• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Airflow vs static pressure fans. Are they a scam?

no worries, in all likelihood your MB/PSU will not be suppling over 12.5v absolute max.
Too many fans for the Motherboard :laugh:
 
Thanks for the info, appreciate it.
I just checked the operating Voltage of my fans which is 7V - 13.2V
Remember that speed variance is static per fan though. I.e. if the spec says 1500rpm, any fan can top out at anywhere from 1350-1650 at its nominal voltage and 100% PWM and will go no lower/higher unless either of those values are changed. So it's essentially a lottery, mostly down to the specific physical properties of the fan motor and overall manufacturing variance. So if your fans actually run at 1500rpm, that's where they top out, and they won't be rated to tolerate any higher. But they are likely to be overbuilt for longevity, so chances of a slight overvolt to increase speeds actually doing any damage are slim. But the higher the voltage (above the rating), the faster the motor risks burning out.

Too many fans for the Motherboard :laugh:
The 12V comes pretty much straight from your PSU though. So unless your PSU is way out of ATX spec or you're using a boost converter, 12.something is what you're getting.
 
And a little nudge again — try removing pci brackets to further reduce GPU noise. You may even be able to remove one set of fans after that mod
Not going to happen, I don't want it to look ghetto. I doubt that it will make any difference, Below graphics card, I have wifi card, so I probably only have one PCI cover left. It won't make any difference.

Not so sure, capacitors and fans seem to be the first thing to wear out for power supplies and one must catch a dry fan early for reoiling or the supply will overheat.

That said, I tend to recap what I have with top quality replacements and oil the fans with a lot more oil than they had originally; at that point I'd tend to agree with you.
I never seen a fan fail myself. I even have 14 year old case fan , it works. I have two cards with fans from 2004, they work. I have old AMD Athlon 64 cooler, fan still works. I only got defective fan from CM with water cooler, but after bypassing PWM pin it works fine. My dad had dead graphics cards fan on Sparkle FX 5200 twice, but it wasn't my computer and maybe it was dust build-up, which obviously he didn't clean. From my experience fans works a long time, very long time. Meanwhile the least reliable thing in computers is clearly motherboards. I have seen many failures and deaths of those things, they simply don't last. I don't know why exactly they die, but capacitors may be a suspect. But then, I haven't seen a failure of old PSU or graphics card, which also have tons of capacitors...

I believe the less blades on a propeller the more efficient.
I don't know about that, it doesn't seem correct. Right now, the best computer fans are Scythe Kaze Flex and they have 9 blades. I only seen 9800 rpm 3 blader, so maybe at those very high revs more blades is worse, but at lower revs there's no reason to think that.

Because we're talking about fans i have a question: if my fans are rated 1500rpm but i make them run at 1600rpm what could it happens to it?
Likely nothing, but if you raise their voltage too much, fans can burn.

Hm? I don't see how that's relevant. mITX is a standard. Heck, if anything, the SFF PC market is an excellent demonstration of how breaking standards leads to problems - try to google how many people have issues fitting the DTX Asus Crosshair 8 Impact into various SFF ITX cases, for example. SFF cases also tendt to be very meticulous in documenting their clearances and limitations for non- or not-quite standardized things like radiator/fan fitment (not the size or mounting patterns, but if any can fit, how large, etc.), CPU cooler clearances, GPU size, etc. As for trying to make PCs smaller, the whole point is doing so while maintaining compatibility. Look at the Dan A4 for example - the entire purpose of that case is allowing you to build a powerful PC with all standard components - ITX motherboard, dual-slot full-size GPU, SFX PSU. All accepted standards. There are plenty of proprietary SFF solutions from the likes of Zotac, but you can't build those yourself. DIY requires adherence to standards to work.
Sure, I meant that mITX may be standard, but when it comes to cooling or graphics cards, it soon becomes wild west. That stuff causes tons of clearance issues and what? People still buy mITX boards and SFF cases. I don't see why thicker fans couldn't be sold.

But you're not accepting the inherent consequences of that, in (likely much) lower sales and limited compatibility. Of PC gamers (to pick a sizeable group with high performance PCs), only a small minority DIY build their PCs. Of those, only a small minority buy additional fans beyond those that come with their case. Of those, a large portion likely just buy whatever is cheapest as they discover free fan slots and are given advice online to just stuff it full of fans. So the people doing targeted purchases of brand-name fans? That's already a very small group. Adding "use cases where a thicker fan will fit and provide tangible benefit" is likely to make the target market so small as to be near meaningless.
But somehow clearly luxury stuff like AIOs, RGB and other stuff sell. Even RTX 3090 and it doesn't make any logical sense either.

