Six 120mm fans compared by MadShrimps

wundi · Post by **wundi** » Tue Jan 10, 2006 3:06 pm

120mm Fan Roundup @ MADSHRIMPS

I figured this roundup might be of interest to this forum.

Fans included are: Antec Tricool / Coolink SWiF-1201 / Papst 4412 F/2GLL / AcoustiFan AFDP-12025 / Titan TFD-12025SH12C / Arctic Fan 12
No Nexus, unfortunately.

Their test methods and conclusions are actually quite reasonable, much better than I would've expected from a "usual" HW site anyway.

But that 32.5dBA ambient noise is just absurd..

jmke · Post by **jmke** » Tue Jan 10, 2006 3:19 pm

not really "absurd" at all my friend

you see, to get <30dBA readings you have to be in a soundproof room, and to get below <20dBA were talking about rooms designed for recordings (like special recording studio's and such).

I know at SPCR the sound recordings are done differently so don't compare the results per dBA here and over at Madshrimps, rather compare the result relative to one another at each site. So the papst scores 34dBA@5v, that's ~2dBA over ambient at 50cm.

now about the fan's motor noises, if I put my EAR next to each fan, I can hear it humming, and the tone (pattern of sound?) is different for each fan, but at 20-30cm that noise is completely gone (as in: not audible). so unless you sleep next to your fan @ 10cm, each fan which scored 34-35dBA = silent.

I have also received suggestions for including other fans; my TO DO list can be found here: http://www.madshrimps.be/forums/showthr ... adid=20600

please don't hesitate to suggest more

Other fans sugested so far:
- Nexus D12SL-12
- Orange Yate Loon
- Vantec Stealth
- Silverstone FM121

if you have suggestions on how to improve testing, add more information, or just want to say what "sucked" about it, I'm all ear.

Felger Carbon · Post by **Felger Carbon** » Tue Jan 10, 2006 3:54 pm

jmke wrote:I have also received suggestions for including other fans; my TO DO list can be found here: http://www.madshrimps.be/forums/showthr ... adid=20600

please don't hesitate to suggest more

if you have suggestions on how to improve testing, add more information, or just want to say what "sucked" about it, I'm all ear.

I own four 120mm fan models that I think deserve to be added. All fall in the ~40-48CFM range at 12V:

SilenX "11 dBA" 1200RPM nominal
Global Win ceramic bearing NCB 1202512L (Coolerguys.com)
Cooler-Master SUF-S12-EB (SVC.com)
Sunbeamtech Silent LED fan (SVC.com) <-- the sleeper! Lotsa CFM.

All four have very similar sound levels. None have obvious mechanical bearing noises.

You might also consider the Akasa AK-183-L2B ball bearing fan.

kenyee · Post by **kenyee** » Tue Jan 10, 2006 3:58 pm

also the Akasa Amber fan in the 120mm size. The smaller ones seem noisier than the competition according on some reviews on this forum. The 120mm version supposedly pushes more air at the same loudness as the Nexus...

frankgehry · Post by **frankgehry** » Tue Jan 10, 2006 4:08 pm

jmke,

Would you provide a link for the SmartSensor SL4001A. I haven't been able to find anything about it. Thx - FG

jmke · Post by **jmke** » Tue Jan 10, 2006 4:30 pm

I bought it in December 2004 from eBay, from somebody working for Inesun; http://www.inesun.com

but it seems they no longer carry the product, and the product does indeed not pop in Google searches

eBay Item Won! NEW-Electronic Digital Sound Level Checking Meter (Item #3859500823)
Dear madshrimps,

You have committed to buy the following eBay item from surptime (Q Tang - ShenZhen, GuangDong China) on Dec-13-04 13:25:00 PST

wundi · Post by **wundi** » Tue Jan 10, 2006 4:45 pm

jmke wrote:I know at SPCR the sound recordings are done differently so don't compare the results per dBA here and over at Madshrimps, rather compare the result relative to one another at each site. So the papst scores 34dBA@5v, that's ~2dBA over ambient at 50cm.

I understand that comparing dBA values directly between different sites isn't very practical. And given the higher ambient noise than eg. in SPCR labs, and smaller measurement distance (30cm, not 50cm I think?) than the usual 1m, the noise will rise over ambient level roughly the same amount, I think in the end it all makes pretty good sense. I probably said that in a pretty complicated way but never mind.

jmke wrote:if you have suggestions on how to improve testing, add more information, or just want to say what "sucked" about it, I'm all ear.

Well here goes.

