Which drive is clicking? Failing?

A confusing problem, which I hope you all won't be as confused by as I
am. Running XP SP3 on a Dell (whose model I think is 4600 or 4700).

I have in my desktop computer 2 hard drives.

One of them -- don't know which -- has clicked about 4 times, ONLY at
startup^^, the last 3 times I've started the computer, but the computer
starts and runs normally. And will run for hours and hours and hours
with no clicking, after it starts. Even when I use Win Explorer to
look at the list of files in either drive C or D.

^^I'm sorry to say I didnt' notice exactly when during start up. Maybe
when the progress bar goes across the very first GUI screen, but maybe
not.

So I backed up drive D using XXClone, which said that the backup was
made bootable (because drive D was.) and I backed up drive C also with
normal copying using Windows Explorer.***

Windows and pretty much everything, all the programs and data, is on
Drive D, but doesn't the mere fact that it's called D mean that the boot
drive is C, the other hard drive?

But Windows XP Disk Management calls D the boot drive. And calls C the
or a system drive. How can D be the Boot Drive?

Both Drive C and D contain files used for booting.

Last night I made the mistake of hibernating the computer, instead of
using Standby. When starting it this morning, it wouldnt' start the
first time because it couldn't find Drive 0 the first time. Drive 0**
is the one that holds Drive D, which runs all the time when the computer
is running, but never clicks.

So it seems to me that drive C is the drive that does the booting and
the one that's clicking, since all my working files are on D and if D
were clicking it would click all the time.

So, which drive is failing??


FWIW Disk Management describes C and D and the hidden partition with
win98 as Healthy. That refers to file structure etc. right, and
doesn't mean a drive has no mechanical problems with the tone-arm,
right? So for me it doesn't mean anything, right?




**FWIW, Drive 0 also has a hidden partition that holds win98. I still
have dual boot, and iirc it works.

*** The problem with using Windows Explorer as you all know is that it
stops when it gets to a bad file, but drive backup programs are designed
not to do that. I plan to back up C with XXClone also, as soon as buy
the pro-version, today.


I backed up my two volumes to one hard drive, meant to be mounted
internally, but I had it in a BlacX drive caddy, connected by USB. I
think that means I can use the backup drive to replace whichever
internal drive fails, and if necessary use Disk Management or a 3rd
party program DiskAssign, iirc, to assign the needed letter to what had
been the backup. Right? If the harddrive fails right now and I can't
run Disk Management from it, I can do so by booting from a Hirem's CD???
Or something similar?


Thanks for any help you can give.

micky wrote:

I have in my desktop computer 2 hard drives.

One of them -- don't know which -- has clicked about 4 times, ONLY at
startup

snippage

Check the power cable. The 3.5" drives use 12V for the motor.
If the voltage on the yellow wire drops below 11V, the drive
will spin down (to protect the heads). It will go into a loop,
and spin down and spin up, if the power cable has a bad
connection or is overloaded.

Some ATX power supplies have two SATA, two IDE (four pin Molex), plus
a floppy connector. It may look like they want you to put five
electrical loads on the wire, but in fact they only really
expect three loads. You can put two SATA and a floppy on the cable.
Or, you can put two IDE and a floppy on the cable. The five
connectors are intended to cover two generations of hardware.

The weak link, can be "Y" cables with cheap connectors on the end.
Some people extend the length of their distribution cable, using
"Y" cable extenders. The cheap connectors can make ohmic connections
and the drive on the end of your power chain, goes into a
spindown/spinup loop at power up. Removing the Y cables
can fix it.

People who work on servers, use "staggered spin" to make
the startup current flow more manageable. I've never seen
any mention of staggered spin for desktop IDE drives, and I
don't know of a way to engage it. On SCSI drives, the drives
can be told to take turns drawing spinup current, which means
it takes longer for your SCSI chain to come up, but it also
avoids power overloads on cabling or on the whole ATX PSU.
I don't generally address this issue in a posting, unless
a user indicates they have more than about four hard drives
connected. Above that level, you're supposed to be doing
the math on power loading. The motor on the drive draws
3 amps for ten seconds, which accelerates the platter up
to the required 7200 RPM level (10000 RPM on a Raptor).

