Data Recovery Case File · Portable Drives · A Timed Failure
The Light That Stays On for Two Minutes
His enquiry included a duration, which is rarer and more useful than it sounds. A Toshiba Canvio 1TB external drive that "suddenly stopped working. My laptop doesn't recognise it any more. Shortly after plugging in, the light remains on for two or three minutes and then stops." Most people report that a drive does nothing. He has reported that it does something, for a measurable period, and then gives up — and that is a considerably more informative symptom. A timed failure describes a sequence being attempted rather than a device that is dead, and the length of the attempt narrows what is failing.
| Media | Toshiba Canvio 1TB portable external hard drive — powers and maintains activity for approximately two to three minutes on connection, then ceases; never presented to the host |
| Reported situation | Abrupt loss of recognition · consistent timed activity on connection followed by cessation · no volume presented at any point · no impact or liquid reported |
| Fault class | Readiness sequence attempted and abandoned — head calibration or reserved-region reads failing within the drive's own retry budget |
| Equipment used | Removed from enclosure and assessed on a native connection · Atola Insight Forensic readiness and current-signature diagnostics · PC-3000 Express technological-mode service-area assessment · Data Extractor imaging under strict timeouts |
The decode: what two minutes of trying means
Why the duration matters: a drive that is genuinely dead — no power, a failed board, a seized motor — does nothing at all, or produces a single brief attempt and stops. His drive works for minutes. That means power is reaching it, the board is functioning, the motor is turning, and the drive is engaged in something. What it is engaged in is the readiness sequence: unparking and calibrating its heads, and reading the reserved region holding its firmware, defect maps and translation tables. None of that is audible or visible except as the activity he can see.
Why it gives up when it does: drives have internal retry budgets. Faced with a read it cannot complete, the drive will retry — recalibrating, repositioning, trying alternate copies of the data it needs — for a bounded period, and then abandon the attempt rather than continuing indefinitely. Two to three minutes is characteristic of that budget being exhausted. So the drive is not failing instantly and it is not hanging forever; it is trying properly and losing.
What that points at: two candidates, both below the filesystem and both invisible to any software. A head calibration failure, where the heads cannot establish position against the servo patterns on the platters. Or a service-area problem, where the reserved region the drive must read before it can identify itself has become unreadable. The second is the more common cause of exactly this pattern, and it is addressed through the manufacturer's own technological modes rather than by anything a computer can do.
Why none of it touches his data: the reassuring part. Both faults concern the drive's ability to start, not the contents of the user area. The files are sitting where they were written, waiting for a drive that can address them. That is why this class of fault, while requiring bench equipment, generally recovers completely.
What to stop: reconnecting it. Each attempt is the same bounded sequence failing the same way, and there is nothing further to observe — he has already timed it, which is all a computer can tell him.
On the bench
The drive was taken out of its enclosure and assessed on a native connection, since the enclosure's own board had already proved functional by delivering power for the duration he observed. The Atola Insight Forensic read the readiness sequence electrically and by current signature — how far the drive progressed, what it retried, and the precise point at which it abandoned the attempt — converting his stopwatch observation into a located finding. Access followed through the PC-3000 Express in the drive's technological modes, with the service area read directly and its damaged modules repaired until the drive identified itself and would address its media. Imaging ran under Data Extractor with strict timeouts.
The outcome
The abandoned readiness sequence located, the service area repaired and the contents imaged and verified. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode, for anyone whose drive works briefly and stops: a timed failure is far more informative than a dead one, because it proves power, board and motor are all working and the drive is attempting its readiness sequence; drives retry against an internal budget and abandon the attempt when it is exhausted, which is what a consistent two or three minutes describes; the causes are head calibration or the reserved firmware region, both below anything software can reach; and neither touches what is written in your files. Report the duration — it is worth more than most descriptions.
Drive that works for a minute or two and then stops
Time it once and then stop reconnecting — that duration is genuinely useful information and you've already gathered all of it. A drive that shows activity for minutes isn't dead: power is reaching it, its board works and the motor is turning, and what it's doing in that time is trying to become ready — unparking and calibrating its heads, and reading the reserved region that holds its own firmware. Drives retry against an internal budget and then give up rather than trying forever, which is exactly what a consistent couple of minutes looks like. That points at head calibration or the firmware region, both of which sit below anything software can reach, and neither of which touches your files. So don't run recovery programs at it, don't try more cables, and do mention the timing when you ask for help.
That timing locates the fault — call Glasgow Data Recovery on 0141 404 0294; read on a native connection, readiness measured by current signature, service area repaired and imaged out.
Request a quote online →
Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.