No noise, no slowdown, no warning. It worked on the Friday and on the Monday it did not, with a whole body of work on it and no copy anywhere else. Sudden failure with no preamble is, oddly, one of the better things that can happen to a drive. A disk that degrades slowly has usually been shedding sectors for months and the damage is spread through the data. A disk that stops dead has generally had one specific thing fail — firmware, the service area, the board, occasionally the mechanism — while everything written on the platters sits exactly as it was left. The task is not to repair the data. It is to get past whatever is stopping the drive from presenting itself, then read it once, carefully, and never write to it.
| Media | Hard drive that stopped without warning, holding a complete body of work with no other copy in existence. |
| Reported situation | Hard drive · failed out of the blue · no prior symptoms · a whole body of work on it · needed back so it could be backed up properly · recovery sought. |
| Fault class | Sudden loss of the drive's ability to present itself — firmware or service area, board, or mechanism. The platter data is normally intact. Reached by reading the drive at its own level on equipment that talks to it below the interface, then imaging read-only in controlled passes where it reads weakly. |
| Equipment used | Read through a hardware write-blocker on PC-3000 UDMA, with firmware, service area, board and mechanism each assessed to localise the fault · fault addressed accordingly: service-area repair, board repair carried out with this drive's own ROM, or head or motor work under laminar flow in filtered air where mechanical · imaged read-only on DeepSpar with weak regions revisited in controlled passes · filesystem rebuilt and validated in R-Studio. |
A hard drive has to do a fair amount before a computer sees it at all. It has to spin up, calibrate its heads, read its own firmware out of a reserved area of the platters — the service area, which the operating system never touches — and only then announce itself. A failure anywhere in that sequence produces the same symptom: nothing. Which is why not being detected tells you very little on its own and needs equipment that can interrogate the drive underneath the interface.
The service area is the most common culprit in a clean, sudden failure. It holds translation tables, defect lists and calibration data, and it lives on the platters like everything else, which means it wears like everything else. When a module in it becomes unreadable the drive stops at that step and never gets as far as presenting a capacity. Repair that module and the drive comes up as though nothing had happened.
The other possibilities are ordinary. Board faults, where the electronics have failed and the mechanism is fine. Head faults, which usually announce themselves with a noise. Motor faults, which announce themselves with silence and warmth. The assessment is there to separate them before anybody spends money, because the four have wildly different prices and there is no honest way to guess between them down a phone line.
The one rule that applies to all four: whatever it turns out to be, the drive gets imaged read-only and the work happens on the copy. Nothing is repaired in place, nothing is written back, and the original is left in whatever state it arrived in. That is not caution for its own sake — a drive that failed once will fail again, usually part-way through the recovery, and the image is what makes that survivable.
It went straight onto PC-3000 UDMA through a hardware write-blocker, where its power behaviour, the firmware, the service area, the board and the mechanism could each be examined separately rather than guessed at. The fault was localised and dealt with according to what it was — service-area repair where the firmware modules had gone, board repair carried out with this drive's own ROM where the electronics were at fault, head or motor work under laminar flow in filtered air had the mechanism been the problem. Once the drive would present itself, it was imaged read-only on DeepSpar, with any weak regions left for controlled later passes rather than hammered at on the first run. The filesystem was reassembled in R-Studio off the image and the working files opened and checked.
Drive read at its own level, the fault addressed, and the body of work recovered onto fresh media — this time with the explicit instruction to keep a second copy somewhere else. The assessment cost nothing, finished within 2 working days of arrival, and produced one fixed written figure, VAT included. Which band it fell into followed from what had actually failed: logical and firmware-level work carries no fix, no fee, while anything requiring the casing to be opened takes half the quote before that stage with the balance owed only on success. The customer knew which applied before agreeing to anything.
Leave it switched off. The temptation is to keep trying it on different machines, and it is worth understanding what that costs: if the fault is mechanical, every attempt is doing physical damage, and you cannot tell whether it is mechanical from the outside if the drive is silent. Do not run repair utilities, do not let Windows initialise it, and do not accept any offer to format. If this drive was the only copy of work you depend on, take the lesson while it is fresh — one copy is not a backup at all, it is a single thing that can fail while you are relying on it. Send it tracked and insured, say plainly what happened and when, and you get the fault named and a price before a penny changes hands.
Each file below comes from a genuine enquiry taken from households and firms across Aberdeen, Aberdeenshire and the wider north-east, anonymised so nobody can be identified. Each one sets out how the fault was reasoned through, what was done about it, and which equipment did the work.
Appearing in Device Manager only proves something is plugged in — the clicking is the mechanism failing, and every retry costs data
A computer that hangs the moment the drive is plugged in is waiting for answers the drive cannot give — and waiting is doing damage
Power cycling with warmth and no spin is a board fault or a seized motor — and twenty years in a drawer does not fade the recording on the platters
Assessment is free and takes 2 working days at most once it lands, the figure is fixed in writing, and logical work carries no fix no fee. Begin online, or lift the phone.