"Static Pressure Fans" Don't Exist

"Static Pressure Fans" Don't Exist - Part 3

Рет қаралды 1,781

Күн бұрын

Can fans be optimised for static pressure? Yes...and no, depending on what fans you're talking about. In this video I discuss and respond to a cheeky Noctua article, and I also consider the subject of optimisation when applied to different kinds of fans.
Read the article here:
noctua.at/en/nf-a12x25-perfor...

Пікірлер: 33

@luisitopalustre5947 5 ай бұрын

Instead of same rpm, noise-normalized result would be better no?

@or2kr 5 ай бұрын

Which Noctua did, as stated by the "Fan speed adjusted to similar on-application noise levels". As airflow scales linearly with RPM we can directly calculate at which RPMs the fans were running at to get those P-Q graphs, which are 1350RPM for the F12, 1230RPM for the S12A (above spec) and 1765RPM for the A12x25. Now the system impedance for the noise level determination was definitely chosen to suit the A12x25 and not the F12 or S12A, but I'd also regard it being a much better fan for real usecases and not just static pressure per RPM (F12) or just air flow with no resistance per RPM (S12A), the latter stalls with the slightest "load" for what it's worth

@smitty7510 5 ай бұрын

This isn't testing for performance, it's more about reviewing manufacturer claims

@FanPhys 5 ай бұрын

@@or2kr This is why I called Noctua disingenuous - rather than actually state the speed of each fan, which would explain the disparity in the graph, they put a cryptic note that gives us no actual information.

@or2kr 5 ай бұрын

@@FanPhys They also totally didn't do that again with the next gen 140mm marketing and chose totally different impedance curves compared to the 120mm marketing sheet :^)

@pixel2182 5 ай бұрын

really enjoying the test methods and validations, would love to see a video compiling all the results at the end of this series. It'd probably help the channel performance a lot as well, this is offering something LTT, GamersNexus or other channels don't ever dive into, and tests they do perform are often pretty poorly designed.

@FanPhys 5 ай бұрын

I'm glad you're enjoing it! Ideally, I'd like to keep adding info to the charts with each set of fans that I test, and there are fans that I'd like to test that I don't yet have (Toughfan 12, Phanteks T30, UNI Fan P28, one of Noctua's PPC fans, etc.). One thing that few others seem to realise is that fans & coolers perform quite differently with different thermal loads, and that's something I'd like to have a look at... but it is a monumental task and would take a lot of time (and money!). Thanks for your comment!

@qlum 5 ай бұрын

I don't think their fan tests are that bad, talking about GN and hwbusters. End of the day the performance that matters to the end user is how well do they cool a system at a given noise level. Real world tests on a somewhat standardized test load do offer that, be it at not the greatest measurement resolution.

@skidnik 5 ай бұрын

You said it at the end of the video. PC fan manufacturers nowadays use a baseline an average consumer actually cares about, noise. And in places you need static pressure there will be noise. Axial fans have a nasty feature: they completely lose pressure when flow resistance gets too high. For example if you have a CPU radiator with an axial fan strapped to it inside a PC which the owner never cleans, eventually the fan facing radiator side will gather so much dust the fan will completely loose pressure and will generate close to zero airflow through the radiator. It's interesting to see where this point is for different fans.

@FanPhys 5 ай бұрын

The thing with noise is that it's subjective, and that's because "loudness" is not really a measure of how noise is actually perceived by different people - the exact same sound might be annoying to some people and almost inaudible to others. This is why IMO testing fans based on noise is a fruitless endeavour. I'd much rather base it on fan speed, because in my experience virtually all 120mm fans are essentially silent when you get down to about 1,000rpm, at which point you've basically nuked the performance that the fan can offer. We shall see what the results say!

@athmaid 5 ай бұрын

@@FanPhys maybe there are free spectrum analyzers out there that can "quantify" the sound quality of a fan, Ansys Sound for example mentions fans on their product page. A psychoacoustic analyzer basically. I also found a paper by University Siegen in collaboration with ebm-papst on the topic. Might we worth looking into. Edit: I just found what looks like code for Matlab for exactly that, accessible for free.

@LogioTek 5 ай бұрын

Nice test & analysis. Shots fired?!

@FanPhys 5 ай бұрын

Haha yep I will definitely be firing shots because I've been sitting on some of this for years xD

@florian7460 5 ай бұрын

Great Videos! Keep them coming

@FanPhys 5 ай бұрын

Thank you!

@1marcelfilms 5 ай бұрын

i need a fan that can PULL well. through restrictive air filters

@FanPhys 5 ай бұрын

I will be testing the pull configuration as well :)

@or2kr 5 ай бұрын

You measure pressure at a set system impedance with your setup, you could theoretically calculate airflow from that though

@FanPhys 5 ай бұрын

It is possible to measure airflow from pressure, however getting an accurate reading would be difficult. Perhaps somewhere down the line I may be able to look into it!

