Data Recovery Case File · Portable Drives · Appearing Is Not Reading
It Gets as Far as Introducing Itself and No Further
His enquiry describes a drive doing half its job. A photographer's drive that "stopped working whilst carrying out a backup. I've tried with a new docking station — it's very slow, the computer finds the drive but then loses it." He has already eliminated the dock, which most people never do — and the cycle he describes has a specific shape: the drive completes its introduction, cannot sustain what follows, and is removed by a host that has run out of patience.
| Media | External hard drive — failing during a sustained backup operation; enumerating and subsequently dropping; transfer rates heavily degraded; alternative docking station eliminated by the owner |
| Reported situation | Drive failing during a backup operation · replacement docking station tried by the owner without improvement · drive detected by the host on connection · drive subsequently lost by the host · operation very slow throughout · professional photographic content held |
| Fault class | Severe read failure with enumeration succeeding — host timeout removing the device; failure revealed by sustained sequential access |
| Equipment used | Owner's docking station substitution accepted as interface elimination · Atola Insight Forensic error-rate assessment · imaged write-blocked under strict per-sector timeouts with retries capped · healthy expanse captured first · losses mapped per file |
The decode: two different things, one of which works
What "finds the drive" establishes: the whole introduction succeeded. The drive powered up, completed its own start-up, read its internal configuration, reported its capacity and identity, and was registered by the host. Every one of those steps requires the drive to be functioning, and they all worked.
What "then loses it" describes: what happens next. Once a device is registered, the host begins reading — structures first, then content. A request that is issued and never answered holds a slot in the system's queue while the drive retries, repositions and escalates through an internal recovery sequence taking seconds per sector. Eventually the host concludes the device has gone and removes it.
So the two observations are not in tension: introducing yourself is a brief exchange of small amounts of information from a reserved area. Reading content is sustained work across the whole surface, and a drive can manage the first indefinitely while failing the second.
Why "very slow" is the same finding again: the time is retries. Nothing about a drive gets tired or gradual — it either returns a sector promptly or spends seconds failing to. Slowness is failure measured in seconds rather than reported as an error, and it is the most reliable early indicator there is.
Why the backup revealed it: a backup reads sequentially across the entire drive, including regions ordinary use never touches. It is the first operation to visit everything, which is why marginal drives fail during backups so consistently — and why the backup did not cause it.
Why his dock substitution matters: a docking station supplies power and interface, and a marginal one produces exactly this pattern. Trying a second one eliminates the whole interface path in a single step, which is a proper control and one hardly anybody performs. The fault is the drive's.
What must stop: further connection attempts and any further backup runs. Each is sustained reading across a failing surface, and the professional consideration is that time spent proving the fault is time not spent capturing.
On the bench
The owner's docking station substitution was accepted as interface elimination, a second dock producing identical behaviour ruling out power delivery, bridge and connector in one step. The Atola Insight Forensic assessed error rates, and imaging ran write-blocked under strict per-sector timeouts with retries capped — enumeration requiring only a brief exchange from a reserved area while content reading is sustained work across the surface, so a drive can complete the first and fail the second indefinitely. The healthy expanse was captured first, with losses resolved per file.
The outcome
The interface eliminated by the owner's own control, error rates measured and the drive imaged under capped timeouts with the healthy expanse taken first. 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: introducing itself is a brief exchange from a reserved area, and reading content is sustained work across the whole surface — a drive can manage the first and fail the second. The slowness is retries, which is failure measured in seconds.
Drive that appears and then disappears
Stop connecting it, and stop running the backup again. Those two observations aren't in conflict: being found means the drive completed its whole introduction — powered up, read its own configuration, reported its capacity — which is a brief exchange of small amounts of information from a reserved area. Being lost afterwards is the host giving up while reading content, which is sustained work across the entire surface. A drive can manage the first indefinitely and fail the second. The slowness is the same finding: nothing about a drive gets gradually tired, so time spent is time retrying. And a backup revealed it because a backup is the first thing to read everything.
Stop retrying — call Oxford Data Recovery on 01865 593000; your dock substitution accepted as interface elimination, error rates measured, imaged under capped per-sector timeouts with the healthy expanse taken first.
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.