*******

Check your SMART statistics, for info about an upcoming failure.
This program has a SMART readout tab. There are other programs
that can do this as well.

http://www.hdtune.com/files/hdtune_255.exe

The two boxes circled in red, are the ones I check. Having
zero in those fields, is the "good" indication. The yellow
marks are bogus, and I see those yellow ones highlighted even
on brand new hard drives. So I ignore those.

http://oi62.tinypic.com/2wghquh.jpg

Every drive that leaves the factory, has defects on it. The
SMART statistics are thresholded, so that the user cannot
see the small number of initial defects. If users were allowed
to see every defect (measured linearly from zero defects),
users would "cherry pick" drives and keep sending them back
to Newegg, if the readout was not zero. As a result, the
SMART readout isn't very "honest" about what is going on.
When it does show a non-zero value, you know you're well
on your way... downhill.

*******

If the problem is reproducible, you can use "stethoscope"
techniques to identify the affected drive. Pressing
a sound-conductive rod against the drive housing, and
listening during startup, may help you find the bad one.

When your problem happened to me, I didn't need to use
the stethoscope. I could tell from inspection, which drive
had a dodgy power chain connected to it. One of my drives
was on the end of a chain of Y cables. And removing the
Y cables and connecting the drive directly, stopped
the problem. I had some warning about the Y cables,
after a connector on one of them burned up. That was
a connector feeding an ATI9800 AGP video card Aux power
input. So now I treat the Y cables with more suspicion.
The early Y cables used good connectors, and I didn't
have a problem with those. Some of the later ones I bought,
just aren't as good. So far, the connectors on the
power supplies themselves, have been OK. I'm a bit
surprised by that. I've been waiting for one of those
to burn up on me...

Paul

On 11/08/2014 11:28 AM, micky wrote:
A confusing problem, which I hope you all won't be as confused by as I
am. Running XP SP3 on a Dell (whose model I think is 4600 or 4700).

I have in my desktop computer 2 hard drives.

One of them -- don't know which -- has clicked about 4 times, ONLY at
startup^^, the last 3 times I've started the computer, but the computer
starts and runs normally. And will run for hours and hours and hours
with no clicking, after it starts. Even when I use Win Explorer to
look at the list of files in either drive C or D.

^^I'm sorry to say I didnt' notice exactly when during start up. Maybe
when the progress bar goes across the very first GUI screen, but maybe
not.

So I backed up drive D using XXClone, which said that the backup was
made bootable (because drive D was.) and I backed up drive C also with
normal copying using Windows Explorer.***

Windows and pretty much everything, all the programs and data, is on
Drive D, but doesn't the mere fact that it's called D mean that the boot
drive is C, the other hard drive?

But Windows XP Disk Management calls D the boot drive. And calls C the
or a system drive. How can D be the Boot Drive?

Both Drive C and D contain files used for booting.

Last night I made the mistake of hibernating the computer, instead of
using Standby. When starting it this morning, it wouldnt' start the
first time because it couldn't find Drive 0 the first time. Drive 0**
is the one that holds Drive D, which runs all the time when the computer
is running, but never clicks.

So it seems to me that drive C is the drive that does the booting and
the one that's clicking, since all my working files are on D and if D
were clicking it would click all the time.

So, which drive is failing??


FWIW Disk Management describes C and D and the hidden partition with
win98 as Healthy. That refers to file structure etc. right, and
doesn't mean a drive has no mechanical problems with the tone-arm,
right? So for me it doesn't mean anything, right?






If the drive is clicking I would /not/ take a chance...I'd replace it.


