I have a strange 7200.11 drive. Wonder if anybody came across something similar before. Don't know if this is cache problem, didn't try to swap PCB yet..
Firstly it looked as a oridinary logical failure, two partitions unreadable, but drive seems to be physically healthy. Smart - no bad sectors
Drive started to behave unnaturally when I launched the raw partition scan for files. On higher speeds it was like repowering, and throughout all reading the heads are like repositioning very fastly, something like loosing the path but only heard as a very short click (like reading, but not linear), anyway, i tried to clone the drive with DC
UDMA off, block size 2, or 4 (higher simply clugged the drive). extensive nor intensive reading didn't get good results (when looked some sector from destination drive, and from source drive, they were not quite exact, many sectors, randomly, where like mispositioned (WTF?)
in hex editor, i figured out, that the mechanism is somehow corrupted - some sectors, or part of the sector, are filled with sector read before/later, didnt exactly figured out no algorithm but seems to be random - i cannot in no way realign nor recover misread parts - they are all over the drive.. files are naturally hard to recover..
Now since 2 days i thaought i found the way to clone the drive - i left it to copy with "look-ahead" turned off, and was sure that it goes well - suprise was when i saw, that now there is no misalignment, but, part of every third or fourth sector, is somehow changed
Drive seems to change data after reading sector from source drive, and before sending it to the computer/screen/destination drive while cloning..
uff.. hope anybody understands my problem, and any suggestions welcome..
maybe i will find some donor PCB, and try.. but PCB seems to be OK, drive spins perfectly

Peter

Hi,

It's not so easy to understand your problem just for that explanation.
It could be that your drive has shifts, but I really don't think that's the issue with it.

What is the file system?

It's not "shift" to me.

Are you sure someone else didn't play with PCB AND/OR firmware before sending to you ? It is important.

Solved. Didn't have time to write here, now let me share some thoughts.
I knew for sure, that nobody messed up with PCB, nor firmware. I spend few days trying to copy, and making simple read tests. When I was sure that in every possible mode and speed, every possible block size, read ahead mode, reverse or forward - there always was difference between two separate reading results of same group of sectors. Best I managed to achieve, was setting small read block size, and turning off look-ahead.. But it turned out that even that was not correct - few sectors were OK, but every few sectors there was some little change in last part of sector. It was not possible to correct manually nor any script. Well, I was pretty sure that It has to be somehow related with PCB or maybe translator.. Drive was repowering when reading in higher block size mode, even in PIO mode.
Maybe important fact is that while trying to recalculate the translator, the drive repowered twice, before finishing with standard report (no errors)
That was strange, but seemingly legit
It took some time to find a donor drive, but i managed to get a good PCB, and It worked.
Didn't have possibility to check the damaged PCB with other HDD, couse the donor hdd was damaged, only PCB was OK. I wonder , how would the suspicious PCB behave with another healthy drive after swap. Maybe It was some combination of unfortunate firmware AND pcb fault. Anyway, case closed, data recovered 100%

Probably a dodgy SATA connection or maybe a suspect cache ram chip

Well done for solving it

_________________
PC Image Data Recovery
http://www.pcimage.co.uk

New!! HDD-PCB.COM for all your PCB and donor HDD requirements!

Happy to have that done, especially since I'm wearing Cervical Collar right now!
The case looks now like an easy job, but I'm sure I'm not the only one who wouldn't suspect the PCB in the first shot
All the best 4 all