Data Recovery Case File · Portable Drives · Two Failures, Not One
Intermittent Recognition and Unreadable Files Are Separate Faults on the Same Drive
Her enquiry describes a drive failing in two distinct ways and it is worth keeping them apart. A portable drive the laptop "has trouble recognising, often displaying that the drive is not accessible" — and then, "when able to access the drive, files are inaccessible." Those are different problems at different levels: one is the device failing to present itself, the other is content failing to read once it has.
| Media | 1TB portable external hard drive — intermittent enumeration with a host inaccessibility error; individual files failing to open when the volume does present |
| Reported situation | External drive frequently not recognised by the host · host reporting the drive as not accessible · drive occasionally presenting its volume · files inaccessible when the volume is available · file browser becoming unresponsive during access attempts · content required |
| Fault class | Intermittent enumeration with surface read failures — connection or supply instability alongside regional degradation; each requiring separate handling |
| Equipment used | Intermittent enumeration and read failure assessed as separate faults · drive powered independently of the host interface with draw measured on a controlled bench supply · imaged write-blocked under strict per-sector timeouts once a stable connection was established · degraded regions mapped and revisited across later passes · content carved independently where structures did not reconcile |
The decode: why two faults matter more than one, and which to solve first
Why the distinction is not pedantry: they are solved differently. An enumeration problem is addressed at the connection and the supply; a read problem is addressed at the surface, and confusing them wastes the drive's remaining life.
What intermittent recognition points at: the connection or the power. A drive that appears sometimes has a physical or electrical inconsistency rather than a data one — a fatigued connector joint, a marginal cable, or a port supplying barely enough current.
Why the inaccessibility message fits that: it is what a host reports when a device stops responding partway. The volume was there and the drive ceased to answer, which is a connection event rather than a filesystem one.
What the second fault points at instead: the surface. Files failing to open on a volume that has mounted successfully means the structures were read and the content behind them was not.
Why the browser becoming unresponsive confirms it: a read that never returns. An application waiting indefinitely on an unanswered request takes the interface down with it, which is what an unreadable region produces.
Why the two are probably related in cause even though they are separate in kind: a drive struggling to read draws power irregularly. Extended retries increase demand, and a marginal supply can collapse under them — so a read problem can present as a connection problem.
Why that ordering determines the approach: the connection must be made reliable before anything else can be judged. A capture attempted over an intermittent link will fail for reasons that say nothing about the surface.
How that is achieved: the drive is powered independently of the host interface and connected through equipment that imposes its own timeouts. That removes both the port's supply limits and the host's habit of waiting forever.
Why capture then proceeds differently from a copy: healthy regions are taken first at speed. Difficult areas are deferred and revisited, rather than stalling the whole operation at the first one that will not answer.
What must not happen meanwhile: no further browsing and no repair. Every attempt is another abrupt disconnection with the heads over the surface, and a consistency check on a drive with unreadable regions discards references it cannot verify.
On the bench
Intermittent enumeration and read failure were assessed as separate faults — the former indicating connection or supply inconsistency, consistent with a host reporting inaccessibility when a device stops responding partway, and the latter indicating structures read successfully with content behind them unreadable, consistent with an application hanging on a request that never returns. Extended retries raise current demand, so a read problem can present as a connection problem. The drive was powered independently of the host interface.
The outcome
The two faults assessed separately, the drive powered independently of the interface, and healthy regions captured first under capped timeouts. 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: you are describing two faults, not one. Not being recognised is a connection problem; files not opening once it is recognised is a surface problem — and the connection has to be made reliable before the surface can be judged.
A drive that is unreliable in two different ways
Stop browsing it and don't run a repair — every attempt ends in an abrupt disconnection with the heads over the surface, and a consistency check on a drive with unreadable regions discards references it can't verify. Keep the two symptoms separate when you describe it, because they're solved differently: intermittent recognition points at the connection or the power supply, while files failing to open on a volume that did mount points at the surface. They can be linked, since long retries raise current demand and a marginal supply collapses under them.
Those are two faults — call Cambridge Data Recovery on 01223 655015; assessed separately, powered independently of the host interface, healthy regions captured first under capped timeouts.
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.