Easy way to tell....try one drive at a time

In message , philo*
writes:
[]
If the drive is clicking I would /not/ take a chance...I'd replace it.


Easy way to tell....try one drive at a time

(Yes, I wondered about that, disconnecting one at a time then powering
up; even if the PC doesn't boot, you should hear the clicks if they're
there. The only concern might be that if the BIOS reconfigures itself in
some way that doesn't restore when you put both back in, such as if you
have a non-default arrangement.)

I'd agree about changing the drive - though I'd try Paul's checks first,
since if the supply chain has gone high resistance, changing the drive
might not cure it, or even might cure it temporarily just by the act of
disconnecting and remaking one of the connectors.

I suppose it's also possible that the power supply isn't well. (If
that's the case, disconnecting one drive might lighten the load
sufficiently that the clicks don't happen, too!)
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

I'm a gay man in a woman's body - and I love it! - Sheridan Smith (actress),
in Radio Times, 3-9 April 2010

J. P. Gilliver (John) wrote:
In message , philo
writes:
[]
If the drive is clicking I would /not/ take a chance...I'd replace it.


Easy way to tell....try one drive at a time

(Yes, I wondered about that, disconnecting one at a time then powering
up; even if the PC doesn't boot, you should hear the clicks if they're
there. The only concern might be that if the BIOS reconfigures itself in
some way that doesn't restore when you put both back in, such as if you
have a non-default arrangement.)

I'd agree about changing the drive - though I'd try Paul's checks first,
since if the supply chain has gone high resistance, changing the drive
might not cure it, or even might cure it temporarily just by the act of
disconnecting and remaking one of the connectors.

"if the supply chain has gone high resistance"???
You mean if there are some connectors that aren't making good contact?
(I assume that's what you're getting at).
I'm guessing if he hears those clicks, it's probably the drive itself,
however.

I suppose it's also possible that the power supply isn't well. (If
that's the case, disconnecting one drive might lighten the load
sufficiently that the clicks don't happen, too!)
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

I'm a gay man in a woman's body - and I love it! - Sheridan Smith
(actress),
in Radio Times, 3-9 April 2010

In message , Bill in Co
writes:
J. P. Gilliver (John) wrote:
[]
I'd agree about changing the drive - though I'd try Paul's checks first,
since if the supply chain has gone high resistance, changing the drive
might not cure it, or even might cure it temporarily just by the act of
disconnecting and remaking one of the connectors.

"if the supply chain has gone high resistance"???
You mean if there are some connectors that aren't making good contact?
(I assume that's what you're getting at).
I'm guessing if he hears those clicks, it's probably the drive itself,
however.
[]
If you read Paul's post (mine came after it but wasn't a direct followup
to it), he was indeed casting suspicion on connectors - especially
Y-cables which people might have bought/used to extend the reach of
drive power cables.

Paul and I weren't talking about the connectors making the clicks -
unless they were actually arcing, I don't think you'd hear that!
However, if the power supply route is slightly high resistance due to
deteriorated connectors/tions, then it is possible that the spin-up
surge drawn by the drive motor might cause the voltage seen at the drive
to drop sufficiently that its electronics cause an abort-and-park. If
this then drops the current drawn, the voltage would then rise back to
the point at which the drive electronics consider it safe to do a
spin-up, which would draw spinup current which would cause the voltage
(after the dodgy connectors) to drop, which ... repeated cycle could
well make a clicking noise. At the drive - without there actually being
anything wrong with the drive itself (in fact its self-protection
circuitry is operating well).

(As I said, it's probably worth considering that the power supply may be
deteriorating, too - fairly easy to check, I'd say.)
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

what - recession? Up north? What we gonna have - more nowt?
(News Quiz 2013-7-26)

J. P. Gilliver (John) wrote:
In message , Bill in Co
writes:
J. P. Gilliver (John) wrote:
[]
I'd agree about changing the drive - though I'd try Paul's checks first,
since if the supply chain has gone high resistance, changing the drive
might not cure it, or even might cure it temporarily just by the act of
disconnecting and remaking one of the connectors.

