Desk drives · case notes · CDF-2025-0315
It Beeps, and the Capacity Changes Each Time.
A d2 sat under a desk in a Caerphilly studio with four terabytes of masters on it — choirs, brass bands — and one day it stopped spinning and took to beeping. Its size depended on when you asked: Disk Utility said 750GB and refused to mount it
, later the size column was simply empty
, and on a colleague's machine 0 bytes, needs initialising
. New cables and another dock changed nothing.
Does that match yours? Call us.
0800 6890668
Why that happens.
A hard disk does not know its own size until it reads it. Capacity, model string and defect tables all sit in service areas written on the platters themselves, and none of that can be collected while the platters are still. Pin them and the drive has nothing truthful to hand over, so every connection produces a fresh piece of fiction and not one of them will mount. The beeping is that same stalemate made audible: the motor heaves, cannot break the pack loose, rests, then heaves again. What holds it is nearly always the head stack, sitting down on a surface instead of parked at the edge.
What we used on this one.
Every step, in order →| The tools | What it did here | Why it helps |
|---|---|---|
| PC-3000 Data Extractor | Captured the whole image within that one stretch of steady rotation | Imaging driven from inside the drive's own firmware, by the PC-3000 |
| ACE Lab PC-3000 Express | Recovered the true model string and its defect tables with the heads clear | Opens the maker's service commands, which a plain ATA link will not pass |
| R-Studio Technician | Mended the HFS+ volume so each session returned to the folder the studio used | Reads nearly every file system, and rebuilds arrays you can rely on |
How it went.
The beep was telling the truth about the drive
He had already tried other cables, another dock and a second computer, and every one of them produced the same beep, which put the fault inside the aluminium housing and nowhere else. A current trace settled it: a couple of seconds drawing hard while the motor strained at something that would not shift, then the protection tripping, then the same cycle over.
Lift the heads clear, and do not stop after that
In filtered air the stack was lifted away and parked, and each slider examined under the scope; the platters below them were unmarked, which is never safe to take for granted beforehand. Powered up again, the drive announced itself correctly — the full 6TB, the proper model string, the firmware we had expected — and the imager, already configured and waiting, was set going.
One spin-up only; all the rest done on the image
The read ran from one end to the other with power kept on the whole way, because a stack that has landed once may not lift a second time. As soon as the image was secure the drive was set aside; repairing the volume and putting the folders straight happened entirely on the copy.
Signing it off.
All four terabytes went onto fresh media with filenames, dates and folders exactly as the studio had left them. He stopped at the first beep, and that one decision is the reason: nothing had touched the surfaces, and every mix played back as it sounded on the day it was cut.
Pages people often read next.
Filed beside Desktop externals.
Is that what yours is doing?
Power it down, get it to us, and hold off on any decision until the diagnosis says what can still be read.