Seagate ST360020A 60gb with bad sectors

[WebClaw]

Aside from the sector-count confusion, is your question simply this:

Can the remaining 35 damaged files be recovered by either:

A. Sending the drive to another professional who may use different PC-3000 imaging, head-mapping, or reread strategies?

Or:

B. Installing another donor head assembly and matching donor PCB yourself?

If that is accurate, I think you would have difficulty finding a reputable lab willing to take on the case given its history and the limited amount of data still missing.

That said, almost anything is possible if you are willing to spend enough money. A further head replacement, particularly a DIY attempt, is more likely to cause additional data loss than improve the recovery.

A. Sending the drive to another professional who may use different PC-3000 imaging, head-mapping, or reread strategies


Can the professional use different PC3000 imaging and change the heads mapping to read these sectors and get the data back??

Think of it from the lab’s perspective.

A previous data-recovery company already replaced the heads and recovered approximately 99.9% of your data. You are now asking another lab, potentially under a no-data, no-charge policy, to invest significant bench time, donor parts, and PC-3000 resources chasing the final fraction of a percent.

What exactly would count as “data recovered” in those terms? One additional sector? One partially repaired file? All 35 files?

This is a high-risk, low-return case with no guarantee that another set of heads, different reread settings, or more time will recover anything further. The head-mapping comment also makes little sense here. Head mapping can control which heads are used during imaging, but it does not magically make unreadable sectors readable.

[Hardcore Games]

Given this vintage of the Seagate ST360020A, the power on hours suggests this disk was not used extensively since it was new.

Older ATA class disks tended to be less reliable as the platters used, while good, tended to suffer from excessive bitrot. Bitrot tends to occur when the disk is powered down for extended periods.

So the disk shows a lot of reallocated sectors (512 byte) so it is possible files will have 512 byte blocks of damage if CHKDSK cannot read the bad block(s).

[ora]

Could a company using a PC-3000 Portable Pro with head mapping and different imaging strategies adjust the head configuration to obtain better reads from those sectors?

No. The previous data recovery lab guaranteed to use the PС-3000 when cloning the hard drive. If they weren't successful, why would another lab with the same equipment get a different result?

If they were talking about 50 or so sectors, those 50 sectors represent only 25 kilobytes of data out of the 60 gigabytes of disk capacity, which is an excellent result for a drive that had some reading issues.

The lab could, of course, manually read the remaining 50 sectors, albeit with error correction disabled. This could yield anything from a few incorrect letters in a text file to no data at all, but it certainly doesn't guarantee a correct recovery.

P.S. When the hard drive firmware detects an unreadable sector, it tries its best to read the data, including adjusting various internal parameters - nothing better than that can exist. You can only give the drive more time or try reading the unreadable sectors several times, intermittently, but if the drive can't read data from several sectors with its original heads, it usually won't be able to read them at all.

I have got the logs so do you want me to paste it on here?

Since you made a copy in HDDSuperClone, you should also have the project file saved. Take a screenshot of the number of copied and uncopied sectors (as in the photo below) to make it clearer what we're dealing with.

My question is whether another data recovery company might have a better chance of recovering data from those remaining bad sectors.

Since I only need about 35 files (ZIP, EXE, TXT, and URL files), I'm wondering whether it's worth getting a second opinion.

Theoretically, yes. The problem is that these remaining files can be corrupted for four reasons: currently unreadable sectors, damaged (and relocated) sectors in the past - part of the data within the file was lost in the process, lost information in the file system about the location of fragments of these files, and corruption/overwriting of parts of these files by the CHKDSC.

My point is that recovering these remaining files can be a hundred times more difficult than cloning the hard drive.

[fzabkar]

There is one thing you try, but it will most probably be fruitless. The ATA SCT READ LONG command reads a sector, ignoring ECC.

https://www.hddsuperclone.com/hddsupertool/manual

Code:

   ata_sct_readlong

     Perform an SCT read long command if supported.

     This command requires the number variable "sector" to be set on the command line.

     This also requires the string variable "file" to be set.

     "file" is the name of the file the data will be written to.

     Note that even if this command is supported, it may still be of no value.

     The ATA documentation states that the data returned may be encoded.

     And my experience so far is that the data is usually encoded and totally useless.

     So don't get your hopes up about using this command to get any worthwhile data.

     It is mainly a demonstration of using the SCT commands.

     Example: hddsupertool -t /dev/sdb -f ata48_sct_readlong sector=100 file=="sector100.bin

This blog tells you what you can expect:

https://www.deepspar.com/blog/read-ignoring-ecc

[ora]

it will most probably be fruitless

Unfortunately, the thread's author can't do anything on his own because his hard drive no longer has heads.

I can’t clone these 35 data I want because the DR firm have modified on the firmware when they installed the new heads before they removed it.

Although it's very interesting, how can a data recovery lab return a drive without heads installed? There's a hole where the contacts should be.

The ATA SCT READ LONG command

As far as I know, these classic Seagate hard drives, and apparently the F3 as well, can be used to directly read the surface via diagnostic terminal commands using a kind of physical CHS, ignoring both errors and translation altogether, for cases where a long read returns zeros.