Devices · spinning hard disks
A spinning drive has four things that break: heads, spindle motor, circuit board, firmware. Not one of them improves while a computer sits there retrying, and retrying is what turns a disk that could still be read into a disk with scoring on it. Take the power away and keep it away, wherever in Plymouth it sits. Which of the four went is a question for a bench.
You will be talking to an engineer
0800 6890668
| What shows up | What that points to | What to do |
|---|---|---|
| The drive is not accessible. Windows offers to format it before it will open. | The file table cannot be read, so the volume shows as RAW. Everything written to it remains | Take the power off. Refuse every format offer |
| Data error (cyclic redundancy check) | A sector failed its checksum on read; the surface is normally on the way out | Abandon the copy. Image the disk entire |
| 0x8007045D — the request failed, I/O device error | The disk ceases to answer commands inside the timeout window | Power it off and write down the full model number |
| Windows detected a hard disk problem | At least one SMART attribute has passed its threshold | Image everything. Do not go picking files off it |
| 0x800701E3 — a fatal hardware error on the device | Halfway through a copy the disk leaves the bus | Retry nothing further. This is bench work |
| SMART Status BAD, Backup and Replace | The disk's own self-test has already written it off | Never boot that machine from it again |
| 0GB capacity reported, or a model name that is gibberish | An unreadable service area leaves the drive misreporting itself | Firmware work, done on the bench |
Whatever arrives is booked in under a case number on the day it reaches the bench. An engineer then works out the actual fault, and that part is free. You get a plain answer on what can come off the drive and what cannot, followed by one price in writing. Nothing further happens until you have read it and agreed.
No repair happens before a copy exists. Your disk goes onto a hardware imager that reads at whatever pace the mechanism will tolerate, and it sets the difficult regions aside for a final pass instead of labouring over them. Work begins when that image is secure and not before.
Donor heads come from a drive matched to yours. A locked spindle is released. A failed board surrenders its ROM chip to the one taking its place.
Every bit of that runs against the image and never against your disk. The file system is reassembled inside it, and you see the list of what came back before an invoice exists.
Thinking it over costs you nothing. Everything the drive gave up is listed for you before any invoice exists, and the bill only follows your go-ahead. Files travel back on media bought in for your job, with return carriage paid at this end, and the case stays on the bench until you confirm they open on your own computer.
How likely is a drive to fail this year? Backblaze's Drive Stats is the one substantial published set of numbers, and it gives 1.36% annualised across the 344,000-odd disks the firm ran during 2025, against 1.55% the previous year. The drive beside a desk in Plymouth is nothing like that fleet: it gets bumped, pulled out mid-write and carried around, and its SMART data goes unread. Treat 1.36% as the lowest it could be.
Weeks had gone by since the disk last turned, and two of its heads had gone with it. A donor stack matched to it went in. The opening imaging pass lifted 96%, and the ledgers, the invoices and a good many family photographs were checked over and posted back.
Generally no. Clicking tells you the heads have stopped locating the servo pattern that guides them. It says nothing whatever about the platters below, and the platters hold your files. Fit donor heads, read the surfaces slowly, and most clicking drives hand over the great majority of what they carry. Each further power-up lowers that proportion.
Usually. Tell us precisely what was done, and by whom. The air in an ordinary room carries dust, and dust that has landed on a platter can take a job beyond no-recovery-no-fee. Nothing gets a price until an engineer has examined the surfaces.
Diagnosis is free of charge. What follows is one written price, and no billable work starts until you have agreed to it. Purely logical faults sit at the low end; donor parts and hours in filtered air sit at the high end. For a single drive, reckon on £300 + VAT.
Yes. IDE, early SATA and SAS turn up on the bench most weeks. All that age changes is which adapter and which imager get used, and nothing else.
The diagnosis is free, and it comes back as a list: which files read, which do not, and what getting them off would take. Until then, leave the drive unplugged.