Hard Disk Drive Myths Debunked! - Page 2

crackerjack · 18th Mar 2004, 02:31 AM

Pitty there are several myths emerging from this article. How so?

Well, for a start, there is no 'spring' to return the head assembly to the 'landing' region on power fail. There is no spring at all. The heads are actively driven to the landing zone by dumping the energy from the spinning platters via a few fets into the coil. This ensure that after power is removed, the heads are always run home, and the platters are breaked with in a reasonably short time.

There is also no physical lock to keep the heads in the landing zone. Its all done with magentism.

If you popped the top of a defunct drive and moved the heads out of the park region to break the magnetic lock, they will just sit there. If you unplug the spindle motor and rotate the platters at some speed with the heads out of the landing zone, the heads will fly and may float in one direction due to air pressure, but thats it.

There are other technical errors in the article, but someone else can post them.

ppg · 18th Mar 2004, 05:22 AM

Hi, I found this really interesting but I'd like to add a few things...

Quote:

Myth #8 :
Cheap power supplies will "slowly kill" your hard disk.

Truth :
Cheap power supplies will NOT "slowly kill" hard disks. If a cheap power supply fries and sends a power surge to your hard disk, it kills the hard disk instantly.

If it cannot provide enough power, your hard disk won't run properly or just plain won't run at all.

That may be true, if the power supply is undervoltaging the hard drive. If the power supply is giving the hard drive 4 volts, instead of the 3.3 or 6 volts instead of the 5, the hard drive will fail over time. It's the same as putting a christmas light up to a nine-volt battery. It will glow brightly, but it won't last long.

Quote:

Myth #20 : If you want to use a hard disk in the vertical position, you must first reformat it in the vertical position!
Truth : Hard disks will work in any position. You do NOT need to reformat it before using it in vertical position or even upside down!

This could not be done with the stepper motor hard drives. The heads would change their position by a siginificant amount that they might start to read/write data off nearby sectors, thereby causing corruption. A format would help re-aligning the sectors.

Good job otherwise!

power · 18th Mar 2004, 04:47 PM

occasional reader first time poster :)

the article is good however there is one error i would like to correct, it is not an urban legen that you can actually restore bad sectors now this does not work with ALL bad sector problems and occasionally the bad sectors MAY return however there is a peice of software called hdd regenerator which WILL fix bad sectors and this program is particularly useful with drives which end up having somewhere between 1 and 100 or so bad sectors http://www.dposoft.net/ try it and see it really does work! i work in a tech support dept in a large computer store and we use this program occasionally (most drives are just stuffed :( ) if you don't believe this program works try it and see i personally have recovered dozens of drives most of which do not return and i repaired a drive of my own with it... it had 1 bad sector :( the proof of the pudding is a drive failing DFT ( http://www.hgst.com/hdd/support/download.htm ) then running a regen and the drive passing dft the best thing is it doesn't mess with the data unlike the bad sector recovery in dft which tends to be not so kind... anyway hope this helps :)

Adrian Wong · 18th Mar 2004, 05:03 PM

Quote:

Originally Posted by crackerjack

Pitty there are several myths emerging from this article. How so?

Well, for a start, there is no 'spring' to return the head assembly to the 'landing' region on power fail. There is no spring at all. The heads are actively driven to the landing zone by dumping the energy from the spinning platters via a few fets into the coil. This ensure that after power is removed, the heads are always run home, and the platters are breaked with in a reasonably short time.

There is also no physical lock to keep the heads in the landing zone. Its all done with magentism.

If you popped the top of a defunct drive and moved the heads out of the park region to break the magnetic lock, they will just sit there. If you unplug the spindle motor and rotate the platters at some speed with the heads out of the landing zone, the heads will fly and may float in one direction due to air pressure, but thats it.

There are other technical errors in the article, but someone else can post them.

Hello crackerjack,

You are right if you are talking about drives that use the load/unload mechanism of parking the heads. However, the load/unload mechanism are only used in mobile and server hard disks. It is not used in desktop hard disks.

In desktop hard disks, there IS a spring that retracts the heads to the landing zone.

