2 Fans with same Noise Level, more noise ?

Status
Not open for further replies.

ytka

Distinguished
Aug 9, 2002
6
0
18,510
Will 2 fans, with noise level 32Db both, make more noise then 1 fan with noise level 32Db ? I mean will noise level increasing of more fans with the same noise level or it will be same level. (Actually, i was not good in physics classes :) )

* CPU not found. Press any key to continue... *
 
Somebody posted on another thread that having 2 30db fans would end up being 32db of noise, so there is a small increase. It shouldn't be too much of a difference.
 
Don't know the exact noise levels but it's much more sensible to have say 4 x 25dB fans than 2 x 50dB fans.

<b><font color=blue>~ Gotta question? Tried searching the boards first? Good! Ask away! ~<font color=blue></b> :wink:
 
Each fan increases the noise level by 10%. Two 32dB(A) fans would create about 35dB(A) of noise.

There is some question about whether it increases by 3dB(A) every time, regardless of the original fan in question, but I'm not totally sure. The 3dB(A) figure came from an overclocking web site. But to me it wouldn't make sense that if you hypothetically had two 2dB(A) fans the noise level would end up being 5dB(A).

Regardless, adding more fans of the same noise output only marginally increases the combined noise. Two 30 dB(A) fans do not create 60dB(A). What does increase in equal units is the sound pressure, or radial kinetic energy from the spinning of the fan. 10 fans of the same speed and size will create 10 times the vibrations as just one of those fans. Since this sound pressure (or vibration energy) can also be a catalyst for overall computer noise this is important to keep in mind.

When looking at fan noise ratings, my general rule of thumb is that the actual noise is about 1.5x-2.0x the rated noise. This is done to account for the noise that is caused by the above stated sound pressure.

1° of separation between my monopoly and yours. That's business with .NET
 
OK guys you're confusing a fairly straight forward problem. If you have 2 identical waves moving through a medium in phase the net result will be exactly doubled. Therefore 2 identical fans is basically twice as loud as one and to double it again would need 4 fans. Back to decibel, I won't go into great detail here but 0 dB is still noise, to go to no noise would be -infinity dB. The equation is a logarithmic scale. In short the net result of one fan at 32 dB and two fans at 32 dB is an increase of 3 dB or double the sound pressure, twice as loud or 35dB. For more details see,
http://www.phys.unsw.edu.au/~jw/dB.html

😱 Who needs heatsinks and fans, I have an igloo 😱
 
Sounds good. I've heard "logarithmic" scale but didn't know exactly what it was (not the best at math). The data I got was from overclocking sites. They say that silence is actually 20 dBA right?

The actual importance is how loud it sounds to "us". Based on the logarithmic scale it increases a flat 3 dB(A) each time. Then the question is, what is quiet and what is loud? What frequencies bother us? I work with practicality and experience and not math, which is why I haven't paid much attention to this. The dB(A) scale was created for human ears, so while two identical sounds will double each other under the dB(V) scale I would imagine it would mean something different under the dB(A) scale. Is 30 dB(A) tolerable? I wrote a post a while ago outlining the difference in dB(A) and dB(V), but frankly, I'm not smart enough to do analysis. :smile: Anyone care to try?

1° of separation between my monopoly and yours. That's business with .NET<P ID="edit"><FONT SIZE=-1><EM>Edited by cakecake on 08/11/02 11:05 PM.</EM></FONT></P>
 
The difference between dB and dB(A) is a filter because the human ear picks up middle frequencies better than low or high ones but won't matter at all in our example because the fan frequency is right in our range.
0 dB Threshold of hearing for healthy youths
10 dB Quiet recording studio
20 dB Quiet living room
30 dB Quiet office
40 dB Quiet conversation
50 dB average office noise
60 dB Conversational speech from 1 ft.
65 dB Busy Street
75 dB Average factory
80 dB Typical stereo listening volume
100 dB Subway
110 dB Power Tools
135 dB Threshold of pain
So there you have it. It does make more sense and every 3dB increase is twice as loud!! Complete silence to us is 0dB, complete silence is actually -infinity dB.

😱 Who needs heatsinks and fans, I have an igloo 😱
 
Wow that's amazing.

I forget what the noise levels were but I think in some instances of extremely loud noises (like bombs exploding) people's ears can actually explode and bleed out.

