Dead IBM 60GXP solutions?

[default]

Hi,

Today one of my IBM 60GXP 41gb drives spontaneously died. It now just makes
a beep-click, beep-click, beep-click sound only. It appears to have gone
offline and is no longer seen by the drive controller. Drive Fitness Test
cannot see it either. The drive does still spin up however and doesn't make
any unusual mechanical noise so the motor/bearings seem intact. It would
appear that it is the electronics that has gone haywire.

The drive is still under warranty, but I would like to get the data off of
it if possible. The data is backed up, but it would still be easier to
clone the drive and keep going than to have to reinstall windows, software,
restore backups etc.

I have 5 others of these drives (41gb gxp60's), 4 of which have failed in
various different ways including one that keeps getting bad sectors, but the
DFT erase drive function will repair for a week or so before it fails again.
Is it reasonable to take the electronics pcb from this drive and mount it on
the newly failed drive so that it could be maybe be run one more time so I
can just copy the drive to a new one? They have the same firmware revision.
Has anyone been successful in doing this?

Thank you for your responses.

 

[Rod Speed]

Put a stake thru its heart to make sure it really is dead.

Today one of my IBM 60GXP 41gb drives spontaneously died.

They aint called DeathStars for nothing.

It now just makes a beep-click, beep-click, beep-click sound only.

Thats the drive recalibrating when it cant read the platters anymore.

It appears to have gone offline and is no longer seen by
the drive controller. Drive Fitness Test cannot see it either.

That means its very unfit indeed.

The drive does still spin up however and doesn't make any
unusual mechanical noise so the motor/bearings seem intact.
Correct.

It would appear that it is the electronics that has gone haywire.

Or the connection to the heads etc.

The drive is still under warranty, but I would like to get the
data off of it if possible. The data is backed up, but it would
still be easier to clone the drive and keep going than to have
to reinstall windows, software, restore backups etc.

Pointless doing anything other than a reinstall on the replacement.

I have 5 others of these drives (41gb gxp60's),
4 of which have failed in various different ways

Told you they aint called DeathStars for nothing.

With a record like that, I'd personally have cloned the drive
before it died and put a stake thru its heart to ensure that no
one, including yourself, would be silly enough to try using it.

As you have discovered, even if you're fully backed up,
death of the boot drive is a complete pain in the arse.

including one that keeps getting bad sectors, but the DFT erase
drive function will repair for a week or so before it fails again.

Pretty typical failure mode for IBM GXP drives.

You should put a stake thru its heart too.

Is it reasonable to take the electronics pcb from this drive and mount it on
the newly failed drive so that it could be maybe be run one more time so I
can just copy the drive to a new one? They have the same firmware revision.

Has anyone been successful in doing this?

No reports of anyone succeeding in doing that with an IBM 60GXP drive.

She's dead Jim. You into necrophilia ?

 

[default]

Thank you Rod. I may try swapping the board anyway. There is not much to
lose. This was the last of the original 60gxp "DeathStars" still in use and
as this one had never given trouble, I thought it may be reliable. And
since I was 100% backed up, I only lost a few hours of work. No more IBM
drives here again. The 75GXP's died, the 34GXP died spectacularily with a
loud screach and crunching and grinding sounds, the 60GXP's died in various
ways often ending up offline. All of these drives were kept very cool with
a fans blowing directly onto them too. DFT was run at least monthly to
check for errors and still...

Now I am worried about the 1 120GXP in use. I would have hoped that IBM
would have started making them reliable after all the warranty returns that
they must have had with the previous versions.

I am going to send a big box of failed gxp's in to Hitachi for repair
shortly. Can the refurbished units that come back from Hitachi be trusted
or should I give them out for Christmas gifts for the less liked?

Which drives are now most reliable in the consumer market. Is it the
Western digital JB series with the 8mb cache and the 3year warranty?

Thank you.

 

[Rod Speed]

Thank you Rod. I may try swapping the
board anyway. There is not much to lose.

True, nothing to lose except a bit of time, particularly
if you dont care about getting the drive RMAed.

This was the last of the original 60gxp "DeathStars" still
in use and as this one had never given trouble, I thought
it may be reliable. And since I was 100% backed up, I only
lost a few hours of work. No more IBM drives here again.

Yeah, I wouldnt personally take one as a free gift.

My time has real value to me.

The 75GXP's died, the 34GXP died spectacularily
with a loud screach and crunching and grinding sounds,
the 60GXP's died in various ways often ending up offline.
All of these drives were kept very cool with a fans blowing
directly onto them too. DFT was run at least monthly to
check for errors and still...

Hell of a record.

Now I am worried about the 1 120GXP in use.

Yeah, I'd bin it personally before it bites.

I would have hoped that IBM would have started
making them reliable after all the warranty returns
that they must have had with the previous versions.

Looks very like they just gave up on making reliable
hard drives and flogged the entire operation off to
the biggest sucker they could find.

It remains to be seen if Hitachi can actually fix the problem.

I am going to send a big box of failed gxp's in to Hitachi for repair shortly.
Can the refurbished units that come back from Hitachi be trusted
Nope.

or should I give them out for Christmas gifts for the less liked?

Yeah, give them to people you hate |-)

Which drives are now most reliable in the consumer market.

Hard to say. I'm personally going with Samsungs now, but they
are well down the list volume of sales wise so its not at all clear
just what their reliability is like. They do have full 3 year warrantys
on all their drives, which may not be a coincidence.

