Twenty years of parish records on a drive nothing in the building could connect to. The disk was fine; the interface had gone extinct, and a drive that age gets one careful read rather than a series of experiments.
A local historical society had two decades of transcribed parish records on a Compaq desktop retired around 2003 and left in a cupboard since. The machine no longer powered on, and nothing in the building had an IDE connector. They had assumed the material was simply gone.
It spun up cleanly, sounded steady, and reported its true capacity. Age had produced a modest reallocated-sector count and nothing worse.
The obstacles were practical rather than mechanical, and they are specific enough to be worth listing, because they are why a great many people conclude an old drive is dead when it is not.
The physical interface is gone. Parallel ATA uses a 40-pin ribbon and predates SATA entirely. Two drives share one cable, distinguished by a jumper setting them as master or slave — or by cable select, which decides by position on the ribbon. Get the jumper wrong and the drive does not appear at all, which is the first thing that convinces someone a disk is dead.
Modern adapters make assumptions this drive predates. A USB-to-IDE bridge is built for the tail end of the PATA era. Drives from earlier report their geometry differently, respond to identification commands with fewer fields populated, and in some cases predate the 48-bit addressing that modern bridges assume. The result is a bridge that reports the wrong capacity, truncates the drive, or refuses to enumerate it — which is exactly what had happened here and exactly what had convinced the society the disk had failed.
The cable itself matters. Later PATA modes require an 80-conductor cable with additional ground lines between the signal lines. Attach a drive expecting one to a 40-conductor ribbon and it will negotiate down or produce read errors that look like media faults.
None of the above damages anything. Three things actually can, and they are why an old drive gets one careful read rather than repeated experiments.
Stiction. On drives that park their heads on the platter surface rather than on a ramp, long storage can let the head and the surface bond. The motor then cannot break them apart, and the drive either fails to spin or spins with the heads dragging.
Lubricant migration. The bearing lubricant redistributes over years of stillness. A drive that has not turned in two decades may spin up perfectly and then develop bearing noise within hours of running.
Loss of magnetic signal. Slow, and much slower than the internet suggests — but the older recording densities were also written at lower field strengths, and a drive of this age sits at the point where marginal sectors start needing more than one attempt.
Which is why the plan assumed one opportunity rather than several.
A period-appropriate IDE controller was used rather than a modern USB bridge, which resolved the enumeration problem immediately — the drive was recognised at its correct capacity on the first attempt.
From there the approach was patience. Read-only imaging in a single planned pass with conservative timeouts, taken sequentially rather than jumping around the surface, on the working assumption that the drive might not tolerate a second full read. Nothing was mounted and nothing was written.
The volume was FAT32 with long filenames, which is its own small complication: FAT32 stores the short 8.3 name in the directory entry and spreads the long name across additional entries chained ahead of it. Damage anywhere in that chain and the file survives with a truncated name, so the directory structure was rebuilt from the image with the chains reassembled rather than accepted as read.
The records were in a word processor format long out of production. Recovering a file nobody can open is half a result, so alongside the original files each document was confirmed to open in a modern converter, and a converted copy supplied beside the original.
That is worth doing on any archival recovery. The client’s problem is not really that the disk stopped working; it is that twenty years of work is inaccessible, and a folder of files in a dead format is still inaccessible.
Complete — roughly twenty years of transcription work, returned on current media in both original and converted form.
Worth saying plainly, because people assume otherwise: an old drive is not automatically a failed drive, and the fact that nothing in your office can connect to it says nothing whatsoever about the state of the platters. What it does mean is that the drive should be read once, properly, rather than experimented on with whatever adapter arrives next. Single drives are from £300 +VAT after a free 48-hour diagnostic.
Post it or bring it in. Either way the diagnostic costs nothing and commits you to nothing, and you get one figure in writing before a single screw is turned.
Nothing can begin until the device is on the bench, which makes this the only step that needs anything from you. Pack it properly, put your details in with it, and send it over. What follows is a free diagnostic and a written figure, in that order.
Posting it? Use something tracked and insured — whatever is on the drive is worth considerably more than the postage. Bringing it in? Weekdays, 9am to 5:30pm, and it still wants packing as above for the journey.
Tell us what the device is, what it is doing, and anything already tried — an engineer will come back with a realistic view of the odds and a price band, before you commit to posting anything.
We’ll get back to you soon. Anything pressing, call 0161 871 0788.
Usually, and often in full. The drives themselves are frequently fine — the obstacle is reaching them. Modern USB-to-IDE bridges assume addressing and identification behaviour that predates them, so they report the wrong capacity, truncate the disk or refuse to enumerate it, and the owner concludes it has died.
Almost certainly not. A period-appropriate controller handles these correctly where a modern bridge will not, and the jumper matters too — master, slave or cable select, set wrong, means the drive simply never appears. None of that says anything about the platters.
Genuinely, yes, and it is why these get one planned read rather than several experiments. Bearing lubricant migrates over years of stillness, heads parked on the surface can bond to it, and board components age. A drive that spins up perfectly can develop bearing noise within hours of running.
From £300 plus VAT for a single drive after a free 48-hour diagnostic, with most jobs no fix, no fee. Donor scarcity on unusual legacy models can affect timescales rather than feasibility.
Kick off with an instant online quote, or ring us and talk it through first. Either way you’ll know a clear, fixed price before any work starts.