An external drive that has stopped has four possible culprits: the lead, the port or supply, the small board translating USB, and the disk itself. Three of those cost nothing to rule out, and hardly anybody gets round to it. This page reads the symptoms off the screen instead, explains what RAW and unallocated actually mean, says exactly where to stop, and sets out what the single-drive band covers.
An external is one disk: £300 + VAT, whichever of the four suspects turns out to be guilty. Send the unit whole and sealed in its own case, because the board inside is half of what is being diagnosed. Two disks or more in one shell is the array band, from £500 + VAT.
Four suspects hide inside any external drive problem: the lead, the port or power feeding it, the small board translating USB, and the disk itself. What the machine puts on screen eliminates most of them in about a minute, which is the fastest free progress available anywhere on this site.
The drive appears in the list, the capacity looks right, and no letter has been given to it. Best position on this page. The hardware is answering perfectly and the fault is filing rather than physical, so nothing needs opening and the odds are excellent — provided nobody now accepts an offer to initialise it.
The same good news underneath. RAW means the identity of the volume cannot be read; unallocated means the map of partitions cannot; and in both cases the contents normally sit exactly where they were left. What must not happen is a format, a repair or an initialise, since each of those lays a fresh structure over the damaged one.
You can feel it running and no operating system will admit it exists, on any machine you try. That pattern points at the translator board inside the case rather than the disk behind it, which is the cheaper of the two outcomes. Try one other computer, ideally a different operating system, and then stop trying things.
No light, no spin, no faint vibration under the fingers. Frequently the case rather than the drive: a dead board, a worn socket, a failing lead, or a mains adapter borrowed from something else in the house. An inexpensive fault in an expensive disguise, and the free examination separates the two without you paying to find out.
Weak areas on the disk, a bridge board running hot, or a bus-powered unit on a port that cannot feed it properly. Starting the copy again is the worst option available, since it sends a struggling drive back over the same difficult ground repeatedly. Stop copying and have it read in stages by equipment built to expect the drop-outs.
The volume mounts, the size is correct, and the contents are missing or showing as shortcuts. A damaged directory structure on working hardware, usually left behind by an interrupted copy, an unsafe removal or malware. The files are almost always still on the platters, unlisted rather than gone anywhere.
Every portable and desktop external is a standard disk with a small board between it and the socket, and that board draws the power, translates the protocol and on some ranges does considerably more besides. So a unit that has stopped might have a dead disk, a dead board, a bad lead, or a port that cannot supply what it needs. All four look identical to somebody standing in front of the machine, and three of them cost nothing at all to eliminate.
Start with the lead, which fails more often than anything else on the list and gets blamed less than anything else on it. Sockets wear, cables fray inside the moulding where nothing shows, and a lead that charges a phone perfectly well may not carry data reliably. Swap it for one you know works. Then change the port, moving to a socket on the back of a desktop rather than a hub or a front panel, and after that change the computer entirely. A drive invisible to Windows sometimes appears without complaint in Disk Utility on a Mac, and the fact that it does is itself the answer: the hardware is alive.
The small bus-powered units take everything they need through the socket, so a port that cannot quite deliver produces symptoms indistinguishable from a dying disk: spinning up and down, dropping out during a large copy, appearing and disappearing between reboots. A rear socket or a powered hub is among the very few free tests genuinely worth running, and it settles a good share of these before anybody has spent anything.
The larger desktop units carry their own mains adapter, and the adapter is usually where they go wrong. Borrowing one from a different device is an efficient method of putting a voltage through a board that was never expecting it, taking out protection components and occasionally rather more. If a unit died at the moment somebody plugged in a spare supply, say so, because it points at the electronics rather than the mechanics and changes what gets looked at first.
Many large consumer drives overlap their recording tracks to fit more onto the same platters. That arrangement reads perfectly well and writes acceptably in small doses, and it becomes extremely slow the moment the drive is asked to sustain heavy work. Add a handful of weak sectors and the familiar picture appears: a copy that begins at a sensible speed, collapses to a crawl, and finally stalls or disconnects. Starting it again is the natural response and the wrong one, because each attempt spends more of a failing disk remaining hours on ground already covered.
