WD guy on hard drive suspension: "It's horrible!"

Architectonic · Post by **Architectonic** » Wed Jun 11, 2008 7:56 am

http://www.thetechlounge.com/article/52 ... of+Caviar/

With all the talk about reducing vibrations, I asked Ted how hard drive suspension affected performance. "It's horrible!" Not only does it increase the chances that the heads sidetrack, in some cases, the drive hits a resonant frequency inside the case, causing massive and repeated failures. If you're Hell-bent on suspending the drive, at least benchmark it before and after: see for yourself.

Of course Ted is just a director of marketing..

m^2 · Post by **m^2** » Wed Jun 11, 2008 8:15 am

Maybe somebody should really do some benchmarks?

Cryoburner · Post by **Cryoburner** » Wed Jun 11, 2008 9:02 am

I'm not so sure I'd buy it. As mentioned in the preceding paragraph, he's trying to sell their new anti-vibration features, which won't really provide much benefit to those who soft-mount their drives. I suppose it's possible that suspending a drive could cause problems, but it also seems likely that he's fabricating information for the purposes of marketing. I'm sure hard drive manufacturers prefer their drives to be properly secured, as there's less of a chance they'll get damaged following installation, and therefore fewer warranty claims. They're certainly not going to publicly endorse soft mounting.

jhhoffma · Post by **jhhoffma** » Wed Jun 11, 2008 9:27 am

If a drive is securely suspended it shouldn't be any more prone to damage than if it's secured to the case. In fact, if proper care is done it could potentially be safer. My HDDs are soft mounted in my SLK3000B, and if my case were dropped (the horror) or moved violently, the HDDs would be the last thing I'd worry about. However, if they were hard mounted to the case, any shock vibration will be directly transmitted. Sloppy suspension is a different story however.

A couple years ago after my first HDD suspension attempt, I did do a couple of benchmarks before and after (due to cooling changes, not hardware changes) and did not notice any real difference (<1%) in Sandra's HDD benchmark, but I can't offer any proof as those results are long gone. Note, the results were with older hard drives: WD Caviar 120GB and Maxtor DM9 80GB.

nutball · Post by **nutball** » Wed Jun 11, 2008 9:30 am

WD chap trying to sell more WD drives wrote:If you're Hell-bent on suspending the drive, at least benchmark it before and after: see for yourself.

That sounds like a challenge to our trusty SPCR reviewers. Someone needs to use the "sword of truth and the trusty shield of fair play" to validate these claims and see whether they're true

jaganath · Post by **jaganath** » Wed Jun 11, 2008 10:47 am

http://www.wdc.com/en/library/sata/2579-001079.pdf

What is Rotational Vibration?
...
RV can also be induced by external forces on the rack or chassis containing the hard drives. Even linear vibrations applied on a chassis may get converted to RV at the drive level if a chassis is not designed appropriately. An example of this would be a drive bay structure that is not rigidly attached to a chassis.

hard to know if this is a real issue, or just marketing fluff invented by WD, but at least they're consistent.

oberbimbo · Post by **oberbimbo** » Wed Jun 11, 2008 10:53 am

Make a few markings on a drive and use a stroboscope on it. If the markets move in the way described they are right, otherwise it's probably marketing.

If rigidly securing the drive to the case fixes the issue, then so would adding some (a lot maybe) deadweight to the suspended drive

spookmineer · Post by **spookmineer** » Wed Jun 11, 2008 3:10 pm

There have been so many people suspending their hard drives, if this would be an issue there would have been a thread about failing hard disks by now.

As far as rotational vibration: drives are spinning at a constant speed after start up. The spinning platter will act as a gyroscope - if not cancelling out any force, at least it will decrease them.

In their .pdf it is stated "HD performance is degraded when an HD is exposed to vibration induced by one or more of its neighbors in the chassis. This induced vibration shakes the head off a track that is currently being read from or written to, resulting in retries and serious performance consequences".

Apparently, the vibration is of mechanical nature (slight imperfections in balance that will cause the hard drive to roll, pitch, I'm not familiar with the words).
What better way to get rid of them then by... suspension?

It's not an effect caused by a drive itself, it is caused onto a drive by other drive(s). Users who only have 1 drive will not have a problem because there is no second drive to exert force on the first one (or vice versa). Users who do have more than 1 drive will be able to cancel out the effect one drive has on another by using suspension of the drives.

breunor · Post by **breunor** » Wed Jun 11, 2008 10:19 pm

Umm, what I'm reading there is that external vibrations can be transmitted to the drive to cause the problems, so suspension helps rather than harms. If the drive itself has harmonic vibrations that the manufacturer hopes a secure chassis mount would absorb, that sounds hokey. If the external HDD body is vibrating so much that direct chassis mounting is needed, the tiny parts in the HDD are already ruined.

