Need cheap hard drive recovery service - Any suggestions?

[Vanguard]

My aunt's second hard drive died. It has all her data. The system
boots okay using the first hard drive where the OS was installed. Yeah,
she's one of those typical users that never backup anything and thinks
hard drives last forever. The second hard drive is a Maxtor 40GB (I
won't know the model number until I yank it out of the host).

The problem started yesterday when she heard a double chirp noise every
20 seconds. I disconnected the internal speaker and powered down the
external speakers and sure enough the 20-second double chirp noise would
still occur. After some testing and listening around, I found it was
the Maxtor hard drive. The voltages on the power tap for it are good
(+12.26V and +5.08V). When I unplug the first hard drive and just have
the Maxtor plugged in, I powered up cold to see if I could hear it spin
up. I didn't hear the typical increasing pitch in whine that a hard
drive makes as it spins up but I could still hear some mechanical
sounds, like the heads trying to move. From her description of the
first noise she heard and the fact that I cannot hear it whine up on
powering it, it probably seized up. It is not recognized using the
Promise Ultra100 controller card or by the motherboard's IDE controller,
so there is no way to get at the drive using software recovery tools.
Since the hard drive isn't spinning up, some physical recovery method
will be needed.

The Maxtor hard drive died. My mention of backups brought on that
familiar deer caught in headlights look from her. I doubt it would be a
problem with the PCB on the hard drive, plus I wouldn't know where to
find another exact model match on this 4- or 5-year old Maxtor product
to try swapping out the PCB. My guess, and this is just a guess, is
that the bearings wore out and the spindle seized up. However, the hard
drive obviously doesn't have a speaker so I'm not sure what mechanicals
inside the drive could generate the chirping noise. That is every 20
seconds could be the drive trying to spin up again after waiting that
long (maybe there's some thermal override on the motor to prevent it
from burning out). She mentioned that there was a burning smell right
after hearing some other shooshing noise yesterday after which the
20-second chirping started.

Without backups but because she wants to recover her data off the dead
drive, I'm thinking she will have to send it to a recovery service with
the clean room facilities to open the drive, remove the platters, and
read off of them anything they can retrieve of her data files. This
isn't cheap. What are the typical prices at the recovery centers?
Which ones are good? She doesn't want to spend over $1,000 to recover
her files (i.e., that's her limit). Of the sites that I've visited so
far, none provide online quotes for pricing. They all want you to call
them or send them an e-mail to give you a rough estimate. I thought
awhile back that I saw someplace that said they charged $20/gigabyte,
but I suspect there was also an initial fixed charge.

Amazing how users never think of or bother with doing backups until they
finally get burned for being lazy.

 

[Pen]

I've seen several people who have had some
success freezing the drive for 4 hours in a
ziplock bag and getting it to run long enough to
get the data off.

 

[Vanguard]

I've seen several people who have had some
success freezing the drive for 4 hours in a
ziplock bag and getting it to run long enough to
get the data off.

Well, I could give that a try. Sounds weird, but if it works long
enough to copy off the files then my aunt will be happy.

 

[kony]

On Sun, 26 Jun 2005 13:37:11 -0500, "Vanguard"

she mentioned that there was a burning smell right
after hearing some other shooshing noise yesterday after which the
20-second chirping started.

I would suspect the motor control chip, which was often a
SO8 chip. If you can find a replacement it might be worth a
shot. If there is circuit (component) damage I would wait
on the freezer attempt, as it is less likely to work.

Without backups but because she wants to recover her data off the dead
drive, I'm thinking she will have to send it to a recovery service with
the clean room facilities to open the drive, remove the platters, and
read off of them anything they can retrieve of her data files. This
isn't cheap. What are the typical prices at the recovery centers?
Which ones are good? She doesn't want to spend over $1,000 to recover
her files (i.e., that's her limit). Of the sites that I've visited so
far, none provide online quotes for pricing. They all want you to call
them or send them an e-mail to give you a rough estimate. I thought
awhile back that I saw someplace that said they charged $20/gigabyte,
but I suspect there was also an initial fixed charge.

I don't think you'll find anything under $1000. Might be
better to try finding a suitable replacement circuit board.

 

[--tom--]

Which ones are good? She doesn't want to spend over $1,000 to recover her
files (i.e., that's her limit).

www.retrodata.co.uk

UK firm but deal internationally. Shouldn't cost you more than $40 or so to
ship the drive insured to the UK.

Very good. Used them a number of times and couldn't question the
workmanship, speed, and overall service.

