Most failed drives arrive here with the same story: it clicked a few times, it vanished from the computer, somebody tried it once more to be sure. We've heard every version and recovered most of them. York Data Recovery runs a hard drive recovery service for households and businesses from Acomb to the Minster quarter — the diagnostic costs nothing, and the figure you approve is fixed in writing before anyone opens the drive.
Every hard drive job is diagnosed free. The quote turns up fixed, in writing, before a screwdriver is lifted.
No fix, no fee all jobs except electronic and mechanical failures, chip level work, DVR and Forensic jobs. Full pricing is on the data recovery cost page.
First job on any hard drive is putting the symptoms against the fault — after twenty-odd years, these twenty-five account for nearly everything that comes through the door.
Heads sweeping for servo marks they can no longer resolve. Head failure announcing itself — and each replay is billed in platter surface.
A seized motor, heads parked on the platter where they shouldn't be, or spiked electronics. One silence, three suspects; the rig separates them.
Power in, no name given. That's the firmware or the controller board sulking — recoverable, though never over a USB lead.
No speaker on board: that's the spindle straining at jammed heads or a locked bearing. Retries wind it tighter, so stop retrying.
Decaying surface that freezes Explorer and chokes copies. Left running, it compounds hourly — so it doesn't get left running.
The address map inside gives out; the drive holds at 'busy' or announces 0GB. Seagate's best-known trick, mended via the service terminal.
Heads and platters meet at speed; the matter is settled in milliseconds. A second power-up renegotiates nothing.
Lightning season and bargain PSUs cook drive electronics. Board-level repair with the ROM carried over is a standing service here.
Under a quick format most data survives; the deciding factor is what got written next. Aim for nothing at all.
Bookkeeping gone, library intact. Partition tables rebuild — unless somebody obliges the prompt, which is the one way to lose.
The 3TB Seagate that made head-stack failure famous still limps in monthly. Its donors have never left our racks.
Sustained writes back the shingled ST2000DM008 up until it seems dead. Mostly it's queueing. SMR-aware pacing gets the image out.
Boxed since the last house move, woken today: stiction. Separating heads from platter is patient hood work, not percussion.
MQ04ABF100 and ST1000LM048 drives ride in machines that hit floors, and the heads pay. Donor stack, filtered hood, then the clone.
Answering at walking pace means failing in instalments. The strategy is one line long: image now, worst regions last.
Relocation records pile up in the service area until a Blue reads like treacle. Cleared down the diagnostic port; the platters were fine all along.
Toshiba's P300 is the self-assembler's default and a frequent head-wear patient. Its donor slot on the rack stays warm.
Age or a static jolt kills the tiny amplifier riding the actuator, and every head goes deaf at once. Only a complete donor stack answers it.
Powering a clicking drive 'one more time' engraves arcs into the surface. What's left afterwards is imaged around the grooves.
Exos 7E8s, X16s and Ultrastar 7K4000s land here wearing five-figure hour counts. Better metallurgy postpones the reckoning, never cancels it.
He8-class and X16 units float their heads in helium; a weeping seal alters fly height and the error rate climbs. Imaged promptly — and opened by nobody outside a lab.
The fluid bearing announces retirement as a hum under load, then runout, then servo grief. Take the image while the wobble is still losing.
Bumpers and gaskets of that era outgas as they break down, varnishing heads and landing zones. A loft-archive speciality the hood deals with.
Interrupt a shingled drive reorganising its media cache and it can misplace the plot — endless 'busy', or writes that never arrived. Resolved from a complete image.
Lid off at home, thumb on the top platter, then posted here with a sheepish note. Often still winnable — though every fingerprint takes its cut.
A hard drive is a stack of polished platters spinning under heads that fly closer to the surface than a human hair is wide — precision that works for years, then stops all at once. Most people reach a hard drive data recovery service the long way round: a freeware scan first, a nervous reboot or two, then the search box. No harm done if the fault is logical; real harm if the heads are damaged, because each retry is another lap of a failing mechanism over your data. Seagate, WD, Toshiba, HGST and Samsung mechanisms are the daily bread here, laptop drives to surveillance stock, and physical hard drive recovery — head swaps under filtered air, seized motors freed, donor parts from matched stock — makes up the bulk of it. If yours clicks or has been dropped: power off, leave off. Drives don't heal.
Storm surges along the A19 and A59 corridors cook controller boards every winter; the cure is a matched donor PCB with the drive's unique ROM carried across, never a guess, followed by sector-by-sector imaging on hardware built for unstable media. Firmware that lost its own translation tables, bad sectors spreading like damp, deleted hard drive data recovery after a format that seemed a good idea at the time — all standard bench work. And while York and North Yorkshire fill most of the diary, tracked post effectively makes this hard drive recovery UK-wide: a properly boxed drive from any county is on the bench the next working day, which is why our hard drive data recovery services carry postcodes from well beyond the city walls. A hard disk drive recovery service is mostly patience, matched parts and knowing when to stop — not glamorous, but the files come back, which is the point.
