As a completely passive system (which idea pleases me noise-wise) has the disavantages of high temperatures, and might limit the choice of system to a lower-performance one, I wonder...

How relevant is silencing the airflow (lowering fan speed / removing fans) beyond a certain point? And if it that point exists, where is it?(at what fan speeds)


*Consider, hypothetically:
-- a room with very low ambient noise;
-- a system with:
- custom case designed specifically to minimize noise and maximize cooling performance;
- ridiculously high quantities of noise&vibration damp treatment in such a way that fans would be the only source of noise;
- Only fans in the system: 8 high-quality fans (12cm, @ 300 RPM);

Please tell:
At what distance had a high-quality microphone to be put in order to not capture any audible noise from the system?
Or
Down to what distance do you think the system would be inaudible?

(Comments are welcomed)
____________________________________

Addendum: The voting is secondary, most important is to discuss about a hypothetical point "between passive and active cooling", just before very slow fans would stall, to get the best out of both worlds:
- inaudible noise(not 0dB/silent, but not audible for human ear),
- good cooling efectiveness,
- low temperatures on all components (even capacitors and voltage regulators)
- ability to have a high end or at least fairly high end system
- etc.

I suspect it is possible, if we:
- take out the bearing noise by using "noiseless bearing fans" and "vibration dampening";
- use fans at lowest possible speeds (i think it is around 300, what do you think/know ?);
- use many fans, distributing the airflow through more space, and so decreasing the "air movement noise";
- seal the space between the fan and surface into which it is mounted, in order to maximize the fan's work by not wasting any of it...

So more than voting, please tell what you think about this and/or any experiences you had/have that confirm or deny this.

Thanks,

8 fans? That's sure a butt-load. I'd vote 2 meters at least.

2 or 3 fans would probably do a bang-up job on the vast majority of systems, and would (of course) be orders of magnitude quieter.

Are there fans that will start consistently at only 300rpm? (especially after they get dusty or worn)

It's going to depend on the ambient noise. 120 mm fans get really quiet below 500-600 rpm if they have good bearings. My system has six low speed fans which idle between 500-750 rpm. The 750 rpm fan is the "fast" one and is in the Seasonic 430 PSU. I can't make this one idle any slower, though I have to put my ear less than a foot away from the back of the case hear it at all. The system also has 4 fairly quiet hard drives, which are louder than the fans. It's hard to hear this machine at all, especially when the machine next to it is running.

I'm an advocate of running several low speed fans as opposed to running fewer fans at a higher speed.

edit: My machine with six fans and 4 hard drives is quieter than my other machine with 2 fans and one hard drive. The fans in the 2 fan machine need to run faster in order to keep the machine cooled. The decrease in noise from the slower speed more than makes up for having more fans.

Jason

RalfHutter wrote:


Yes 8 fans, but spinning at ~300 RPM !

I think that 2 fans @300RPM make much less noise than 1 fan @600RPM, am I wrong?
And I'm talking about high-quality (but non-ball-bearing) fans, in order to take bearing noise out of the equation.

xan_user wrote:

- The fans i'm considering are NoiseBlocker-M12-S1, which @5v would spin at 313RPM. It's specifications say that it can start at 5.5v, but i read in a review it could start at lower, although I must get one and confirm - but if it doens't work, there are other options for not much higher rpm;

- about getting worn, these fans have a MTBF (at 25°C) of "160 000" @stock speed;

- about dust, the custom designed case i'm idealizing would have Extra-large filters sufficiently big to not present significant resistance.

I am indeed planning (and have been idealizing it for very long) to build such thing (a complex case with separate chambers for hot components, extreme noise/vibration dampening, including a "self-made innovative psu cooler", that would be inaudible(not silent, but not audible for human ear) down to 25cm or so, capable of cooling well (with low temps) a fairly-high end system or so i hope! (one can dream!!!... )).

@nomoon:

Nomoon's right, it depends entirely on ambient.

My PC has an SSD, PicoPSU, and one Yate Loon fan @ 5V. At night, from 3m away, I can hear it faintly clicking away. It's inaudible over the hum of the monitor, which is also inaudible during the day.

If your question is "At what distance does a fan at 300RPM make 0dBA?", then you need a really, really quiet anechoic chamber to find out. Also, I think the answer is "pretty far". Like, 10-100m is my guess.

