HD failure...

Do you have a question? Post it now! No Registration Necessary.  Now with pictures!

Threaded View
I have a 5(?) yr old Dell dimension that the HD started making clicking
Since it would still boot up, I tried to clone it, but it crapped out for
I replaced with a new WD 80G HD, reloaded & all was fine for about a month.
Now, same original symptoms are back, clicking sounds & not booting up.
I downloaded the diagnostic software from WD's site & burnt to a bootable
CD, but it wouldn't see the drive.
Since this is still under warranty, I called WD's tech support, who
suggested I put in another machine as a slave & try to run the diagnostics,
but it's still not seeing it.
Rather than drop another HD in there, I am wondering if there could be
something with the computer that could be trashing the drives.
Any suggestions?

Mr. Simpson, there are thousands of people like you with no discernable
Yeah, they're called Congress.

Re: HD failure...

M'Balz Es-Hari wrote:
Quoted text here. Click to load it

The clicking sound implies the drive is spinning at speed (so
has enough 12V on the supply, to be in a position to try to
work). The clicking or clunking happens when the disk controller
is having trouble finding a particular track, and then it commands
the head assembly to go back to track zero. It may hit a mechanical
stop while doing so. That suggests to me that the drive itself has
a problem.

The drive is sensitive to the voltage given to it. Some BIOS
have a hardware monitor screen in the setup, and you can read the
main voltage values there. So you may be able to tell whether
the +5V and +12V are within 5% of their normal value. If the 12V
is a little low, the drive may sense that and refuse to spin up.
And then there would not be a clunk from it. Sometimes you can
get a hum or buzz from a drive, if the motor is stuck and
it is unable to start the spindle spinning.

If the voltage occasionally goes out of spec, the clunking
may even only happen when the voltage is low. And work
normally at other times.

A disk will not respond, give its identity string in the
BIOS and the like, unless it was able to spin up, reach
full speed, and read a bunch of stuff during its internal
test and initialization. If it cannot be seen on a second
machine, you've double checked the jumper settings and
they're correct for what you've got connected to the
cable, then you have to consider the possibility that
it is a drive failure.

A future option, would be to purchase a PCI IDE or SATA
card, as a means to host disk drives. To use it with a
brand new disk, and install an OS from scratch, you'd press
F6 during the beginning of the install, and present a
floppy with drivers for the card. One of the files on
the floppy would be TXTSETUP.OEM, which tells the OS installer
what the driver is.

If the computer was still working, what you'd do is install
the drivers for the new PCI card, and then move the ribbon
cable from the motherboard connector, to the IDE connector
on the card. And then the system should be able to boot. I
used that method (PCI card), to move my Win2K disk from my
old computer, to the new one, and boot from it. But to do
it, I had to install a driver before moving the ribbon cable
to use the new card.

You could try a PCI card, when connecting the next replacement
drive. I don't know if a disk diagnostic will work with
such a card, and the maximum compatibility comes from
using the motherboard connector for that. (At least on
a five year old system it would.) Using the PCI card, would
eliminate some kind of problem with the signals coming
from the current connector. And really shouldn't be necessary.
But if you want to change something, your choices are a
new power supply, or a PCI IDE card. The PCI IDE card is
cheaper than a new motherboard, and is also a handy thing
to have in your parts box. I have three of them here, I


Re: HD failure...

Quoted text here. Click to load it

Have you checked the cooling fans/filters for blockage (fluff/dust). It just
could be that they are being overheated.

Re: HD failure...

On Wed, 17 Dec 2008 00:22:42 -0600, "M'Balz Es-Hari"

Quoted text here. Click to load it

It could be that you're the 1 in ~ 100 people unlucky enough
to have had an infant mortality, that the new drive failure
isn't directly related to the old one failing.

Otherwise, overheating can cause damage but this was a short
period of time and presuming you're in Texas, it shouldn't
be very warm there this time of year.  I'd think a failing
power supply is most likely.  Several parts can fail in a
PSU but given that it runs still, vented capacitors would be
my best guess... I mention this because if you pull the PSU
and leave it unplugged for a few minutes, this kind of
failure is often visible with the cover off.

Beyond that, if you don't have a spare compatible PSU,
assuming it takes a standard PSU so the replacement cost
isn't excessive, I'd buy a PSU from someplace with a good
return policy in case it's not the problem.

If it takes a proprietary form factor PSU, it may not be
worthwhile to repair this system.  In that situation you
might be able to remount the parts in a standard ATX case
that accepts as many of the original parts as possible plus
a new standard PSU, or it might be time to start over with a
new case, PSU, motherboard, CPU, etc... at least it'd be an

Site Timeline