@gubimann8092 5 ай бұрын

Just so you know, LED stands for "Light Emitting Diode" not "Light Emitting Dynamics" But Great Video:)

@FanPhys 5 ай бұрын

Hehe thanks, I do already know this - I was poking fun at people who think that RGB fans are hi-tech xD

@gubimann8092 5 ай бұрын

@@FanPhysOhh well that completely flew over my head

@Leonard_MT 5 ай бұрын

Just found this series, now I'm staying. I am going to be really _really_ interested in the next part since blower fans have been demonized in graphics cards. Also do you have any recommendations for PC case fans for someone who really doesn't care that much about noise and cares more for performance? (I currently run 2 NF-A12x25 fans [they were just recommended for me] at maximum RPM and can still tolerate more noise. And hey, now I now that the "curved fin" design kind of sucks, err... blows for performance)

@FanPhys 5 ай бұрын

Thank you! Performance fans are generally either bigger (well...thicker), or faster, or both. Virtually any 120mm fan at 2000+rpm should match or exceed the performance of the A12x25, and at the high end a fan like the NF-F12 PPC 3000 or the Phanteks T30 would do a great job. Blowers are not so great because they produce a lot of noise relative to the volume flow that they provide, but I'll look into that soon!

@Leonard_MT 5 ай бұрын

@@FanPhys Thanks for the recommendations! Can't wait for the next video in the series!

@pluggedfinn-bj3hn 4 ай бұрын

"PC fans will never get this level of obstruction in real world use." Meanwhile me here with a niche use case just wanting to find out what'll be the best fan to reallllly force air through. Although, I've already realized the answer is just, high rpm. I know these videos are addressing talking points said on forums, but still it makes me roll my eyes.. kinda feels like you're talking down on us from time to time.

@thethiagosc 5 ай бұрын

I miss the blower graphics cards that didn't make a furnace inside the case and in the past you did not need 20 fans inside the case.

@FanPhys 5 ай бұрын

Yeah, those were different times indeed!

@hellraserfleshlight 5 ай бұрын

I don't really get the point of this series... You say right in the video that noise is the priority with home PC fans. Are you just upset that some KZfaqrs popularized the term "static pressure" when it doesn't apply relative to industrial fans where noise isn't a consideration? When I'm picking fans for a home PC, I know I could get some $150+ industrial fans and enjoy my gaming sessions with earplugs in... It's just not an option I think it's a good one in that scenario. Most people putting any real thought into fans look at testing of their use case in noise normalized scenarios, or as you also note, the RGB. I don't think anybody thinks their Noctua will compare to a Delta on performance, but they aren't gonna put that thing in a PC next to their face on their desk.

@FanPhys 5 ай бұрын

I just find this stuff really interesting, and of course when I heard the term "static pressure fan" a few years ago I laughed a lot and didn't really think much about it, but now people actually make purchase decisions based on this nonsensical misinformation so I thought I'd make some videos about it. The main purpose of the series is to demonstrate that static pressure performance doesn't necessarily translate to actual cooling performance, and hopefully people will be better informed about which fans are actually the best cooling fans, which will take a fair amount of testing to ascertain - which, for me, is the fun part.

@hellraserfleshlight 5 ай бұрын

@@FanPhys I absolutely agree that it's interesting, in a purely academic sense, but as someone with years in the PC building and customization space, "static pressure" has a very specific place in how I make a decision. I wouldn't expect it to make much difference in a low-resistance application like an tower air cooler, but I will consider it when paired with a radiator. Even there, I won't be looking so much at the static pressure, but the real-world performance of a fan through a radiator. That's somewhere where I really wish there were more complete and objective data. You can find scattered info about airflow through a radiator, but that generally won't test more than a couple of radiators that only differ in thickness, not fin density or geometry. I haven't seen any testing of the difference fan design might make between two different radiators in terms of actual thermal dissipation, but that would be incredibly fascinating.

@zoson 5 ай бұрын

This testing does not prove what you are saying. Your fan box does not accurately replicate the kind of resistance a fan encounters on a heatsink, and your last video proved that your testing methodology is poor. If youre test environment was a reliable analog, it would produce results that map directly to the results you saw when the fans were used on your heatsink in your computer. But, the results from your test environment didn't even remotely resemble what you saw using a heatsink. You're also disregarding all the testing results from known reputable testers, which very clearly prove that certain fan designs maintain more airflow through a heatsink or radiator. E.G. They maintain more pressure in the face of heatsink/radiator type restriction. It's disingenuous to try and say that your test results prove there is no difference between fans marketed as higher static pressure, when you have clearly shown yourself that your test setup does not replicate the results seen when a fan is on a heatsink.

@FanPhys 5 ай бұрын

Again, you are assuming that pressure is the only consideration when we measure fan-on-heatsink performance, and that's not how fluid mechanics works - it is vastly more complex than that. And for what it's worth, the vast majority of "known reputable testers" have a poor grasp of physics when it comes to this sort of thing, which is precisely why I am making this series. Why would I regard test results that are obtained from poor, uninformed methodologies?!