Slashdot Mirror
Silencing a Hard Drive Using Household Items
Reader Justblair recommends his blog entry detailing how he made a hard drive silencer for a pittance. "This article demonstrates a very easy-to-make hard drive silencer that not only outperforms most commercially available devices, but is cheaper to implement as well. Requiring very little in fabrication skills, it is an ideal addition to a media PC or HTPC. It may even suit you if your head is aching after many hours of being whined at by your hard drive."
the fans are the noisiest part of my computer, and always have been on any computer ive ever had. However i do occationally hear the click-click of the head moving, but never the whine of the platters, will this sort that out too?
ive seen another hard drive silencing technique elsewhere that's even cheaper, although possibly not quite as effective, which is simply to mount it with rubber bands in a 5.25" bay rather than screws.
http://www.spodesabode.com/archive/content/article/hddnoise
He's lucky his drive lasted that long. I've yet to see a maxtor or a seagate inside of one of their enclosures last that long. Having taken them apart, I saw that the seagate one was completely covered, multiple times, with no airflow.
Those things get way too hot. My mom has a new hard drive (as of this summer) with three directories of files recovered from signatures. Nasty.
Drives should be covered with moving air. They should also be mounted to the ground plane (which is the PC case.)
There is a reason why Apple uses (used to use) FEA programs to design the cooling systems of their computers, and it is not marketing. In the good old days, you often found bad engineering practices in cheap PCs - such as the hard drive being screwed wrong side down to the chassis - and it was then not unusual for them to work OK as a desktop but fail quickly if used as a server, because the HDD was now actually doing some work.
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
I've got about 6 years worth of cat hair coating my hard drive and it's very quiet now... also heats my feet during the winter... you get used the smell after the first year...
Here are some more do-it-yourself tutorials about hard disk drive silencing techniques as well as about selfmade cooling techniques. The ideas are ranging from an acoustic cabinet, switching off the HDD when not in use to cure vibration (the main cause of noise) with some rubber and others.
I simply cut some pipe insulation foam in halves, and rolled both parts around the disk, one near the front side, one near the rear side. I used some electric wire around the foam to hold it in place. Now my 3"5 disk fits perfectly in a 5"25 slot in front of the case's fan, and the foam's thickness prevents it from moving. I can't hear it *at all* now, eventhough it's a SCSI 15k rpm, because the noise from the motor normally conducts through the metal and the fixations only.
It requires very little material, skills and time to do this, and the disk can be
extracted at any moment without hassle.
And 9 times out of 10, a person who can hear those things is in their twenties or younger. The ability to hear those very high pitched sounds goes away with age. Of course, there are exceptions. I'm in my 40's and can still hear them.
My old roommate had a television that emitted a horrendous (to me anyway) screaching sound which he had never noticed before somehow. He claims now that I've pointed it out to him it's unbearable to him too, so he gave the TV to his sister, and nobody in her family knows about it. I'm very curious if they can be "trained" to hear it as well.
Actually you can have a silent PC with silent fans. The standard is 20dB at one metre (3 ft). Those 3 feet are what enable a quiet PC to become a silent PC.
Turn just about any good fan down to 800rpm or less, and it becomes nearly inaudible in free air. Once you combine such a quiet fan with the PC's chassis resonance (on a good chassis), the chassis' acoustic properties will effectively shape the noise (like a bandpass speaker box). Some of that noise gets muffled internally, some of it gets dispersed at the vents, and ideally very little sound will reach your ears.
Making a quiet PC is easy, because off-the-shelf components have gotten very quiet over the years. Making a silent PC is more like building an awesome loudspeaker - there's a lot of planning, acoustic measurements and math involved to meet your sonic goal.
-Billco, Fnarg.com
The problem with that technique is that it won't dissipate heat very well. I had a similar setup until one day I checked the temperatures of the drives which were running at 50+C. Fans help a little but then you're negating the point of silencing the drive.
Drives dissipate heat through the sides which works best if there is a heatsink either in the form of a the metal computer case or some sort of harddrive specific heatsink.
That's a UI design issue. If there was an on-screen indication that "things are happening" or even a "magic keystroke" that overrides the normal gui and pop in and out of some kind of general system activity display in all cases where the kernel isn't frozen, then you wouldn't need to rely on design flaws of other systems to give you the necessary feedback.
But since that doesn't really exist, at least not conveniently, I've also used the HDD noise as a valuable diagnostic tool. Now, if only I'd bought better PSU and CPU fans, I'd be able to actually hear it without straining...
Can you be Even More Awesome?!
I did the same thing five years ago, at the height of noisy desktops. My bedroom computer is now in my closet, with the monitor/keyboard/speakers/mouse/etc on a desk on the other side of the wall.
The only problem is that my new computer is so quiet that the whole arrangement seems silly.
Hard drives get pretty hot, and high temperatures will shorten their lifespan.
Google did some pretty comprehensive testing and found this not to be true. The well cooled drives actually failed more than moderately hot ones (at really high temperatures the failure rate started to climb again.)
... also, I can kill you with my brain.
Embedding in gel looked like a pretty bad idea. Hard drives get pretty hot, and high temperatures will shorten their lifespan.
That's a very good point. The trick I've used to quiet ordinary 5400RPM or 7200RPM IDE and EIDE drives by wrapping them in plush carpet would be bad for a drive that tends to run hot and the carpet will not conduct the heat away from the drive to the case.
Some of the Miniscribe and Microscience SCSI drives I've had in the past ran *VERY*hot...I mean hot enough that they were uncomfortable to hold for very long while or just after they'd been busy for awhile. IRRC, it was the pair of old (brand new and state-of-the-art at the time) 9GB 10,000RPM Miniscribe drives I had in one machine that died of head crashes due to heat death. An engineer I had reason to believe told me that in order to cram 10GB of capacity onto what was then a very fast SCSI drive, Miniscribe had to use platters that were so large that their edges were almost rubbing the inside of the drive case. Apparently, Miniscribe didn't take thermal expansion into consideration and one many of these drives the platters would expand enough that their edges scrapped against the inside of the drive case, creating a fine dust which would eventually find its way between the heads and the platter surfaces, causing a head crash.
While I can't verify that explanation, it fits what I observed. One of the pair of drives was mounted directly over the other and it was from that drive I heard the distinct sounds and noticed the erratic drive performance that precede a lot of head crashes. The at drive died first in a very noisy way (I'd been making frequent made backups of both since I first heard the strange whining sounds), followed a couple of weeks later by the drive below it. The lower drive exhibited the same failure mode but died rather suddenly, unlike the upper drive which went though noisy death throes for many weeks.
That stands to reason as the lower drive's waste heat was rising and thus increasing the temperature of the drive above it. The upper drive was used as a "data" drive while the lower one held the WinNT OS and all the software programs.
I suppose the problem was partly my fault because I did nothing special to keep the drives cool since I had no idea they tended to run so much hotter than the lower performance drives I'd been using up to that point. Adding an extra case fan or even just a better case fan and separating the drives by an open bay might have kept them running for a long time instead of about a year or so.
I can say one thing for sure: the sound of a 10,000RPM head crash is truly annoying, almost agonizing, especially when combined with the noise from another one that is imminent.
"You're young, you're drunk, you're in bed, you have knives; shit happens." -- Angelina Jolie