Desktop externals · bench record · MHD-2025-6315
Beeping, and a Capacity That Changes.
Four terabytes of choir and brass-band masters sat inside a d2 that had stopped turning and started beeping under a desk in a Bracknell studio. Asked what it was, it answered differently each time: Disk Utility showed 750GB and refused to mount it
, then a line with nothing in the size column
, and on a colleague's PC uninitialised, zero bytes
. New leads and a different dock changed nothing.
Much the same fault? Call us.
0800 6890668
In everyday terms.
A drive does not know its own capacity from memory. It reads that, with its model string and defect tables, off service areas written on the platters, reachable only once they turn. A pack held still has nothing to declare, so what reaches the host is a fresh guess each time, none good enough to mount. The beeping is that standstill heard from the other side: a spindle motor loading, failing to free the platters, resting, trying again. What holds them is almost always the head stack, resting on a platter rather than parked at the disk's edge.
What the bench used here.
How a job is handled →| Tool | Why we used it | What it brings |
|---|---|---|
| PC-3000 Data Extractor | Captured the whole image within that one window of steady rotation | The PC-3000's imaging module, working under firmware-level control |
| ACE Lab PC-3000 Express | Read back the true model identity and defect lists once the heads were clear | Reaches the vendor-specific commands that ordinary ATA has no way of issuing |
| R-Studio Technician | Rebuilt the HFS+ volume so each session sat where the studio had filed it | Its file-system coverage is wide and its array rebuilds hold up |
The lab work.
Believe the beep — the fault lay inside the case
He had already changed the cable, the dock and the computer, and every substitution produced the same beep, which eliminated everything outside the aluminium shell. The current trace settled the rest: a heavy draw for two or three seconds while the motor strained against something immovable, then its own protection cutting in, then the same cycle again.
Take the heads off the surfaces, then do not stop
In filtered air the stack was walked back onto its ramp and the sliders examined; the surfaces below them proved unmarked, which is never something to take for granted. Powered again, the drive reported itself correctly — full 6TB capacity, right model, expected firmware — and the imager, already configured, was set going.
The drive gets one sitting; the rest happens on the image
The read ran end to end without the supply being cut once, because a stack that has landed already may well not lift a second time. With the image secured the drive played no further part; repairing the volume and sorting folders happened on the copy.
How the job closed.
All four terabytes reached fresh media with names, timestamps and folder structure exactly as recorded. Credit for that belongs to the decision to stop at the first beep: the surfaces were undamaged, and the mixes played back as they had on the day.
More of the jobs we've closed.
Filed beside Desktop externals.
Does any of that match your case?
Nothing needs deciding until the diagnosis is back: switch the device off, send it in, and let the findings settle it.