HDD silencing

matty c · Joined: 03 Jul 2003 Posts: 1191 Location: NW London, UK

I waws intereste to notice the article on HDD silencing, especially as I did a similar one myself last night!

My approach was very similar, and so was the reason; to cut down vibrations.

Since I only use the one 3.5" drive, I removed the drive cage from the case, and suspended the HDD on elastic tied to an empty 5.25" bay. The front edge of the drive was just touching the insaide front of the case, so I stuck an old mouse matthere to cushion it. It doesn't look brilliant, but it does the job!

Oh, I might attach an image mlater, but it''ll have to wait; I'm in the process of rebuilding my one working machien ATM after Windows went down the pan.

Spode · Posted: Tue Feb 17, 2004 10:32 am Post subject:

What a coincidence!

Mike · Joined: 05 Apr 2003 Posts: 1037 Location: Southampton, UK

i can't say i've ever had really noisy hard drives, although my western digi 8mb cache drives only run at 7200rpm..... you can't hear them over my fans anyway - i need a quieter PSU....

Spode · Guest

I'm a bit worried about the rubber band solution. Fast-running hard drives have to dissipate a fair bit of heat. Some is convected off, but part is sunk through the drive cage.

Rubber bands aren't exactly the best thermal conductors known to science. If in a poorly-ventilated or hot-running PC, I can see the drives melting down. (Ever seen a drive with melted seals? 'Cos I have.)

Jon

Spode · Posted: Tue Feb 17, 2004 3:47 pm Post subject:

I don't think it's a solution for all. People with poorly ventilated cases are generally not going to be the type of people who would attempt this.

But in saying that, people wanting really quiet machines often have minimalistic cooling.

I think, as with everything, you have to take it on a case by case basis.

Kaitain · Posted: Tue Feb 17, 2004 3:55 pm Post subject:

In the images in that review, it's fairly clear that the HDDs are placed right in front of ventilation fans, which should keep the drives nicely frosty. I doubt conduction through the case accounts for very much of the heat dissipation: think of the amount of contact the drive has with the case. 2 strips 10cm by 2cm isn't really a lot. The majority is convected away.

I think for most IDE/ATA drives this solution will be fine, provided the drives don't get hot enough to perish the elastic of the bands (ooh, sticky rubbery mess). My concern would be for larger, heavier or faster-spinning drives (think: scsi). When things are rotated, there's a tendency for both the disc and the chassis to rotate in opposed directions until the relative motion between them is e.g. 13000rpm. When constrained by the case, only the drive plates rotate (this is the same logic as putting the tail rotor on a helicopter). At steady state, when the drive's at max speed, this is all fine - chassis is held by rubber bands, plates rotate. At startup, though, because the elastic is not solid, the whole setup is going to twist. Imagine having a particularly hefty drive that manages to rotate hard enough to knock against the side of the case...

Wonder whether Allan would power the machine up once or twice with the bay covers off and see whether this happens to a significant extent - perhaps measure the deflection if possible?

Spode · Guest

Spode · Posted: Tue Feb 17, 2004 5:12 pm Post subject:

You know, that was an accident! Seriously!!

Allan · Joined: 07 Apr 2003 Posts: 185 Location: Denmark

matty c · Joined: 03 Jul 2003 Posts: 1191 Location: NW London, UK

I ended-up monkeying a colling solution, as always.

1x enormous slot 1 HS, 2x old 60mm fans, all selolotaped to the top Very Happy

ClubBarf · Joined: 26 Apr 2003 Posts: 2135 Location: London, UK

Can I suggest that anyone doing something like this uses a small sliver of something conductive to earth the drive to the chassis?

SCSI drives especially. I dunno if drives still earth via the chassis, but a while ago a cheap PC (olivetti mini-desktop P120, I think) came with plastic rails to mount drives. The machine stopped working after a while, and when cracked open, the drive just needed earthing.

A chewing gum wrapper was used for a few days before a proper wire was screwed in, and the drive carried on working with no loss of data - but without the earthing, it seemed to build up a charge that eventually stopped it working (temporarily).

On second thoughts, I *think* it was a packard bell...

Anyhew, you get the point. If you're gonna soundproof your HDD, think about earthing it.

loop · Joined: 06 Apr 2003 Posts: 386

Rubber bands perishing is a problem. I know Dave did this some time a go and after a while all the rubber bands snapped. He didn't notice for ages until he bothered to open his case one day and saw the hdd lying on the floor of his case with some snapped rubber bands next to it.

ClubBarf · Joined: 26 Apr 2003 Posts: 2135 Location: London, UK

Thought about using rubber grommit's instead? I'm told they're just as effective at reducing noise, but don't have the same problem with degrading, and allow you to use a 3.5" bay, usually.

Basically they insulate the drive bay screws from the drive bay, which is all that's needed, assuming that the drive isn't actually touching the drive bay. It's a tight fit, but I've seen it done.

I also had the idea (which I've not bothered trying) of getting 4 rubbers (NO! Not THOSE rubbers, I mean pencil erasers...) gluing or screwing them to the drive, then screwing into the rubbers with fairly long screws. That one would definately need a 5.25" bay.

Spode · Posted: Thu Mar 25, 2004 2:10 pm Post subject:

Doesn't the Molex into the harddrive have 2 earth wires on? Or are they used as a ground for the 5 and 12v and therefore not useable? I've always been unsure on this.

ClubBarf · Joined: 26 Apr 2003 Posts: 2135 Location: London, UK

AFAIK, the yellow wire is +12v, the red is +5v and the two black are ground, so yes - the drive should ground itself through the molex connector, and I should imagine the better designed ones do just that.

Mostly peeps who do the rubber band trick should be fine without extra grounding wires, I *think*. I'm just making a suggestion for those who start getting weird behaviour from their drives, or don't wanna risk data loss for lack of a tiny wire. The machine I saw with this problem worked fine for a couple of months without grounding the drive...