I'm not sure what voltages the Antec Tricool presets are configured to, but it might make a fairer comparison to set it at full speed and adjust the voltage to identical levels as the other fans.

Also, I'm slightly confused by the "PWM" values in the charts. What are they? I don't think it was mentioned in the review. I know what PWM stands for but these values seem to follow the CPU temps quite closely.

Perhaps some more accurate subjective analysis would also be nice in the next revision of the roundup. Especially for the SPCR crowd the slightest difference between two quiet fans could mean a lot. Simply measuring close to each other doesn't mean they sound the same, even from a distance...

Other than that, it's a very nice roundup and even gives SPCR a little competition. To think about it, it's kinda weird but funny to comment on it here on another site's forum.

jmke · Post by **jmke** » Tue Jan 10, 2006 4:56 pm

good catch on the PWM, I've included more information on the last page about this;

"PWM temperature through SpeedFan, this represent the area around the CPU socket, the power management caps which you see on a motherboard, they are there to make sure the power which is fed into the motherboard coming from the PSU is filtered and delivered to the CPU and other components. Too high temperature will cause Vcore fluctuations which in turn causes system instability."

I'll check up on the little Antec Tricool presets, to see how they are configured;

when you say subjective analysis.. are you referring to the "put fan next to ear" part? because at at the lowest of noise levels (the one where SPCR readers are most aiming for) they all sound the same, or better, they all lack noise:)

at 1 meter I could not hear if the system was running, or turned off.

wundi · Post by **wundi** » Tue Jan 10, 2006 5:11 pm

jmke wrote:when you say subjective analysis.. are you referring to the "put fan next to ear" part? because at at the lowest of noise levels (the one where SPCR readers are most aiming for) they all sound the same, or better, they all lack noise:)

at 1 meter I could not hear if the system was running, or turned off.

Heh, yeah I probably couldn't tell the difference either unless I was in a very well soundproofed room together with the fans to be compared. But hey, the crazy person isn't the one who asks, it's the one who delivers.

Then again the cold fact is that if you have 5 of the quietest fans, of different brands, all running at 5V, still they will sound different to some people in their various environments.

frankgehry · Post by **frankgehry** » Wed Jan 11, 2006 12:55 pm

I would like to see 120x32mm and 120x38mm fans tested. The silverstone fm-122 is interesting (120x32mm, 9 blade) - http://www.silverstonetek.com/products-fm122.htm

and the panaflo FBA12G12L (120x38) would make for a good comparison.

jaganath · Post by **jaganath** » Wed Jan 11, 2006 1:15 pm

the cold fact is that if you have 5 of the quietest fans, of different brands, all running at 5V, still they will sound different to some people

I can guarantee that they will sound different. One fan may be doing 1500rpm at 5V, and another might be doing only 780rpm; the only true "apples to apples" comparison is to measure sound emission from the same size of fan at the same rpm. This is the problem with measuring the performance of fans that don't come with a tacho signal; I don't know of any way to detect the speed of a fan if it doesn't report rpm. Anyone know?

Aleksi · Post by **Aleksi** » Wed Jan 11, 2006 1:20 pm

One way (OK, this isn't something you can just take from your top drawer) is to use a strobe light. You set it behind the fan and start increasing the frequency. When it seems the fan blades are still, you can count the RPM from the frequency.

jaganath · Post by **jaganath** » Wed Jan 11, 2006 1:28 pm

Alternatively, you could colour one of the blades bright yellow (or orange or black or whatever, anything that stands out), and then mark a point on the frame, and count it manually? However this almost certainly wouldn't be possible with rpm higher than about 1000-1200, that's more than 20 revolutions per second!

Aleksi · Post by **Aleksi** » Wed Jan 11, 2006 1:37 pm

Jaganath,

Sorry, but being finnish I really don't know when someone is being sarcastic. I hope you were...

Anyway, I actually know people who deal with fans and use that method if they need to count the RPM without the tacho. That's how I know of it. I guess it could be an easy solution for an internet review site.

jmke · Post by **jmke** » Wed Jan 11, 2006 2:41 pm

so you want me to stare at a strobe light for extended period of times.. uhm.. what if I just use my anemo meter?

jmke · Post by **jmke** » Wed Jan 11, 2006 4:58 pm

frankgehry wrote:jmke,

Would you provide a link for the SmartSensor SL4001A. I haven't been able to find anything about it. Thx - FG

Don't know if you are also the same person asking the same question at the madshrimps forums:) but this might be information for more people

http://www.madshrimps.be/forums/showthr ... post116618

it's the Smart Sensor AR-824 (AR824);

frankgehry · Post by **frankgehry** » Wed Jan 11, 2006 5:19 pm

Wasn't me, but thx. - FG

There is a B&K 2204 on ebay with a starting bid of $295.00. It should be interesting to see what it goes for.

