Blow in or Blow out?

[pcslookout]

I got a new case now with the fans should I put them where they blow air in or blow air out please?

Thanks.

 

[pcslookout]

BTW I have five 120mm fans total with 3 at top and 2 at front.

 

[Valantar]

No rear fans? That's odd. I always advocate for positive pressure - keep more air flowing into the case that out or it; this keeps dust out of the case too. Difficult to achieve with your setup though. What case are you using? And what cooling setup? All air? Or any water cooling?

 

[aigomorla]

id have them all blow in.

Ignore the entire possitive pressure vs negative pressure debate, because a case is not air tight sealed, so any pressure is quickly dissipated.

But top fans will blow down to give your air sinks fresh air, and keep the mosfets on your board cool.

The front fans will typically keep your hdd / gpu with fresh air as well.

 

[Ajay]

Go with Mother Nature (convection) - fronts blow in, tops blow out. Moving as much air as is reasonably possible through the case (subject to your noise requirements) has always worked for me. In that regard - I kinda made a poor case choice last go around - c'est la vie.

 

[Valantar]

id have them all blow in.

Ignore the entire possitive pressure vs negative pressure debate, because a case is not air tight sealed, so any pressure is quickly dissipated.

But top fans will blow down to give your air sinks fresh air, and keep the mosfets on your board cool.

The front fans will typically keep your hdd / gpu with fresh air as well.

If the case has loads of vents, that would work. If not, the fans will just be creating turbulence, pushing against air pressure and not really creating air flow. Not a good idea in a relatively sealed off case (even though the temp differences will most likely be in the ~5C range, as airflow in general isn't that important).

The whole point of positive pressurization is for it to dissipate through cracks and openings, thereby ensuring controlled airflow (i.e. no air coming in where you don't want it to). This keeps cool air where you want it, while keeping dust out.

Go with Mother Nature (convection) - fronts blow in, tops blow out. Moving as much air as is reasonably possible through the case (subject to your noise requirements) has always worked for me. In that regard - I kinda made a poor case choice last go around - c'est la vie.

If he has two front intakes and three top exhaust, assuming all fans are the same and running at the same speed, that gives you negative pressure, effectively sucking in dust and debris through every unfiltered opening in the case. Generally not a good idea. Besides, the role of convection in an actively cooled system is negligible unless you're actively fighting it (e.g. exhausting through the bottom of the case).

 

[Ajay]

If he has two front intakes and three top exhaust, assuming all fans are the same and running at the same speed, that gives you negative pressure, effectively sucking in dust and debris through every unfiltered opening in the case. Generally not a good idea. Besides, the role of convection in an actively cooled system is negligible unless you're actively fighting it (e.g. exhausting through the bottom of the case).

This is a fantasy, unless you want to run fans with high static pressure (like high speed Deltas). The 1-2 mm (H2O) static pressure of most case fans is insignificant unless you take the time to make your case nearly air tight. Just put filters on the intakes and blow out your case once a year, if necessary. Besides that, do whatever makes you happy.

 

[Valantar]

This is a fantasy, unless you want to run fans with high static pressure (like high speed Deltas). The 1-2 mm (H2O) static pressure of most case fans is insignificant unless you take the time to make your case nearly air tight. Just put filters on the intakes and blow out your case once a year, if necessary. Besides that, do whatever makes you happy.

That's simply not true. Sure, the pressure differential will be small, but it will be there. And yes, it will equalize due to the case not being sealed. How will it equalize? By drawing air in from any and all available openings. That's how physics works, after all. If you're arguing that there won't be enough pressure for it to make a difference, that's essentially the same as saying the exhaust fans won't actually be doing their job, simply spinning and creating turbulence without creating any flow. Which isn't going to happen unless you're talking about an entirely sealed box. More air being pulled out of a more-or-less enclosed case than what is pushed in creates a negative pressure differential between the case and the surrounding room. That's not something you can deny. The static pressure rating of the fans only really affects this if they encounter resistance - after all, static pressure isn't a fixed value, but depends on the resistance to airflow. In a relatively unrestricted situation, such as a normal case, airflow volume matters far more for pressurization (adding radiators or restrictive filters or case panels of course changes this). If you have three fans exhausting 65cfm each, and two taking in 65cfm each, that remaining airflow needs to come from somewhere. In a sealed case, depending on the static pressure of the exhaust fans, they will either perform less efficiently due to the lower pressure, or will pull more air through the case than the intakes deliver. In a non-sealed case, the remaining air will seep in whereever it can, to equalize the negative pressure differential.

