Damage by connecting bad hard drive

[Jim (Barry) Roche]

How much damage can be done by connecting a bad hard drive to an EIDE
interface? Last night I received a panic request to extract data from
the hard drive of a "PC that had just stopped working". I connected it
to a known good machine which subsequently refused to boot with and now
without the second drive. Machine now powers up, BIOS takes a long,
long time to see drives and then reports no system found etc. My
suspicion is that the MB/EIDE interface may have been damaged in the
first test. Any thoughts? What is the "correct" way to test a
suspected bad drive?

Thanks.

Jim

 

[Guest]

How much damage can be done by connecting a bad hard drive to an EIDE
interface? Last night I received a panic request to extract data from
the hard drive of a "PC that had just stopped working". I connected it
to a known good machine which subsequently refused to boot with and now
without the second drive. Machine now powers up, BIOS takes a long,
long time to see drives and then reports no system found etc. My
suspicion is that the MB/EIDE interface may have been damaged in the
first test. Any thoughts? What is the "correct" way to test a
suspected bad drive?

Thanks.

Have you double checked your hardware: drive cable(s),
jumper(s) etc? How was the second drive jumpered when
you attached it to the known good system?

Also check the boot order in your system bios to make sure
it wasn't changed.

 

[Mike T.]

How much damage can be done by connecting a bad hard drive to an EIDE
interface? Last night I received a panic request to extract data from
the hard drive of a "PC that had just stopped working". I connected it
to a known good machine which subsequently refused to boot with and now
without the second drive. Machine now powers up, BIOS takes a long,
long time to see drives and then reports no system found etc. My
suspicion is that the MB/EIDE interface may have been damaged in the
first test. Any thoughts? What is the "correct" way to test a
suspected bad drive?

Thanks.

Jim

If your suspect drive actually damaged a mainboard, that would be the first
I've ever heard of THAT happening. But anything's possible.

As others have said, check your jumper settings, reset your mainboard BIOS
(clear CMOS)...

But in the future, if you want to get the data off that drive, I'd probably
play it safe and hook it up to an external hard drive enclosure that
interfaces with the PC by USB or something. If it really damaged a
mainboard, it might damage the IDE circuit in an external enclosure also.
But that's a helluva lot easier to replace. -Dave

 

[Rod Speed]

How much damage can be done by connecting a bad hard drive to an EIDE interface?

Its theoretically possible to kill the controller, but its very very rarely seen.
In fact I dont recall ever seeing anyone report that happening, ever.

Last night I received a panic request to extract data from the hard drive
of a "PC that had just stopped working". I connected it to a known good
machine which subsequently refused to boot with and now without the
second drive. Machine now powers up, BIOS takes a long, long time
to see drives and then reports no system found etc. My suspicion is
that the MB/EIDE interface may have been damaged in the first test.
Any thoughts?

Its MUCH more likely to be something as basic as a bad
ribbon cable that got damaged by reefing it off the bad drive
so it now cant communicate with the original drive properly.
The obvious way to try that possibility is to try a different cable.

Its also possible that the bad drive has partially killed the power
supply but thats not very likely. More likely than killing the controller tho.

The long delay before the bios decides that there isnt
any bootable drive visible is normal, that happens
when it cant poll the drives for their basics properly.

What is the "correct" way to test a suspected bad drive?

The way you did it, obviously with a PC you
dont care about if you're a complete neurotic.

 

[Jim (Barry) Roche]

Its theoretically possible to kill the controller, but its very very rarely seen.
In fact I dont recall ever seeing anyone report that happening, ever.


Its MUCH more likely to be something as basic as a bad
ribbon cable that got damaged by reefing it off the bad drive
so it now cant communicate with the original drive properly.
The obvious way to try that possibility is to try a different cable.

Its also possible that the bad drive has partially killed the power
supply but thats not very likely. More likely than killing the controller tho.

The long delay before the bios decides that there isnt
any bootable drive visible is normal, that happens
when it cant poll the drives for their basics properly.


The way you did it, obviously with a PC you
dont care about if you're a complete neurotic.

Tks for the responses. I'm feeling a little more reassured after all
your comments. The PSU is an Enermax Liberty which has so far proven
to be bullet proof (fingers crossed, touch wood). In general, the
cables I use are those that come packaged with MB's and stored
carefully. The one mistake I did make last night was initially using a
40-pin cable instead of an 80. The BIOS picked this up and alerted me.
Could this have upset things?

