Data Recovery Case File · Desktop Externals & Aging Drives · Supply, Not Surface
A Small Board Cannot Deliver What a Desktop Drive Draws
His enquiry contains an experiment that produced a result worth re-reading. A set of old drives, some from a much earlier generation, where his computer recognises them but reads nothing, and "I think I managed to see some data using a single-board computer once, but it was so slow." Seeing data at all is the important finding — and the slowness may say more about what that board could supply than about the condition of the disks.
| Media | Four hard drives of two generations — two of 1TB on a modern interface and two smaller on an earlier one; recognised without readable content on a desktop host; partially read at low speed on a single-board computer |
| Reported situation | Four drives retained from machines of an earlier era · two larger on a current interface and two smaller on an older one · drives recognised by a desktop computer without content being readable · some content previously visible using a single-board computer · read performance described as extremely slow · content required |
| Fault class | Content demonstrably readable under some conditions — supply adequacy and interface configuration to be excluded before drive condition is assumed |
| Equipment used | Prior partial success accepted as evidence content is present · supply adequacy assessed independently of the drives · earlier-interface configuration jumpers checked before connection · imaging on a controlled bench supply under strict per-sector timeouts · each drive assessed independently |
The decode: three things to exclude before blaming the disks
Why his single-board result is the most useful sentence: he saw data. Whatever else is true, the content exists and the drives can be read under some conditions — which rules out the worst possibilities before anything is opened.
The first thing to exclude — power. A desktop drive draws considerably more than a small board can supply, particularly on start-up when the motor has to bring the platters to speed. An underpowered drive spins up slowly, drops out under load, or reads at a fraction of its speed — which is precisely what he describes.
Why that is so commonly missed: the drive works, so the supply appears adequate. Marginal power does not produce a clean failure; it produces poor behaviour that looks like a failing disk, and the difference is invisible without measuring.
What it means practically: larger drives need their own supply rather than drawing from a host connection, and a controlled bench supply removes the question entirely.
The second thing to exclude — interface configuration. Drives of the earlier generation he describes are configured by small jumpers setting their role on a shared cable. Two drives set to the same role will not both work, and a drive set wrongly for the adapter it is attached to may be recognised without being readable.
Why nobody checks this any more: the practice disappeared twenty years ago. People connecting old drives through adapters have no reason to know those settings exist, and the pins are unlabelled without the diagram that was printed on the drive.
The third — recognised but not readable, on a modern system. That combination frequently means a filesystem the current system does not mount rather than an unreadable drive. Drives from that era carry filesystems modern systems may recognise and decline, and a single-board computer running a different operating system would read them where a desktop would not, which fits his experience exactly.
Why each drive is assessed separately: four drives of two generations have nothing in common but an owner. They may have entirely different faults, or three may have none at all.
On the bench
Prior partial success was accepted as evidence content is present. Supply adequacy was assessed independently of the drives — desktop drives drawing more than a small board delivers, particularly during spin-up, which produces slow reads and drop-outs indistinguishable from disk failure without measurement. Earlier-interface configuration jumpers were checked before connection, role settings on that generation determining whether a drive is readable, and imaging performed on a controlled bench supply.
The outcome
The prior partial read accepted as evidence, supply and interface configuration excluded before any assumption about the disks, and each drive assessed independently. Free assessment, one fixed written figure including VAT per drive, 50% of parts and labour upfront with the balance only on successful recovery. The decode: seeing data once is the finding, and the slowness may have been supply rather than surface. A desktop drive draws more than a small board delivers, especially spinning up — which produces exactly the behaviour you saw.
Old drives that read slowly or not at all
Give them their own power supply before concluding anything — desktop drives draw considerably more than a small board or an unpowered adapter can deliver, especially during spin-up, and marginal power produces slow reads and drop-outs that look exactly like a failing disk. If you saw data even once, that's the important finding: the content is there and readable under some conditions. On drives from the earlier interface generation, check the configuration jumpers too, since role settings determine whether a drive is readable and the practice vanished twenty years ago. And recognised-but-unreadable often means a filesystem your current system declines to mount.
Try a proper supply — or call Oxford Data Recovery on 01865 593000; supply and interface configuration excluded first, imaged on a controlled bench supply under strict per-sector timeouts, each drive assessed independently.
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.