The fault · ticks, beeps and grinding
Your files are usually still intact at the moment the noise starts, whichever part of Cardiff the drive has come in from. What takes them is the next half hour. A head that cannot find its place retreats to the ramp, sets off across the surface again, misses again, and goes round that circuit hundreds of times while you watch. Every pass is another chance to touch the platter. Cut the power and the count stops.
Describe the symptoms to an engineer
0800 6890668
| Warning signs | What has gone wrong | Do this now |
|---|---|---|
| A steady tick, about one a second, on and on | The arm steers by servo marks on the platter; unable to read them, it parks on the ramp and begins the approach all over again | Take the power off |
| A faint high beep, usually from a 2.5" portable | Nothing is turning: either the spindle bearing has seized or the heads have stuck flat against the platter surface | Do not tap it or shake it |
| A rasping, grinding noise | A head is scraping the platter — the crash is happening now | Every second powered costs files |
| A brief chirp each time it starts | Calibration is failing, often after a fall | Power down, leave it |
| Several clicks, then silence | It abandons the attempt, resets itself and starts over | Unplug it and break the cycle |
| Clicking on certain files only | One head is failing, or one band of the surface is bad | Abandon the copy; have it imaged |
The diagnosis costs you nothing. Every item that arrives is logged under a case number of its own, and an engineer works out what has actually failed — then tells you plainly which files can come back and which cannot. The price follows that: one fixed figure, in writing, and no work begins until you have read it and said go ahead.
The mechanism and the platter surfaces are examined before any power reaches the drive. If a noisy disk does have to run, an engineer stands over it, in bursts of a few seconds, and it is never left going.
Out comes the damaged head stack and in goes a matched donor set. A locked spindle is freed. Heads lying on the platter are eased off it without the surface being touched. After that the drive is asked for one thing only: a complete image.
A hardware imager reads with a light touch, stepping over anything that stalls and coming back to the stubborn areas at the end, until the copy holds every sector the disk can still give up. File system work then happens on that copy, never on the drive.
Nothing is billed while you are still deciding. A complete listing of what came off the drive reaches you first, and the invoice only follows your yes. Files travel back on media bought new for the job, return postage ours, and we do not close it here until every one of them opens on the machine you will use.
There is one instruction here, and it is to stop. The gap between a head and the surface below it is a few nanometres, at thousands of revolutions a minute, and nothing on that scale rights itself. Across every noisy disk that reaches the bench, the difference between a good result and a poor one is how fast somebody got to the socket.
It came in from a Cardiff shop, ticking one beat a second, with five years of takings on it. A pair of donor head stacks, then an image allowed to take its time, and the books were whole again, in time for the VAT return.
Normally they are not. A tick means the heads cannot work out where they are. It says nothing about the platters below, and that is where your files live. Matched donor parts and a patient image usually get nearly all of it back. The harm comes from the hours a drive spends running once the noise has begun.
No. That idea belongs to drive designs that stopped being made decades ago. Chill a modern disk and moisture condenses inside the casing while the metal contracts, so now you have corrosion and new mechanical damage on top of the original fault.
The heads are searching for servo marks written at the factory, failing to find them, retreating to the ramp and beginning the whole sequence over. A bent head, a failing preamplifier and a scored platter all sound much alike from outside, and only a look inside the sealed chamber tells them apart.
No. Every read it asks for still goes through the heads that have already failed, so the program spends its run demanding exactly the work that finishes the drive off.
A failing drive has only so many starts left in it. Spend none of them. Keep the power off, and the free diagnosis will tell you what is still readable.