Adrian Wong · 18th Mar 2004, 05:11 PM

Quote:

Originally Posted by ppg

Hi, I found this really interesting but I'd like to add a few things...

Quote:

Myth #8 :
Cheap power supplies will "slowly kill" your hard disk.

Truth :
Cheap power supplies will NOT "slowly kill" hard disks. If a cheap power supply fries and sends a power surge to your hard disk, it kills the hard disk instantly.

If it cannot provide enough power, your hard disk won't run properly or just plain won't run at all.

That may be true, if the power supply is undervoltaging the hard drive. If the power supply is giving the hard drive 4 volts, instead of the 3.3 or 6 volts instead of the 5, the hard drive will fail over time. It's the same as putting a christmas light up to a nine-volt battery. It will glow brightly, but it won't last long.

Good point, ppg!

The truth is I wrote that in response to someone who claims that the hard disk will spin down whenever there is "less power" from the power supply and spin up again when full power is restored. There was never any mention of overvolting so it slipped my mind.

Thanks for reminding me! I will add that to the guide.

Quote:

Originally Posted by ppg

Quote:

Myth #20 : If you want to use a hard disk in the vertical position, you must first reformat it in the vertical position!
Truth : Hard disks will work in any position. You do NOT need to reformat it before using it in vertical position or even upside down!

This could not be done with the stepper motor hard drives. The heads would change their position by a siginificant amount that they might start to read/write data off nearby sectors, thereby causing corruption. A format would help re-aligning the sectors.

Well, since hard disks using stepper motors are extinct, I don't think there is a point in mentioning them!

Adrian Wong · 18th Mar 2004, 05:22 PM

Quote:

Originally Posted by power

occasional reader first time poster

the article is good however there is one error i would like to correct, it is not an urban legen that you can actually restore bad sectors now this does not work with ALL bad sector problems and occasionally the bad sectors MAY return however there is a peice of software called hdd regenerator which WILL fix bad sectors and this program is particularly useful with drives which end up having somewhere between 1 and 100 or so bad sectors http://www.dposoft.net/ try it and see it really does work! i work in a tech support dept in a large computer store and we use this program occasionally (most drives are just stuffed

) if you don't believe this program works try it and see i personally have recovered dozens of drives most of which do not return and i repaired a drive of my own with it... it had 1 bad sector

the proof of the pudding is a drive failing DFT ( http://www.hgst.com/hdd/support/download.htm ) then running a regen and the drive passing dft the best thing is it doesn't mess with the data unlike the bad sector recovery in dft which tends to be not so kind... anyway hope this helps

Hello power,

Actually, I covered this in Myth #17.

It is possible to format or use some software to "hide" bad sectors but it doesn't really repair them. All a low-level format or that software you mentioned do is replace bad sectors that they find with free sectors on spare tracks found in every hard disk. This is essentially what happens to every hard disk at the end of the manufacturing process.

As you mentioned above, there is a limited number of bad sectors that the software can "repair". That is because the number of spare sectors is limited. Once the spare track is used up and there are no longer any spare sectors, then low-level formatting or that software will not be able to hide additional bad sectors.

GoatCheez · 19th Mar 2004, 12:35 AM

Myth #19 :
The hard disk can only be installed in the horizontal position.
Truth :

Hard disks can be installed in any position - horizontal, vertical, even upside down!

Even though I completely agree with this 100%, I had a single instance where this proved to not be true 100% of the time. With a quantum atlas V scsi drive. The drive would work horizontal and vertical, but when I would tilt the drive so that it was an angle where none of the faces of the drive would be flush with the case (30degrees at any angle for example), the drive would fail to detect. Since I thought this myth was a myth I thought there was something wrong with the drive. As it turns out, the only thing wrong with the drive was the angle at which it was tilted. Weird.

pauldriver · 19th Mar 2004, 01:01 AM

Quote:

Myth #19 : The hard disk can only be installed in the horizontal position.
Truth : Hard disks can be installed in any position - horizontal, vertical, even upside down!

Myth #20 : If you want to use a hard disk in the vertical position, you must first reformat it in the vertical position!
Truth : Hard disks will work in any position. You do NOT need to reformat it before using it in vertical position or even upside down!

