Home / Case Studies / Desktop and Ageing Externals
Desktop and Ageing Externals · case file

It Started With Write Errors

A lecturer in Childwall noticed write errors and did the sensible thing, which was to start copying everything off. He got about seventy per cent across before Windows stopped being able to read the drive at all. What was left behind was a single folder of recorded lectures, around 50GB, and it was the folder he could least afford to lose.

External DriveCorruption / FilesystemNot Detected
// case at a glance
MediaA WD Elements 3TB external drive that began reporting write errors, from which most content was rescued before Windows lost the ability to read it, with one 50GB folder still outstanding.
What was reportedWD Elements 3TB, write errors reported, roughly 70% copied off before failure, Windows now unable to read the drive at all, one folder of about 50GB still needed.
Fault classLogical, at sector level. The drive had developed a growing population of weak and unreadable sectors, and ordinary retries were making the read take longer rather than making it work. No mechanical fault was present and the board was healthy.
On the benchRead behind a hardware write blocker so nothing could be written, imaged read only on DeepSpar in controlled passes with the good areas taken first and weak sectors approached gently under a head health watch, filesystem parsed in R-Studio from the image, the outstanding folder located, validated and written to fresh media.
// the decode

The decode

Write errors are the first symptom you get, not the first fault you have. By the time a drive is failing writes it has usually been quietly reallocating sectors for a while. The error is the point at which it runs out of spare capacity to hide the problem behind.

Copying everything off was the right instinct executed the hard way. A file copy reads in the order the filesystem hands it out, and when it hits an unreadable sector it stops and retries until it gives up. On a drive with a growing bad area, that turns into hours spent hammering the worst part of the disk while the healthy part goes unread.

Imaging inverts the order and that is the whole point. The bench takes the readable areas first, in a single pass, and only then goes back for the difficult ones, in short controlled attempts. That way the bulk of the data is safely captured before any time is spent on the sectors most likely to finish the drive off.

No opening, no board work. Bad sectors are not a mechanical failure and they are not an electronic one. This was a read job, done read only, which is why it ran on the logical terms rather than the physical ones.

// on the bench

On the bench

The drive was put behind a hardware write blocker so nothing, including Windows itself, could write to it. It was imaged read only on DeepSpar in controlled passes. The first pass skipped anything slow and captured everything that came back cleanly, which took the majority of the surface.

Later passes went back over the weak areas in short attempts with head health watched throughout, so the drive was never left grinding at a sector it was not going to return. The filesystem was parsed in R-Studio from the finished image, the outstanding folder located, and the recordings validated by playing samples from across the set before being written to fresh media.

// the outcome

The outcome

The folder came back complete, recordings playable end to end. The seventy per cent he had already rescued stayed where it was, which kept the job small.

One drive, £300 + VAT. The fault was at sector level and the work was a read only image with nothing opened and no board touched, so it ran on no fix, no fee terms.

At the first write error

Stop copying and stop retrying. Every retry against a failing sector is more running time spent on the part of the disk least able to take it, and the clock you are burning belongs to the healthy part you have not read yet. If a copy has already stalled, cancel it rather than leaving it to grind overnight. Do not run a check disk, because repair tools write. Unplug the drive, note roughly how much you managed to rescue and what is still missing, and send it in. Imaging in controlled passes gets far more back than a file copy ever will.

Something like this on your own desk? Every job starts the same way. The diagnostic is free and takes 2 working days from the parcel reaching the bench, never faster, and it ends in one written figure that does not move afterwards. The fastest a completed recovery leaves here is 2-4 working days. The published rule is this: no fix, no fee all jobs except electronic and mechanical failures, chip level work, DVR and Forensic jobs. So logical faults, firmware and service area work included, run on no fix, no fee. Anything physical — a drive opened in clean air, a board or chip-level repair, water or fire damage — takes 50% of the quoted price upfront, with the balance only on success. CCTV and DVR work, BitLocker volumes and forensic jobs with a report are settled in advance. Cards and USB pen drives are £250 + VAT, any single hard drive or SSD including NVMe is £300 + VAT, arrays, NAS boxes and servers start at £500 + VAT, CCTV and DVR is £400 + VAT, BitLocker is £400 + VAT and forensic work with a report is £800 + VAT. Where the data sits inside a laptop, a desktop, a Mac or a server, take the drive or SSD out and post that on its own: internal drives are not removed from machines here, and flash soldered directly to a logic board, as in ultraslim laptops, tablets and Apple Silicon Macs, is outside what this lab takes in. If it unplugs or unscrews, it is our sort of work. Post it tracked and insured to Manchester Data Recovery, Peter House, Oxford Street, Manchester M1 5AN, or book and pay for a courier yourself. There is no collection service and no counter in Liverpool. Freephone 0800 689 0668, Mon–Fri 9:00am–5:30pm. Packing advice and the shipping form live here.
Start a free diagnostic

Every case file here is written up from a real enquiry handled for households and businesses across Liverpool, Merseyside and the wider North West, anonymised so that nobody is identifiable. Each one gives the reasoning behind the diagnosis, the method the fault actually called for, and the equipment it was worked on.

// related case files

Other files in the same vein

Browse all case studies →

Recognise your own drive in this one?

The diagnostic is free and takes 2 working days from arrival, the figure is fixed in writing, and logical faults run on no fix, no fee. Start online or ring the freephone.