Data Recovery Case File · Second Fixes & Trade Handoffs · A Repair That Bought Days

Chkdsk Fixed It, Then the Drive Died

His enquiry came from someone doing a favour, and it describes a sequence that plays out constantly in the trade. "I was asked to fix an unresponsive PC earlier in the week and was able to get it booting again after running chkdsk from a recovery CD. This morning it froze and the hard drive is now not being recognised. It still seems to spin up, does not make any clicks, but is not found by disk manager when attached to another PC." His observations are good and his testing is sound. What deserves examining is the middle of that story — because the repair genuinely worked, the machine genuinely booted, and the drive genuinely died a few days later. Those are not three unrelated events. A filesystem repair that succeeds on a failing drive usually buys days rather than fixing anything, and understanding why changes what anyone should do the next time one works.

The decode: why the repair worked, and why that was the warning

What a filesystem repair actually does: it examines the structures describing files and folders, finds inconsistencies, and resolves them by writing corrections — reconciling records that disagree, discarding entries it cannot make sense of, and marking bad sectors so the filesystem stops using them. On a healthy drive with a corrupted filesystem, that is genuinely a fix. The important word is filesystem: the tool operates entirely at that layer and knows nothing about the drive's physical condition.

Why it worked, and why that was the warning: here is the part worth taking away. If the underlying problem was media degradation — weak sectors, deteriorating surfaces, a failing head — then the filesystem inconsistencies were symptoms rather than the disease. The repair cleaned up the symptoms, marked the currently-unreadable sectors as unusable, and the machine booted, because the filesystem was now internally consistent again. Nothing about the drive's physical condition changed. So the successful repair did not indicate a solved problem; it indicated a filesystem that had been damaged by something ongoing. And a drive that damages its own filesystem once will do it again, sooner and worse. A repair that works on an unexplained failure should be treated as a warning, not a resolution — and the correct next step is to image the drive immediately, while it is briefly cooperative, rather than to hand the machine back.

What his current symptoms say: spinning normally with no clicking means the motor, bearings, power and board are all working — the mechanical stage is clear. Not being found by disk management on a second machine means the drive is not completing the silent part of its start-up: head calibration, and reading its own service area. Since that region is stored on the platters like everything else, degradation reaching it produces exactly this. The progression from filesystem damage to service-area unreadability is a coherent single story rather than bad luck.

The trade note, said fairly: he was doing someone a favour and the repair was a reasonable thing to try on an unresponsive PC. The refinement worth carrying is one of sequence: on a machine whose failure has no clear cause, image the drive before repairing it, and if a repair does succeed, copy the data off the same day rather than declaring victory. That costs an hour and occasionally saves everything.

On the bench

The Atola Insight Forensic read the drive's power-on behaviour directly — power delivery, spin-up, calibration attempts and their outcome, and the exact point at which progress stopped — which turned "not recognised" into a located finding and confirmed the mechanical stage was sound. Where the service area proved to be the obstacle, access followed through the PC-3000 Express in technological mode, with damaged modules repaired until the drive would identify itself and address its media. Imaging began immediately under Data Extractor with per-sector timeouts enforced and healthy regions captured first, since a drive that has already been marking bad sectors has plenty more to find. The volume was rebuilt from the image and the contents verified by opening.

The outcome

The drive read at the level its fault required, imaged under timeout control and the contents verified and delivered. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The lesson, for anyone who has just fixed a machine with a filesystem repair: the tool works at the filesystem layer and knows nothing about the drive's physical condition, so if the real problem was media degradation, a successful repair tidied the symptoms and changed nothing underneath; a repair that works on an unexplained failure is a warning rather than a resolution, and the right response is to image the drive that day while it is still cooperating; and the progression from filesystem damage to a drive that spins but never identifies itself is one continuous story, because the drive's own management data lives on the same deteriorating surfaces.

Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.