Jim

 

[Rod Speed]

Rod Speed wrote

Tks for the responses. I'm feeling a little more reassured after
all your comments. The PSU is an Enermax Liberty which has
so far proven to be bullet proof (fingers crossed, touch wood).

Yeah, that should shut down gracefully if the bad drive overloads
either of the rails. Its only the cheap crap that can die when overloaded.

In general, the cables I use are those that come
packaged with MB's and stored carefully.

One problem with ribbon cables is that there is a small chance
that it wont be made properly in the first place. The little prongs
that bite into the ribbon can get bent over when its made and the
cable can work fine for quite a while and then the act of inserting
and removing from a hard drive can short to the adjacent etc.

Just try another cable.

The one mistake I did make last night was
initially using a 40-pin cable instead of an 80.
The BIOS picked this up and alerted
me. Could this have upset things?

Not in the sense of being a problem when you use the correct cable now.

Bet its just a bad cable.

 

[Ghostrider]

How much damage can be done by connecting a bad hard drive to an EIDE
interface? Last night I received a panic request to extract data from
the hard drive of a "PC that had just stopped working". I connected it
to a known good machine which subsequently refused to boot with and now
without the second drive. Machine now powers up, BIOS takes a long,
long time to see drives and then reports no system found etc. My
suspicion is that the MB/EIDE interface may have been damaged in the
first test. Any thoughts? What is the "correct" way to test a
suspected bad drive?

Thanks.

Jim

Probably no damage at all unless the onboard bios chip in the bad
hard drive were to write back to the motherboard bios and corrupt
the bios table containing HD information. But the solution to this
is also very easy...clear the bios and re-flash. Physical damage, as
others have mentioned, involving the cable, connectors, etc., is more
of a possibility.

In the future, to extract material from any suspect HD, use an external
USB or Firewire enclosure or dock.

 

[Jim (Barry) Roche]

Probably no damage at all unless the onboard bios chip in the bad
hard drive were to write back to the motherboard bios and corrupt
the bios table containing HD information. But the solution to this
is also very easy...clear the bios and re-flash. Physical damage, as
others have mentioned, involving the cable, connectors, etc., is more
of a possibility.

In the future, to extract material from any suspect HD, use an external
USB or Firewire enclosure or dock.

Thanks all for the accurate diagnosis - the IDE cable to the hard drive
(master) and CD/DVD burner (slave) on the working system had become
partially disconnected at the CD/DVD connector. Of course, I didn't
even look at this as it had not been touched until I went to reseat
last night.

Interesting side note: I went to buy some round IDE cables this morning
and they had very few in stock.

Thanks again.

Jim

 

[Rod Speed]

Thanks all for the accurate diagnosis - the IDE cable to the hard
drive (master) and CD/DVD burner (slave) on the working system had
become partially disconnected at the CD/DVD connector. Of course, I
didn't even look at this as it had not been touched until I went to
reseat last night.

Interesting side note: I went to buy some round IDE cables this morning

I dont use those, they flout the ATA standard.

 

[Jim (Barry) Roche]

I dont use those, they flout the ATA standard.

So round ATA cables should be avoided then? Thats OK as they did
appear to be ridiculously expensive anyway.

 

[Rod Speed]

Rod Speed wrote

 

[Jim (Barry) Roche]

So back to the beginning: I tried this setup last night - known good HD
(master) and CD/DVD (slave) on Primary IDE and suspect drive (master)
on secondary IDE. BIOS recognized everything but Windows filed to
load. Got splash screen and then black screen with cursor flashing in
upper left hand corner. Question is what would prevent XP from booting
in this situation? Is booting Linux from CD the only way forward now?

 

[Rod Speed]

Rod Speed wrote

So back to the beginning: I tried this setup last night - known good
HD (master) and CD/DVD (slave) on Primary IDE and suspect drive
(master) on secondary IDE. BIOS recognized everything but Windows
filed to load. Got splash screen and then black screen with cursor
flashing in upper left hand corner. Question is what would prevent
XP from booting in this situation?

Most likely XP isnt able to communicate with the drive properly
but its not so bad that it can see that the drive is dead. Likely
you are seeing lots of timeouts on drive activity.

Is booting Linux from CD the only way forward now?

Yes, if you cant get XP to look at the drive, thats the only way to get the
data you care about off that drive while you can. That does work with some
flakey drives, essentially because linux isnt as fussy about the NTFS structures.