Meaning that with any significant R&D cost, these fans would become ridiculously expensive. High end 25mm fans like the NF-A12x25 are already ridiculously expensive.
That's only because Noctua overprices their ripped off Gentle Typhoons. Those fans are certainly not even worth 10 Euros. On fun note, in some tests Arctic P12 beat Noctuas in noise and performance and they sell their fans for about 5 Euros. Do you honestly think that Noctua isn't ripping off?

And BTW what RnD? Is just stretching already existing designs so expensive?

Starting from scratch with literally every single part of the fan for a >35mm fan? I wouldn't be surprised if that doubled the costs from there.
You keep talking like thicker fans don't exist. Servers, cars, fridges and other things already have thicker fans. It's certainly not a Herculean task to make 40mm fans. And I said that someone has already done that.


You clearly have a highly optimistic view of how informed and rational people's purchase decisions are. Heck, even lots of people on these forums make poorly informed purchase decisions (excluding those who come here after the fact to seek help, which is not an insignificant group). But most people don't frequent tech/PC forums. Trying to inform your way out of breaking compatibility is a fool's errand, will inevitably mostly reach the people who already understand the issue (as those are far more likely to seek out these channels of information anyhow), and ignores entirely the game of telephone that is public discourse on any topic. Things get lost or transformed in transmission, and there's no amount of marketing or clear communication that can realistically overcome this. It can make a positive impact vs. doing nothing at all, but the problem will by no means go away.
With such thinking, you won't achieve anything. That's all I can say here. You know, there wouldn't be PCs if someone hadn't broken standard.


That doesn't really make sense. The main benefit of a thicker fan is allowing for more pressure, as the thickness minimizes the effects of back-pressure resisting flow or flowing back between the blades. Also, if you look at the specs, it isn't really a pressure design - the 2200rpm 25mm Vardar is rated for 3.16mmH2O, while the 1800rpm 38mm Meltemi is rated for 2.75mmH2O.
Well I look at blade shape and they clearly look like Silencio FP's, which is pressure oriented design. BTW you said that pressure is measured while fans are completely blocked, Meltemis may not lose as much pressure as generic 7 blader at higher speed, which is what you want in radiator.
 
I never seen a fan fail myself. I even have 14 year old case fan , it works. I have two cards with fans from 2004, they work. I have old AMD Athlon 64 cooler, fan still works. I only got defective fan from CM with water cooler, but after bypassing PWM pin it works fine. My dad had dead graphics cards fan on Sparkle FX 5200 twice, but it wasn't my computer and maybe it was dust build-up, which obviously he didn't clean. From my experience fans works a long time, very long time. Meanwhile the least reliable thing in computers is clearly motherboards. I have seen many failures and deaths of those things, they simply don't last. I don't know why exactly they die, but capacitors may be a suspect. But then, I haven't seen a failure of old PSU or graphics card, which also have tons of capacitors...

Consider yourself fortunate. Very few fans have 100% failed on me, but a significant proportion started making unpleasant noises a few years (or sometimes a few months or even weeks!) into their life. CM, Cougar, Sunon, Yate Loon, Arctic, Enermax, Delta, it can happen to any of them. My dataset isn't large enough to propose any trends, though.
 
Consider yourself fortunate. Very few fans have 100% failed on me, but a significant proportion started making unpleasant noises a few years (or sometimes a few months or even weeks!) into their life. CM, Cougar, Sunon, Yate Loon, Arctic, Enermax, Delta, it can happen to any of them. My dataset isn't large enough to propose any trends, though.
I probably don't even notice any new noise and I wouldn't consider that a complete failure either. Meh. The only noisier fan that I have is ATi X800 Pro, which made awful grinding noise, when I got it from eBay. Once I cleaned it up and reoiled, it works a lot better.

I'm probably no-lifing again, but I decided to watch some fan comparisons and see what they found out:

I honestly give up. This is bullshit. I don't get it, how couldn't they all find one definitive winner. This doesn't make any sense. Most fans no matter how different their blade shape is, perform very similarly anyway. I hope that I'm missing something and somebody will correct me, but currently I'm just confused.
 
If there is more than one criterion there may not be a definitive winner; makes sense to me.

One may get better air flow at the expense of noise; perhaps one needs the flow and are not worried about the noise, or it could be the other way around.
 
If there is more than one criterion there may not be a definitive winner; makes sense to me.
They don't even agree on quietest, radiator, air cooler and case fans. Other than Corsair ML and BQ Shadowings are poor.
 