 

[Ajay]

Ok, so you want to use physics. So, we have 5 low speed, very low static pressure 'case fans'.
First up we have a rigid sealed box with three intake fans on one side and two exhaust fans on the opposite side. Turn them all on until steady state is achieved. The flow rate is the same both in and out of the case (since we can't create or destroy the air in the box*). If the flow is the same at both openings with fans, the change in pressure in the box is zero.

Now put a fan sized hole in another part of the box. Turn the fans on until steady state is achieved. The flow rates at each opening with fans are different, but the change in pressure in the box is still zero because the sets of fans are decoupled due to the extra opening. Air is flowing out of the opening because it's nonrestrictive.

You say that if I have just the right amount of total open area, I will see a positive air flow. What I'm saying is good luck finding that number and if you find that number, good luck finding a case that matches that magical open area. In any case - do whatever makes you happy.


*Also, because under these very low pressures the air won't compress. One needs to compress air to increase its pressure.

 

[ch33zw1z]

in thru front

out thru top

 

[ScottAD]

Discussion for the ages... Ultimately natural convection doesn't matter. The article is aged but given efficiency improvements mount your fans how you want.

https://www.pugetsystems.com/labs/articles/Vertical-vs-Horizontal-Case-Cooling-89/
https://www.pugetsystems.com/labs/articles/Vertical-vs-Horizontal-Case-Cooling-89/
That being said you can use top as intake if you want but filter it if possible. Otherwise you'll have to clean a bit more often but right now I have a front mounted 280mm rad with fans pushing across the rad in to the case. Two top mounted Corsair AF 140mm fans. Front fan is intake, rear is out take and then the stock 120mm rear fan as an out. My R5 1600 CPU runs at 30 celsius and my GPU runs at 42 celsius, both at idle. Under workload my CPU goes up about 15 degrees and GPU doesn't go over 66.

 

[aigomorla]

By drawing air in from any and all available openings. That's how physics works, after all. If you're arguing that there won't be enough pressure for it to make a difference, that's essentially the same as saying the exhaust fans won't actually be doing their job, simply spinning and creating turbulence without creating any flow. Which isn't going to happen unless you're talking about an entirely sealed box.

no that is exacty what i mean...

Even the rear fan in most cases is pointless unless were talking about a duct server chasis or a dell which come with ducts.

At best the rear fan will keep air moving over the motherboard for turbulence, but to regulate pressure inside a case is just wishful thinking without a working duct.

In 95% of the cases a consumer can purchase, im fairly sure 100% in your case, your case is not sealed / ducted / designed to have pressure.

This is why i always tell people, it doesnt really matter, except the source of air.
You want air being pulled inside from the coolest point.
You dont want to pull stale air from the rear for example.

 

[Valantar]

You do have a point, but I don't entirely agree with you. While the effect on cooling between positive, neutral and negative pressure setups is pretty much negligible, dust retention and directed airflow are somewhat different matters. While you're right that the most important thing is to draw in air from a cool (and preferably dust free/filtered/both) place, any fan will create a pressure differential. Heck, that's what a fan does, essentially - a pocket of low pressure on the intake side, and a pocket of high pressure on the exhaust side. Now, depending on the number of intake fans, their speed and airflow, the restrictions these have to fight, the open/closedness of the case, and the number of exhaust fans (and their speed, airflow and the restrictions these have to fight), that pressure will dissipate at various rates - but it will be there nonetheless. In a normal, not open-air case, there is the potential for a small pressure differential with an intake-weighted fan setup simply because the case is small and enclosed enough to offer some resistance to the pressure dissipating quickly. While this difference will be tiny (it's not like I have ever seen a measurement of this, but I'd bet it's in the range of one hundredth or thousandth of a bar), it will be there, meaning that any opening in the case will have some air coming out of it (and none going in) keeping dust out.

As for directing airflow, that's far more complex, usually requiring ducting or at the very least very high airflow fans. I would argue this is the key point of having a rear exhaust fan - it ensures that the cool air being taken in through the front is at least partially passing over the CPU socket.

 

[Ajay]

The pressure created by the typical case fan is around 2 ten thousandths of a bar. Seriously, minuscule. Any resistance will just slow down the fans. You can have a net positive air flow when there is very little restriction, but positive pressure - not really.

 

[Ratman6161]

Air goes in the front and sides, down low and out the back and top up high. This keeps the air all moving in one direction and the way it naturally wants to go i.e. heat rises. On one of my systems I have a 140MM top fan so I turned the tower cooler on the CPU sideways. What I mean is that in the normal mounting, this cooler would be blowing the air through the cooler and towards the rear 120MM exhaust fan. Instead I have it turned so the fan blows upwards towards the top 140MM exhaust and the 140MM top fan also blowing out.