Drives without servo tracks have these issue (yes I know, old stuff) and the myth is born from the reality of the IBM 6 and 8 MHZ ATs.

If the machine was relocated from the desktop into the optional tower enclosure (a case that held the case) or tilted to it's side, the drive needed to be low level formatted, or data loss was gaurenteed.

There are also, STILL drives that will suffer main bearing failures when mounted upside down.

LOL, I just noticed that I joined in Dec, 2002 and this is my first post!

crackerjack · 19th Mar 2004, 01:43 AM

Quote:

Hello crackerjack,

You are right if you are talking about drives that use the load/unload mechanism of parking the heads. However, the load/unload mechanism are only used in mobile and server hard disks. It is not used in desktop hard disks.

In desktop hard disks, there IS a spring that retracts the heads to the landing zone.

Sorry, you are incorrect on this. Whilst most desktop hard disks do not actually load / unload from the media (deskstars are an exception), they all use the same method for running the heads into the landing zone. There is no spring. Infact a spring that would run the head assembly home rapidly would cause the positioning servo to chew quite a bit of power fighting it, which is something all manufacturers are avoiding. Tell you what, I'll make it easy on you. Grab a defunct hard disk (a rotary VC one, not an ancient linear one!), pop the top off and snap a photo. Post it here clearly marking out the spring. For the purpose of the arguement, a spring can be either the classic coiled wire or simply a flat strip of metal (or plastic) that is tensioned in some way.

Ifoot2 · 19th Mar 2004, 05:58 AM

Great guide!

I have a couple of questions.

Myth #14 : It is better to spin down the hard disk whenever you can to reduce stress on the spindle motor.
Truth : Normally, the platters are spun up at start up and kept spinning ................Therefore, it is better to keep the hard disk spinning for better performance and to reduce stress on the spindle motor.

What effect does this have on the bearings?
Is it better to shut down the disk for longer bearing life, or is better to keep them spinning since more/heavier contact between shaft and bearing occurs during startup/shutdown?

Myth #18 :
You must format your hard disk every <insert duration of choice> to improve performance.

Truth : This is yet another common fallacy. Formatting your hard disk regularly will NOT improve your hard disk's performance.

If you notice a significant degradation in your hard disk's performance after several months, this is because the data in the hard disk has become so fragmented that the read/write heads have to seek all over the hard disk while reading or writing data.

Is it possible that parts of the boot sector that does not get re-written can fade over time? IE: loose the magnetism and become unreadable.

I am making a few assumptions with this question that I just don't know enough about, such as:
The boot sector is written only once, during formatting.
It's possible for the magnetism to fade over time.

I have an 8 year old hard drive that is still working fine. I know, I know, it's on borrowed time!
At about the 3 yr mark, the machine started having boot problems. Sector not found. Sometimes it would boot fine.
There was very little chance this was due to a virus. The only fix I performed was to reformat the hard drive.
Norton disk utilities showed no bad sectors before the reformat.
I have never been able to leave it that long again between reformats to confirm that this was the problem.
Is there any truth to this? Or am I just propagating a new myth?

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Hard Disk Drive Myths Debunked	Dashken	Reviews & Articles	58	24th Dec 2008 07:22 PM
Seagate Unleashes 750GB Desktop PC Hard Drive!	Dashken	News	50	25th Jun 2007 03:29 PM
NEW 12GB 1-INCH HARD DRIVE FROM SEAGATE ENABLES MORE AUDIO AND VIDEO IN MOBILE DEVICE	Dashken	News	0	19th Feb 2006 04:05 AM
Seagate Momentus Hard Disk Giveaway Contest!	Dashken	Contests!	280	30th Sep 2005 12:40 AM

18th Mar 2004, 02:31 AM	#11 (permalink)
crackerjack Newbie Join Date: 18 Mar 2004 Posts: 3 Reputation: 0 Rep Power: 0	Pitty there are several myths emerging from this article. How so? Well, for a start, there is no 'spring' to return the head assembly to the 'landing' region on power fail. There is no spring at all. The heads are actively driven to the landing zone by dumping the energy from the spinning platters via a few fets into the coil. This ensure that after power is removed, the heads are always run home, and the platters are breaked with in a reasonably short time. There is also no physical lock to keep the heads in the landing zone. Its all done with magentism. If you popped the top of a defunct drive and moved the heads out of the park region to break the magnetic lock, they will just sit there. If you unplug the spindle motor and rotate the platters at some speed with the heads out of the landing zone, the heads will fly and may float in one direction due to air pressure, but thats it. There are other technical errors in the article, but someone else can post them.