Edwood · Post by **Edwood** » Thu Jun 12, 2008 11:03 am

They're just trying to justify that stupid Icepack Noise Amplifier.

plympton · Post by **plympton** » Sun Jun 15, 2008 6:46 am

breunor wrote:Umm, what I'm reading there is that external vibrations can be transmitted to the drive to cause the problems, so suspension helps rather than harms. If the drive itself has harmonic vibrations that the manufacturer hopes a secure chassis mount would absorb, that sounds hokey. If the external HDD body is vibrating so much that direct chassis mounting is needed, the tiny parts in the HDD are already ruined.

That's similar to something going through my brain - I'm sure the drive is DESIGNED to be rigid mounted, so the designers take that into account as the design is tested, etc.

I would also venture that they have some test protocol that vibrates the system being tested at various frequencies for 30 days at a time (real-world accelerated testing), or it scales up and down frequencies (not real-world, but..). I also guess that during one of these tests, some engineer soft-mounted a drive, and the vibration 1 of the case, led to a resonant vibration of the soft-mounted drive, which then hit a resonant frequency in the drive, and bingo - failure!

I'm sure if you slowly shake any soft-mounted drive, there is a frequency it won't like.. no idea what it is, but just disrupting the gyroscopic effect of the spinning platters in JUST the right way...

I still have my 3 drives suspended, though...

-Dan

bexx · Post by **bexx** » Sun Jun 22, 2008 11:20 am

LOL funny thing is pretty much any time I have a customer complaining about early HD failures it is due to too much vibration. Its usually the 40mm 15krpm fans in 1U rackmounts that due to case design are vibrating the crap outa the HD and its failing early. Usually we'll get it as a percentage of systems are having early failure... like 15% of 100 system cluster... and then when you closer inspect the failing systems their fans (due to variation) vibrate more than the working systems. Usually resolution is just getting rubber washers to isolate either fan or hd depending on which is easier.

I believe these statements by WD to be false do to the fact that a suspended hard drive vibrates very little. A hard mounted HD in a regular desktop system might vibrate less than the suspended drive but that doesn't mean its worse off. 1U rackmounts in particular vibrate like crazy and this vibration is MUCH greater than any suspended hard drive-- yet HDs function just find hard mounted in a 1U chassis.

Actually I'm pretty sure this could be tested. Hard drives spec sheets have vibration limits on them. Suspend a hard drive, measure the vibration and see where it falls. The best tool I got is my hand and according to it theres no vibration issue with suspended HDs

. hehe

Post by **MikeC** » Sun Jun 22, 2008 12:46 pm

TechLounge interview with Ted Deffenbaugh of WD wrote:With all the talk about reducing vibrations, I asked Ted how hard drive suspension affected performance. "It's horrible!" Not only does it increase the chances that the heads sidetrack, in some cases, the drive hits a resonant frequency inside the case, causing massive and repeated failures. If you're Hell-bent on suspending the drive, at least benchmark it before and after: see for yourself.

My knee-jerk reaction (with BS-antennae quivering at full alert): This is total hogwash. I'd be happy to do w/ and w/o elastic suspension performance tests on a bunch of drives... when I have a bit of time. Just remind me if you don't see it here or elsewhere in SPCR in the nest week or so.

FartingBob · Post by **FartingBob** » Sun Jun 22, 2008 1:37 pm

I'll do some benchmarks tomorrow, but i highly doubt this will have any effect, and if it does it will be within the margin of error for the benchmark programs.

whiic · Post by **whiic** » Tue Jun 24, 2008 10:24 am

Cryoburner: "I'm not so sure I'd buy it. As mentioned in the preceding paragraph, he's trying to sell their new anti-vibration features, which won't really provide much benefit to those who soft-mount their drives. I suppose it's possible that suspending a drive could cause problems, but it also seems likely that he's fabricating information for the purposes of marketing. I'm sure hard drive manufacturers prefer their drives to be properly secured, as there's less of a chance they'll get damaged following installation, and therefore fewer warranty claims. They're certainly not going to publicly endorse soft mounting."

Cryo has a good point there. It's either marketing talk slash bullshit, or it's about misunderstanding what soft-mounting really is.

Points made by Deffenbaugh are based on real-world issue... especially for a HDD engineer. Probably not as much to the end-user as the devices are designed to tolerate external and self-induced disturbance.

Spindle isn't the main source of rotational vibration. Usually the worst kind of rotational vibration is rotation around axis of actuator assembly... not rotation around axis of spindle assembly axis. And rotational vibration around spindle axis is kinda low ... since it's about the jerks applied to keep platters spinning. Imbalance of spindle assembly causes peak frequency at rotational frequency but it's axis isn't the same as spindle axis.

What causes the most concerns is seeking... of neighbouring drives. Self-created seek vibration is predictable and can (and should) be compensated for. Soft-mounted HDDs have no issues with neighbouring drives! (That is, unless you build a HDD cradle with several HDDs bolted rigidly together, and soft-mount that cradle to the case. But individually soft-mounted HDDs that is not the case.)

spookmineer already mentioned these facts and I completely agree on his/her/its opinion.

