Devices · disks with platters in them
Four parts of a spinning drive break: the heads, the spindle motor, the circuit board and the firmware. None of them recovers while the machine keeps retrying, and the retrying is what scores a surface that was still imageable an hour earlier. So take the power off, and leave it off, wherever in Nottingham it is sitting. Working out which of the four went is bench work, it costs you nothing, and one drive is £300 + VAT — a figure fixed in writing before anything chargeable happens.
Ask an engineer, and the first look is free
0115 8220606
| What is happening | What usually causes it | What that means for you |
|---|---|---|
| Drive not accessible, and Windows offers to format it | The file table will not read, so the volume shows as RAW — although what was written to it has not moved | Take the power off, and refuse every format offer |
| Copying stops: Data error (cyclic redundancy check) | A sector has failed its checksum, and usually the surface under it is going | Abandon the copy and image the whole disk |
| 0x8007045D — an I/O device error, and the request fails | Commands go unanswered before the timeout runs out | Power it down and write the full model number down |
| A pop-up: Windows detected a hard disk problem | One SMART attribute has crossed its threshold | Take an image of the whole disk; no picking files |
| 0x800701E3 — the device reports a hardware error | Part-way through a copy the disk falls off the bus | Retry nothing. It is bench work now |
| At boot: SMART Status BAD, Backup and Replace | The drive has already failed its own self-test | Never boot the machine off it again |
| A reported 0GB, or a model name that is plainly gibberish | With the service area unreadable, the drive describes itself wrongly | Firmware work, and only on the bench |
Your device picks up a case number the day it arrives, and an engineer then establishes what has genuinely gone wrong — at no cost, and ahead of everything else. Two things come back to you together: a plain account of what can be lifted and what cannot, and one fixed figure in writing. Agree to that figure, or turn it down and pay nothing.
A copy comes before any repair. Imaging hardware reads your disk at whatever pace the mechanism will stand, putting anything difficult to one side for a last pass instead of grinding at it. The job proper starts once that image is safe.
Heads come from a donor matched to your own drive. A seized spindle is freed. Where the board itself is the fault, the ROM chip off it moves onto the replacement.
Your disk is out of it by then; the rest is done to the image. Inside that copy the file system goes back together, and a list of what came back reaches you well before any invoice does.
The file list reaches you before any invoice does. Say yes and it is billed; say no and it is not — and on most jobs, if nothing comes back there is nothing to pay. Whatever comes off goes home on media bought in for your case, carriage at our end. Nothing here is closed until you have opened those files on your own machine.
How often do drives actually fail? The only substantial published set of figures is Backblaze's Drive Stats: an annualised 1.36% over some 344,000 disks through 2025, where the previous year gave 1.55%. Those sit in racks. Yours sits under a desk in Gainsborough or Melton Mowbray, gets knocked, gets unplugged mid-write, travels about, and nobody has ever read its SMART data. Treat 1.36% as the floor.
Weeks had gone by with the drive sitting idle before it arrived, two of its heads already gone. Matched donor heads went in, and the first imaging pass alone returned 96%. Ledgers, invoices and a couple of decades of family pictures went home on new media once the owner had seen them listed.
Usually not. Clicking is a head fault, and your files are not on the heads — they sit on the platters underneath, about which the noise says nothing at all. What it does say is that the heads can no longer follow the servo pattern they steer by. Donor heads go in, the surfaces are read slowly, and a clicking drive normally gives up the greater part of what is on it. That share falls away with every further power-up.
Usually, yes — say who opened it and what they did. Dust travels in ordinary room air, and once it has settled on a platter the job can move beyond the terms that make most work no-fee. No price is set until an engineer has had the surfaces under a light.
One drive: £300 + VAT. You pay nothing for the look that comes first, you get the figure in writing, and nothing chargeable starts before you say yes. A purely logical fault sits at the bottom of that band, while donor parts and hours in filtered air lift it. On most work, no data back means nothing to pay.
IDE, early SATA and SAS all turn up here most weeks. What age alters is the adapter and the imager we put behind it — never the job itself.
The look costs nothing. What comes back to you is a file list — what opened, what did not — plus one figure to finish the job, in writing, before anything chargeable begins. Most jobs carry no fee at all unless the data comes back. Leave the drive switched off until you have that list.