?

RoGuE · Post by **RoGuE** » Fri Jul 02, 2010 10:25 pm

baffling is usually unnecessary...the p180 and cases like it are designed to push and pull air, with a fans in the front, back and on top

the reason they use a push-pull strategy, it is the most efficient way to change out the air in your case. you want cool air being sucked into the system (front fan) and hot air being exhausted (rear fans). Don't baffle fans from your CPU to the case fans..it's straight up pointless. The air will go where you send it..via pressure differences. NOTE: some people swear by only positive / negative pressure cases, but from an engineering standpoint, this configuration is far from ideal. I recently read a book on electronic cooling (for work), and the reason for this ultimately comes down to fluid dynamics of air and the mechanics of fans.

The main thing you can do to silence the beast, is buy good quality fans and change out the tri-cools that come stock with antec p-series cases. Get some speed controllers while your at it, and then you can experiment with the cooling and sound levels in your own home. also, when considering your fan situation, don't forget about your power supply pushing air out of the case as well...

danimal · Post by **danimal** » Sat Jul 03, 2010 1:37 pm

i think that it could be difficult and/or disadvantageous to use neutral pressure inside of a pc case, so positive/negative pressure are the only things left to argue about

put a real slow-turning fan on the top of the p182, underneath the vent shroud... it should be pretty quiet.

i removed the rear fan, put two fans on the mugen2, and built an exhaust shroud from the cpu cooler to the back fan grill.

RoGuE · Post by **RoGuE** » Mon Jul 05, 2010 12:17 pm

danimal wrote:i think that it could be difficult and/or disadvantageous to use neutral pressure inside of a pc case, so positive/negative pressure are the only things left to argue about

This is actually very wrong. People never talk about 'neutral' pressure because they are too busy talking about positive/negative.

Fans work by creating a pressure differential across them. On the intake side, there is a higher* pressure, while on the exhaust side there is a lower* pressure. Fluids always try to flow from high to low pressure, so fans achieve airflow. The problem with fans, is this pressure differential is really only local..meaning the further you get from the fan, the less of an effect this pressure differential has on the surrounding air. By NOT having all your fans on one side of the case (i.e. the back), you increase the effective area in which fans are actively moving air. The front fans help bring in the air, while the exhaust fans help push the air out. If* you have a case which filters the air, like the antec cases, having a 'neutral' pressure configuration is what one should strive for.

Simply put, it moves the most air, the quietest. And in silent computing, this is what we are all about..efficient cooling, so as to have very slow spinning, quiet fans.

It is a myth that positive/negative pressure cases cool better..the only advantage goes to a positive pressure case which tends to cut down on dust buildup. However, if your intakes are properly filtered, this is a non-issue.

danimal · Post by **danimal** » Tue Jul 06, 2010 6:00 pm

RoGuE wrote:The problem with fans, is this pressure differential is really only local..meaning the further you get from the fan, the less of an effect this pressure differential has on the surrounding air.

if the pressure differential was only local, there couldn't be any positive or negative pressure, because it would mean that the fan wasn't moving any air... but we know that even the weakest fans pump a certain volume of air, they are rated by cfm.

i think that your definition of neutral pressure is that the volume of air going in equals the volume of air going out, which is what i was referring to... the reason that i said that it's difficult to achieve neutral pressure inside of a case is because, among other things, power supply fans pump a variable amount of air out of the case, that's not equally matched by the steady rpm intake fan.

so you can tout the concept of neutral pressure all you want, but it's seldom achievable... even the p18x, with seperate compartments, does not perfectly seal the upper chamber from the lower... in the same vein, it's difficult to seal a case off to the point where no dust can enter, which is why you want a positive pressure situation.

RoGuE · Post by **RoGuE** » Wed Jul 07, 2010 7:57 am

danimal wrote:so you can tout the concept of neutral pressure all you want, but it's seldom achievable...

Did I say it was achievable all the time?

RoGuE wrote:If* you have a case which filters the air, like the antec cases, having a 'neutral' pressure configuration is what one should strive for.

Stive being the operative word here. All I'm saying is, one can achieve the most efficient (and in turn quiet) cooling, if you employ a push/pull configuration with the case fans. (There are some exceptions for eccentric cases, however). This is a proven and well documented design goal, and it's why you see almost 99% of cases have a front AND rear fan.

Also, you missed my point about the pressure differential. All I meant by the word 'local' was the further you get from the fan, the less of a transient pressure differential you see as a result of the fan. In simpler terms..the further you are from the fan, the less the volumetric flow rate. This is due to what's called 'head loss' aka friction. Fans are only good at applying a *local* pressure differential, meaning close to the fan. The actual distance we're talking about here varies drastically based on the fan obviously.

So where am I going with all this? Well, if you have a couple fans in the back of a case, but none in the front, they have to work harder to move the air through the case! However, if you have a fan(s) pulling air in simultaneously, you can then run the fans much slower to achieve equivalent airflow.

Sure, you won't ever be able to perfectly match your intake and exhaust flow rates, but the case will always make up the difference because if you recall, matter cannot be created or destroyed ;)

danimal · Post by **danimal** » Wed Jul 07, 2010 8:09 pm

i always like reading your engineering theories.

neutral pressure is more like rarely achievable...

if you look at the spcr-awarded case, the two intake fans don't match the two exhaust fans... plus, the intake side of the front fans are blocked with filters, on the weakest side of the fan... in terms of neutral pressure, it's a design that was destined to fail.

RoGuE · Post by **RoGuE** » Thu Jul 08, 2010 4:20 am

If you take the example you gave, note there are 2 intakes and 2 exhaust. Maybe they aren't the same size, and you have impedances like a filter..they are still in push/pull configuration. They do this because it's much more efficient than having fans only* in the back of the case.

That's all I'm saying.