Fan choice: positive air pressure and sufficient cooling

[Yonatan_1]

I have an NZXT H440 case which I'd like to keep fairly ventilated, and with a slightly positive air pressure (to keep dust from coming in through cracks and the back panel). I'm currently planning to have:

• 
  Intake: 3 NF-P12 fans as intake in the front (behind dust filters and a rather restrictive front panel).
  
  Exhaust: 3 NF-P12 fans as on an EK-CoolStream SE 360 radiator on the top of the case (slim radiator due to MB clearance; the top panel is also a somewhat restrictive) & a single NF-A14 fan on an EK-CoolStream CE 140 radiator on the back.

As you see, all exhaust fans will be mounted on radiators, which I figured will restrict airflow to the degree that the case will have a positive pressure (even though 4 fans are operating as exhaust and only 3 as intake). Does that make sense?

As an aside, do you think the radiators will be effective in this scenario, dispersing heat from an i7 CPU (I'm waiting on the i7-7700k) and a single GPU (GTX1080)? (Note that mounting a radiator on the front fans will block my HDD trays--which I'm not sure I'll be using but would like to keep my options open. Also, those are the only fans which are protected by dust filters.)

 

[Mikel_4]

You didn't mention anything about open loop water cooling or AIO water cooling,

CPU only AIO water cooling
The biggest heat source will be from graphic card, especially the open air (EVGA ACX, Gigabyte windforce, ect) type, meaning without front panel air burst moved the dispersed warm air, the warm internal temperature may reduce top-side radiator performance.

Open loop CPU GPU water cooling
So you're looking to use 360 mm rad + 140 mm rad in series? or 140 for CPU and 360 for dual GPU? anyway, pulling air through filter scenario is better than pushing air inward through radiator, which lead me to a question, what pump/pump+top/pump+top+res you are using now? I see no room left for mounting if drive cages intact + GTX 1080 is ~ 277 mm long & 115 mm tall.

 

[Yonatan_1]

You didn't mention anything about open loop water cooling or AIO water cooling,

CPU only AIO water cooling
The biggest heat source will be from graphic card, especially the open air (EVGA ACX, Gigabyte windforce, ect) type, meaning without front panel air burst moved the dispersed warm air, the warm internal temperature may reduce top-side radiator performance.

Open loop CPU GPU water cooling
So you're looking to use 360 mm rad + 140 mm rad in series? or 140 for CPU and 360 for dual GPU? anyway, pulling air through filter scenario is better than pushing air inward through radiator, which lead me to a question, what pump/pump+top/pump+top+res you are using now? I see no room left for mounting if drive cages intact + GTX 1080 is ~ 277 mm long & 115 mm tall.

Thanks! I'm planning to do the latter (although I haven't decided on an order yet; note, however, that the 140mm radiator is more than twice as thick as the 360mm radiator, so their thermal performance won't differ by a lot, I don't think).

I was looking at the EK-XRES 140 pump+res combo, and I believe I'll be able to only take out some of the HDD bays to accommodate it, not all. I don't have it yet, though, so I can't be sure; Was I mistaken?

Also, would you care to address pressure and cooling performance?

 

[Mikel_4]

Pressure related to how much flow resistant, the CPU and GPU water block is the biggest contributor to pressure drop.
Most vendor enlist their pump specification i.e the original Laing D5 pump is rated 1200 liter per hour flow rate and with 12 meter head pressure, this numbers will decrease when coolant is on water block, they have to because they're suppose to exchange heat where coolant will carry whatever temperature to be cooled by radiator.

PUMP
When the pump is flowing coolant to the entire loop there can be only single dynamic pressure/flow rate at pump's inlet and outlet. most AIO such as corsair hydro doesn't need high flow rate and head pressure because their engineers have calculated optimal cooling ability for certain model.

RADIATOR
In theory, the wider the radiator core size, then makes blown air equalize more warm coolant temperature (flowing inside radiator's core) with ambient temperature. I forget the formula but some vendor such as Hardwarelabsdo enlist their radiator performance (that's why they use "labs" I guess they hire thermo-dynamics scientist whatever).

WATER BLOCK
Can't really comment on this because this is the part that doesn't degrade and happen to be I only have six of them, one for my i7 6900K and one for i7 6700K and four for GTX 780Ti. I use EKWB Supremacy MX and EKFC CSQ,

COOLING PERFORMANCE
Good place for references: www.martinsliquidlabs.com , www.extremerigs.com , www.thermalbench.com

• 
  ■ The material quality, design, and orientation (regular and goofy for CPU only) of water block.
  ■ Radiator's core width and sometimes thickness.
  ■ Fan's RPM and configuration (push only, push-pull, and sandwhich).
  ■ Pump's max pressure and dynamic flow rate

 

[blinding]

I doubt anyone could accurately predict whether that combination will give positive or negative case air pressure. Why don't you just fire it up and check? If you open a gap some where then you should be able to find out if air is leaking out or into the gap. Use some thin plastic and see if it is sucked into the gap or blown away.

If case pressure is negative then cut back on the exhaust or increase the intake. If your fans are changing speed based on chip temperature then it might be one way at idle and different under load.

 

[Yonatan_1]

I doubt anyone could accurately predict whether that combination will give positive or negative case air pressure. Why don't you just fire it up and check? If you open a gap some where then you should be able to find out if air is leaking out or into the gap. Use some thin plastic and see if it is sucked into the gap or blown away.

If case pressure is negative then cut back on the exhaust or increase the intake. If your fans are changing speed based on chip temperature then it might be one way at idle and different under load.

It's as I feared

Seriously though, I didn't think anyone could predict with certainty, but I do believe/hope some of you guys know how far a fan's (particularly a well-known fan such as the NF-P12) CFM drops behind a dust filter / when mounted on a radiator.

You'll note there really isn't much I can do to cut down on exhaust in this setting (even capping the fans' speed would negatively affect cooling performance), so if anyone knows of another model / brand of fan that operates better as intake (particularly in the H440's restricted build), that would be a valuable tip.