Why mechanical failure causes HDD being undetectable by bios or OS ?

[andy]

Hi!
Could someone please explain why in the case of *mechanical* failure HD
becomes sometimes undetected by BIOS and/or the operating system (e.g. win xp
or linux)?
If it was an electronic failure then such behaviour would be obious, but why
the same happens with some mechanical failures? When electronics is working in
my opinion it still should be detected by bios and/or the system (win xp or
linux), but often it is not.
I could recover about 80% of the data from my HDD (which apparently has a
mechanical failure - plates spin up and down, heads create bad noises) if only
the disk could be seen by the system all the time. But often during copying of
the data heads hit with a loud sound so badly that sometimes even the plates
stop rotating, and the disk then dissapears from the system. It is then very
difficult to make it detectable by the system again, sometimes the sytem can
detect it but only after several minutes of copying it freezes and then
dissapears again.
Recently, I was unlucky, and even after several dozens of retries it's still
undetectable by the system.

Could you please advice what to do to make the disk detectable by the system
all the time?
What causes that it is not detectable although the failure is in mechanics not
electronics?

BTW, if someone has the same disk model (Quantum Fireball ST64A011), please
let me know.

andy

 

[Ron Reaugh]

Hi!
Could someone please explain why in the case of *mechanical* failure HD
becomes sometimes undetected by BIOS and/or the operating system (e.g. win xp
or linux)?

Consider the options available to the HD designer. What do you want the
user to see during POST about a HD that knows itself that it can't possibly
work. The HD designer knows that many BIOSs have no ability to detect and
display a HD error status during POST. The BIOS may only be able to report
'there' or 'not there'. If you were the HD designer would you want the HD
to report 'there' during POST even when the HD itself really knew that it
wasn't there? How would you answer the query of a poster in this NG who
wanted to know why a HD reported 'there' during post but was not there for
all intents and purposes for any booting steps after POST?

Did the HD designer make the correct design choice in the first place which
would likely cause a competent user to try another HD and assume that the HD
was dead dead which in fact it is?

If it was an electronic failure then such behaviour would be obious, but why
the same happens with some mechanical failures? When electronics is working in
my opinion it still should be detected by bios and/or the system (win xp or
linux), but often it is not.

What on earth for? Such would be highly misleading and a very poor design
choice.

I could recover about 80% of the data from my HDD (which apparently has a
mechanical failure - plates spin up and down, heads create bad noises) if only
the disk could be seen by the system all the time.
Huh?

But often during copying of
the data heads hit with a loud sound so badly that sometimes even the plates
stop rotating, and the disk then dissapears from the system.
Duh!

It is then very
difficult to make it detectable by the system again, sometimes the sytem can
detect it but only after several minutes of copying it freezes and then
dissapears again.
Recently, I was unlucky, and even after several dozens of retries it's still
undetectable by the system.

Could you please advice what to do to make the disk detectable by the system
all the time?
What causes that it is not detectable although the failure is in mechanics not
electronics?

The drive is DOA!

 

[Apollo]

Could you please advice what to do to make the disk detectable by the
system
all the time?
What causes that it is not detectable although the failure is in mechanics
not
electronics?

For recovery only;
I've had luck in the past by freezing dead hard drives, leave it over night
in the freezer, have everything ready on your pc, i.e. ide and power cables
loose so that you can plug it straight in and side panel off.

Plug the drive in as soon as it comes out the freezer and boot your pc,
don't bother fixing it into the case. You won't have long to recover your
data (if it works at all) so before trying the procedure, make a mental note
of what is most important to you on the drive and try and grab that first.
Don't expect to be able to recover a large amount of data using this method,
as the drive warms (quickly) you'll soon be back to the original problem.

Good luck, I bet you'll start backups now

HTH

 

[andy]

Thanks for your reply.

What on earth for? Such would be highly misleading and a very poor design
choice.

But I could then recover 80% of my data, and now I can recover 0% of my data.
Does it make sense for you now?

Is there any way to disable that feature? (I mean to make the malfunctioned
HDD visible to the system again?)

The drive is DOA!

Don't get DOA.

a.

 

[andy]

For recovery only;
I've had luck in the past by freezing dead hard drives, leave it over night
in the freezer, have everything ready on your pc, i.e. ide and power cables

Thanks for the advice, but I tried that already to no avail.

a.

 

[Joep]

Hi!
Could someone please explain why in the case of *mechanical* failure HD
becomes sometimes undetected by BIOS and/or the operating system (e.g. win xp
or linux)?

*Sometimes* ... So, maybe it's the nature of the problem that prevents the
disk from being detected.

