[SOLVED] Which Noctua Case fans configuration is best for airflow vs noise balance.

May 13, 2020
5
0
10
0
Hi all. I wonder if I can get help on the math of my case fan upgrades. Basically, I'm thinking of changing my PC case to use Cooler Master Silencio S400 case and upgrading my CPU cooler to use Noctua DH-15 for my i9-9900K. So I'm thinking of also installing these case fans:
  • 2x Noctua NF-A14 PWM 140mm 1500 RPM for the front intake.
  • 2x Noctua NF-S12A PWM 120mm 1200 RPM for the top exhaust.
  • 1x Noctua NF-S12A ULN 120 mm 800 RPM for the rear exhaust.
Do you think this is a good combination for the perfect airflow? I know that doing the math I'm creating -8.9 m³/h difference. Which means it's slightly negative airflow. But there's going to be a dust filter on top of the case which may I hope might mitigate the slightly negative pressure in the case.
Otherwise, If I have room for:
  • 2x slots in front for either 120/140 mm case fans.
  • 2x slots on top for either 120/140 mm case fans.
  • 1x slot in the rear for a 120 mm case fan.
Which is the best combination for good airflow and noise if I want to fill all the slots with Noctua case fans?
See below for placement. White larger squares are 140mm, smaller white squares are 120 mm.

 

Paperdoc

Glorious
Ambassador
Do not worry about top intake. We hear all the time "hot air rises". ONLY in VERY limited circumstances that do NOT apply to fan-cooled computer cases. Hot air rises ONLY when there are NO other forces contributing to air movement. Even a single fan circulating air around inside a case will completely dominate the way air flows inside. (In some industrial applications, cooling electronics inside a closed metal case is done simply by having a fan inside move the air around, and letting the heat carried from heat sources to exterior walls escape though those case walls even though there are no air escape routes.) A case with multiple fans in intake and exhaust functions ensures lots of air movement inside and through the case, so by far the most important factor is number of fans and the resulting total air flow, NOT where the fans a positioned. The only role of fan position is to ensure that the flow pattern over components and through the case is reasonably smooth, not complex and circuitous.

More fans does mean more noise generated but it also means much more heat removed. Having several fans operating at modest speeds is likely quieter than one of two at full speed. But your best bet is to choose fans that generate less noise than others due to their designs, and for that you have done the best, perhaps. Noctua fans generally are reputed to be quieter than most, to push slightly more air flow and to last longer than most. If you want to check, see the table of specs on the Noctua site which includes max air flows and noise produced at max speeds. (To convert their m³/hr to CFM, 100 m³/hr = 58.86 CFM.) Compare those to similar fans and see what you think.

A note on this. Noctua supplies with their fans little items called Low Noise Adapters that you can insert into the connectors when you plug them in. These simply are resistors that reduce the voltage supplied to the fan, and thus reduce speed and air flow. They are useful only if you are plugging your fan into a power source with NO ability to alter fan speed, and you need the fans never to run full speed. When you ARE using automatic fan speed control based on sensor temperatures, these items will severely limit the max cooling you need at heavy gaming workloads, so do NOT use them at all.
 
Reactions: Rameyukk

Paperdoc

Glorious
Ambassador
What you plan is reasonable, so I'll add a couple comments and alternatives to consider.

I presume you are aiming for slight positive pressure inside to reduce intake of unfiltered air carrying dust. That is my preference, too. In fact, it is impossible to predict from fan count only (and their max air flow specs) the pressure balance because various other elements in the air flow paths affect real world performance. The most obvious of those is dust filters (and their effects can change as they clog up, so you have to clean from time to time). On that note, the dust filters supplied with that case for the TOP are optional - you do not need to use them, and most people would not when the top is used for exhaust fans. Anyway, with the filters in place for the intake at front, the imbalance will be more than your numbers suggest because those fans will deliver less intake than their max.

One way to change the imbalance would be to switch the plan for the exhausts. Put two of the ULN models at top, and one of the regular models at rear. This changes exhaust max capacity from 289 to 256 m³/h yielding a slight positive pressure, but that will be reduced by the effect of the front panel filters. Whether you do this or not, you can use the test-and-tweak process below to optimize as you like.