"if the supply chain has gone high resistance"???
You mean if there are some connectors that aren't making good contact?
(I assume that's what you're getting at).
I'm guessing if he hears those clicks, it's probably the drive itself,
however.
[]
If you read Paul's post (mine came after it but wasn't a direct followup
to it), he was indeed casting suspicion on connectors - especially
Y-cables which people might have bought/used to extend the reach of
drive power cables.

Paul and I weren't talking about the connectors making the clicks -
unless they were actually arcing, I don't think you'd hear that!

Right. I didn't think you were suggesting that. :-)

However, if the power supply route is slightly high resistance due to
deteriorated connectors/tions, then it is possible that the spin-up
surge drawn by the drive motor might cause the voltage seen at the drive
to drop sufficiently that its electronics cause an abort-and-park. If
this then drops the current drawn, the voltage would then rise back to
the point at which the drive electronics consider it safe to do a
spin-up, which would draw spinup current which would cause the voltage
(after the dodgy connectors) to drop, which ... repeated cycle could
well make a clicking noise. At the drive - without there actually being
anything wrong with the drive itself (in fact its self-protection
circuitry is operating well).

But I thinking that's the (much) more unlikely scenario, in most cases. :-)

(As I said, it's probably worth considering that the power supply may be
deteriorating, too - fairly easy to check, I'd say.)
--
J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf

what - recession? Up north? What we gonna have - more nowt?
(News Quiz 2013-7-26)

Hi, all.

Background. Please skip to the line of caps if background bores you.

I'm not sure whether to start a new thread, or continue with this one.
If your news reader is set up like mine and you keep posts for years,
tacking on to a post from Nov 8, 2014, will cause the new post to sort
way off the screen. But some people say to do this.

Quoted below this new text, you'll see most of the details of my
problem, a drive that clicked on startup.

After this thread in early November, I made sure I'd backed up the
partition on each drive I also stopped turning off the computer since
Nov 8 and turned off the option to go into sleep, because it once
clicked coming out of sleep.

I also checked with two methods, one of them Windows Disk Management,
and all partitiions, 2 drives in the computer, and one bare 1.5T drive
in a BlacX caddy, with a backup partition for each of the two partitions
I'm using, are listed as healthy, (still today after the problem got
worse).

And everything has been fine for the last 7+ weeks until AVG said it was
now inactive because I was supposed to restart about 2 weeks ago (to
update AVG) and I didn't.

So yesterday I backed up again, restarted, used it, hibernated, and
today it wouldnt' start!!

HERE IS THE CURRENT SITUATION.

The computer would not start. The Dell XP startup screen showed up and
the computer clicked about 6 times when the progress bar (a series of
green boxes) was 1/4 of the way across, then sometimes 2 times 1/2 of
the way and maybe 6 more times 3/4s of the way. When the progress bar
got to the end, it just stayed there, until a black & white screen came
on with "SATA secondary dfive 0 not found".

In the BIOS, SATA Primary Drive is listed by model number, the word
"auto" and the size of the drive, about 250G.
However the SATA Secondary Drive has Unknown for the model number,
"Auto", and a question mark for the size. It is actually the same model
as the other drive.

I used Hirem's CD and its Mini XP to look at my drives. Using that, I
could look at the file list, edit text files, modify them and save them
on both drives!!!! It's mostly*** only startup that it gets stuck
on.

Boot.ini is on the C: drive, and the default partition was also on the
C: drive, I think. The default line in boot.in said
Default multi (0) disk (0) rdisk (1) partition (2)
I guess it's 2 because there is a hidden win98 partition on that drive,
copied from y previous win98 computer.
The other choice in the list was
multi (0) disk (0) rdisk (0) partition (1)
and I made this the default in C:\boot.ini .

