Data Recovery Case File · Desktop Externals & Aging Drives · Use the Window
A Drive With a Working Minute In It
His enquiry describes a fault with a clock on it, and he has already done the elimination properly. A 1TB drive in a USB caddy: "tried multiple caddies, pretty sure it's drive related not caddy. Windows detects the drive and partitions, but after a minute or so I get a hardware error and the drive is not accessible." That consistency is the useful part. A drive that works for a predictable interval and then stops has a window — and what happens inside that window decides everything, because browsing folders spends it and imaging banks it.
| Media | 1TB hard drive read through a USB caddy — detected with partitions listed, failing with a hardware error after approximately one minute; behaviour consistent across multiple caddies |
| Reported situation | Drive detected on connection with partition structures visible · hardware error occurring after approximately one minute · drive inaccessible thereafter · behaviour reproduced across several different caddies |
| Fault class | Time-limited operation — drive functioning briefly before failing under load or temperature; recovery dependent on capturing within repeated windows |
| Equipment used | No browsing or exploratory access · imaged write-blocked with resumable multi-pass capture (PC-3000 Express with Data Extractor) · sector map retained across power cycles · thermal behaviour observed between passes · files validated by opening |
The decode: what a consistent minute means, and how to spend it
What his caddy test established: more than most enquiries manage. Multiple caddies means different bridge chips, different cables and different power arrangements, all producing identical behaviour. That eliminates the interface entirely and places the fault in the drive — which is exactly the conclusion he reached, and it is correct.
Why detecting the partitions matters: reading a partition table means the drive spun up, became ready, positioned its heads and successfully read the very start of the disk. So the mechanism works, the electronics work, and the beginning of the media is readable. Whatever fails, fails after all of that has already succeeded.
What a consistent one-minute failure usually is: two candidates, and they are distinguished by observation rather than repetition. Thermal — a marginal component works from cold and fails as it warms, which produces a very repeatable interval because the warming is repeatable. Or load-dependent — the drive is fine until sustained reading begins, and a failing head or a struggling power circuit gives out under continuous work rather than at idle. Both present as a hardware error and a device that drops off the bus.
Why this is the crucial part: he has a finite, repeating budget of working time, and most people spend it the way anyone would — plugging it in, opening folders, looking for the important files, watching it die, trying again. Every one of those minutes is gone and returns nothing. The window should be spent capturing sectors, not looking at them.
How imaging exploits a window like this: an imager keeps a persistent map of what it has already captured. When the drive drops, it waits, allows the device to re-enumerate after a power cycle, and resumes from where it stopped rather than starting again. So a drive with sixty usable seconds at a time can be read in full across many passes, each one adding to the same image. A file copy cannot do this, because it restarts the file it was working on and reaches the same point each time.
What helps if it is thermal: allowing the drive to cool fully between passes lengthens each window. That is not the same as freezing it, which introduces condensation onto the platters and should never be done.
On the bench
No browsing or exploratory access was performed, since the drive offers a finite repeating budget of working time and folder navigation spends it without capturing anything. The drive was imaged write-blocked with resumable multi-pass capture on the PC-3000 Express under Data Extractor, the sector map retained across power cycles so that each window added to the same image rather than restarting it. Thermal behaviour was observed between passes to establish whether cooling lengthened the working interval, and files were validated by opening.
The outcome
The interface eliminated, the drive imaged across repeated windows into a single accumulating image and files validated. 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 whose drive works briefly and then errors: that consistency is a resource. Detecting the partitions proves the mechanism, electronics and start of the media are all fine, so the failure comes after everything difficult has succeeded — usually thermal or load-dependent. Spend the window capturing rather than browsing, because an imager keeps a map and resumes across power cycles where a file copy restarts and dies at the same point.
Drive that works briefly then throws an error
Stop opening folders on it — that predictable minute is a resource and browsing spends it for nothing. What you have is a drive with a repeating window of function, and the way to use it is to capture sectors rather than look at files. An imager keeps a persistent map of what it has already read, so when the drive drops it waits, lets the device come back, and resumes from where it stopped; across enough passes a drive with sixty usable seconds can be read completely. A file copy can't do that, because it restarts whatever file it was on and hits the same failure point every time. Testing several caddies was the right move and rules out the interface. If the failure is thermal, letting the drive cool fully between attempts lengthens each window — but never freeze it.
Don't spend the window browsing — call Newcastle Data Recovery on 0191 406 1051; imaged write-blocked with resumable multi-pass capture, sector map retained across power cycles, files checked by opening.
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.