Whichever way you go, I suggest you do a simple tracer test when it's all done, and then there is a way to tweak it. For a start, I will assume that you configure the various fan headers to do their normal automatic fan speed control using the pre-set fan "curves" provided. I suggest further, IF your mobo has at least two case fan headers, that you group them into the front fans on one, and all the exhaust fans on the other. The tracer test requires a small source of smoke like a cigarette or a smoldering incense stick. With the system running normally, you move that smoke source near case small cracks where air can leak. If the smoke drifts away from the case at modest speed, you have a small positive pressure as you and I would like. If it is draw into the case, it has negative pressure. If the smoke moves rapidly, the imbalance may be too large. If you want to get very thorough, you can repeat this a different system workloads to force fan speed changes. Or, rather than changing actual workloads, you can simulate that by using the options for custom fan "curves" to force the fans to specific speeds and test for the smoke flow patterns.

You have not told us what mobo - maker and exact model number - you will have. If you do, we can help by looking up its manual for details. But most mobos today offer in the options for each fan header a choice to set your own custom fan "curve" of fan speed versus measured temperature, rather than using the pre-set one. If you have that, you can tweak your fans' speeds so that the smoke test tells you that you have the pressure balance you need over a range of fan speeds.
 

Phaaze88

Splendid
Ambassador
The Silencio S400, eh?
Do you think this is a good combination for the perfect airflow? I know that doing the math I'm creating -8.9 m³/h difference. Which means it's slightly negative airflow. But there's going to be a dust filter on top of the case which may I hope might mitigate the slightly negative pressure in the case.
There's a little more to it than fan specs - it's not that cut and dry, unfortunately. You also have to take into account the 'obstacles'. Some have minor impact on airflow, others are major:
Front panel design: major
Side inlets: minor
Filters: major or minor depending on thickness
Psu shroud: minor, but can also be major depending on the gpu model and proximity to the shroud

In a case like this, a positive pressure setup may not be ideal due to the major airflow choke point in the front. A negative pressure setup makes more sense in a case of this caliber.

To do a successful positive pressure setup - which I don't believe is ideal - in there would be:
2x NF-A14 front intake
1x NF-A14 or S12A top intake - that's why there's a filter for the top panel.
1x NF-S12A rear exhaust

Negative pressure is easy enough:
2x NF-S12A top exhaust - remove the top filter
1x NF-S12A rear exhaust
That's it. I wouldn't bother putting any fans in front, since they won't accomplish much other than add more noise.
That, and they might have the undesirable effect of recirculating air inside the case - some solid front cases have a funky manner of doing that with front intake fans.

Silent cases have the downside of poor thermal performance, and are a poor match for high power consuming hardware on air.
You would have to switch to liquid for more reasonable temps - even if the fans can't move that much air through the radiators, the superior thermal tolerance of liquid over air would mitigate that.
Noticing that you have a 9900K with a NH-D15 on top of it, you probably have at least a 2070 Super to go along with it...
If you want silence above all else, you'll definitely get it, but don't get too hung up on temps though.
 
May 13, 2020
5
0
10
0
Thank you for both of your replies. To add more information, this is my current build that I want to switch the case. I'm not changing any of the components below:

Motherboard: MSI Z390i
CPU: Intel i9-9900K
GPU: INNO3D GeForce RTX 2080 Ti Gaming OC X3
RAM: 2x 16GB Corsair Vengeance LPX 3200 MHz
SSD: 1TB Samsung 970 Pro M.2 NVMe
PSU: Silverstone SX800-LTI 800W SFX-L

I am changing the following items:
CPU Cooler: Noctua NH-L12S
Case: Silverstone Raven RVZ03 Mini-ITX

I've had it for 18 months now, and it's useful for most scenarios. I use this PC for gaming in my living room on my 55" Sony X900F 4K HDR TV and Sony HT-Z9RF 7.1.2 Dolby Atmos Soundbar. So I'm not changing this either.

