This is actually a transplant of a mini-thread started by CRT_Leech in the Hard Drive Forum. I'm reposting it here, as suggested by Crisspy to see if anyone else has any opinions they'd like to weigh in with.

CRT_Leech's original comment was:




To which I responded:



Anybody have an expertise in this area?

that's good idea and all but for us, i don't think that would be practical...

firstly, only noise you'll be cancelling would be the noise directed at the cancellation device. any noise directly directed outside of the case will be emitted freely such as noise from psu, air turbulance noise from exhaust fans, etc.

secondly, that is with the assumption that the device would work perfectly.

as of the fan solution, his best bet really is panaflo @ 5v. i thought of every possible configurations and let my imagination wonder freely, but that seems to be the most logical and practical solution at the moment.

post directed to that crt guy.

hehe my bad rusty... didn't realize you asked for an "expertise". but hey! it's america. over here anyway

Hello,

On the subject of noise cancellation, I just wonder if the Sony noise cancellation headphones (MDR-NC11) will work with PC noise. It is said to work best between 50Hz - 1,000Hz, is this in the range of PC noise is?

Sound wave experts let me know..

If this works, I can possibly sleep in peace, without a PC humming away.

I am not a "sound expert", but I can comment a bit from memory of my days of studying DSP and communication systems as an EE Masters student (having defected from physics and math as an undergrad). I'd say the idea is sound (ok, bad pun) but the practice is where you run into problems. The problem is that sound can come from any direction (in general), and to cancel out a particular sound you need your 180-degree-out-of-phase cancelling waveform to come from the same direction (or close to it). Or, more generally, the point in space at which the sound is received is where the cancelling wave needs to be.

You'll notice in practice that noise cancelling systems, like high-end head phones (and I think some luxury car audio systems), are taylored to a specific environment. They rely on the fact that your ear is located in a well-known position, so they know where to sample the ambient noise and direct the cancelling sound.

The trick is to get the cancelling wave to your ear and have it match (minus the phase inversion) of the incoming noise. You could put a microphone in your case and send the inverted wave out of the speakers. Assuming the phase inversion processing is fast enough (which is a separate issue), I suspect you'll still run into the problem that if you move your head around much the cancelling will stop working. Or if move the speakers (or mic), it may not work as well. This is rather analagous to the 3D parallax barrier displays that have a "sweet spot"; you move too much to either side and the effect stops working.
Also, you'd have the issue that the noise the microphone "hears" inside the case is probably going to be different from what the user hears outside the case, which means the cancelling wave isn't going to be an exact (opposite) match. I would guess the best scenario is to have the mic at the user's ear, or at least somewhere in the direct line of the noise source (maybe on the desk between in the user and computer?).

I don't think an array of microphones would help much here, except perhaps to have two for stereo cancellation (one for each ear). The array would likely get noise from different sources that won't match what the user hears. A single microphone (per ear) should work just fine IF it's sampling the noise the person is hearing.

My two cents is that reducing the noise produced in the first place is going to give you much better returns on the invested money and effort. But this could still be a fun project if someone wants to give it a try. Or maybe just try some noise-cancelling headphones (I haven't tried them myself, so I don't know how well they work).

Hey, I just had an interesting thought along a slightly different line of thinking. I knew an EE/music double major in grad school who built his own electrostatic speakers. For those who don't know what these are, take a look at

http://www.audiovideo101.com/dictionary ... peaker.asp

Anyway, here's where this is relevent to the current discussion. The speakers he built looked like plexiglass sheets sandwiched by wire mesh. I was just thinking about the new acrylic computer cases you can buy. What if instead of acrylic panels, each side of the case was an electrostatic speaker? And furthermore, you had a microphone just inside each case side that recorded the interior noise at that point and then sent a phase-inverted copy to the corresponding speaker (nevermind what's required to the audio processing for now)? Now you could have the case itself "broadcasting" an anti-noise wave to try to cancel the noise from the components.

Of course, this wouldn't be very practical for a number of reasons (including some of the ones in my previous post). But I just thought it was an interesting idea that I'd throw out.

I did not have any intention to attach speakers to them or modify it any way to make it work with PC to cancel noise. The headphone I am talking about has built in mic, and I just wanted to wear the headphones as it is, when I was sleeping. Would it cancel out the noise of a PC humming away?

I have read many reviews of this unit and most say it works wonders on trains and planes. Obviously there was no mention about PC noise, thats not there intended purposes of this unit. But I just wondered if this unit will cancel out the noise from the PC. Is the 50-1000Hz frequency in the range of PC noise?

to be honest, only viable application of this technology for us would be an earphone with the noise cancellation device.

since sound will be pretty much unidirectional as entering the ear via earphone, earphone can take the sound out....

then again, i doubt ANYBODY in here can hear panaflo @ 5v with simple 5cents earplugs on.... maybe i'm wrong

well i don't see this being done to any of our systems anytime soon.

to be honest, only viable application of this technology for us would be an earphone with the noise cancellation device.

since sound will be pretty much unidirectional as entering the ear via earphone, earphone can take the sound out....

then again, i doubt ANYBODY in here can hear panaflo @ 5v with simple 5cents earplugs on.... maybe i'm wrong

well i don't see this being done to any of our systems anytime soon.

to be honest, only viable application of this technology for us would be an earphone with the noise cancellation device.

since sound will be pretty much unidirectional as entering the ear via earphone, earphone can take the sound out....

then again, i doubt ANYBODY in here can hear panaflo @ 5v with simple 5cents earplugs on.... maybe i'm wrong

well i don't see this being done to any of our systems anytime soon.