If it was an electronic failure then such behaviour would be obious, but why
the same happens with some mechanical failures? When electronics is working in
my opinion

It is hardly a matter of opinion ...

it still should be detected by bios and/or the system (win xp or
linux), but often it is not.
I could recover about 80% of the data from my HDD

How do you know? How did you come up with the 80%?

(which apparently has a
mechanical failure - plates spin up and down, heads create bad noises) if only
the disk could be seen by the system all the time. But often during copying of
the data heads hit with a loud sound so badly that sometimes even the plates
stop rotating, and the disk then dissapears from the system. It is then very
difficult to make it detectable by the system again, sometimes the sytem can
detect it but only after several minutes of copying it freezes and then
dissapears again.

You should try to clone it as long as you can see it. However, every read
may worsen the condition of the disk, in general it is advised to cease DIY
recovery attempts (if the data is important to you) when a disk is maing
unusual and scary noises.

Recently, I was unlucky, and even after several dozens of retries it's still
undetectable by the system.

Could you please advice what to do to make the disk detectable by the system
all the time?

Your issue is a psychological one. You can not accept that there are
situations you can not resolve and have no control over. Apart from
contacting a data recovery lab, you also need to work out this problem.

BTW, if someone has the same disk model (Quantum Fireball ST64A011), please
let me know.

Why do you want to know?

 

[Apollo]

Thanks for the advice, but I tried that already to no avail.

a.

You're out of luck then I'm afraid. If the data is that critical then don't
do anything else to it, contact a professional data recovery operation. Be
prepared to pay a lot of money (2nd mortgage) though, these people are not
cheap, and for good reason. They will probably strip your drive and mount
the physical disks in their own apparatus under clean room conditions - this
is what costs.

Or write it off to experience and learn from it. Do regular backups, and
remember backups are useless unless you test them and ensure they work.

 

[andy]

Or write it off to experience and learn from it. Do regular backups, and
remember backups are useless unless you test them and ensure they work.

I was doing backups of critical data, and that is why I'm not prepared to pay
a lot of $$$ for data recovery.
However, I'm prepared to pay a little bit for recovery of the rest of the
data. I think that if I could purchase another such disk even with bad
electronics, but with good mechanics, and then I could recover the data, even
if I open the disk not in a virtually dust free conditions.

a.

 

[andy]

news:[email protected]...

*Sometimes* ... So, maybe it's the nature of the problem that prevents the
disk from being detected.

Failure to mechanics seems to be the problem causing the non-detection
problem.
Unfortunatelly the "sometimes" is now "nearly always".

How do you know? How did you come up with the 80%?

When the disk was detectable then about 20% of files could not be read.
This was not because bad sectors (the disk did not have any AFAIK), but
because of the mechanics failure (when it started to have the symptoms of the
failure also 20% of data became unavailable).

You should try to clone it as long as you can see it. However, every read
may worsen the condition of the disk, in general it is advised to cease DIY
recovery attempts (if the data is important to you) when a disk is maing
unusual and scary noises.

Yes, it seems that the condition very quickly became much worse.

Your issue is a psychological one. You can not accept that there are
situations you can not resolve and have no control over. Apart from

How can I know that? If they designed it that way that it should not be
detectable when mechanics fails, then maybe also for the service purpose they
designed it also to be possible to disable that feature, making the disk
visible despite mechanical failure.
I hoped that someone knows how to disable that feature.

Why do you want to know?

If someone has such disk with bad electronics, but good mechanics, then I
could use the mechanics to recover my data. Just for one time recovery even
opening the disk in not sufficiently clean condition possibly could work.

a.

 

[CJT]

Hi!
Could someone please explain why in the case of *mechanical* failure HD
becomes sometimes undetected by BIOS and/or the operating system (e.g. win xp
or linux)?
If it was an electronic failure then such behaviour would be obious, but why
the same happens with some mechanical failures? When electronics is working in
my opinion it still should be detected by bios and/or the system (win xp or
linux), but often it is not.
I could recover about 80% of the data from my HDD (which apparently has a
mechanical failure - plates spin up and down, heads create bad noises) if only
the disk could be seen by the system all the time. But often during copying of
the data heads hit with a loud sound so badly that sometimes even the plates
stop rotating, and the disk then dissapears from the system. It is then very
difficult to make it detectable by the system again, sometimes the sytem can
detect it but only after several minutes of copying it freezes and then
dissapears again.
Recently, I was unlucky, and even after several dozens of retries it's still
undetectable by the system.

Could you please advice what to do to make the disk detectable by the system
all the time?
What causes that it is not detectable although the failure is in mechanics not
electronics?

BTW, if someone has the same disk model (Quantum Fireball ST64A011), please
let me know.

andy

Maybe it stores part of its own software on the platters.

