Preparing Automatic Repair, Forever

The decode

His cross-machine test is the single most useful thing he could have done, and it should be recognised. The most common uncertainty with a non-booting drive is whether the fault lies in the computer or the drive. By connecting it to a second machine and seeing the same failure, he has already answered that — the drive carries its fault with it, so the problem is the drive. That is proper diagnostic reasoning, and it saves a step.

"Preparing automatic repair" that never finishes is Windows meeting unreadable data. The repair process launches when the system files needed to boot cannot be read correctly, and it hangs when it cannot read them either — an endless loop because the underlying data is physically inaccessible, not because the repair is slow. The loop is a symptom of drive degradation, exactly as with a bad-sector fault.

The external-connection failure confirms the depth of the problem. When a drive attached as external storage to a healthy machine still cannot be read, its filesystem structures or the media holding them are damaged at a level no host can work around. Detected-but-unreadable, or not usable when externally connected, is a drive whose own data-bearing surfaces or firmware are failing — precisely the hardware conclusion he reached.

Every attempt so far has drawn down a limited budget of reads. A degrading mechanical drive has a finite number of reliable reads left, and each boot attempt, each repair loop, each external-connection try spends some of them. The protective move now is to stop trying to access it for use and hand it to controlled imaging — reading it once, carefully, rather than repeatedly and hopefully.

Imaging separates the salvageable bulk from the damaged remainder. A hardware imager reads under strict retry control, mapping what each head can still manage and banking the readable majority first — his personal files among them — before straining for the difficult regions. The Windows installation itself is of little interest; it is the documents, photos and personal data that the imaging is aimed at securing first.

The honest ceiling is set by the surface and reported per file. Where bad regions hold personal files, those come back damaged or incomplete, and the result is an honest listing of what recovered cleanly. A mechanical 2TB drive caught before total failure typically yields the bulk of its personal data — but the drive's physical condition, read at assessment, sets the real limit, stated before the work.

On the bench

With the fault confirmed drive-side, the drive's health was assessed on PC-3000 UDMA — sector errors, reallocations and firmware state. It was imaged under a hardware write-blocker on DeepSpar Disk Imager with per-head mapping and tight retry limits, the readable bulk and personal files secured first and degraded regions revisited last. NTFS structures were rebuilt from the image in R-Studio, damaged areas carved, and the recovered personal files validated by rendering.

The outcome

Drive-side fault confirmed and health assessed, imaged under retry control with personal files banked first, and the data rebuilt from the image. 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 decode: your second-machine test already proved it — the fault travels with the drive, so it is the drive. The repair loop is Windows meeting unreadable data, and imaging secures your personal files before the drive is asked to strain for the rest.

Our case files are written up from genuine enquiries our lab has handled for customers across Sheffield and South Yorkshire, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery approach our engineers apply to that fault, using the equipment listed.