The problem I've been having lately is thermal shutdown when I play new games in full 4K HDR. I do know this current case is the problem, not enough room for airflow to move around. I don't do overclocking though, so I keep everything at stock. But I do want better cooling.

I'm limited in my space on my living room. So I'm planning to choose the Coolermaster Silencio S400 because it's still bigger than my current case with more room for air to move, but compact enough to fit on my TV cabinet.

But I don't like having Tempered glass or RGB to show off. My style is to have an opaque minimalistic box that looks like nothing more than a mere box, but packs a punch inside. I don't mind a little hum of a noise coming from the fans, as long as it gives an efficient and effective cooling. But I don't want it to sound like a jet engine, either. So I need to make the PC as quiet as possible without compromising cooling too much. an Open aiflow case may be good, but it might be too noisy.
1x NF-A14 or S12A top intake - that's why there's a filter for the top panel.
I don't know if I'm comfortable with having a top intake.. That seems to go against the physics law of thermodynamics.
 

Paperdoc

Glorious
Ambassador
Do not worry about top intake. We hear all the time "hot air rises". ONLY in VERY limited circumstances that do NOT apply to fan-cooled computer cases. Hot air rises ONLY when there are NO other forces contributing to air movement. Even a single fan circulating air around inside a case will completely dominate the way air flows inside. (In some industrial applications, cooling electronics inside a closed metal case is done simply by having a fan inside move the air around, and letting the heat carried from heat sources to exterior walls escape though those case walls even though there are no air escape routes.) A case with multiple fans in intake and exhaust functions ensures lots of air movement inside and through the case, so by far the most important factor is number of fans and the resulting total air flow, NOT where the fans a positioned. The only role of fan position is to ensure that the flow pattern over components and through the case is reasonably smooth, not complex and circuitous.

More fans does mean more noise generated but it also means much more heat removed. Having several fans operating at modest speeds is likely quieter than one of two at full speed. But your best bet is to choose fans that generate less noise than others due to their designs, and for that you have done the best, perhaps. Noctua fans generally are reputed to be quieter than most, to push slightly more air flow and to last longer than most. If you want to check, see the table of specs on the Noctua site which includes max air flows and noise produced at max speeds. (To convert their m³/hr to CFM, 100 m³/hr = 58.86 CFM.) Compare those to similar fans and see what you think.

A note on this. Noctua supplies with their fans little items called Low Noise Adapters that you can insert into the connectors when you plug them in. These simply are resistors that reduce the voltage supplied to the fan, and thus reduce speed and air flow. They are useful only if you are plugging your fan into a power source with NO ability to alter fan speed, and you need the fans never to run full speed. When you ARE using automatic fan speed control based on sensor temperatures, these items will severely limit the max cooling you need at heavy gaming workloads, so do NOT use them at all.
 
Reactions: Rameyukk

Phaaze88

Splendid
Ambassador
:whistle:
Nice setup!
Silverstone SX800-LTI 800W SFX-L
I found a review of this unit: https://www.jonnyguru.com/blog/2017/05/22/silverstone-sfx-sx800-lti-800w-power-supply/
It's a good one, but the fan is a sleeve bearing type: https://www.gamersnexus.net/guides/779-computer-case-fan-bearing-differences
If it's starting to fail, or has failed, then the psu could overheat under load and shut off - refusing to let you power it back on for a set period of time.

I don't know if I'm comfortable with having a top intake.. That seems to go against the physics law of thermodynamics.
Naturally speaking, yes, heat rises, and in a passively cooled case, the chimney style is one of the most effective for case cooling.
BUT! Against something unnatural, like fans, which defy the natural law of thermodynamics, that does not apply, until the heat is expelled out of the chassis.

So I'm planning to choose the Coolermaster Silencio S400 because it's still bigger than my current case with more room for air to move...
I'm not so sure about that due to the chassis' design, the targeted audience for said model, and the high power consumption of your hardware.
Well, time will tell... I hope it works out for you.
 

ASK THE COMMUNITY

TRENDING THREADS