Do suspended hard disks break sooner?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
Do suspended hard disks break sooner?
I have been suspending hard disk for just about as long as this website exists. I have always suspended my hard disk with a rubber elastic band just as many of you have done here.
About 6 months ago, I bought a 150gb WD Raptor. It is a great disk but I have just RMA'ed it the second time. I have also done some research on NAS devices and came across a statement where the manufacturer prided itself in low disk vibrations due to the study frame of the device. This supposedly results in much lower disk failures.
Based on personal experience, not only with the Raptors, but also some IBM and then Hitachi disk, I am beginning to wonder if there is any truth.
Does anyone else have problems with suspended Raptor (or other) disks?
About 6 months ago, I bought a 150gb WD Raptor. It is a great disk but I have just RMA'ed it the second time. I have also done some research on NAS devices and came across a statement where the manufacturer prided itself in low disk vibrations due to the study frame of the device. This supposedly results in much lower disk failures.
Based on personal experience, not only with the Raptors, but also some IBM and then Hitachi disk, I am beginning to wonder if there is any truth.
Does anyone else have problems with suspended Raptor (or other) disks?
Anecdotal evidence only, but my suspended 150G Raptor worked well from when I bought it in early 2006, and was still working when I sold it last week. My three suspended Samsung 500's are still working too. As is my suspended Velocioraptor, and the Samsung 160 which was suspended in my previous PC for a couple of years (now lying on the sofa, but it still works!).
The hard-mounted 80GB Samsung in my server died after a couple of weeks though, the first h/d failure I've had in nearly ten years.
To be honest I think it's fruitless trying to figure out such things. The organisations which consume hard-drives in sufficient quantities to be able to gather enough data to detect a trend don't suspend their drives, and generally don't release their data. When they do release their data people don't believe the conclusions unless they're aligned with their own prejudices. All you'll gain from people posting in forums is a small-scale, uncontrolled, biased sample, which is useless for drawing meaningful conclusions from.
The hard-mounted 80GB Samsung in my server died after a couple of weeks though, the first h/d failure I've had in nearly ten years.
To be honest I think it's fruitless trying to figure out such things. The organisations which consume hard-drives in sufficient quantities to be able to gather enough data to detect a trend don't suspend their drives, and generally don't release their data. When they do release their data people don't believe the conclusions unless they're aligned with their own prejudices. All you'll gain from people posting in forums is a small-scale, uncontrolled, biased sample, which is useless for drawing meaningful conclusions from.
Re: Do suspended hard disks break sooner?
Same exact thing happened to me with my 150, had to RMA it twice after only a year. I stopped suspending drives after that.michaelr wrote:I have been suspending hard disk for just about as long as this website exists. I have always suspended my hard disk with a rubber elastic band just as many of you have done here.
About 6 months ago, I bought a 150gb WD Raptor. It is a great disk but I have just RMA'ed it the second time. I have also done some research on NAS devices and came across a statement where the manufacturer prided itself in low disk vibrations due to the study frame of the device. This supposedly results in much lower disk failures.
Based on personal experience, not only with the Raptors, but also some IBM and then Hitachi disk, I am beginning to wonder if there is any truth.
Does anyone else have problems with suspended Raptor (or other) disks?
-
- Posts: 628
- Joined: Sat Dec 13, 2008 1:25 pm
- Location: Finland
I just say ten thousand and noting else. I've had the joy to play with a Raptor X and it vibrates as much as my previous 10000RPM SCSI HDD. I think everything above 7200RPM should not be suspended. I've had no success with SCSI HDDs eiher, especially 15000RPM drives will start to have malfunctions ( or I suspended them wrong and that explains why 10% of the time the HDD would get recognized at all in the BIOS ). Of course hardmounting that HDD there was no problems what-so-ever.CyberDog wrote:well it is possible if you suspend drives on too soft/loose they vibrate too much... Don't know how vibrant those raptors are...
-
- Posts: 628
- Joined: Sat Dec 13, 2008 1:25 pm
- Location: Finland
Det är korrekt Kaleid, but when suspending the HDD it is in relation to the case not 100% in horizontal or vertical position. With fluid dynamic bearing it shouldn't be a problem for most HDDs, but malfunctioning can happen. You can never be 100% sure about something, but suspension is as safe as one wants it to be.