Is it the Western digital JB series with
the 8mb cache and the 3year warranty?

I just went for the Samsung P80 drives myself.

Loverly quiet drives.

 

[Matt]

Hi,

Today one of my IBM 60GXP 41gb drives spontaneously died. It now just makes
a beep-click, beep-click, beep-click sound only. It appears to have gone
offline and is no longer seen by the drive controller. Drive Fitness Test
cannot see it either. The drive does still spin up however and doesn't make
any unusual mechanical noise so the motor/bearings seem intact. It would
appear that it is the electronics that has gone haywire.

The drive is still under warranty, but I would like to get the data off of
it if possible. The data is backed up, but it would still be easier to
clone the drive and keep going than to have to reinstall windows, software,
restore backups etc.

I have 5 others of these drives (41gb gxp60's), 4 of which have failed in
various different ways including one that keeps getting bad sectors, but the
DFT erase drive function will repair for a week or so before it fails again.
Is it reasonable to take the electronics pcb from this drive and mount it on
the newly failed drive so that it could be maybe be run one more time so I
can just copy the drive to a new one? They have the same firmware revision.
Has anyone been successful in doing this?

Thank you for your responses.

Call me lucky, but I am using various IBM drives, 2 75GXPs as a RAID 0 (!)
for my o/s, and two 120GXPs as data. I do keep them cool, which is key. I
did have a problem with a 120 GXP similar to what you describe and was able
to get it to work temporarily, long enough to copy the data, by gently
bending, and I mean GENTLY, the part where the data cable plugs into the
drive down away from the top of the drive. This seemed to restore whatever
contact was faulty so that I could get my data off the drive. IBM then sent
me a new Hitachi 180 GXP. I think this area of the drive is flimsy and the
circuit board is not strong enough to handle the loads associated with
plugging and unplugging the cable.

 

[default]

Hi Matt,

Thank you for your response. These drives were kept cool for their entire
lives. They did run 24x7 but that should be good for them compared to
starting and stopping all the time I have heard.

I tried replacing the PCB with one from an identical model, but this was
unsuccesful. I even reflowed the head assembly contacts on the PCB with a
hot-air rework center that I have access to. I also inspected every solder
joint of every IC with an industrial inspection stereo microscope. There
was nothing visibly wrong. After replacing the PCB, the drive stopped
making the beep-click sounds but I still could not get it to move any data.
Oh well. At least everything was backed up and I only lost a few hours of
work.

I originally used 4 60gxp's in a raid zero array. This was way too
unreliable, although it was a very good performance between drive failures.
Later I switched them into Raid1 mirrored arrays, but even this was quite
unreliable. Often both drives would get bad sectors at once and I would
have to repair both drives.

I won't be buying another ibm hard drive again.

 

[Rod Speed]

Call me lucky, but I am using various IBM drives, 2 75GXPs as a RAID 0 (!)
for my o/s, and two 120GXPs as data. I do keep them cool, which is key.

You dont know that that is key.

I did have a problem with a 120 GXP similar to what you describe
and was able to get it to work temporarily, long enough to copy the
data, by gently bending, and I mean GENTLY, the part where the data
cable plugs into the drive down away from the top of the drive. This
seemed to restore whatever contact was faulty so that I could get my
data off the drive. IBM then sent me a new Hitachi 180 GXP. I think
this area of the drive is flimsy and the circuit board is not strong enough
to handle the loads associated with plugging and unplugging the cable.

Classic dry joint/cracked solder problem.

 

[Baron Von Black]

You dont know that that is key.


Classic dry joint/cracked solder problem.

Ripped my GXP60 appart last night as it failed again!

Rather than just binning it I decided to have a nosey.

Once I removed the cover to inspect the Platters, everything appeared to be
clean
with no signs of Head contact. so I proceeded to removed the 2 screws which
hold the
Connector block which mates with the PCB below.

Loan behold dry joints all along where it solders to the orange cable which
goes to the head assembly.

Touched the up with a small soldering iron and tried it.

Formatted with a few errors, as expected but certainly explains a few probs
hey people.

Naturally I won't be using the drive for serious stuff, but may use it to
see if the probs reoccur.

 

[Rod Speed]

Ripped my GXP60 appart last night as it failed again!

Rather than just binning it I decided to have a nosey.

Once I removed the cover to inspect the Platters, everything appeared to be
clean
with no signs of Head contact. so I proceeded to removed the 2 screws which
hold the
Connector block which mates with the PCB below.

Loan behold dry joints all along where it solders to the orange cable which
goes to the head assembly.

Interesting.

Hell of an inditement of IBM that they didnt fess up to that
problem if that is what has produced most of the problems tho.

 

[Baron Von Black]

Interesting.

Hell of an inditement of IBM that they didnt fess up to that
problem if that is what has produced most of the problems tho.

Yep! sure is Rod, as you and no doubt many others are aware, most
manufacturers
like seagate, WD etc mate the two components via a small PCB.

Pin contacts below to interface with a PCB, and connector on top for the
head cable.

Shame on IBM of it was to be proved that skimping on the pennies has cost
them pounds.

Never mind the reputation.

 

[chrisv]

Never mind the reputation.

How about your's? If you insist on using broken software to post to
USENET, how about deleting the mangled mess that it, and you, leave
behind? Sheesh.

 

[Baron Von Black]

How about your's? If you insist on using broken software to post to
USENET, how about deleting the mangled mess that it, and you, leave
behind? Sheesh.

Sorry Dad