Why not just go down to the local hardware store and buy a pair of gun-muffs or heavy duty hearing protectors? Might be all that you need, and 1/10th the price.

Balgarion, that's an awesome idea. I laughed when I read it. Brilliant. You make it work, and I'll buy one.

Ah, sorry, I missed the original context. Oddly enough, I just noticed a "review" of noise-cancelling headphones today at http://www.barefeats.com/bose.html. They say that the $300 Bose QuietComfort headsets work really well, at least on airplanes. (For that much money, they'd better work well!) The author, later in the article, does comment that he used them to take a nap in a noisy "office/home" environment, although there is no specific mention of any computers involved.

As far as frequency response goes, an outfit like Bose might have specs available on their website. I believe good headphones can go down to around 5-10 (?) Hz or so, which is probably good for covering computer noise. I'd think the limiting factor on the frequency response would be the microphones they use. Of course, any audio device is going to have a gradual drop-off near its limits, so a device rated to 10 Hz, for example, might not attentuate (or emit) at as well as in the middle of its design range. But I'd think most computer noise, coming from fans, would be higher frequencies that should fall in the working range of a unit like the Bose. That's just my guess, though.

Say, wouldn't sleeping with bulky headphones on be problematic? Earbud-style units would be better, but I haven't seen any noise cancelling versions of those yet. I second the idea of earplugs or something similar; cheap and proven effective, with good frequency response. Or, if you have to wear the big over-the-ear headphones, you could hold them in place with a liberal application of duct tape.

On a less related note, I noticed that the Bose headphones in the review require two AAA batteries. I wonder if they could work up a kinetic drive system or a solar recharger like the newer watches... Having lots of gadgets that all require batteries is a real hassle.

The only thing that can work along these lines is a headphone type setup (and the Bose units used to cost a lot less, BTW -- like $70-80 IIRC). Because in order the cancel a noise, the "antidote" MUST be exactly 180 degrees out of phase, and since the sources of the various noises in a computer case -- are in various places, there is no practical way that a single device could cancel all the noise!

And as you move around, your ears change the distance from those noise sources -- and from the cancelation device, so again, there is no practical way that a cancelation device can work EXCEPT if it is attached to your head!

Case closed.

As an example of how hard it is to cancel sounds, try this: take two stereo speakers and wire one out of phase with the other, buy inverting the positive and the ground at ONE end or the other. Place them face to face, and play some "white noise" from between some FM stations, or play a mono sound sound source -- and anything that you can hear is what is not being canceled. This could be from any non-linear variations in the drivers, crossovers, or anything in the signal path, or any distance between the corresponding drivers. Remember that the wavelength of higher frequencies get pretty small. Bass is much easier to cancel, but the high stuff gets very tricky... Now pull the speakers 1" apart. It gets a lot louder, huh?

The Sony unit I mentioned is of the earbud type.

Oh by the way this unit cost around £100 in the UK, so I guess it cost about $100+ in the US.

Thanx for all your comments

I think Rusty missed my attempt at transplanting this topic over into Fans a few days ago. Maybe General Discussion is a better place for it anyways. After seeing what has been brought up so far about headphones, I find myself cringing at the idea of wearing ear protection when using my computer, odd that... I like the sound of silence. Anyways, here's what I've thought up so far.


Seems like ARM and Antec are getting nearly as quiet what we can fuss out ourselves at home, but the big problem is still how to push lots of air silently/quietly. Any thoughts?

To answer everyone's question. I have a noise cancellation headphone from Brookstone that I use in the office. It does a nice job w/ the printer hum and some other noice around the office, but not so will w/ the computer fan noise.

We have Dell Optiplex GX240's. It is a very quiet system already. With all the background noise around the office, I could hear the computer at all. I took my headphone place it next to the exhaust and there was no different between having it on and off. It could be that it isn't design to remove noise at that range. I might do better w/ louder fans? I have to take the headphone home and see if it will do anything w/ louder fan.

Another experience here... I've used Sennheiser and Bose noise-cancelling headphones, and they both do a good job of reducing fan noise. I don't know that they would make much difference now that I've replaced most of the noisy fans in my systems, but for your run-of-the-mill-noisy-fan-having PC they did a great job.

not possible to have a speaker fixed to the inside of one of the side panels? to create a vibration out of phase with the unwanted vibration..

dont the side panels act as amplifiers making them a major cause of noise?

I also tried a pair of the Brookstone noise cancellation headphones on my last job where I was working with large numbers of systems that were industrial rackmounts with multiple large fans moving lots of air and sound waves. Since this was a production floor, they weren't even as insulated as they would have been on a customer site. I don't know the exact Db level, (I asked, but nobody could / would tell me) but it was probably about as loud as one could be without OSHA requiring hearing protection. (I know that in the course of my working there, we introduced a MK II ventilation box that was much quieter, due to customer pressure)

I found that the Brookstone phones (which are of the 'Walkman' style small speakers with a metal spring headband) helped some. You could hear the difference if you turned them on and off, but the improvement wasn't a big one. I'm guessing about 6Db worth, it was easily noticeable, but nowhere near 50% reduction. It was almost marginal as to whether the advantage was worth the hassle of having to explain to every manager that walked by that I was NOT listening to a Walkman or equivalent...

Gooserider