I would probably aim for somewhere around 10dB. The way I've figured I would do this is by using a water cooled:

1)CPU
2)VPU
3)PSU

And setting up my computer against a wall and drilling holes in the wall so that the pump, radiator + fan, and hard drive can be placed on the opposite side of the wall, thereby isolating the computer completely from the noise. The pump, radiator, and hard drive could even be placed outside for good cooling.

I read an article recently about how people cheat on the ANSI tests for noise levels for their fans, hard drives, and just about anything that people don't really want to hear. My guess so far has been that the average person has a 40-45dB computer--average means they have a stock cooler, a PSU fan, a VPU fan, and a hard drive. Obnoxiously this goes way past the specs of "ultra quiet" fans so I guess we now know why. The main problem is that there's currently no way to standardize sound pressure.

1° of separation between my monopoly and yours. That's business with .NET
 
ya 40-45dB is probably right for stock (1 HSF and 2 case fans). My own system (2 HSF, 2 VCF, 4 case fans) is probably around 55dB but could be as high as 65dB. My personal opinion is that if it doesn't hurt my ear drums when I turn it on then it's ok. So I can add alot more fans yet!!!

😱 Who needs heatsinks and fans, I have an igloo 😱
 
Two fans DO NOT sound twice as loud as one fan. (Try it. It takes about 5 fans to double the sound).

Audio perception is not as simple as sound pressure level. See article below.

<A HREF="http://hyperphysics.phy-astr.gsu.edu/hbase/sound/loud.html#c2" target="_new">http://hyperphysics.phy-astr.gsu.edu/hbase/sound/loud.html#c2</A>


<b>I have so many cookies I now have a FAT problem!</b>
 
cakecake, when we sheetrocked my room we made a hole in the wall so I can now run my computer in the outer room and the monitor, KB and mouse inside. I only have to go out there to switch cd's and stuff.

Extremely quiet.

In the winter, I let that room get to about 5C and have a 22" fan blowing into the case. My processor temps have been as low as 15C I think. This year I have a volcano 7+ and some arctic silver 3, can't wait for it to get cold out again.
 
The article said it wasn't completely conclusive as to how actual sounds affect signals sent to the brain. I wonder if the variety of different noises affects us as much as the actual volume, and if so by how much? For example, would a person be irritated more by having 8 fans of the same pitch or 4 fans of the same pitch and 1 hard drive? My guess would be that having a variety of noises would be more irritating than having sounds of just one type.

Bardic - you got great low temps there it sounds like. A computer's dream home. Some places never cool down, like SE Asia and Africa.

Maybe one day I will be able to make this computer system I'm thinking about. Having the CD/DVD-ROM outside might be a little bit extreme but probably I'd want everything else outside. Maybe I can put the entire case on the other side of a wall or something and thread the monitor, mouse, and keyboard through like you did with your own system.

1° of separation between my monopoly and yours. That's business with .NET
 
The article said it wasn't completely conclusive as to how actual sounds affect signals sent to the brain. I wonder if the variety of different noises affects us as much as the actual volume, and if so by how much? For example, would a person be irritated more by having 8 fans of the same pitch or 4 fans of the same pitch and 1 hard drive? My guess would be that having a variety of noises would be more irritating than having sounds of just one type.
4 fans and a hard drive? Don't know about that. This Maxtor hard drive is nearly silent. I can't hear it over one case fan.

Fan noise is really interesting. I just re-did a quick experiment and hooked up 4 case fans (all I have on hand, 28 dBA I think). At first I thought I was wrong. 4 sounded more than twice as loud as one but that was with my ear about a foot away. At normal distance, about 3 feet, 4 fans sound nearly the same as one. Not sure how it would change in closed case. (I can only fit two in my case).

Next I turned on a CPU fan, a 42-46 dBA 60mm Delta. With that high pitched whine I can just barely detect the sound from the 4 case fans. The Delta is just so irritating. It's the high pitch.

My next impromptu test was my current CPU fan, a 70mm, 32 dBA, Evercool fan. This sounds about 2/3 as loud as the Delta when both are auditioned alone. Put it together with the 4 case fans and the whole combination only sounds half as loud as the 4 case fans + delta. The Evercool kind of just blends in with the case fans but Delta drowns them out. It's loud!

