Perhaps there's something I don't understand about hard drives, but is it normal for a drive, when idle, to be making a noise?
I have a Samsung HD501LJ 500GB HDD. When i stop all fans in my PC, I can still hear some noise coming from the hard drive, eventhough there isn't much (if any) disk activity happening. Is there a way to stop this noise?
I can understand the hard drive making a noise when it's seeking, but why should it make any noise when it's idle? It's not irritating, but it's the noisiest component in my system. For what it's worth, the hard drive is secured using the silicon grommet supplied with the Antec NSK3480. I guess the only thing I haven't tried is elastic suspension.
I also have a 1TB Western Digital Caviar Green HDD enclosed in a Vantec 3.5" CX Enclosure. I can hardly hear it and it's closer to me than my PC. Are these "green" drives built differently to the older drives? Do they power down when idle?
Idle HDD noise
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
The Samsung spins faster than the GP (7200 vs 5400rpm) so even though the Samsung is considered quiet it'll never match the GP for idle noise, being slightly louder and higher in pitch because of its greater rotational speed.
Is the Samsung your system drive? If so (and assuming your OS is XP) then it's unlikely ever to spin down since XP won't let it idle for long enough. Suspension will go a long way towards silencing it. If you go down that road and it still gets on your nerves you might also want to try some sort of absorbtive and/or barrier material on your case.
If the Samsung isn't your system drive and you know for a fact it's idle for long periods, then you could adjust the power-down settings in Start>Settings>Power Options>Power Schemes and change "Turn off hard disks:" to an appropriate figure.
Is the Samsung your system drive? If so (and assuming your OS is XP) then it's unlikely ever to spin down since XP won't let it idle for long enough. Suspension will go a long way towards silencing it. If you go down that road and it still gets on your nerves you might also want to try some sort of absorbtive and/or barrier material on your case.
If the Samsung isn't your system drive and you know for a fact it's idle for long periods, then you could adjust the power-down settings in Start>Settings>Power Options>Power Schemes and change "Turn off hard disks:" to an appropriate figure.
some people are more sensitive to idle whine than others. the general advice around here is elastic suspension and for most people that is enough. I'm apparently more sensitive than most, I have to use suspended aluminium enclosures to get it down to tolerable levels. there are some threads around here with some excellent homebrew enclosures plus there are some commercial ones. i've tried silentmaxx and scythe so far. silentmaxx hd-silencer is my favourite of the two but i can't get it in UK last time i looked. my alternative was the scythe quiet drive which keeps the hd cooler but is not as effective for noise. at the end of the day i'm gonna have to bite the bullet and get a Smart Drive as they are reputed to be the best but they cost the same as a hd!!!
Yes, the Samsung is my system drive and XP is my current OS. However, I previously had Arch Linux installed and I can't recall the hard drive being much quieter when idle.blackworx wrote: Is the Samsung your system drive? If so (and assuming your OS is XP) then it's unlikely ever to spin down since XP won't let it idle for long enough. Suspension will go a long way towards silencing it. If you go down that road and it still gets on your nerves you might also want to try some sort of absorbtive and/or barrier material on your case.
I tried the power option settings that you suggested above. I did notice the hard drive powering down and the idle whine going away. However, after a short period of time, the hard drive powered back up, eventhough I didn't touch the keyboard or mouse.If the Samsung isn't your system drive and you know for a fact it's idle for long periods, then you could adjust the power-down settings in Start>Settings>Power Options>Power Schemes and change "Turn off hard disks:" to an appropriate figure.
Anyway, thanks for your suggestions. I think I will try HD suspension on my next build. Not going to bother with this build as I can live with idle whine.[/quote]
[rant]
HDD keeps spinning at full rpm whether it's idle or not. (Sure, there's exceptions like STAND-BY mode when platters are spun down to 0 rpm either by immediate command or by timeout value. And Hitachi's APM implementation with low-rpm mode after 10+ minutes of idling.) There's fundamental reason for it to keep spinning.