Now , at least once, it booted just fine, and everything works just as
it used to.

I don't understand. It still has to go to the Crive to read boot.ini,
but I guess that's the last thing it has to do with the C: drive. Now
boot.ini points to the D: drive, which iiuc is on rdisk0, which must be
okay! But even with the previous default to a partition on rdisk0, the
computer used the Windows directory on the D: partition. How is that
possible????? There is a windows directory on the C: drive,, but none
of the subdirectories ever get new files. (I forget how all this arose
because it's been maybe 8 to 10 years!!)

It's a DELL Dimension 4600, at least 10 years old, but it's been working
fine. I'm trying to fix up a much newer one, but always short of time.
This weekend is an example. I should be searching for a surgeon for my
just-diagnosed hyperparathyroidism, but instead I'm doing this.

I turned on the drive caddy, when I was last in the BIOS but the BIOS
didn't recornize it. I should have turned it on BEFORE I opened the
BIOS, right????? If I did that, and made the right partition of the
external drive the first hard drive in the boot order, that would also
have gotten me started, probably, right?

But what would be better, am I right, is to install the 2n'd partition
of the backup drive -- call the drive BBB- -- as the primary SATA drive,
an internal drive, and then back that up to a brand new drive? Rght???

Now the backup drive, BBB, has two other partitions on it, two copies of
the XP partition, one of them from yesterday, and the other older. It
will sitll work as the internal drive, right?

But maybe it's not so good to back up two different partitions to the
same drive, because of the situation I'm in now. Maybe what I shoudl do
is buy one or two more bare drives. The BlacX caddy has two slots and
it says you can copy a partition from a drive in one slot to a drive in
the other.

Any advice will be appreciated.

As it stands now, it seems if the second internal drive fails, I can
just install the backup drive in place of both current internal drives??

***One more thing. It doesn't click anymore since Windows is past
startup, but it pings quite a bit. 2 times, 4 times, 10 imes . It
didn't do that until the last hour yesterday. I'm in the basement and
it sounded like a thin screwdriver striking a water pipe, maybe from the
townhouse next door. When I put my hand on the drives, I don't feel
anything when I hear the noise. This is like clicking, right? It
means the drive is about to fail??




On Sat, 08 Nov 2014 12:28:37 -0500, micky
wrote:

A confusing problem, which I hope you all won't be as confused by as I
am. Running XP SP3 on a Dell (whose model I think is 4600 or 4700).

I have in my desktop computer 2 hard drives.

One of them -- don't know which -- has clicked about 4 times, ONLY at
startup^^, the last 3 times I've started the computer, but the computer
starts and runs normally. And will run for hours and hours and hours
with no clicking, after it starts. Even when I use Win Explorer to
look at the list of files in either drive C or D.

^^I'm sorry to say I didnt' notice exactly when during start up. Maybe
when the progress bar goes across the very first GUI screen, but maybe
not.

So I backed up drive D using XXClone, which said that the backup was
made bootable (because drive D was.) and I backed up drive C also with
normal copying using Windows Explorer.***

Windows and pretty much everything, all the programs and data, is on
Drive D, but doesn't the mere fact that it's called D mean that the boot
drive is C, the other hard drive?

But Windows XP Disk Management calls D the boot drive. And calls C the
or a system drive. How can D be the Boot Drive?

Both Drive C and D contain files used for booting.

Last night I made the mistake of hibernating the computer, instead of
using Standby. When starting it this morning, it wouldnt' start the
first time because it couldn't find Drive 0 the first time. Drive 0**
is the one that holds Drive D, which runs all the time when the computer
is running, but never clicks.

So it seems to me that drive C is the drive that does the booting and
the one that's clicking, since all my working files are on D and if D
were clicking it would click all the time.

So, which drive is failing??


FWIW Disk Management describes C and D and the hidden partition with
win98 as Healthy. That refers to file structure etc. right, and
doesn't mean a drive has no mechanical problems with the tone-arm,
right? So for me it doesn't mean anything, right?




