Data Recovery Case File · Desktop Externals & Aging Drives · The Adapter That Wasn't Its Own
Three warnings in one enquiry: the substituted adapter decoded as a hazard, "not the usual noise" taken seriously, and the repeated host crashes read for what they were — a failing drive stalling the bus, six terabytes at a time
His enquiry is a sequence of reasonable decisions that each made the next one harder — which is why it earns a page rather than a shrug. The 6TB G-Drive "wouldn't power up, so I changed power adaptors and it started up — but it isn't making the usual noise it makes and won't connect to my desktop Mac. I have managed to get my laptop to 'see' it and started trying to create a backup using recovery software, but it restarted my laptop several times. I need to try and fully recover the 6TB of data." Three separate warnings are stacked in those lines, and none of them is obvious unless someone tells you: a substituted power adapter is a real hazard, not a neutral workaround; "not making the usual noise" is the single most diagnostically valuable thing he wrote; and a host that restarts repeatedly during an imaging attempt is not a software problem — it is a failing drive dragging the computer down, and every crash is another unmanaged power cut delivered to a mechanism that can't afford one.
| Media | 6TB G-Drive external hard drive — initially dead to power; started on a substituted power adapter; abnormal running sound; won't mount on the desktop Mac; visible to a laptop |
| Reported situation | Power adapter swapped to obtain start-up · running noise changed from normal · consumer recovery software attempted an image and crashed the host repeatedly · full 6TB recovery required |
| Fault class | Mechanically distressed drive under questionable supply — instability stalling the host bus; compounded by repeated uncontrolled power interruptions |
| Equipment used | Correct-specification power on a controlled bench supply · Atola Insight Forensic current-signature and head diagnostics · DeepSpar USB Stabilizer 10Gb / native path with hardware-managed instability handling · PC-3000 Express with Data Extractor prioritised head-mapped imaging · volume rebuilt from the image and verified |
The decode: the adapter, the noise, and the crashes
Why a borrowed adapter is a hazard, not a fix: desktop-class external drives need a specific supply — a defined voltage, a defined polarity, and enough current to spin a large mechanism up under load. A substituted adapter that fits the socket may differ in all three. Too little current and the drive brown-outs: it starts, struggles, and behaves erratically under demand — which fits his symptoms precisely. Wrong voltage or polarity, and the damage can be immediate and permanent to the drive's own electronics. His swap achieved the appearance of success — it powered up — but the appearance is the trap: a drive that starts on the wrong supply is not being helped; it is being run outside its design. None of this is a scolding, because the hazard is genuinely obscure: no label warns you, and the plug fits. It is simply the first thing to undo.
"Not making the usual noise" — the sentence that matters most: owners know their own hardware's sound better than any instrument knows it, and a change in that sound is real evidence. A drive that used to spin up briskly and now runs differently — quieter, rougher, hesitant, or cycling — is reporting one of two things: it isn't getting the power it needs (the adapter), or the mechanism itself has changed (heads, bearing, spin-up under strain). Both are reasons to stop, and both make the next decode inevitable.
Why the laptop kept restarting: the crashes are the case's clearest signal, and the most misread. When a drive stalls mid-read — retrying a bad region, resetting its interface, dropping and re-announcing itself — the host's storage subsystem waits, then wedges, then falls over; on an under-supplied external, the drive may also be sagging the bus as it strains. So the repeated restarts weren't the software's incompetence and weren't his laptop's fault: they were a failing drive taking the computer down with it. What makes them serious is the arithmetic that follows. Each crash is an abrupt loss of power to a mechanism mid-operation — heads unparked, platters spinning down uncontrolled — and every one of those is a roll of the dice on the very surfaces holding six terabytes. Consumer imaging tools have no defence here: they cannot enforce timeouts at the hardware level, cannot absorb resets, cannot prioritise. That is precisely what a hardware imager exists to do — which is why the correct next step was to stop the attempts entirely, put the drive on a known-correct supply, and let it be read once, under control.
On the bench
The borrowed adapter came off first: the drive placed on a controlled bench supply at its correct specification, so that every subsequent observation described the mechanism rather than the power. The Atola Insight Forensic then read what the changed noise had been reporting — current signature through spin-up, head behaviour, the distress mapped as measurement rather than impression. Imaging ran with instability handled in hardware rather than hoped away: the drive's stalls and resets absorbed inline, timeouts enforced, and the PC-3000 Express under Data Extractor sweeping head-mapped and prioritised — healthy expanse first, weak regions budgeted last — until the six terabytes stood banked without a single further host crash. The volume was rebuilt from the image, verified by opening, and delivered on fresh media.
The outcome
The full 6TB imaged under correct power and hardware-managed control, rebuilt on the copy, verified and delivered. Free assessment, one fixed written figure including VAT, no recovery, no fee. Three warnings, decoded for everyone who has ever reached for a spare adapter: a substituted supply is a hazard — voltage, polarity and current all matter, and a drive that starts on the wrong one is running outside its design; a change in a drive's familiar sound is real diagnostic evidence and a reason to stop, not to persist; and a host that restarts during recovery software isn't a software problem — it's a failing drive stalling the bus, with every crash delivering another uncontrolled power cut to the surfaces you're trying to save. Correct power, one managed pass, no more crashes.
Drive revived with a different power adapter — and now behaving strangely
Undo the adapter first: desktop externals need a specific voltage, polarity and current, and a substituted supply that merely fits can brown-out the drive (erratic behaviour under load) or damage its electronics outright. Take the changed sound seriously — you know your drive's normal noise, and a difference means either insufficient power or a mechanism that has altered; either way, stop using it. And read repeated host crashes correctly: when a computer restarts during recovery software, the drive is stalling the bus, and each crash cuts power to a spinning mechanism mid-operation — some of the most damaging events a failing drive can suffer. Consumer tools can't enforce timeouts or absorb resets, so don't retry with a different program. Get the drive onto correct power and let it be imaged once, under hardware control, with the healthy regions captured first.
Stop the retries — call Leeds Data Recovery on 0113 322 3083; correct supply, current-signature diagnostics, one hardware-managed prioritised image, verified out — one written figure, no recovery, no fee.
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.