If spun down at every idle period, time to spin up will take several seconds (in addition to seek time). It would cause extra wear to spin up frequently. And why do the platters have to revolve? It's because otherwise the capacity would be severely limited. Even with densities of modern HDDs, a single track doesn't contain much data. A spinning platter HDD with stationary arm would be like this. A stationary platter with moving actuator would store even less data as distance travelled is roughly 2*Pi*R(actuator)*(angle between inner and outer track)/360 instead of 2*Pi*R(outer diameter of platter).
A 7200rpm HDD revolves 120 rounds per second, and if it has a transfer rate of around 100 MB/s, it means that it has less than 1 megabyte per track. If actuator was statuary it would have less than 1 megabyte of total capacity multiplied by number of heads. And if the platters were statuary, the capacity would be roughly 200 kilobytes per head with a moving actuator. Obviously, we are ignoring that head needs a cushion of air under it to be moved around by the actuator... to reduce friction significantly.
Obviously if we wanted a media that had only one dimension in which bits are stored there would be no idle noise, and it wouldn't be in form of a platter but in form of a magnetic tape. Then again, seeking with magnetic tape requires fast forwarding/rewinding meters and meters of tape to reach wanted data, thus tape would be much, much slower. No general purpose file system can be used as such mundane tasks as seeking for a specified text file and doing a single character edit would require to seek FAT/MFT, seek and read file, rewind and rewrite bits, seek back to file metadata and change date altered information... some of the longer seek distances taking several minutes to perform.
So, just accept the fact that HDD has two "axis" of movement, one being constantly scanned whether needed or not and the other being used only when data is accessed.
1-dimensional storage is dead. Some fools still think tape drives are competitive but I don't think they can be cured from their delusions. Even solid state is 2-dimensional array of charge holding transistors. And some time in future, storage will be 3-dimensional. It's inevitable. The details on how it's made 3-dimensional are uncertain. (Dual-layer DVD is only one example and it's number of layers doesn't give it much advantage. Even solid state IC memory can be stacked over each other to save surface area.)
[/rant]
HDD keeps spinning at full rpm whether it's idle or not. (Sure, there's exceptions like STAND-BY mode when platters are spun down to 0 rpm either by immediate command or by timeout value. And Hitachi's APM implementation with low-rpm mode after 10+ minutes of idling.) There's fundamental reason for it to keep spinning.
If spun down at every idle period, time to spin up will take several seconds (in addition to seek time). It would cause extra wear to spin up frequently. And why do the platters have to revolve? It's because otherwise the capacity would be severely limited. Even with densities of modern HDDs, a single track doesn't contain much data. A spinning platter HDD with stationary arm would be like this. A stationary platter with moving actuator would store even less data as distance travelled is roughly 2*Pi*R(actuator)*(angle between inner and outer track)/360 instead of 2*Pi*R(outer diameter of platter).
A 7200rpm HDD revolves 120 rounds per second, and if it has a transfer rate of around 100 MB/s, it means that it has less than 1 megabyte per track. If actuator was statuary it would have less than 1 megabyte of total capacity multiplied by number of heads. And if the platters were statuary, the capacity would be roughly 200 kilobytes per head with a moving actuator. Obviously, we are ignoring that head needs a cushion of air under it to be moved around by the actuator... to reduce friction significantly.
Obviously if we wanted a media that had only one dimension in which bits are stored there would be no idle noise, and it wouldn't be in form of a platter but in form of a magnetic tape. Then again, seeking with magnetic tape requires fast forwarding/rewinding meters and meters of tape to reach wanted data, thus tape would be much, much slower. No general purpose file system can be used as such mundane tasks as seeking for a specified text file and doing a single character edit would require to seek FAT/MFT, seek and read file, rewind and rewrite bits, seek back to file metadata and change date altered information... some of the longer seek distances taking several minutes to perform.
So, just accept the fact that HDD has two "axis" of movement, one being constantly scanned whether needed or not and the other being used only when data is accessed.
1-dimensional storage is dead. Some fools still think tape drives are competitive but I don't think they can be cured from their delusions. Even solid state is 2-dimensional array of charge holding transistors. And some time in future, storage will be 3-dimensional. It's inevitable. The details on how it's made 3-dimensional are uncertain. (Dual-layer DVD is only one example and it's number of layers doesn't give it much advantage. Even solid state IC memory can be stacked over each other to save surface area.)
[/rant]