External drives · bench notes · BHD-2025-9124
I Ran DMDE All Night, Then Came the Grinding.
He meant the scan to be the rescue. Twenty years of a mobile DJ's working life sat on that Expansion — every set I have ever played, and everything I recorded myself
— but Disk Management called the whole volume unallocated
and offered to initialise it for him. He said no, found a forum thread, and started DMDE last thing before bed
. By morning the log was error after error, thousands of them
, and there was a sound the drive had never made before: a grinding you could hear out in the hallway
. His email finished I should have killed the power long before I did
.
Sounds like yours? Give us a ring.
0800 6890668
What that means.
The trouble with that word is the action it invites. When a disk can no longer find its own partition table, the usual reason is that it has stopped reading dependably, and tired heads sit near the top of that list. A full-surface scan then sends those same heads over every sector on the disk, one after another, for hours with no rest. Not many failing heads come through that errand. At some point before daylight the weakest of the four lost its cushion of air and came down on the platter it is supposed to fly over. That contact, heard from another room, is the grinding. Every minute after that removed more coating and returned nothing.
What did the work here.
The order of the work →| Equipment | The job it did | What it adds |
|---|---|---|
| DeepSpar Disk Imager 4 | Surveyed every surface and fenced off what the night's scan had wrecked | Head mapping first, then one surface at a time, with resets, timeouts and power all governed |
| PC-3000 Data Extractor | Recovered in long unbroken passes, taking the sound areas first | Reads the disk from inside its firmware, using the PC-3000's imager |
| UFS Explorer Professional Recovery | Reassembled the file system so every track landed back in its own folder | Handles the file systems other tools give up on: APFS, ReFS, XFS, ZFS, Btrfs |
What we did.
Work out the damage before promising anything
The head stack was beyond use when we opened it, and the platters showed banding — faint over much of the disk, far deeper anywhere the scan had lingered. What saved the job was the hour he switched it off. Give that drive one more night instead of a plug pulled at breakfast and there would be no case to write up.
Heads first; no figures until the survey is done
A donor stack matched to the model gave the drive its reading back, and the survey that followed marked out honestly where the disk still worked. Big stretches lifted quickly and without argument. The scored bands would not; over those the signal grew thin and then stopped altogether. He was quoted nothing until that picture existed.
Secure the good ground first, then edge into the damage
The read was arranged as long sweeps with the actuator kept as still as the job allowed, because every seek over a scored platter can drag loose material into ground that is still sound. Clean areas went across whole and onto safe media first. Only then did the banded regions get their turn, in short passes at reduced speed, taking whatever came.
How it closed.
Most of the collection went back with the folder structure he had built. Everything absent fell within the banded areas, and the report listed it line by line: a stack of commercial releases he can purchase over again, and four sets of his own recording that are simply gone. We wrote it up as it stood, without shading it our way.
Other pages worth a look here.
Others under Desktop externals.
Is yours behaving the same way?
Switch the device off, post it in, and decide nothing until the diagnosis comes back to tell you where you stand.