 

[Vanguard]

On Sun, 26 Jun 2005 13:37:11 -0500, "Vanguard"



I would suspect the motor control chip, which was often a
SO8 chip. If you can find a replacement it might be worth a
shot. If there is circuit (component) damage I would wait
on the freezer attempt, as it is less likely to work.



I don't think you'll find anything under $1000. Might be
better to try finding a suitable replacement circuit board.

But would a fault in the PCB cause the platters to not spin up on a cold
start? I know you can send a command to the drive (as part of the power
saving feature) to have it stop spinning its platters but when the drive
first comes up from a cold boot then it hasn't received any commands
yet, especially since the ribbon cable isn't connected during the tests.

The reason that I think the spindle seized up is the description of the
sound that my aunt mentioned when the problem started. Plus the fact
that I cannot hear the motor start spinning up the platters when power
is first applied. I strongly suspect that finding a replacement PCB
won't help if the spindle or bearings seized.

I had one drive a long ago also have a bearing wearout. I took it to
the computer shop and the guy was trying to test it but couldn't get the
drive recognized. He had it hooked up to power and was spinning when I
came around the counter into this work area because it was taking him so
long. He couldn't figure it out until I touched the drive. I knew it
was running way more hot than I thought it should (it was toasty when I
removed it from the computer but it had been spun down for awhile). I
almost burned my hand. It was so hot you could cook eggs on it.
Problem identified: spindle or bearing wear. What I had noticed is that
it wouldn't spin up on a cold boot unless I gave it a whack. That was
back when Seagate used to have a plastic coating in the rust dust and it
got soft at high heat. The heads would eventually land on the platters
and cause a tiny divot that was still too large to get the weak motor to
start the platters spinning until the heads got unstuck. It worked long
enough to get off the data but I got rid of the drive before it
eventually siezed up.

Unless the drive actually makes noise, it could wear and run extremely
hot for a long time and then sieze up. The "shunk" noise verbalized by
aunt sounded a lot like a drive that got stopped way too quick.
However, my aunt also mentioned the plastic smell at the time of the
noise and when the chirping started, and I don't recall a hot siezed up
hard actually having a smell (other than the stale smell of hot dust
coming off of it).

 

[kony]

But would a fault in the PCB cause the platters to not spin up on a cold
start?

Yes, the spindle motor is not directly connected to the
power rails but rather through a controller, which
facilitates the precision RPM control.

The reason that I think the spindle seized up is the description of the
sound that my aunt mentioned when the problem started. Plus the fact
that I cannot hear the motor start spinning up the platters when power
is first applied. I strongly suspect that finding a replacement PCB
won't help if the spindle or bearings seized.

It's possible the bearings froze. However, a circuit board
failure could prevent the drive from spinning up. Perhaps
it's even possible that circuit board failure caused the
motor to "violently" short and damaged the bearings due to
the sudden load- or perhaps it isn't possible, I dont' know.

I had one drive a long ago also have a bearing wearout. I took it to
the computer shop and the guy was trying to test it but couldn't get the
drive recognized. He had it hooked up to power and was spinning when I
came around the counter into this work area because it was taking him so
long. He couldn't figure it out until I touched the drive. I knew it
was running way more hot than I thought it should (it was toasty when I
removed it from the computer but it had been spun down for awhile). I
almost burned my hand. It was so hot you could cook eggs on it.
Problem identified: spindle or bearing wear. What I had noticed is that
it wouldn't spin up on a cold boot unless I gave it a whack. That was
back when Seagate used to have a plastic coating in the rust dust and it
got soft at high heat. The heads would eventually land on the platters
and cause a tiny divot that was still too large to get the weak motor to
start the platters spinning until the heads got unstuck. It worked long
enough to get off the data but I got rid of the drive before it
eventually siezed up.

Unless the drive actually makes noise, it could wear and run extremely
hot for a long time and then sieze up. The "shunk" noise verbalized by
aunt sounded a lot like a drive that got stopped way too quick.
However, my aunt also mentioned the plastic smell at the time of the
noise and when the chirping started, and I don't recall a hot siezed up
hard actually having a smell (other than the stale smell of hot dust
coming off of it).

If you can reverse engineer the PCB a bit, you may be able
to determine if applying a very low, incrementally rising
voltage would result in the motor spinning. I would imagine
then a tach signal could be sensed- even if you can't
determine if RPM range is right, if the tach signal varies
it should indicate rotation (which at very low/safe voltage
might not cause the patter rotation to be loud enough to be
audible. Alternatively if you can find the specs for the
motor, you might be able to increase the voltage a bit more.
I've never tried this but the method seems solid enough.

