120GB PATA drive changes model number

[larrymoencurly]

What would cause an old Western Digital PATA IDE 120GB drive to change its model name from WD1200BB-22FTA to WD1210CB-23-23FUA1 !!!!!!!!!! (exclamation marks given as part of the name)?

Also the maximum 48-bit LBA for the drive went from 234,441,648 to 72,058,693,783,997,360, and Paragon backup reported the drive's capacity at around 600GB.

This happened with the drive connected to the motherboard's built-in PATA controller (Intel G-41 chipset) or a Silicon Image SiL680A PCI controller card.

The drive seems to be back to normal now, with the original model number and LBA.

 

[Franc Zabkar]

What would cause an old Western Digital PATA IDE 120GB drive to change its model name from WD1200BB-22FTA to WD1210CB-23-23FUA1 !!!!!!!!!! (exclamation marks given as part of the name)?

Also the maximum 48-bit LBA for the drive went from 234,441,648 to 72,058,693,783,997,360, and Paragon backup reported the drive's capacity at around 600GB.

This happened with the drive connected to the motherboard's built-in PATA controller (Intel G-41 chipset) or a Silicon Image SiL680A PCI controller card.

The drive seems to be back to normal now, with the original model number and LBA.

It looks like a stuck bit on the IDE interface. The IDE interface is
16 bits wide, and each character occupies 8 bits. Therefore a bad bit
would result in an error in alternate characters.

WD1200BB-22FTA
WD1210CB-23FUA
^ ^ ^ ^
12121212121212

As you can see, there is an error in the first byte of each pair. The
bit difference points to an error in the least significant bit.

C:\>debug
-e 100 "WD1200BB-22FTA"
-e 110 "WD1210CB-23FUA"
-d 100 11f
100 57 44 31 32 30 30 42 42-2D 32 32 46 54 41 WD1200BB-22FTA
110 57 44 31 32 31 30 43 42-2D 32 33 46 55 41 WD1210CB-23FUA
-q

Therefore I suspect that DD8 (pin #4) in the IDE interface has a
problem. I would reseat the IDE cable at each end, and I would examine
the pin for a dry solder joint. Also examine the associated resistor.

http://pinouts.ru/HD/IdeInternal_pinout.shtml

- Franc Zabkar

 

[Franc Zabkar]

As you can see, there is an error in the first byte of each pair. The
bit difference points to an error in the least significant bit.

BTW, the bit difference between an ASCII space and an exclamation mark
is also in the lowest bit, ie 0x20 and 0x21.

- Franc Zabkar

 

[Arno]

What would cause an old Western Digital PATA IDE 120GB drive to change its
model name from WD1200BB-22FTA to WD1210CB-23-23FUA1 !!!!!!!!!!
(exclamation marks given as part of the name)?

Also the maximum 48-bit LBA for the drive went from 234,441,648 to
72,058,693,783,997,360, and Paragon backup reported the drive's capacity
at around 600GB.

This happened with the drive connected to the motherboard's built-in PATA
controller (Intel G-41 chipset) or a Silicon Image SiL680A PCI controller
card.

The drive seems to be back to normal now, with the original model number
and LBA.

Flaky firmware or a flaky interface.

If you look closely, you find that every second character from a
point on has been increased bu 1 form even to odd, i.e. consistent
with a bit flipped from 0 to 1. The "!!!!" look like the terminating
zero being changed to a 1. Capacity was likely mangled in a similar
fashion.

You should discard this drive.
Arno

 

[larrymoencurly]

It looks like a stuck bit on the IDE interface. The IDE
interface is 16 bits wide, and each character occupies 8 bits.
Therefore a bad bit would result in an error in alternate
characters. WD1200BB-22FTA WD1210CB-23FUA ^ ^ ^ ^ 12121212121212
As you can see, there is an error in the first byte of each pair.
The bit difference points to an error in the least significant bit.
C:\>debug -e 100 "WD1200BB-22FTA" -e 110 "WD1210CB-23FUA" -d
100 11f 100 57 44 31 32 30 30 42 42-2D 32 32 46 54 41
WD1200BB-22FTA
110 57 44 31 32 31 30 43 42-2D 32 33 46 55 41
WD1210CB-23FUA -q
Therefore I suspect that DD8 (pin #4) in the IDE interface has a
problem. I would reseat the IDE cable at each end, and I would
examine the pin for a dry solder joint. Also examine the
associated resistor. http://pinouts.ru/HD/IdeInternal_pinout.shtml

Great explanation. I had been under the mistaken impression that
parity or even ECC was sent along with the data and commands.
Changing cables didn't help, so I feared the drive had failed (would
be my first drive failure), but I think the problem was due to using
the PCI drive controller for the hard disks and the motherboard's
PATA controller for the CD drive, even with programs that didn't
use Windows. Every PCI HD controller I've tried has revealed a quirk
under some conditions.

 

[Franc Zabkar]

Changing cables didn't help, so I feared the drive had failed (would
be my first drive failure), but I think the problem was due to using
the PCI drive controller for the hard disks and the motherboard's
PATA controller for the CD drive, even with programs that didn't
use Windows. Every PCI HD controller I've tried has revealed a quirk
under some conditions.

I once had a strange case where DOS sound card software (SBINIT.EXE ?)
somehow caused a particular bit to be dropped when accessing the HD.
The HD was OK until the software was loaded by the AUTOEXEC.BAT file.

This was the thread, but it appears to have disappeared from Google's
archives:

OPTi 82C931 sound card problems:
http://groups.google.com/group/comp.sys.ibm.pc.soundcard.tech/msg/a484246f62937e44?dmode=source

I never managed to properly resolve the problem.

- Franc Zabkar

 

[Joe Pfeiffer]

BTW, the bit difference between an ASCII space and an exclamation mark
is also in the lowest bit, ie 0x20 and 0x21.

Very nicely done. I don't know if I ever would have noticed that and
diagnosed his likely problem.

 

[Franc Zabkar]

I don't know if I ever would have noticed that and
diagnosed his likely problem.

I spent a large part of my career chasing bad bits in digital logic.
One of the first things I was taught was to compare the received and
expected data. I guess it's second nature to me now.

Actually, I have seen the OP's problem several times before, in
several forums, and I've experienced it once personally. In fact I
coincidentally encountered a similar question about 10 minutes before
posting here:

http://forum.hddguru.com/galaxy2d-displayng-corrupt-and-and-datas-t24213.html

- Franc Zabkar