Fair warning: capability is decided by tooling. Here is what every single hard drive at York actually goes through:
The firmware rig a lab either owns or pretends it doesn't need. Factory service mode opens up, translators get rebuilt, defect tables mended, locked firmware zones read out. Consumer software cannot begin any of that.
Failing mechanisms are cloned one head at a time while the current draw is watched. Timeouts per sector, retries decided in hardware. A software imager hopes for the best; these plan for the worst.
No lid comes off outside the filtered hood. Room dust on a platter shortens its life, so the two are never introduced.
Jigs, combs and lifters for shifting head stacks and complete platter packs between chassis. No surface ever touches another — a rule with no small print.
Donor racks sorted by model, firmware family and factory code. Nine times out of ten the matching head stack is already in the building rather than on a courier.
Electronics mended under the scope, with the drive's one-off ROM moved to any donor board — leave the adaptives behind and you own a very quiet paperweight.
A physical one-way valve between rig and drive. From booking-in to hand-back, nothing at our end can write a single byte to yours.
An intake log, that, not a brochure. The Exos and Ultrastar units come off small servers around Clifton Moor and Monks Cross; IronWolfs and Red Pluses climb out of household NAS boxes; the DM008 books itself in by playing dead under shingled write pressure; and the ST3000DM001's reputation still delivers us a fresh example most months. Diagnosis is free with the answer inside two working days — and yes, the legacy shelf still reads IDE drives older than our apprentice.
Blue in the family tower, Red Plus minding the NAS, Gold in the rack, WD_BLACK with the gamers — WD's colour chart sorts our week for us, and no single model gets imaged here more than the WD20EARZ. WD's slow-motion ailment is a service area silted with relocation records: detected, then glacial. Housekeeping through the maker's diagnostic mode clears the silt, after which the platters image as if nothing happened. The other steady WD trade is drops and knocks — matched donor heads, fitted beneath the hood, settle those.
No logo lands on the booking sheet more often. The shingled ST2000DM008 stages theatrical deaths under heavy writes yet is rarely dead; DM-family firmware seizures park a healthy mechanism at 'busy' or 0GB and lift via the terminal with the lid never raised; and the ST3000DM001's taste for its own heads keeps a dedicated donor rack busy a decade on. Knock-induced clicking completes the set — hood, donors, image. And the slim ST1000LM048 adds a flourish: crawling so convincingly it gets reported as deceased when it's merely exhausted.
Deskstar and Travelstar have answered to IBM, then Hitachi, then HGST under Western Digital — and every vintage still climbs out of North Yorkshire lofts and office clearances. The house failure is a head stack tiring one surface at a time: errors gather on one platter face, reads slow to a limp, and eventually ticking replaces mounting. We image surface by surface, banking what the strong heads can still reach before donors take the remainder. Ultrastar 7K4000s and helium-sealed He8s arrive off small servers wearing six-figure hours and, seal permitting, image like gentlemen.
Deskstar and Travelstar were IBM lines until the 2003 handover — and the 75GXP model earned itself a nickname the industry hasn't forgotten. IDE-era survivors keep surfacing in retired towers holding the only copy of the old accounts. Twenty years of standing still means seized spindles, stiffened lubricant and heads bonded to platters, so every such unit visits the hood for inspection before receiving a single volt. PATA adapters and period donors hold a permanent place on the legacy shelf; a vintage badge never yet condemned the data underneath.
Take the drive out of the machine before sending it over to us. Not sure how? Ring 0800 689 0668 and we'll point you right — or any local computer shop will have it out for a small charge.
Drive still bolted inside a PC, laptop or server? It must come out before anything is posted — bare drives are all this workshop accepts, and dismantling machines is work we leave at your end. Most cases involve a handful of screws, and any repair shop in York will oblige for loose change. The single hard limit, said once: storage soldered into a motherboard is beyond scope.
Nearly every job on our bench arrived by tracked, insured post — it's the quickest, safest route in. There's no collection service, so the parcel is yours to send or hand in.
Is the drive still inside a computer, laptop, MacBook, iMac, CCTV / DVR or server? Take the hard drive or SSD out first and post the bare drive on its own — removing drives from machines isn't a service we offer. Storage that's soldered to a motherboard (Apple Silicon Macs, certain slim laptops) is the one thing we can't work on: if it doesn't come out, it can't come in.
↓ Print the booking-in & shipping form (PDF)
Address the parcel for the attention of Leeds Data Recovery — about 40 minutes from York via the A64 and A1(M), or next working day by tracked post. You'll hear from us the moment it's booked onto the bench.
Not certain what to pack? Ring 0800 689 0668 first, or run the free online diagnostic.
Free diagnosis, a fixed written figure, no fix no fee on most work — start online or ring the freephone.