What I mean is to check everytime you move the PC that the suspended HDDs are sitting correctly in place and every year or so take apart the suspension system and control the material which was used to hold the HDD. Also, make sure to take extra steps when suspending an HDD if your case allows it, for example put a big block of foam underneath the HDD and the suspension system to ensure that in case of a loose tie or other accident the HDD will not get damaged too severely if and when it will fall out of the suspension system. On top of all make sure to control the HDD temperature and switch off the PC in case you find the temperature to be too high. You might need to reconstruct your PC from the inside to allow even a bit of air to reach the HDD. Last but not least make regular backups to external media if applicable. The least you want is to lose your confidential data and it can happen with or without suspension. If taken great care like this you will ensure to have a joyful computing experience.
What I mean is to check everytime you move the PC that the suspended HDDs are sitting correctly in place and every year or so take apart the suspension system and control the material which was used to hold the HDD. Also, make sure to take extra steps when suspending an HDD if your case allows it, for example put a big block of foam underneath the HDD and the suspension system to ensure that in case of a loose tie or other accident the HDD will not get damaged too severely if and when it will fall out of the suspension system. On top of all make sure to control the HDD temperature and switch off the PC in case you find the temperature to be too high. You might need to reconstruct your PC from the inside to allow even a bit of air to reach the HDD. Last but not least make regular backups to external media if applicable. The least you want is to lose your confidential data and it can happen with or without suspension. If taken great care like this you will ensure to have a joyful computing experience.
i remember reading a long loooong time ago, someone did low-level tests on drives that were hard-mounted vs suspended, and actually came up with different figures for things like random seeks or latency.
someone chimed in that it was due to the drive's center of gravity changing via the head's inertia or some such, causing the drive itself to move by tiny amounts and throw off calibration. then, since there was a quantifiable difference in performance when suspending there was a possible method of reducing drive lifetime. i think someone from a manufacturer even chimed in.
of course nothing ended up being decided. and the fact that there hasn't been a rash of users with dead suspended drives here bumps the safety of suspension up from 'anecdotal' to 'data'. all imho mind you
then there's the issue of temperature having a larger effect on lifetime than any other metric..
someone chimed in that it was due to the drive's center of gravity changing via the head's inertia or some such, causing the drive itself to move by tiny amounts and throw off calibration. then, since there was a quantifiable difference in performance when suspending there was a possible method of reducing drive lifetime. i think someone from a manufacturer even chimed in.
of course nothing ended up being decided. and the fact that there hasn't been a rash of users with dead suspended drives here bumps the safety of suspension up from 'anecdotal' to 'data'. all imho mind you
then there's the issue of temperature having a larger effect on lifetime than any other metric..
I see no correlation between "reduction in life time" and "hasn't been a rash of users with dead suspended drives".incorrect wrote:... there was a possible method of reducing drive lifetime. ...
... and the fact that there hasn't been a rash of users with dead suspended drives here bumps the safety of suspension up from 'anecdotal' to 'data'. all imho mind you
My experience is that the normal technical life time of a regularly mounted HDD is >10 years.
Cut this by, say, 50% due to suspended mounting, and we still have a technical life time longer than the typical time it's in service with the normal disk suspender.
Cheers
Olle
-
- Friend of SPCR
- Posts: 2887
- Joined: Mon Feb 28, 2005 4:21 pm
- Location: New York City zzzz
- Contact:
Eh, sux. I was going to suspend my hd this month too. Looks like it's pointless. I am concerned about putting things on foam blocks as it is raises temperatures on the most sensative areas. looks like a crappy idea for long term stability. It does make the most sense though. It reduces vibrations and not just isolates.Olle P wrote:I see no correlation between "reduction in life time" and "hasn't been a rash of users with dead suspended drives".incorrect wrote:... there was a possible method of reducing drive lifetime. ...
... and the fact that there hasn't been a rash of users with dead suspended drives here bumps the safety of suspension up from 'anecdotal' to 'data'. all imho mind you
My experience is that the normal technical life time of a regularly mounted HDD is >10 years.
Cut this by, say, 50% due to suspended mounting, and we still have a technical life time longer than the typical time it's in service with the normal disk suspender.
Cheers
Olle
-
- *Lifetime Patron*
- Posts: 2269
- Joined: Sun May 21, 2006 9:09 am
- Location: Northern California.
Don't give up on foam blocks, they can work great.
put the drive on its side and carve a notch into the foam.
Two "u" blocks on the bottom and one ore two "u" blocks on top, then wedge it into the lower space below the floppy drive.
The top and bottom of the drive (now the sides) are fully exposed to air flow and stay nice and cool..
|------| < "u" foam
H |X|
D |X|
D |X|
|____| < "u" foam
Im happy with foam over the hassle of suspension. 32 C idle 39 C under full, long term load.
put the drive on its side and carve a notch into the foam.
