Hello everyone,

I found an old hard drive that failed about 10 years ago and I'm trying a DIY recovery out of curiosity and to learn. I work in electronics but have very limited data-recovery experience.

What I did:

The drive was completely dead (no spin/noise). I replaced the PCB with an identical board I bought online. After the swap the drive spins and the heads make noise, so the original PCB looked faulty.

I attempted to read and copy the ROM from the original PCB. Reading the chip in-circuit failed, but after desoldering it I could read the ROM successfully.

I flashed that original ROM image onto the replacement PCB. Result: the drive does not spin (same symptom as the original failure).

If I reflash the firmware that originally came on the replacement PCB (I kept a backup), the drive spins again but makes noises and is not recognised by the host.

I also tried ROM images from identical drives found online; they produce the same "spins + noise, not recognised" behaviour.


My current thinking: sometimes the PCB appears to be the issue (replacement board makes it spin), but moving the original ROM to the replacement board yields no spin at all. This could indicate the ROM image is unreadable/corrupted, but in my experince, it's unlikely the data got corrupted if the chip still works.
Another thought I had is that there are mismatched low-level parameters between the ROM and the replacement PCB (even on "identical" boards), or a mechanical fault that manifests differently depending on microcode/servo parameters.


Any ideas what could be happening and what I should try next?
Thanks in advance, any pointers would be greatly appreciated.

Upload your patient and donor ROM dumps.

_________________
A backup a day keeps DR away.

Here they are

The first 0x1000 bytes of the patient are empty. That's the boot block. Has the drive been anywhere else?

_________________
A backup a day keeps DR away.

What do you mean anywhere else?
You mean a previous repair attempt? No, it has been sitting on a shelf since itdied

Well, that's a very strange failure mode.

I would use a hex editor to patch the first 0x1000 bytes from the donor ROM into the patient ROM and see how that goes.

HxD is free:

https://mh-nexus.de/en/hxd/

_________________
A backup a day keeps DR away.

I tried taking the first 0x1000 bytes from the donor ROM and combining it with the rest of the original ROM, I get the same spin+noise symptoms

I tried doing that, I sticked the first 0x1000 bytes of donor to the original ROM, and I get the same spon+noise symptoms

If the drive spins, then the ROM is most probably OK. However, there is a unique NVRAM section, so the ROM must be original. A donor ROM will spin up the drive but won't work.

If everything is original, then there is probably an internal fault.

_________________
A backup a day keeps DR away.