Brian wrote:


No, I meant to ask: The minimum distance at which it is inaudible for human ear. Am I wrong thinking that the human ear can't notice something like a 1 or 2dB change from a very low ambient noise?

Maybe I didn't explain myself well:
More than the voting, I meant us to discuss about a hypothetical point "between passive and active cooling", just before very slow fans would stall, trying to get the best out of both worlds: inaudible noise, good cooling efectiveness, low temps, etc.

I built an Atom-based WHS system around a fanless Intel mobo... but for the lulz, ordered an inexpensive (US$10) Nexus 120mm fan when buying parts.

I found the fan would start reliably at 7V and was inaudible (in my very quiet country house) at any distance greater than 10-15cm. (Compared say to my pair of WD Caviar Green 1 TBs that are audible at 3-4 m.) I also found that even at that threshold fan speed, CPU temps dropped by 20C or so below what I measured with no fan (from memory, 45C vs 65C).

So while I'd planned on a 'fanless' system, the reality was that a silent-for-all-practical-purposes fan still made a pretty nice diff in operating temp, and I expect I'll continue to build future systems with 'fanless' parts but supplemented by a large, lazy fan. (Unless dust proves to be a huge problem.)

Not sure about 8 of them, though...

Eddykilowatt, thankyou for your post, please tell me what was the speed of that fan (and correspondent voltage) for me to be able to extrapolate information out of it.
Thanks,

Using quality fans running at low speed, I've always found that it is the sound of air movement, rather than the fan itself, that makes the noise. A well-designed system should make the most of minimal airflow.

0dBA is not silence (-10dBA is quieter, for example), but it is approximately the threshold of human hearing. However, even in SPCR's anechoic chamber, ambient is plenty loud enough to drown out 0dBA. So, the answer to your question is, quieter than everything else in the room. Unfortunately, that sound level is different for everyone, and depends on the time of day and time of year.

At night, my practically inaudible system becomes audible if I listen for it. During the summer, I open the windows and the sounds of the neighborhood drown everything else out.

Search Bluefront's posts on this forum. It'll help you.

Thanks all for the input, it helped, I'm going to make some experiments, investigate more and then process it altogether, and then finally build my system that has been waiting for long already. Some time from now when I come to a conclusion, if relevant, I'll post about it.

i just bought a big passive cooler, stuck it onto my core2duo (now it could be a core i5), and put it into enough space that the hot air can move away. works 24/7 without any issue.

passive cooling is not that complicated. no fans at my home. hate them. okay still one (and one in each laptop).

What cooler?

according to my online shop of choice, it's the "Thermaltake CL-P0323 Sonic Tower Rev. 2"

That is an impressive looking cooler.
http://www.overclock.co.uk/product/Ther ... _1185.html

To bad it isn't available in the US

it isn't? wtf.. that sucks.. it's a cool cooler indeed

A couple years ago,a poster,Felton Carbon,did some neat builds using a 250 mm fan. He had it spinning at something like 300 rpm or less but moving air. Rigged up a big filter for it.

I thought about a custom rig where there would be a 250 exhaust on the BOTTOM below a horizontal motherboard with the PSU fan also helping cool the CPU Heatsink .Naturally,there's more spacing between the mobo and the fan below,and there's a base to have the fan above any shelf. i'd create deflectors and ducts to make air do where I needed it. Hard drives would be in a second box at max distance (about 6 ft) on eSATA,though an internal SSD would be nice for a boot/OS drive. Getting the HDD's out of the main box means better circulation,and puts their noise more distant.
You can of course even splice extra wire on and put a quiet fan in the remote HDD case

I think it will never be silent with 8 fans even @300 RPM. In fact... over 2 fans, you can't have silence...

in fact, over 0 fans, you can't have silence

8 fans sounds like too many for any case, never mind their speed. imo, it makes more sense to half the number (or even less) and put them where they really need to be. With such slow rotation, the effective distance of the fan's cooling reach is very short.

For example, there is only one fan in a couple of the SPCR lab PCs, and it is right on the heatsink, blowing out toward the case back vent. These fans run ~500rpm and are inaudible from more than 2' away -- even in a highly reflective (acoustically) converted kitchen.

ONE single underclocked nexus fan is audible to me up to about 4 meters distance.
Go figures...