Switch to full style
Data recovery and disk repair questions and discussions related to old-fashioned SATA, SAS, SCSI, IDE, MFM hard drives - any type of storage device that has moving parts
Post a reply

Barracuda 7200.7 160GB with big chunks of IDNF errors

February 22nd, 2007, 17:03

I have a Seagate ST3160021A 160GB drive with lots of errors.

I tried ddrescue but there were far too many bad sectors, so I brought in the big guns with MHDD. I get fairly regular long patches of IDNF errors(86*255 sector blocks = 21930 sectors each) with 60,000-100,000 good sectors (or so) in between each.

Does this sound like anything in particular to those of you with more experience? Lost a head maybe? I have two questions really:

1. Am I likely to be able to recover the data that's on the disk, even if it needs to go to changing the head stack (I can get access to a clean room if needed, and I'm happy working on surface mount electronics)?

2. Is reinitialising the disk likely to make it useful again or should I just chuck it?

Thanks in advance, it's been really interesting browsing the forums.

February 22nd, 2007, 21:19

Hello,

If the pattern is regular U probably have a defective head.
A clean room is a good idea for changing heads inside, but it is more important to have proper tooling and proper experience and of course head assembly.
BTW some accurate diagnostics should be performed through terminal port.

pepe

February 25th, 2007, 9:48

Thanks Pepe, that's what I thought. It looks as though about 1/4 of all sectors are IDNF, and I see this drive has four heads - which would fit!

I have looked at making the MAX232 cable - as far as I can tell all it does is buffer the PC serial port, or is there some level conversion going on? Is the Seagate diagnostic port running at +/-12V or TTL levels? I guess I just need to put a scope on it...

Alternatively, what is the best program for reading what can still be read from the disk with so many 'bad sectors' on it? I've tried using ATOF in MHDD but it seems to be slower than it could be (because I don't think it uses the 'verify' step first like it does when just scanning, so it retries a lot on each dead sector). Can I cut the retry time in MHDD, or is there something else that will allow me to tell it to timeout very quickly on each bad sector?

When booted to DOS to use MHDD, I need to mount an NTFS disk to have space to store the image files - I'm using NTFS4DOS 1.4 at the moment, is there a better option?

Sorry to ask so many questions...
Post a reply