Data Recovery Case File · Portable Drives · The Diagnostic Trap

Seen until asked: the Passport that vanished the moment its own maker's software tried to test it

The pattern the customer described is one of the most precise fault signatures a home user can report, though it rarely feels that way from the kitchen table. His 500GB My Passport — believed to hold old family photos — would start up, light on, audible; his Mac, a Windows machine and a Linux box all failed to detect it; after a couple of minutes it produced one slightly louder click; and WD's own Drive Utilities could see it sitting there — right up until he selected any check or diagnostic, whereupon it disappeared and the app asked him to plug in a WD drive.

The enquiry

“When plugged in, the drive does seem to start — it makes noise and the light turns on — but it isn't detected. I've tried macOS, Windows and Linux. After a couple of minutes or less, it makes one slightly louder clicking sound. WD Drive Utilities actually detects it initially, but as soon as I select any option — checking for errors, running a diagnostic — it just disappears and asks me to plug in a WD drive. Is this still recoverable?”

Why the diagnostic was the thing making it vanish

Assemble his three observations and the drive has told you everything. It could stand: power, spin, and enough electronic presence for a utility to see a device existed. It could not work: presenting an actual volume requires the heads to read, and three operating systems' silence said they couldn't. And the vanishing act completed the picture — because a "check for errors" is not a gentle enquiry after the drive's health; it is a demand for sustained reading, the one thing a drive with marginal heads cannot supply. Asked to perform, it collapsed off the bus; the single louder click after a couple of minutes was the sound of it abandoning its own retries. The trap in this pattern is that the manufacturer's own software feels like the responsible next step — and on a healthy drive it is. On a failing one, every diagnostic run is an unsupervised stress test on the exact component that's dying, and each vanishing act it triggers is flight time nobody logged. He ran it once, noticed, and stopped: the reason this page ends well.

The recovery

The lab equivalent of his experiment, run safely, confirmed it: heads too weak to sustain their opening reads, platters — spared further diagnostics — clean. A donor head-stack from the matching family went in under laminar airflow; the drive's firmware and self-maintenance were secured and stood down over the PC-3000 through the USB Stabilizer; and Data Extractor imaged the rebuilt drive with the patience its predecessor was never shown — strong surfaces banked early, the frailer territory negotiated across later passes, the family's hardware encryption resolved with the drive's own keys as the image closed. The "believed" photos proved gloriously real: years of family albums, extracted with structure intact, verified, delivered.

Outcome

Effectively complete recovery of a collection its owner only half-remembered was there — the quiet argument for rescuing every "I think there are old photos on it" drive before deciding they don't matter. And the portable lesson: when a drive is already misbehaving, observation beats examination. Note what it does, note what makes it worse, and bring the notes — his were as useful as any diagnostic log, and they cost the drive nothing.

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.