Stop at the moment anything wants to write. Do not initialise a disk when Windows offers. Do not create a partition, assign a volume or run a repair tool. Do not let recovery software save its results onto the drive it is reading, and do not install it onto the machine whose disk you are trying to save. And if the unit is making a mechanical noise — clicking, beeping, a rhythmic scraping — none of the above applies, because the correct action is the power switch. Every further minute of that noise is surface being lost for nothing.
This page is about establishing which part stopped. What happens once the answer turns out to be the disk itself — heads, motors, boards, donor matching — belongs to external hard drive repair, and the quirks of the two biggest ranges are on the Western Digital and Seagate pages.
An external is one disk, so it is £300 + VAT — the same as an internal drive, and the same whether the culprit turns out to be the disk, the translator board or the file system. Nothing is added for capacity and nothing is added for a dead bridge. A desktop unit holding two disks or more is a different animal and sits in the array band from £500 + VAT, climbing with the member count.
Briefly, the rest: cards and sticks £250 + VAT; recorder disks and encrypted volumes with the key available £400 + VAT; forensic casework written up £800 + VAT, or £400 + VAT for the verified image with deletions extracted and no report. The cost page sets all five out together.
Looking is free and closes on the second working day after the unit is logged in, with the figure fixed in writing at that point. Logical faults — a RAW volume, a lost partition, deleted folders, a failed bridge with a sound disk behind it — carry no fix, no fee. Chip-level work, DVR jobs, forensic jobs and mechanical or electronic failure sit outside that, and invasive work takes half the agreed figure upfront.
An external drive is one of only two things that travel complete rather than stripped down, the other being a network box. Leave the case shut. The board inside is half of what is being diagnosed, and taking it apart on a kitchen table removes the evidence about which of the two components stopped — quite apart from what a screwdriver tends to do to a disk that has already been dropped once.
Pad it properly, since these units are fragile before they set out and the platters inside a dropped drive do not want a second impact in a sorting office. Leave the mains adapter and the lead at home unless the unit takes an unusual supply, in which case mention it in the note. Add a line about what happened — dropped, unplugged mid-copy, stopped being seen, started clicking — because the history genuinely changes the order of work at the far end. Oxford takes drop-offs on weekdays from 9:00am to 5:30pm, about fifty-five miles from Coventry down the M40, and there is no Coventry counter and no collection round.
Recovered data comes home on separate media with a file list to check against what you were expecting, so any gaps are spotted at the start rather than discovered in six months. Nothing is ever written back onto the original, which is both a principle and the reason a second pass stays available if something turns out to be missing.
The failed drive comes back too if you want it, though it should be treated as finished rather than reused. A disk that has needed a laboratory once has already told you what it thinks of your data. The uncomfortable question is worth asking afterwards: why was this the only copy. Externals end up in that role by accident rather than decision in most households and most small offices around Kenilworth, Solihull and Stratford, and the cheapest hour anybody spends is the one that puts a second copy somewhere else entirely.
Three of the four suspects cost nothing to rule out. The fourth is the only one worth a laboratory, and the examination that identifies it is free.
Something rigid, some padding and a tracked label treat an ailing device far better than a week of being carried about in a bag while its owner decides what to do next. Handed in at a Coventry counter during the afternoon, it generally reaches the Oxford bench the following working morning, which beats what most people manage by clearing a day for the M40.
As a rule the storage comes out and the machine stays where it is. That applies to a laptop, a tower, an iMac and to the recorder sitting under a counter. Taking equipment apart is not something this bench does, and a repair shop will free a drive in a few minutes. Two things go the other way: an external drive stays sealed inside its own case, and a NAS travels as a complete unit with its disks still in their bays. A Fusion Mac is a third case — both of its drives come out and travel together, each one labelled. The single situation nobody can work around is memory soldered flat onto a mainboard, which is how Apple Silicon Macs and a good many slim laptops are built: if the storage will not unbolt, there is no parcel to send.
↓ Print the shipping & booking-in form (PDF)
Put Oxford Data Recovery on the label. From Coventry it is roughly fifty-five miles straight down the M40, about an hour if you would rather drive it in than post it. Either way you are told the moment it is logged, and the free diagnostic finishes two working days later.
Unsure what ought to go in the box? Ring 0800 689 0668 before you seal it, or work through the free online diagnostic and let it do the asking.