Of course there is the other option of attempting to provide
a clean-room environment inside a plastic baggie- hold the
lid down and take out the screws, then seal it inside a
plastic bag along with a probe. With entire thing sealed
and grasping the cover through the bag, take the cover off
and carefully use the probe to manipulate the top of the
center mounting plate (on the center of the platter), seeing
if it turns easily.

 

[Shep©]

Well, I could give that a try. Sounds weird, but if it works long
enough to copy off the files then my aunt will be happy.

Weird Maybe,but it works.BTSTGTTS.

Put the drive inn your deep freeze in a sealed plastic bag overnight
and try again

 

[Vanguard]

Yes, the spindle motor is not directly connected to the
power rails but rather through a controller, which
facilitates the precision RPM control.



It's possible the bearings froze. However, a circuit board
failure could prevent the drive from spinning up. Perhaps
it's even possible that circuit board failure caused the
motor to "violently" short and damaged the bearings due to
the sudden load- or perhaps it isn't possible, I dont' know.



If you can reverse engineer the PCB a bit, you may be able
to determine if applying a very low, incrementally rising
voltage would result in the motor spinning. I would imagine
then a tach signal could be sensed- even if you can't
determine if RPM range is right, if the tach signal varies
it should indicate rotation (which at very low/safe voltage
might not cause the patter rotation to be loud enough to be
audible. Alternatively if you can find the specs for the
motor, you might be able to increase the voltage a bit more.
I've never tried this but the method seems solid enough.

Of course there is the other option of attempting to provide
a clean-room environment inside a plastic baggie- hold the
lid down and take out the screws, then seal it inside a
plastic bag along with a probe. With entire thing sealed
and grasping the cover through the bag, take the cover off
and carefully use the probe to manipulate the top of the
center mounting plate (on the center of the platter), seeing
if it turns easily.

Geez, you gotta be right all the time, don't you! Take a look at a
picture of the PCB on the Maxtor drive at:

http://home.comcast.net/~vanguard.2/pictures/maxtor_bad.jpg

Yep, there's a blown chip. If you zoom in on the identified area on the
PCB (you might have to download the picture and using the built-in image
viewer, IrfanView, XnView, or whatever you like), you'll see the plastic
casing for the chip labelled as "U503" has a divot where the pin blew
out and a burnt solder pad. I couldn't see this with my naked eye and
only spotted it after using a lighted magnifier glass. I checked the
voltages at the power tap used for the drive and they were +12.26 and
+5.08, so I'm not sure why the component would've blown on the PCB.
Maybe there was enough friction with a worn drive to up the amount of
current that it was trying to regulate.

So now I'm trying to hunt down a replacement PCB for the drive. I've
sent off a tech support e-mail (via their web form) to Maxtor but I
don't know if they'll let me buy just the PCB or even deal with me. The
warranty for it expired back on March 2003. I'm still hunting around
for a USA online vendor that still carries this discontinued model
(Maxtor Diamond Max 40GB UDMA-66, model 94098U8).

 

[Vanguard]

Weird Maybe,but it works.BTSTGTTS.

Put the drive inn your deep freeze in a sealed plastic bag overnight
and try again

That didn't work. See my other post as a reply to Kony. I found a
blown chip on the PCB which was what my aunt probably smelled at the
time the chirping started.

 

[Noozer]

My aunt's second hard drive died. It has all her data. The system boots
okay using the first hard drive where the OS was installed. Yeah, she's
one of those typical users that never backup anything and thinks hard
drives last forever. The second hard drive is a Maxtor 40GB (I won't know
the model number until I yank it out of the host).

isn't cheap. What are the typical prices at the recovery centers? Which
ones are good? She doesn't want to spend over $1,000 to recover her files
(i.e., that's her limit). Of the sites that I've visited so far, none
provide online quotes for pricing. They all want you to call them or send
them an e-mail to give you a rough estimate.

So... your aunt was too lazy to do backups and you're too lazy to call
someone for a rough estimate?

This data doesn't sound very important to me.

 

[Noozer]

Geez, you gotta be right all the time, don't you! Take a look at a
picture of the PCB on the Maxtor drive at:

http://home.comcast.net/~vanguard.2/pictures/maxtor_bad.jpg

Looking at that picture, that chip seems to be a "driver" chip for one of
the "steps" on the spindle motor. I haven't looked at the specs for that
part number, but suspect that this pin is a power input, since these three
chips seem to be tied together at that pin by the winding trace.

Most likely the bearings have seized up and caused the motor to short,
popping that chip in the process. If you can get the platters to spin AND
replace that chip you might be able to get the data of the drive. If you
just swap the PCB, it will probably blow again due to the siezed bearing.

