Data Recovery Case File · Formatted & Logical Faults · When Tools Disagree, Believe the Errors
Hard Drive Showing as RAW: When Tools Disagree, Believe the Errors
His report described a drive giving three different answers at once, and ended with the word most people reach for at that point. "I have a hard drive that shows some file structure. Windows drive tools show it as expected, but other tools show it as RAW and cannot find partitions. Some directory structure is there, but I am getting basically continuous read errors on the disk. Strange." It is not strange — it is a specific and very readable picture, and the resolution is in the last observation rather than the first two. When software tools contradict each other about a drive, the disagreement itself is the finding: it means each tool is reading a different partial view of a surface that is not answering reliably. And the continuous read errors say plainly which layer the fault lives on. He also noted the live chat was closed when he tried it; everything here was handled in writing instead.
| Media | Internal hard drive — partial directory structure visible; one utility reports the volume normally, others report RAW with no partitions found; continuous read errors during access |
| Reported situation | Contradictory reports between disk utilities · persistent read failures across attempts · owner attempting to diagnose before committing |
| Fault class | Degrading media beneath a partially readable filesystem — inconsistent tool output caused by unreliable reads rather than by structural damage alone |
| Equipment used | Atola Insight Forensic media and head assessment · PC-3000 Express with Data Extractor prioritised imaging under per-sector timeouts · filesystem reconstructed from the completed image · contents verified by opening and delivered |
The decode: why tools disagree, and what continuous errors settle
Why two utilities give two answers: disk tools do not all read the same things. One may consult the partition table and report what it declares; another scans for filesystem signatures and reports what it finds; a third tries to mount and reports whether that succeeded. On a healthy drive these agree, because they are all reading the same reliable surface. On a drive where reads sometimes succeed and sometimes fail, each tool reads a slightly different subset — and each reports honestly on what it happened to get. So "Windows says it's fine, other tools say RAW" is not two opinions about the same evidence; it is two different samples of an unstable disk. The disagreement is the symptom.
What "RAW" actually means: worth defining, because the word alarms people more than it should. RAW is simply a tool's way of saying I found a volume but cannot identify its filesystem. It is a description of a failed identification, not of erasure. Files are not converted to RAW and nothing has been reformatted. On a stable drive that verdict usually means genuine structure damage; on an unstable one it can simply mean the tool could not read the structures cleanly enough to recognise them — which is exactly what his read errors predict.
Why the read errors outrank everything else: here is the decisive point. Filesystem problems are software problems, and software problems do not produce continuous read failures — a torn catalogue over a healthy surface reads perfectly, it just describes things wrongly. Persistent errors mean the drive is failing to return data from the media itself: weak sectors, degrading surfaces, or heads no longer reading reliably. That reorders the whole job. It means the real risk is not that the filesystem is broken but that the drive is getting worse while being examined, and it means every further scan with a consumer utility is spending reads that the recovery needs. The instruction that follows is unambiguous: stop running tools at it and image it first.
Why imaging resolves the contradiction: once a complete image exists, the instability disappears from the problem entirely. Every tool then reads the same stable copy and stops disagreeing, and the filesystem question can be answered properly — structures rebuilt where damaged, files recovered by content where the catalogue is beyond repair. The order matters more than the technique: capture first while the drive can still read, interpret afterwards where nothing degrades.
On the bench
The Atola Insight Forensic settled the layer question by measurement — the media and heads assessed directly, the weak regions his read errors had been announcing mapped as physical fact rather than inferred from software behaviour. Imaging then ran on the PC-3000 Express under Data Extractor with per-sector timeouts enforced and the sweep prioritised: healthy expanse captured first and completely, the failing neighbourhoods visited last and briefly so they could never stall the pass. On the completed image the contradiction evaporated — one stable copy, one consistent reading — and the filesystem was reconstructed, with content carving filling the areas the damage had taken. His files were verified by opening and delivered on fresh media.
The outcome
The drive imaged under timeout control and the volume rebuilt from the copy, verified and delivered. Free assessment, one fixed written figure including VAT; on drives that have to be opened, 50% of parts and labour upfront with the balance only on success — otherwise no recovery, no fee. The decode, for anyone whose utilities are arguing: contradictory tool reports are not a puzzle to solve with a better tool — they are evidence that the drive is returning different data on different reads; RAW means a filesystem could not be identified, not that anything was converted or erased; and continuous read errors settle the argument, because software faults do not cause them. When those three appear together, stop scanning and image the drive while it can still be read.
Disk tools contradicting each other, with read errors throughout
Read the disagreement as the diagnosis: different utilities read different things, and they only disagree when the drive is returning inconsistent results — which points at the media, not the filesystem. Don't be frightened by the word RAW; it means a tool couldn't identify the filesystem, not that your files were converted or wiped. But do take the read errors seriously, because a purely logical fault doesn't produce them — persistent failures mean the surface or the heads are struggling, and every additional scan is spending reads you'll want back. So stop running tools at it, don't attempt chkdsk or a repair utility (they write, and they'll stall on the same bad regions), and get one complete image taken with proper timeout control. Once there's a stable copy, the filesystem question becomes safe to answer.
The errors are the answer — call Leeds Data Recovery on 0113 322 3083; assessed by measurement, imaged under per-sector timeouts, the volume rebuilt on a stable copy.
Request a quote online →
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.