Did you ever have a small fan like a chipset fan or a graphics card fan start to fail? They make that tiny whine yet it rings out like a shot over the louder, more powerful fans

It's strange how noise works.



<b>I have so many cookies I now have a FAT problem!</b>
 
I think psykoikonov's figures are fairly close, I've seen similar lists in the past. Based on the numbers I'd say that the reason twice as much sound pressure doesn't equal twice as loud by human perception is because our hearing is logarithmic. That is why we use dB scale (and a dB scale makes sense) and not a straight correlation to the amount of energy present in the sound.

Maybe there is a doctor in the house that can confirm/deny...
 
Oh right, my point was: to get twice as loud to humans there must be a doubleing of the number of dB's.
 
I think psykoikonov's figures are fairly close, I've seen similar lists in the past
I wasn't arguing against his reference list nor his math only the math does not take into account perception.
I'd say that the reason twice as much sound pressure doesn't equal twice as loud by human perception is because our hearing is logarithmic. That is why we use dB scale
The reason we use a logarithic scale is to simplify the numbers.

From psykoikonov's link..

<i>The ear is capable of hearing a very large range of sounds: the ratio of the sound pressure that causes permanent damage from short exposure to the limit that (undamaged) ears can hear is more than a million. To deal with such a range, <b>logarithmic units are useful</b>: the log of a million is 6, so this ratio represents a difference of 120 dB.</i>

The decibel scale is just a representation for comparing two values (any values and doesn't have to be sound) that can differ largely. The scale itself does not take into account loudness perception. That, as you say, is also a logarithmic function but it is instrinsinic to human hearing and separate from the decibel scale that was arbitrarily chosen to compare sound pressure levels.

I thought my link explained the problem quite clearly.

<b>I have so many cookies I now have a FAT problem!</b>
 
Missed your second reply (because it was to yourself).

No, you don't double dB either.

According to my link, with some sounds (many), 10 dB would be a doubling of perceived sound. However, as psykoikonov stated, 3 dB is mathematically a doubling in acoustic energy. I've seen other articles where the perceived doubling requires only 6 dB.

Judge for yourself. Take some fans and decide how many seem to double the "loudness" of one fan. I say four or five. It's definitely more than two. Try to do the test in quiet room with only fans making sound.

It's not scientific but it should prove my point.


<b>I have so many cookies I now have a FAT problem!</b><P ID="edit"><FONT SIZE=-1><EM>Edited by phsstpok on 08/16/02 07:24 PM.</EM></FONT></P>
 
what can I say you are right, I am wrong. thanks for setting me straight.
 
I checked your link and aside from being a little vague about a few points it does make sense. It boils down to it takes about 5 fans from the distance we are talking about not 2. The closer you get the less fans you need and vice versa for more fans (or noise), also for noise to add "idealy" you need the same frequency characteristics in the components added. I haved looked for more answers but the only real way to test is with your own system in an anachoic chamber, (did I spell that right) and even then that is not the human ear. The same as I need a wind tunnel to test how well my fans actually do there jobs.

😱 Who needs heatsinks and fans, I have an igloo 😱
 
I think my simple test is enough.

I'm sure there is much more at work here. It seems the article describes sounds that are simpler than the noise of a fan. The comment that 10 violins sound about twice as loud as one violin is misleading for our purposes. I think 10 fans would seem to be much more than twice as loud as one fan. I'm guessing that the fact that the sound of a fan is really made up of multiple frequencies and multiple intensities (noise) contributes to the different result than the violin example.



<b>I have so many cookies I now have a FAT problem!</b>
 
In my own system, my hard drive is the loudest component. If I turn off my hard drive by going into standby mode, all I can hear is a soft whoosh of air that is actually very therapeutic. :lol: This therapeutic property is actually very important to what I'm suggesting.

What I'm talking about is the subjective ability of us to focus on specific sounds and identify where they're coming from. Let's say the human brain is some RAM in a computer (simplified for my analogy). If you have 8 fans sounding the same, the brain is only running 1 process and 8 threads. If you have 4 fans and 1 hard drive or more fairly if you have 6 fans and 1 hard drive, the brain is running 2 processes: 6 threads in one process and 1 thread in the other process. Now this needs scientific evidence of course, but until then we can share subjective opinions: does processing for sound differentiation require more brain power than processing for distinguishing volume? My answer in my own opinion/experience is yes. And more brain power means would mean that it's more tiring to listen to. I'm theorizing that it's better to listen to 8 fans of the same type than it is to listen to 6 fans of the same type and 1 hard drive.