18th Mar 2004, 04:47 PM	#13 (permalink)
power Newbie Join Date: 18 Mar 2004 Posts: 2 Reputation: 0 Rep Power: 0	occasional reader first time poster :) the article is good however there is one error i would like to correct, it is not an urban legen that you can actually restore bad sectors now this does not work with ALL bad sector problems and occasionally the bad sectors MAY return however there is a peice of software called hdd regenerator which WILL fix bad sectors and this program is particularly useful with drives which end up having somewhere between 1 and 100 or so bad sectors http://www.dposoft.net/ try it and see it really does work! i work in a tech support dept in a large computer store and we use this program occasionally (most drives are just stuffed :( ) if you don't believe this program works try it and see i personally have recovered dozens of drives most of which do not return and i repaired a drive of my own with it... it had 1 bad sector :( the proof of the pudding is a drive failing DFT ( http://www.hgst.com/hdd/support/download.htm ) then running a regen and the drive passing dft the best thing is it doesn't mess with the data unlike the bad sector recovery in dft which tends to be not so kind... anyway hope this helps :)

19th Mar 2004, 12:35 AM	#17 (permalink)
GoatCheez Newbie Join Date: 19 Mar 2004 Location: Tampa, FL, USA Posts: 1 Reputation: 0 Rep Power: 0	Myth #19 : The hard disk can only be installed in the horizontal position. Truth : Hard disks can be installed in any position - horizontal, vertical, even upside down! Even though I completely agree with this 100%, I had a single instance where this proved to not be true 100% of the time. With a quantum atlas V scsi drive. The drive would work horizontal and vertical, but when I would tilt the drive so that it was an angle where none of the faces of the drive would be flush with the case (30degrees at any angle for example), the drive would fail to detect. Since I thought this myth was a myth I thought there was something wrong with the drive. As it turns out, the only thing wrong with the drive was the angle at which it was tilted. Weird.

19th Mar 2004, 05:58 AM	#20 (permalink)
Ifoot2 Newbie Join Date: 19 Mar 2004 Posts: 2 Reputation: 0 Rep Power: 0	Great guide! I have a couple of questions. Myth #14 : It is better to spin down the hard disk whenever you can to reduce stress on the spindle motor. Truth : Normally, the platters are spun up at start up and kept spinning ................Therefore, it is better to keep the hard disk spinning for better performance and to reduce stress on the spindle motor. What effect does this have on the bearings? Is it better to shut down the disk for longer bearing life, or is better to keep them spinning since more/heavier contact between shaft and bearing occurs during startup/shutdown? Myth #18 : You must format your hard disk every <insert duration of choice> to improve performance. Truth : This is yet another common fallacy. Formatting your hard disk regularly will NOT improve your hard disk's performance. If you notice a significant degradation in your hard disk's performance after several months, this is because the data in the hard disk has become so fragmented that the read/write heads have to seek all over the hard disk while reading or writing data. Is it possible that parts of the boot sector that does not get re-written can fade over time? IE: loose the magnetism and become unreadable. I am making a few assumptions with this question that I just don't know enough about, such as: The boot sector is written only once, during formatting. It's possible for the magnetism to fade over time. I have an 8 year old hard drive that is still working fine. I know, I know, it's on borrowed time! At about the 3 yr mark, the machine started having boot problems. Sector not found. Sometimes it would boot fine. There was very little chance this was due to a virus. The only fix I performed was to reformat the hard drive. Norton disk utilities showed no bad sectors before the reformat. I have never been able to leave it that long again between reformats to confirm that this was the problem. Is there any truth to this? Or am I just propagating a new myth?