Mains-powered externals · logged · HUL-2025-0903
Two Clicks, and the Heads Were Worn Through.
A family in Beverley spent most of a year blaming the USB port. The Toshiba held their two children's first ten years in photographs, and the only sign of anything wrong was that it kept getting slower and slower every month
. A fatal hardware error
from Windows then ruled the port out. A neighbour put it on his own PC and his laptop never saw it at all
. By then it would power up, click twice and shut itself down
.
Sounds like yours? Ring us.
0800 6890668
What causes it.
The adaptive and calibration tables sit in a reserved band on the platters. No drive can offer itself to a computer until it has read them, and heads worn this thin could not lift that data. What comes next is designed rather than accidental. The firmware allows itself two attempts, recalibrates between them, and then shuts down rather than run on numbers it does not trust. Those attempts are the clicks. The slow year beforehand was that same wear accumulating in silence, and the Windows message only said aloud what the disk had been telling them for months.
The tools this one needed.
The steps we follow →| The gear | What this one did | Why we have it |
|---|---|---|
| DeepSpar Disk Imager 4 | Measured every head on the new stack and set the order the surfaces came off | Grades each head first, reads one surface at a time, and keeps resets, timeouts and power under control |
| PC-3000 Data Extractor | Ran the read passes, answering each stall with either a controlled reset or a cut in power | Copying is run at firmware level, with the PC-3000 in charge |
| R-Studio Technician | Rebuilt the folder structure from the finished image | Handles almost any file system, and puts RAID sets back together |
The stages.
The noise was evidence, not an annoyance
Two clicks each time, never three, then a shutdown. That belongs to the power-on read and not to ordinary use, which points at the reserved tracks for the first failure. Magnification agreed. No chipping, nothing that looked like an impact, only even wear on every slider after a very large number of flying hours.
New heads before anything
A stack from the same production batch went in, and the disk got through its start-up read, stayed powered and gave its own identity. Mapping came next rather than being put off: wear that slow hardly ever spreads evenly across the surfaces, and the map is what shows where readable data is still sitting and where it is not.
Image it at once, the strongest heads first
A disk worn this far will not sit still for a second attempt, so imaging started straight after the repair, with the drive still on the bench. Full speed on the sound heads to begin with. The bands in doubt were held back for slow passes once everything else had been copied. With the reserved area readable once more, nearly all of it came off without a fight.
How it finished.
The image came out complete. Everything was pulled from that copy and not from the disk, and the folders were checked year by year before the case was closed. The Toshiba was scrapped: once a drive has ground its own heads away there is nothing left in it for anybody.
What people read after this.
Tagged Desktop externals too.
Is yours doing the same?
Turn it off. Send it to us. The diagnosis comes first. It says which files are still readable and which are not.