 

[ChrisJ9876]

From: andy (e-mail address removed)

Date: 09/03/2004 7:22 AM Eastern Daylight Time
Message-id: <[email protected]>




Failure to mechanics seems to be the problem causing the non-detection
problem.
Unfortunatelly the "sometimes" is now "nearly always".


When the disk was detectable then about 20% of files could not be read.
This was not because bad sectors (the disk did not have any AFAIK), but
because of the mechanics failure (when it started to have the symptoms of the
failure also 20% of data became unavailable).


Yes, it seems that the condition very quickly became much worse.


How can I know that? If they designed it that way that it should not be
detectable when mechanics fails, then maybe also for the service purpose they
designed it also to be possible to disable that feature, making the disk
visible despite mechanical failure.
I hoped that someone knows how to disable that feature.


If someone has such disk with bad electronics, but good mechanics, then I
could use the mechanics to recover my data. Just for one time recovery even
opening the disk in not sufficiently clean condition possibly could work.

a.

It's not a "feature". The mechanical noises and other symptoms you describe
most likely result from a head crash, where the heads, which normally float on
a very thin cushion of air, have contacted the disk surface and scraped the
oxide off the disk, making it unreadable. If it can't be read, it won't come
"ready" (as in "ready to use") and the BIOS won't detect it. Once the heads
crash, it only gets worse. If you open it up, you'll probably find bare spots
on the disks where your data once was.
To repeat, it's not a "feature" that can be disabled. Your drive is dead. Trust
me, I've serviced computer disk drives of various kinds since long before PC's
existed, and I know how they fail.
Chris

 

[andy]

It's not a "feature". The mechanical noises and other symptoms you describe
most likely result from a head crash, where the heads, which normally float on
a very thin cushion of air, have contacted the disk surface and scraped the
oxide off the disk, making it unreadable. If it can't be read, it won't come
"ready" (as in "ready to use") and the BIOS won't detect it. Once the heads

It doesn't make sense to me. Even if part of the surface is destroyed, most of
it is not, why then not to detect the disk?

crash, it only gets worse. If you open it up, you'll probably find bare spots
on the disks where your data once was.

I opened it now (I don't care about the dust, since it's dead anyway), and the
surface of the first plate is in perfect condition, I can't see surfaces of
other plates, though.

To repeat, it's not a "feature" that can be disabled. Your drive is dead. Trust
me, I've serviced computer disk drives of various kinds since long before PC's
existed, and I know how they fail.

I know it's broken down, but still, this behaviour of the disk is mysterious.
If it was like you said, then the disk would be dead all the time, but as I
described before sometimes it was possbile to detect it despite this failure.

Someone designed it that way, but perhaps there are ways to come round it.

a.

 

[Ron Reaugh]

Thanks for your reply.


But I could then recover 80% of my data, and now I can recover 0% of my data.
Does it make sense for you now?

Perfect sense and no you couldn't recover 80% of your data.

Is there any way to disable that feature? (I mean to make the malfunctioned
HDD visible to the system again?)


Don't get DOA.

Dead On Arrival

 

[Ron Reaugh]

Failure to mechanics seems to be the problem causing the non-detection
problem.
Unfortunatelly the "sometimes" is now "nearly always".


When the disk was detectable then about 20% of files could not be read.
This was not because bad sectors (the disk did not have any AFAIK), but
because of the mechanics failure (when it started to have the symptoms of the
failure also 20% of data became unavailable).

No, now 100% is unavailable.

Yes, it seems that the condition very quickly became much worse.

Isn't that what I said.

 

[Ron Reaugh]

Maybe it stores part of its own software on the platters.

Most all current HDs do that.

 

[Ron Reaugh]

heads

It doesn't make sense to me. Even if part of the surface is destroyed, most of
it is not,

Nonsense. Once any part of the surface is destroyed then then rest dies
VERY soon thereafter.

 

[andy]

Perfect sense and no you couldn't recover 80% of your data.

Why?

a.

 

[andy]

Nonsense. Once any part of the surface is destroyed then then rest dies
VERY soon thereafter.

It depends what you mean saying that.
When the disk was detectable always the same data was unavailable, therefore I
assume that if only the disk could be detectable then I could recover 80% of
the data.

a.

 

[andy]

No, now 100% is unavailable.

Only because the disk cannot be detected.

a.

 

[Folkert Rienstra]

Most all current HDs do that.

Nope, "most all current HDs" probably do not.
My IBM DMVS does not and that drive is already ~5 years old.
Flashrom has become cheap enough to take all the firmware, not
just the bare minimum part to spin the drive up and load the rest.