Two "u" blocks on the bottom and one ore two "u" blocks on top, then wedge it into the lower space below the floppy drive.
The top and bottom of the drive (now the sides) are fully exposed to air flow and stay nice and cool..
|------| < "u" foam
H |X|
D |X|
D |X|
|____| < "u" foam
Im happy with foam over the hassle of suspension. 32 C idle 39 C under full, long term load.
so you've been suspending for years, but when a new drive fails after 6 months you think there could be a problem with suspending them? Idunno, me personally i would just assume its the drive. If suspension were really a problem you would see it with the drives suspended for years, not on one you just recently purchased.
The only thing i can possibly think of that would cause drives to prematurely fail due to suspension, is not grounding them with a wire and having the circuitry on them short out to the chassis.
The only thing i can possibly think of that would cause drives to prematurely fail due to suspension, is not grounding them with a wire and having the circuitry on them short out to the chassis.
Probably performance may be affected... see what can happen just screaming in front of your drivers
http://blogs.sun.com/brendan/entry/unusual_disk_latency
http://blogs.sun.com/brendan/entry/unusual_disk_latency
-
- Posts: 78
- Joined: Mon Mar 16, 2009 2:06 pm
- Location: East Midlands, UK
I went back to soft pad mounting of my hard drives in the HD tray in my Antec Solo case. Suspension was ok, but rubber cables lost elasticity and I ended up with a hard drive loose and floating too much.
Noisemagic has a in-between thing that I like.. rubber hold but disk rest on soft pad. Waiting for the 2.5'' to be back in stock to put my velociraptor on that baby.
The 3.5'' version was review by SPCR back in the days...
http://www.silentpcreview.com/article39-page1.html
US seller is frozencpu
http://www.frozencpu.com/products/3038/ ... 44c329s853
Noisemagic has a in-between thing that I like.. rubber hold but disk rest on soft pad. Waiting for the 2.5'' to be back in stock to put my velociraptor on that baby.
The 3.5'' version was review by SPCR back in the days...
http://www.silentpcreview.com/article39-page1.html
US seller is frozencpu
http://www.frozencpu.com/products/3038/ ... 44c329s853
-
- Posts: 2198
- Joined: Thu Feb 10, 2005 11:20 am
- Location: TN, USA
If you are the type to avoid intake fans or even taking it to the extreme of having your only case fan be a rear 120mm at 4v and other variations on that theme you are going to shorten the life of a suspended drive.
In that case it isn't vibration that is the issue it is heat that causes the issue. (old school hard mounting means the drive cage acts as a heat sink for the hard drive thus a hard mounted drive with no fan is cooler than a soft mounted drive with no fan).
I've got a WD6400AAKS suspended in a Solo but I have two 92mm intake fans at 5v in front of it so temps aren't an issue.
I'm considering hard mounting the drive just to reduce boot time as the primary user of that PC keeps asking me if there is any way I can shave another second or two off the boot time.
One of these days I'll do that buy putting the boot partition on an SSD but until then I may just try some tests softmounted vs hardmounted and see if it makes a measurable difference.
I probably will wait until I buy a Phenom II X3 720 or similar so that I can do the testing while I have the PC out of his hands for the rebuild.
In that case it isn't vibration that is the issue it is heat that causes the issue. (old school hard mounting means the drive cage acts as a heat sink for the hard drive thus a hard mounted drive with no fan is cooler than a soft mounted drive with no fan).
I've got a WD6400AAKS suspended in a Solo but I have two 92mm intake fans at 5v in front of it so temps aren't an issue.
I'm considering hard mounting the drive just to reduce boot time as the primary user of that PC keeps asking me if there is any way I can shave another second or two off the boot time.
One of these days I'll do that buy putting the boot partition on an SSD but until then I may just try some tests softmounted vs hardmounted and see if it makes a measurable difference.
I probably will wait until I buy a Phenom II X3 720 or similar so that I can do the testing while I have the PC out of his hands for the rebuild.
Anecdotal evidence here too, I purchased a number of Seagate 7200.10 160gb drives in January 2006 with one in my main PC suspended for the last 3 years. I just replaced it last month with a 640gb WD drive. The Seagate was still running normally at the time of replacement and has since been retired to external backup drive use.nutball wrote:Anecdotal evidence only, but my suspended 150G Raptor worked well from when I bought it in early 2006, and was still working when I sold it last week. My three suspended Samsung 500's are still working too. As is my suspended Velocioraptor, and the Samsung 160 which was suspended in my previous PC for a couple of years (now lying on the sofa, but it still works!).