Attachments

  • asymmetric fan blades.jpg
    asymmetric fan blades.jpg
    9.8 KB · Views: 71
Last edited:
@Valantar I managed to find those very thick fans. Turns out they were Feser Triebwerk TK-121.

You can read about them here:

Turns out that Feser and NoiseBLocker have teamed up with goal to create radiator fan. It seems that nothing good came out out of that. They just made very thick fan and it doesn't particularly perform well. Oh well, perhaps only diameter matters for computer fans.

Quietest is also relative; one trick is to spread the noise across the spectrum and it is then perceived as quieter by the ear.

This can be done with asymmetric fan blades.
Did you really watch those reviews?
 
I don't have an hour to spare; sorry.
 
twin blade prop vs jet turbine for efficiency....any fluid dynamicists here?? i do realise that jets do not directly create airflow over the wings.


probably should be fuel per thrust unit, of the blade and drive unit, or something similar.
propellers much like jets, are there for propulsion to fight drag, the lift is provided as the (whole!) wing goes through the air. NASA has a lot of good stuff if you have the time to get lost over there:

(E:)oops this is more direct https://www.grc.nasa.gov/www/k-12/airplane/lift1.html
cheers
 
I probably don't even notice any new noise and I wouldn't consider that a complete failure either. Meh. The only noisier fan that I have is ATi X800 Pro, which made awful grinding noise, when I got it from eBay. Once I cleaned it up and reoiled, it works a lot better.

I'm probably no-lifing again, but I decided to watch some fan comparisons and see what they found out:

I honestly give up. This is bullshit. I don't get it, how couldn't they all find one definitive winner. This doesn't make any sense. Most fans no matter how different their blade shape is, perform very similarly anyway. I hope that I'm missing something and somebody will correct me, but currently I'm just confused.
If you would look at Arctic P12's air flow to pressure graph in their website, it has a specific impedance spot where air flow decreases and pressure increases. It looks unnoticable to the eye, but if you read up on it, to keep the motor silent and not overload, you have to keep at a specific side of the curve, so the fan works louder unequivocally due to turbulence when impedance contrarily decreases.
That puts the practical limit to silent operation, not some other dynamic variable - impedance is what costs most fans a top spot. You cannot increase fan speed because impedance increases, or some form of it. Don't just listen to streaming reviews, you'll find more substance in written media.
PS:
Above the optimum range of operation, the flow across the upper and lower sides of the impeller blades becomes turbulent and the impeller starts to go into an aerodynamic stall.

Stall_Area.png


At this point, the fan will start to vibrate and create lots of noise. Operating at this point can potentially cause the motor driving the fan to overheat as the turbulent forces on the impeller blade disrupt the motor operating efficiency.
 
Last edited:
As far as I understand this is the formula of fan airflow:
air moved = rpms*volume of air that fan blades catch per revolution*static pressure of fan
Where did you get this from? I am not even sure it is dimensionally correct.
 
You should try to keep impedance low. Fixed that mistake in my previous post. Not all fans are good pressure fans because of it. If you see a good "kink" in the pressure curve, it is going to be a loud "stalling" fan.
 
If you would look at Arctic P12's air flow to pressure graph in their website
What graph? There isn't one anywhere:


Don't just listen to streaming reviews, you'll find more substance in written media.
Any example of that? It's actually difficult to find any written fan test.

You should try to keep impedance low.
Should I put my graphics card in trash bin? :D

Where did you get this from? I am not even sure it is dimensionally correct.
I came up with this formula myself. It's probably not correct and I expect to be corrected here.
 
What graph? There isn't one anywhere:



Any example of that? It's actually difficult to find any written fan test.


Should I put my graphics card in trash bin? :D


I came up with this formula myself. It's probably not correct and I expect to be corrected here.
It is there. Scroll to technical data. It comes in a pdf file.
 
Last edited:
Haha, wrong file. Told you.:)
Okay, but this graph shows mostly that when fan is nearly completely blocked, airflow curve is non linear. And maximum pressure is achieved at either very low rpms or when fan is 70% blocked.
 
But radiator fans don't spin all that fast and having asymmetric fans would cause balancing problems that would affect bearing life at higher RPM's.
This one blade 'fan' is balanced
 

Attachments

  • single blade.jpg
    single blade.jpg
    19.9 KB · Views: 76
Okay, but this graph shows mostly that when fan is nearly completely blocked, airflow curve is non linear. And maximum pressure is achieved at either very low rpms or when fan is 70% blocked.
Please, pay attention to what I said before the discussion convolutes further. You are pushing up a case where I demonstrated the opposite. It will stall and not operate quietly. And it will also increase the load and burn the motor. You are free to try at your own behest. Just don't take it further.
 
Last edited:
Back
Top