Deffenbaugh and Seagate's document both talk about possible resonance scenarion of soft-mounted HDDs. Well, a HDD suspended with rubber bands have a resonance frequency below 10 Hz. That's quite a bit lower than spindle vibration frequency and frequency of intensive seeking. Talking about "wind-up" and "release" is kinda ridiculous if HDD is decoupled with highly elastic material. Release is just too soft to have any effect.

Seagate's vibration related study showed that semi-rigid mounting is WORSE than bolting the HDD into 20 kilogram cube of lead... or to a HDD mounted elastically. So guess if mounting HDDs to 0.2mm thick piss-cheap steel chassis is ideal? It's the worst solution there is! And HDDs still tolerate it with all the vibration they have to endure.

HDD manufacturers don't see a reason to object mounting HDDs to thin steel frames because:
- HDDs can tolarate it (even if it's the worst case scenario)
- HDDs are mounted in a way they are not in danger of falling or short-circuiting against some other object... yeah. I know you say you mount them properly and they are better protected when decoupled. Can you say the same about every single decoupling attempt out there? Decoupling is a big risk if done by incompetent person.

Post by **MikeC** » Tue Jun 24, 2008 10:38 am

whiic -- totally agreed. I posted a quote from Seagate's 2001 white paper on Disc Acoustics here:
viewtopic.php?p=419449&highlight=#419449

Compddd · Post by **Compddd** » Wed Jun 25, 2008 12:06 pm

I don't know if this worth anything, but I purchased a 150GB Raptor in March 06, in March 07 it failed, and then 1 year later almost to the day in March 08 the replacement drive WD sent me failed as well. Both drives were suspended with the elastics that came with my P150 during that two year period.

I am wondering/worried about if what this WD guy says is true, and if I should risk taking my VelociRaptor out of the IcePack and suspending it.

nutball · Post by **nutball** » Thu Jun 26, 2008 12:32 am

Compddd wrote:I don't know if this worth anything, but I purchased a 150GB Raptor in March 06, in March 07 it failed, and then 1 year later almost to the day in March 08 the replacement drive WD sent me failed as well. Both drives were suspended with the elastics that came with my P150 during that two year period.

That's anecdotal at best I'm afraid, and for each story like that you'll find at least one (probably more) folks here who've been running suspended drives for years without failures.

jhhoffma · Post by **jhhoffma** » Thu Jun 26, 2008 8:28 am

nutball wrote:That's anecdotal at best I'm afraid, and for each story like that you'll find at least one (probably more) folks here who've been running suspended drives for years without failures.

Around here, it's more like 10-to-1.

Compddd · Post by **Compddd** » Thu Jun 26, 2008 8:54 am

Guess I just had a reall bad string of luck lol, seeing as how they carry 5 year warranties

SlaveToSilence · Post by **SlaveToSilence** » Sun Jan 04, 2009 1:21 am

what if the drive was suspended with tight elastic from all 4 corners to the case above and below, surely this would eliminate vibration noise but still keep the drive pretty still, reducing the above problems (if they exist) by a good 80%-90%?

SlaveToSilence · Post by **SlaveToSilence** » Sun Jan 04, 2009 1:23 am

sort of like this? (i think) really tight, rather than just dangling from the top

courtesy of google images

dhanson865 · Post by **dhanson865** » Sun Jan 04, 2009 8:06 am

http://www.youtube.com/watch?v=tDacjrSCeq4&fmt=18 shows that noise induced vibration slows down drives.

Is it so far fetched to believe that suspended drives could be slower or less reliable than hard mounted?

Oh and for the record I have a Solo with the HD suspended and a few other cases with drives hard mounted. I don't have a dislike of suspension.

If Mike doesn't do it I might benchmark both ways myself next time I do an install on the SOLO. (That could be months from now)

SlaveToSilence · Post by **SlaveToSilence** » Sun Jan 04, 2009 8:15 am

my system drive is inside an enclosure (giant soundproofed heatsink) and placed on some stiff plastic-foam-like packing material, do you think it would/could suffer like a suspended drive?

thejamppa · Post by **thejamppa** » Sun Jan 04, 2009 8:54 am

SlaveToSilence wrote:my system drive is inside an enclosure (giant soundproofed heatsink) and placed on some stiff plastic-foam-like packing material, do you think it would/could suffer like a suspended drive?

Hardly. My HDD is on Scythe Quiet Drive which is lying on two sorthabne constructed feets ( basicly on sorthobane bad ) and my drive has worked like that for nearly 2Â½ years like it. Its very stead as its in P182.

I personally think whole suspension kills your HDD is hogwash too... If not, thank God there is transition going for SSD's... I think my HDD can live that long that I can get 500 GB + SDD for reasonable price...

SlaveToSilence · Post by **SlaveToSilence** » Sun Jan 04, 2009 8:55 am

define reasonable :p