**FWIW, Drive 0 also has a hidden partition that holds win98. I still
have dual boot, and iirc it works.

*** The problem with using Windows Explorer as you all know is that it
stops when it gets to a bad file, but drive backup programs are designed
not to do that. I plan to back up C with XXClone also, as soon as buy
the pro-version, today.


I backed up my two volumes to one hard drive, meant to be mounted
internally, but I had it in a BlacX drive caddy, connected by USB. I
think that means I can use the backup drive to replace whichever
internal drive fails, and if necessary use Disk Management or a 3rd
party program DiskAssign, iirc, to assign the needed letter to what had
been the backup. Right? If the harddrive fails right now and I can't
run Disk Management from it, I can do so by booting from a Hirem's CD???
Or something similar?


Thanks for any help you can give.

micky wrote:
Hi, all.

Background. Please skip to the line of caps if background bores you.

I'm not sure whether to start a new thread, or continue with this one.
If your news reader is set up like mine and you keep posts for years,
tacking on to a post from Nov 8, 2014, will cause the new post to sort
way off the screen. But some people say to do this.

Quoted below this new text, you'll see most of the details of my
problem, a drive that clicked on startup.

After this thread in early November, I made sure I'd backed up the
partition on each drive I also stopped turning off the computer since
Nov 8 and turned off the option to go into sleep, because it once
clicked coming out of sleep.

I also checked with two methods, one of them Windows Disk Management,
and all partitiions, 2 drives in the computer, and one bare 1.5T drive
in a BlacX caddy, with a backup partition for each of the two partitions
I'm using, are listed as healthy, (still today after the problem got
worse).

And everything has been fine for the last 7+ weeks until AVG said it was
now inactive because I was supposed to restart about 2 weeks ago (to
update AVG) and I didn't.

So yesterday I backed up again, restarted, used it, hibernated, and
today it wouldnt' start!!

HERE IS THE CURRENT SITUATION.

The computer would not start. The Dell XP startup screen showed up and
the computer clicked about 6 times when the progress bar (a series of
green boxes) was 1/4 of the way across, then sometimes 2 times 1/2 of
the way and maybe 6 more times 3/4s of the way. When the progress bar
got to the end, it just stayed there, until a black & white screen came
on with "SATA secondary dfive 0 not found".

In the BIOS, SATA Primary Drive is listed by model number, the word
"auto" and the size of the drive, about 250G.
However the SATA Secondary Drive has Unknown for the model number,
"Auto", and a question mark for the size. It is actually the same model
as the other drive.

I used Hirem's CD and its Mini XP to look at my drives. Using that, I
could look at the file list, edit text files, modify them and save them
on both drives!!!! It's mostly*** only startup that it gets stuck
on.

Boot.ini is on the C: drive, and the default partition was also on the
C: drive, I think. The default line in boot.in said
Default multi (0) disk (0) rdisk (1) partition (2)
I guess it's 2 because there is a hidden win98 partition on that drive,
copied from y previous win98 computer.
The other choice in the list was
multi (0) disk (0) rdisk (0) partition (1)
and I made this the default in C:\boot.ini .

Now , at least once, it booted just fine, and everything works just as
it used to.

I don't understand. It still has to go to the Crive to read boot.ini,
but I guess that's the last thing it has to do with the C: drive. Now
boot.ini points to the D: drive, which iiuc is on rdisk0, which must be
okay! But even with the previous default to a partition on rdisk0, the
computer used the Windows directory on the D: partition. How is that
possible????? There is a windows directory on the C: drive,, but none
of the subdirectories ever get new files. (I forget how all this arose
because it's been maybe 8 to 10 years!!)

It's a DELL Dimension 4600, at least 10 years old, but it's been working
fine. I'm trying to fix up a much newer one, but always short of time.
This weekend is an example. I should be searching for a surgeon for my
just-diagnosed hyperparathyroidism, but instead I'm doing this.