Aleksi · Post by **Aleksi** » Wed Jan 11, 2006 11:38 pm

the strobe was the first thing that came into my head after reading Jaganath's question. I don't see any other functional ideas posted?

jmke,

anemometers aren't exactly best suited for axial fans (it needs laminar airflow which isn't exactly the case with fans). But, in need of better funds that's probably the best tool for giving you hint of the real airflows.

It would be nice to see a fan review where you would have

- free air airflow measured with fans at the same RPM
- these fans run at this same RPM when fitted into a restrictive environment to give hint of pressure capability.

frankgehry: I'm pretty sure people always need good anchors, the price might go high.

jaganath · Post by **jaganath** » Thu Jan 12, 2006 12:52 am

Sorry, but being finnish I really don't know when someone is being sarcastic. I hope you were...

Actually I wasn't, but I do realise that trying to count the revolutions visually is impractical beyond very, very slow fan speeds. I was thinking you could attach a small magnet to one of the fan blades and then have some kind of device which records how many times the magnet passes in a given time period? There must be a better/simpler way of doing this. What signal does the tacho signal in a fan actually report? Is it the revolutions of the motor?

Aleksi · Post by **Aleksi** » Thu Jan 12, 2006 4:26 am

I think the common way to create the tacho is to use a hall sensor that usually gives two pulses per rotation.

Tibors · Post by **Tibors** » Thu Jan 12, 2006 8:39 am

Somewhere burried in the SPCR forums there is a thread which describes/links to some electronic circuit that can generate a tacho signal for two wire fans. IIRC the fan motor generates some noise on the 12V line, which the circuit can use.

jmke · Post by **jmke** » Thu Jan 12, 2006 9:38 am

somewhere not burried at all, this guide on adding RPM monitoring to fans that don't have it.

http://www.madshrimps.be/?action=gethow ... howtoID=26

tested on Papst fan back then;

checked with the Antec Tricool, I do see 3 connections, so very possible to have one small 3-pin connector with one wire wired to that 3rd connection to provide RPM

jaganath · Post by **jaganath** » Thu Jan 12, 2006 10:49 am

Wow, what a great guide JMKE! Thanks! That was just what I was looking for!

Only for people who are skilled with a soldering iron though. So do you think most 2-pin fans have the extra RPM connector on the PCB?

jmke · Post by **jmke** » Thu Jan 12, 2006 11:21 am

jaganath, would not dare to give you an answer on that; with all the different models out there.

Tibors · Post by **Tibors** » Thu Jan 12, 2006 5:50 pm

It wasn't that burried after all. Just searching for RPM and circuit found it.
http://forums.silentpcreview.com/viewtopic.php?t=3565
But the pictures of the original design are lost. On the third page there is an alternative.
http://forums.silentpcreview.com/viewto ... 218#119218

jmke · Post by **jmke** » Fri Mar 24, 2006 6:43 am

I've been absent for a while, but not idling, finished comparing 17 fans and results can be found here: http://www.madshrimps.be/gotoartik.php?articID=421

I tried to make it as complete as possible, dBA measurements, recorded noise samples, CPU temps and subjective noise tests. These are the fans I compared head to head:

AC Ryan Blackfire4
AcoustiFan AF120C
AcoustiFan AFDP-12025
Aerocool Turbine 1000
Arctic Fan 12
Coolermaster ALU AAF-B12-E1
Coolermaster LED Silent TLF-S12
Coolermaster Ultra Silent SAF-S12-E1
Coolink SWiF-1201
GlobalWin 1202512L
mCubed X12
Nexus D12SL-12 (= Yate Loon)
Papst 4412 F/2GLL
Scythe S-FLEX SFF21D
Scythe S-FLEX SFF21E
Scythe S-FLEX SFF21F
Spire FD12025C1E3

your input, as always, is appreciated, the sound recordings are experimental and nowhere near the accuracy of MikeC's excellent work.

I was most surprised by the Globalwin and the Scythe series, their bearings are really what set them apart from the rest for the extreme silence situations. The Nexus and Coolermaster Ultra Silent do well also in that department but not quite as good. The mCubed X12 proved to be very versatile, as it kept performance up even at lower voltages and low noise.