If it were me, I'd be pulling the drive open CAREFULLY and trying to unstick
the platters. I've done it before and the drive lasted years afterwards -
but I never used it for important data after that.

 

[Vanguard]

So... your aunt was too lazy to do backups and you're too lazy to call
someone for a rough estimate?

This data doesn't sound very important to me.

How many recovery centers are open on Sunday? I didn't even bother to
call because, well, it's Sunday. If I don't hear anything back from
Maxtor, I'll probably buy another Maxtor 94098U8. I found one for $70.
If that PCB blows, I'll consider your suggestion in your other reply of
opening the drive and making sure the platters aren't seized, and buy
another drive at $70. I doubt even Maxtor is going to sell me a PCB
replacement, if they have it, for much less than the cost of the 2
replacement drives (to cannibalize their PCBs).

It's a personal computer, and I do mean *personal*. It is her home
computer. Is any data you have on your home computer (assuming you
don't also do real company work on it, which begs the question of why
unless your company provided the computer) worth $1000 on your data-only
hard drive? In a company of 500 employees, I can count on my hands how
many users that I know who do regular backups, and they're in my QA
department or the IT folks. Users don't backup until they've been
taught the hard way. It took a year of bitching to my father to get him
to finally get backup hardware and software and then to actually use it,
and his home computer was used for his home office (i.e., it was for his
business!), and even he didn't start until he lost data for a bid (he's
a mechanical contractor) and had to recompute it all over. Putting a
computer in someone's hands doesn't guarantee they get smarter.

 

[CBFalconer]

.... snip ...

It's a personal computer, and I do mean *personal*. It is her home
computer. Is any data you have on your home computer (assuming you
don't also do real company work on it, which begs the question of
why unless your company provided the computer) worth $1000 on your
data-only hard drive? In a company of 500 employees, I can count
on my hands how many users that I know who do regular backups, and
they're in my QA department or the IT folks. Users don't backup
until they've been taught the hard way. It took a year of bitching
to my father to get him to finally get backup hardware and software
and then to actually use it, and his home computer was used for his
home office (i.e., it was for his business!), and even he didn't
start until he lost data for a bid (he's a mechanical contractor)
and had to recompute it all over. Putting a computer in someone's
hands doesn't guarantee they get smarter.

The problem is that backup takes time and patience. The cure is to
make it quick and easy. The way to make it quick is to mount a
second drive, and software to maintain that as a clone of the
first, without excess copying. The lack of excess copying is what
makes it quick (in my case in the order of 5 minutes), together
with the use of a hard disk.

To do this simply mount xxcopy. Read its instructions, especially
for the /clone switch. Then create a batch file or alias (if you
are using 4dos) to run it. On my machine I simply enter "backup
<enter>" and one further confirmatory <enter> for the destination
drive. It can even go on in the background. This batch or alias
is what makes it easy.

Staples was selling 60G Maxtors for $45 last week. In your case I
would simply buy two of them. Once you get the data that is
savable onto one, install it as the main drive and install the
second as the clone. If you have your drives power down on
inactivity it won't even create heat. When (not if) the main drive
fails, you then just swap drives, replace the failed item, and
carry on.

 

[Noozer]

This data doesn't sound very important to me.

How many recovery centers are open on Sunday? I didn't even bother to
call because, well, it's Sunday.

Touche... got me there.

 

[larry moe 'n curly]

Take a look at a picture of the PCB on the Maxtor drive at:

http://home.comcast.net/~vanguard.2/pictures/maxtor_bad.jpg

Yep, there's a blown chip.

The spec sheet for the chip can be seen at:

www.fairchildsemi.com/ds/ND/NDS8961.pdf#search='nds8961'

The National Semiconductor (Fairchild) and International Rectifier (IR)
parts have very different numbers but are probably the same, and I once
fixed an 80GB Maxtor with IR chips with National chips from a 7.5GB
Maxtor. They were the chips located higher up in your picture and used
for moving the heads (clicky-clacky sound). All six of the chips
normally run hot, about 70C even in 25C surrounding air, but Maxtor
said that was OK, and the spec sheet says that 150C is the limit. Also
any air flow at all over the chips, even from just orienting the drive
vertically, can cool them by 10-20C.

 

[larry moe 'n curly]

Staples was selling 60G Maxtors for $45 last week.

Through tomorrow, Fry's has 120G WDs for $20 after $60 rebates, and a
few months ago OfficeMax charged just $50 for a 120G WD external. I
realize that these can't be used to fix a Maxtor, but it shows just how
cheap backup drives can be.