I turned on the drive caddy, when I was last in the BIOS but the BIOS
didn't recornize it. I should have turned it on BEFORE I opened the
BIOS, right????? If I did that, and made the right partition of the
external drive the first hard drive in the boot order, that would also
have gotten me started, probably, right?

But what would be better, am I right, is to install the 2n'd partition
of the backup drive -- call the drive BBB- -- as the primary SATA drive,
an internal drive, and then back that up to a brand new drive? Rght???

Now the backup drive, BBB, has two other partitions on it, two copies of
the XP partition, one of them from yesterday, and the other older. It
will sitll work as the internal drive, right?

But maybe it's not so good to back up two different partitions to the
same drive, because of the situation I'm in now. Maybe what I shoudl do
is buy one or two more bare drives. The BlacX caddy has two slots and
it says you can copy a partition from a drive in one slot to a drive in
the other.

Any advice will be appreciated.

As it stands now, it seems if the second internal drive fails, I can
just install the backup drive in place of both current internal drives??

***One more thing. It doesn't click anymore since Windows is past
startup, but it pings quite a bit. 2 times, 4 times, 10 imes . It
didn't do that until the last hour yesterday. I'm in the basement and
it sounded like a thin screwdriver striking a water pipe, maybe from the
townhouse next door. When I put my hand on the drives, I don't feel
anything when I hear the noise. This is like clicking, right? It
means the drive is about to fail??

It's possible for a SATA drive "to go nuts". Which
would normally be fine and all, except for the little
detail that pressing the RESET button has no effect
on SATA drives. Similarly, doing a restart from Windows,
would not improve the situation.

The IDE ribbon cable, has an actual RESET signal on the
cable. And that means, the same RESET signal which adds
sanity to motherboard hardware, brings the IDE drive into
line too.

If a SATA drive "disappears" while you're using it, and
shows in the BIOS as ??? or whatever, the SATA drive could
be in need of some sanity. To do that, you need to cycle
the power on the machine. At least, power down using the
button on the front. You don't have to switch off at the back,
but you can if that's the only other option you've got. Wait
30 seconds, before switching on again at the back (to give any
inrush limiter, time to cool off).

When I had this happen (SATA disappears), the problem stopped after
I re-wired the power distribution in the computer, and put fewer
loads on that cable.

Whereas lots of other funny things can happen on the computer
(and I don't follow all of your OS test cases), the BIOS one
is pretty cut and dried. Even a relatively sick drive can manage
to dump a model number and size information into the BIOS. It
doesn't take too many SATA packets to do that. To get the model
number from a SATA drive:

1) Drive must be able to spin up, load the heads, read the extended
firmware off "track -1", "become ready", and listen for commands
from the BIOS.

And to do that much, the drive must be sane. If the CPU or the
comm channel on the drive hang up, then the drive will remain
oblivious to everything you do. And power cycling will fix it.

2) The SATA port must be enabled in the BIOS, for the BIOS code to
consider wasting time probing the port.

3) It only takes a few packets to get the model number and size
information. Even if the channel was flaky and the BIOS did
multiple retries, it is likely to eventually get the info.
Whereas plenty of other command sequences, like writing big files,
could die on a "delayed write failure" due to comm channel problems.
So of the hard drive responses, not seeing the drive in the BIOS
is an indicator that something went nuts - or, that the drive
spun down on a serious error and was unable to come back up.

Rather than follow all the weird "does this with WinXP, does
that with Hiren", I'd sooner just download the Seagate diagnostic
and use that. But beware that these diagnostics are not kept
up to date and maintained all that well. You're a good candidate
for the diagnostics, because you have an old computer, and
they're more likely to work on an old computer. On my two most
modern computers, I can't get either diagnostic to run
(Windows or MSDOS). I think my P4 2.8GHz computer is a good candidate.

Paul