I'm currently building a computer in which I won't be able to mount the hard drives using elastic suspensions (my usual favorite noise killing technique). However, I had an idea for an alternate means of dampening vibrational noise. What if I bolted two identical hard drives into adjacent bays such that they spun in opposite directions -- for example, placing one upside down, the other rightside up? In theory I'm imagining that the drives would tend to dampen one anothers' vibrations, thus reducing noise. Has anyone around here tried this, and what were the results?

Though this may be theoretically possible, one practical consideration is that it is unlikely that both drives will spin at the exact same speed simultaneously (i.e. as the sound from the one drive would not be of equal frequency and 180 degrees out of phase w/sound from the other, the noise from the two drives would not cancel). For more info, see this thread. Take care.

[Edits]:
PS Please note that I am far from being an expert in physics.
PPS Also, for this to have a chance at working, wouldn't the two drives have to be oriented not only upside down but one would have to be backwards as well (i.e. the area for connecting cables would the face front of the case; incidentally, a friend of mine oriented all of his hard drives backwards in the name of cable management and it worked quite nicely)?

Thanks for the link to the thread, which I read with much interest.

Still, the drives will be in close proximity to one another, and we're talking about fairly low frequency noise, for which the margin for error cancellation-wise should be greater than for, say, fan noise. It seems like even with small differences in RPM between the drives, a significant amount of vibrational noise should still be eliminated.

Also consider that the "noise" I'm describing here is the result of the drive's spin vibrating the computer case. If two identical (or in real world terms, highly similar) drives spin in opposing directions, wouldn't they "push" the case in more-or-less opposite directions at any point in time, thereby opposing each others' vibration? I think some automobile engines use this concept (horizontally opposed engines, or perhaps the "boxer" engines in some Subarus) to reduce the engine vibration that gets transmitted to the chassis.

So the question remains, though again I'm thankful for the noise cancellation thread provided by LushMD. Anyone else have some info on this, perhaps firsthand experience with similar efforts?

well think about how much more noise one drive creates when it is seeking. so if one drive is being used whiel the other is idle wouldn't one be giving more vibration then the other? this may cause a battle of teh vibrations and could (possibly) cause more noise.

Just my thoughts.

You can try it, but the likelihood of getting a double the amount of vibrational energy (waves summed in phase) as getting them elimianted (wave summed opposite phase) is equal.

So, I would not try doing it myself, but that's just my opinion.

I have heard what 2 10 000 RPM Fuji SCSI drives do when they 'sing' in unison and the sound is not pretty

You are dealing with so many parameters here that it is imho close to impossible control them all in order to achieve vibrational cancellation of the drives energies.

regards,
Halcyon

raid 0 is hte answer