A distribution firm at Magna Park, Lutterworth, keeps scanned delivery notes and signed manifests on an old Maxtor external, because claims can arrive up to six years after a load moves. The office noticed that some scans opened normally while others hung and then failed altogether. Nothing had happened to the drive. They rang the same week rather than waiting to see whether it settled, which turned out to be the decision that saved most of the archive.
| Media | An ageing Maxtor external drive holding six years of scanned delivery paperwork. Some files open normally while others hang and never load, with no impact or incident behind it. |
| What was reported | Maxtor external drive, some files still open and others will not load at all, no drop or knock, the office wanting the archive off it before more becomes unreadable. |
| Fault class | A failing surface. Parts of the disk still read at full speed while others had degraded to the point where the drive retried endlessly and gave up. The mechanism was intact; the recording layer had worn unevenly, which is ordinary for a drive of this age and use. |
| On the bench | Read behind a hardware write blocker with the health and defect tables examined before anything else, imaged on DeepSpar in passes ordered by condition so intact regions were captured first, weak regions revisited at reduced speed on a narrowed head map with the drive's temperature and error rate watched, the volume parsed from the image in R-Studio, and the recovered documents validated and written to fresh media. |
Patchy access is the most informative symptom a drive produces. A dead drive tells you nothing about how long it had been unwell. A drive where some files open and others do not is showing you the boundary between the parts that still work and the parts that have gone, and that boundary moves in one direction. Acting the same week rather than the same month is worth more than any equipment.
Hanging is the drive trying, not the computer sulking. When a read fails, a drive retries, and by default it will retry for a long time before returning an error. Those long pauses are the disk working extremely hard on one small area, generating heat and wearing at exactly the region that is already weakest. It is the reason a failing drive gets worse fastest while somebody is patiently waiting for a file to open.
Copying folder by folder is the natural approach and the wrong one. Dragging files across makes the drive work through the difficult regions in whatever order the folder tree happens to sit in, spending its remaining good behaviour on the hardest parts early. Imaging by condition takes everything easy first, so that if the drive deteriorates half way through, what is lost is the region already known to be bad rather than an arbitrary slice of the archive.
Retry-happy consumer software is the specific hazard. The tools sold for this purpose read through the operating system, which means they cannot control how long the drive spends on a failed sector, and many of them will pass over the same bad region repeatedly on successive scans. A bench imager can be told to give up after a set time, come back later and approach from a different direction, which is the whole difference in outcome.
The health tables were read first. The grown defect list was long and still growing, with reallocations recorded during the office's own attempts to open files, which confirmed that continued use was actively costing them. Nothing suggested a mechanical fault: the mechanism was healthy and the recording layer was not.
DeepSpar imaged it in passes ordered by condition. The first pass took everything that read at speed, which came to a little over eighty per cent. Later passes returned to the weak regions with the head map narrowed, the timeout shortened and the drive allowed to cool between runs, recovering most of what remained. R-Studio parsed the volume from the image, and the scanned documents were opened as documents before anything was written to fresh media.
The archive came back at about ninety-seven per cent, and the office was given a list of the specific files that had not, so they could tell a customer honestly rather than discovering a gap during a claim. The drive was not returned to service.
One disk is £300 + VAT. The lid stayed on and no parts were transplanted, which puts the job squarely inside no fix, no fee, and the office was billed once it had the archive.
Take the drive out of use the same day. The boundary between what reads and what does not only moves one way, and the fastest way to move it is to keep opening files, because each failed read is the drive retrying at length on the weakest part of itself. Do not run a scanning tool over it, do not run a disk check, and do not try to copy the important folders off first: the important folders are wherever they happen to be, and reading them in that order spends the drive's remaining margin on the difficult regions. If it is still readable enough to be worth imaging, that state will not last, so the useful thing is to unplug it and send it while the good areas are still good.
Each of these is a real enquiry, taken from a household or a business somewhere in Leicester or the surrounding county, with the name of the customer, the firm and anything else that would identify them taken out. What is left is the reasoning: what turned up, what had actually failed inside it, the method that fault required and the equipment it was worked on. One of the thirty-six is a job that was refused, and it belongs here every bit as much as the successes do.
A drive that cannot read its own service area never gets as far as introducing itself
Blaming the enclosure is the natural first move and it is almost never the answer
Beeping is a mechanism drawing power it cannot turn into movement
Examining it is free and takes 2 working days from arrival, the figure goes to you in writing and does not move afterwards, and a logical fault carries no fix, no fee. Start online or ring the freephone.