Data Recovery Case File · Portable Drives · Writing Fails First
It Looked Fine Until Something Was Asked of It
His enquiry reports the failure and, more usefully, the moment he noticed it. An external drive that "some time in the last week has ceased to be detected or visible in the file manager. I found this out trying to save a document. I then found that nothing could be accessed. I purchased a new cable, to no avail." The discovery is the interesting part. He was not investigating a problem — he was working, and the drive had been sitting there apparently normal until the moment something was actually required of it.
| Media | External hard drive — no longer detected or presented by the host; failure discovered on attempting to save a file; replacement cable tried without effect |
| Reported situation | Drive ceasing to be detected at some point within the previous week · failure discovered on attempting to save a document · nothing subsequently accessible · replacement cable purchased and tried without change |
| Fault class | Failure to present with the interval since onset undetermined — readiness or interface fault; write demand exposing an already-failed state |
| Equipment used | No repeated connection attempts · drive removed from the enclosure and assessed on a native connection · readiness sequence read under strict timeouts · imaged write-blocked on restored readiness |
The decode: why writing is what discovers these, and what the cable proved
Why a failed drive can go unnoticed for days: a computer holds a great deal in memory. A folder browsed earlier stays cached, a document already open remains open, and the icon representing a drive can persist after the device beneath it has stopped responding. So a drive that failed on Tuesday may look entirely present on Wednesday to somebody who is not asking it for anything new. Absence of a symptom is not evidence of function.
Why saving is the operation that finds out: a write cannot be served from cache. It has to reach the device, be acknowledged, and be committed — so it is the first operation that genuinely requires the drive to be there. That is why so many people in this archive discover a failure at the moment they try to save something, and it is also why the discovery moment tells you nothing about when the drive actually failed. His "some time in the last week" is honest and probably right.
Why that matters practically: the interval is unknown, which means it is unknown how long the drive has been failing and how much has been asked of it since. It also means the last successful backup — if there is one — may be more recent than he fears, since the drive may have failed earlier than he noticed rather than later.
What the new cable established: a genuine elimination, and a cheap one. Cables do fail, particularly on portable drives that live in bags, and replacing one is the correct first move. Its making no difference rules out the cheapest explanation and leaves the enclosure, the bridge board, and the drive itself.
What is worth doing that he has not: trying a different port and a different computer. Both are free, take minutes, and eliminate the host — which matters, because a failing port produces exactly this and is invisible from the machine it is on.
The instruction meanwhile: not to keep connecting it. If the drive is stalling during its start-up, every attempt is another full readiness cycle on a device that cannot complete one.
The habit worth taking from this: a drive that holds anything important should be opened occasionally rather than trusted to be there. Not a backup regime — just a file opened from it now and then, which is the only thing that distinguishes a working drive from a cached icon.
On the bench
No repeated connection attempts were made, each being another full readiness cycle on a device that may be unable to complete one. The drive was removed from the enclosure and assessed on a native connection, the owner's cable replacement having already eliminated the cheapest cause and left the enclosure, the bridge and the drive. The readiness sequence was read under strict timeouts, and imaging performed write-blocked on restored readiness.
The outcome
The cable eliminated by the owner, the enclosure taken out of the question on a native connection and the drive imaged on restored readiness. 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, for anyone who found out by trying to save: that is the usual way. Computers cache heavily, so a drive's icon and its browsed folders can persist after the device has stopped responding — while a write cannot be served from cache and has to reach the device. So saving is the first operation that genuinely requires the drive, and it says nothing about when the failure began.
Drive that seemed fine until you tried to save something
Try a different port and a different computer before anything else — both are free and they eliminate the host, which a failing port makes impossible to spot from the machine it's on. Replacing the cable was the right first move and its making no difference has already ruled out the cheapest cause. Worth knowing why you discovered this the way you did: computers cache heavily, so a drive's icon and any folder you've browsed can persist long after the device stopped responding, and a document already open stays open. A write is different — it can't be served from memory and has to actually reach the drive. That's why saving is what finds these, and why you can't tell from the discovery when it really failed.
Try another port and machine first — then call Newcastle Data Recovery on 0191 406 1051; assessed on a native connection with the enclosure eliminated, readiness read under strict 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.