To put this in perspective, ask yourself why does a sleeve bearing fan that runs out of oil bother you? Imagine if you have 6 sleeve bearing 80mm DC fans running at once on medium speed. Now imagine just one of those suddenly starts losing oil and makes slight scratching sounds. Why does this bother you? Again, my theory is that the human brain was not made for volume as much as it was made for noise pitch and differentiation. Ask yourself too, when listening to a wide orientation of sounds, why does a person become disoriented? If you were to suddenly be drowned by noise in the next 5 seconds, would it be more disorienting to listen to 8 of the same exact sounds totaling 40dBA or 8 different sounds totaling 40dBA? (All of these scenarios require the noise levels to be the same so they do not apply to computers where the noise level is so high that noises drown out other noises i.e. fans that drown out hard drive noise - referring to your system, phtsstpok, but no harm intended.)

Part of my "experience" as I describe it above comes from playing in orchestras and symphonies, where you learn that sounds can be pleasant and also wretched. I have played in 4-violin concertos as well. When too many instruments are playing different sounds at once, it's very easy to become confused. This can happen when you have 4 violins all playing with different length, pitch, variation (vibrato), and bow speed. The opposite is when these same 4 violins play exactly the same tune at once, which is one of the distinguishing characteristics of a professional symphony or philharmonic orchestra. Orchestras where the 1st violins all play the same way sound better than orchestras where the 1st violins are all trying to solo and stand out from one another. So already we can see evidence that duplicate sounds are more pleasing to the ear than sounds that are different or even varied in minute ways. How sensitive the brain is to this I won't even get into. 😎

Here's another thing to ponder. Often I have had to tune my violin. Question: How do humans tune violins, pianos, and other string instruments? Answer: We do this by harmonizing--playing two strings both at once and adjusting the string's tightness until they "agree". On a violin it would be A first then AE together, AD together, and DG, instead of just A, E, D, G. Remember in music class where your teacher taught you chords and how in medieval times they called some chords "good" and some "evil"? I'm suggesting that computer noise needs to harmonize, and that the easiest way to do that is to eliminate the number of different noises. For example if you have the chance to use a total of 6 fans, you wouldn't use 3 panaflos and 3 enermaxes. You'd use 6 panaflos or 6 enermaxes.

When quieting down our systems, we should be paying more attention to just dBA. We should be paying attention to the number of different sounds we are creating in our computer. Finally, as I said in another post, there is currently no standardized test for sound pressure (vibration energy), only sound power, and maybe we need a standard.
 
OK, you lost me with the violin example.

I think the article which I linked to earlier explains differing sounds quite well. The loudness appears additive when two sounds are not in the same critical range. When they are in the same critical range, such as two nearly identical sounds, then two of them don't sound doubly loud.

I'll have to give some thought about using all identical fans. Right now I have two case fans, different models, and an entirely different 70mm CPU fan. This combination is loud but just barely exceptable to me. Now I wonder how my system would sound if I had only one case fan (more powerful) and an identical fan for my CPU.

There is another consideration to this idea. Blowing air through the heatsink completely changes the sound signature of the fan. So I wouldn't end up with two identically sounding fans.

I'd still like to give it a try but I don't have any powerful case fans on hand nor an 80mm fan adapter for any of my heatsinks.

Phsstpok

By the way, for people with cheap cases (like me) that have those little punched holes for a fan grill instead of a wire grill you can reduce fan noise a lot by cutting out a real hole. Cuts down noise by at least 25% and improves cooling to boot.



<b>I have so many cookies I now have a FAT problem!</b>
 
I agree with cakecake's logic. It is true from my expierence as well, different sounds bother different people and more different sounds seem to bother us more as well.
I currently run 2 identical 80X80 case fans and 1 70X70 HSF, it purrrrrrrrrs like kitten and CPU no load is 42C,perrrrrrrrfect(ambient room temp. 20C).


😱 Who needs heatsinks and fans, I have an igloo 😱
 
Status
Not open for further replies.