The NVMe stick is the fastest consumer storage ever produced, but it dies suddenly: no sound, no warning, just an error message at boot-up where the stick once was. M.2 and PCIe modules from every manufacturer land on this bench from York's offices, gaming PCs and the Science Park alike.
A diagnosis for every NVMe job is completed at no cost. The quote comes back to you fixed, in writing, before we lift a single screwdriver.
In most cases you only pay if we recover your data. Chip level, electronic and mechanical repair is charged at 50% at start-up, and the other 50% only upon successful recovery of your data; all BitLocker, ransomware, forensic and CCTV/DVR recoveries are charged at start-up when you accept our fixed quote. More information can be found on the data recovery cost page.
The first task with any NVMe device is matching the symptoms to the fault. After 20-odd years, these 25 cases account for nearly all of the devices that come into our lab.
Controller or firmware trouble dropped it off the bus overnight. Your data sits intact behind the silence, waiting for better handling.
A bad controller, a failed power stage or corrupted boot code are the common causes of death, while dead NAND is by far the least likely. The odds favour recovery; the module just can't report it.
When the translator fails, the drive answers enumeration with numbers that are off any known spec sheet. Service-mode repair, clearly signposted.
The system module was interrupted in the middle of a sentence. Released from the machine, it gives up its files directly on the rig.
Bare modules throttle, then completely disappear under heavy load, and finally jumble their table data after repeated hot spells. Images are taken with the modules cooled.
Same result from either Windows or the vendor tool: a drive which can no longer tell us its own name. Reflashed here, or read around entirely.
Flash doesn't keep a logbook and never sends a postcard. Immaculate service right up to the moment of failure is the standard biography.
Thin circuit boards lose out to upgrade levers and thin envelopes. Small breaks get track repair, and the rest is done by removing the chips and reading them directly.
Before any type of read is attempted, surge damage to the power circuitry is repaired and rebuilt at board level. Sequence is everything.
Background erasing of freed blocks begins at deletion and stops only without power. Pull the plug; stop the clock.
WD's flagship shipped with firmware that would stall and vanish under certain loads until a patch was applied. The unpatched population keeps our booking sheet interesting.
The write appetite of the 980 / 990 generation spent the life of its NAND early until updates were applied. Examples that have never been updated still arrive exhausted.
Kingston's SNV2S ships with whatever controller and flash were in the dock that month. The controller and flash inside determine where we go from here.
The Steam Deck and other handheld modules bake inside enclosures that have no room to breathe. Tiny format, entirely traditional recovery process.
DRAM-less Blacks use borrowed host memory for their maps of locations on the drive, and one bad interaction can corrupt what's borrowed. Entering service mode puts the ledger back in order.
Crucial's Gen4 throttles hard when run bare and ages quickly. At first there will be dropouts. Later there will be nothing at all, and the imaging window is the dropout phase.
Sustained shader compiles and cache thrashing age a gaming module approximately one year per season. Stuttering is the warning sign; a complete absence is the final deadline.
Host Memory Buffer defects in various budget designs corrupt the mapping after sleep or a crash. It's a documented failure class with a documented way out.
Budget drives hammer their fast cache region until it dies, leading to stalling or bricking. In most cases, the TLC memory behind it will continue to answer.
Over-torqued heatsink screws and mangled thermal pads flex the board and crack joints. These are reworked under the microscope, then imaged.
After a BIOS update, or on a marginal riser, the drive and slot stop agreeing terms and the module "disappears". This is proven out on known-good lab sockets.
If one NAND die fails, the drive can shrink, throw errors on its stripes or lock into read-only mode. We image around the dead die and have its neighbours carry your data out.
Power lost during garbage collection leaves the map half rewritten: the drive mounts, loops, sulks. Rebuilding it in service mode restores order.
Counterfeit "Pro" modules are recovered for what they really are, not what the label advertises. The label, for the record, fools nobody at this end.
Cracked socket latches and dropped PCIe lanes leave the module limping at x1, then failing entirely. It is imaged on an intact lab socket.
An M.2 module rides the PCIe bus straight into the CPU, glorious while it works, abrupt when it doesn't, because there's no mechanical stage to limp along on first. When the controller or its firmware stumbles, the drive simply stops answering, but your data is still stored within the silicon. Manufacturer service modes allow entry where the controller permits it, and direct chip access with a rebuilt translation map where it doesn't. No amount of additional rebooting at home will help either route.
Modules at 2280, 2242 and 2230, add-in cards, Gen3 to Gen5, from WD, Samsung, Crucial, Kingston, Sabrent, Kioxia and Solidigm: all workable as long as the module unplugs so it can be posted. One notable exception is soldered-down NVMe, which we always decline. Heat gets its own mention: if a drive runs hard without a heatsink, it will throttle, drop off the bus and, over repeated episodes, scramble its own mapping. We see it every summer from small-form-factor builds. Whatever started yours, the free assessment says what's recoverable before money comes into it.
Work on NVMe controllers, firmware and chips at PCIe speed. The rigs are specced fast enough that the workshop never waits:
Silent modules are raised in the manufacturer's own service mode (firmware corrected, translator recreated) so the module can be cloned without ever gambling the NAND.
PCIe devices that stall or wander off are captured under hardware-enforced timeouts, with retry behaviour no software tool possesses.
From 2230 stubs up to 2280s, add-in cards and enterprise U.2 modules, all use native connections, with active cooling for modules that "faint" when they get warm.
Where service mode can’t be used, packages are removed from the board one by one and read one at a time. Longer road, same destination.
Interleave, XOR and ECC are solved from the dump files until the mapping stands again, so that your volume can stand on it once more.
Burnt power stages are rebuilt component by component, and cracked boards are reworked under the microscope before any imaging is attempted.
The reputations of these drives were gained the hard way. Early SN850X firmware would often struggle under load until WD's fix was released. Never-updated 980-generation Samsungs still arrive with their endurance spent. The NV2 has no consistent configuration, so what's inside it decides our method. The P5 Plus overheats anywhere a heatsink was treated as optional. From 2230 to 2280, add-in cards, U.2, Gen3 up to Gen5: all are workable as long as the module unplugs for posting. Soldered-down NVMe is the single refusal.
Get a quote first, by calling us on 0800 689 0668 or using our free online diagnostic tool. After receiving your quote, tracked and insured post is how nearly every job reaches us: it is the fastest and safest way to get your drive to our lab in Leeds. There's no collection service, so the parcel is yours to send, or you can hand it in at the Leeds lab's reception.
Is the hard drive or SSD still inside a machine, such as a laptop, MacBook, iMac, CCTV/DVR system or server? Please remove the hard drive or SSD from the machine first and then post just the bare drive by itself; we do not provide a service for removing drives from devices. For a server or RAID, send every disk, each clearly marked with its bay. For a NAS, send its disks labelled by bay if you can remove them, otherwise send the NAS itself. We cannot work on storage soldered to the motherboard (e.g. Apple Silicon Macs, T2 Macs and certain slim laptops): if it doesn't come out, it can't come in. If you are unsure whether your storage is removable, call us before sending.
↓ Print the booking-in & shipping form (PDF)
Once you have your quote, send your package addressed to Leeds Data Recovery (about a 40 minute drive from York on the A64 & A1(M) or next working day if you want to send it using trackable post). We will contact you as soon as we have taken delivery of your package and put it in line to be worked upon.
Can't figure out what to pack? Ring 0800 689 0668 first, or try our free online diagnostic tool.
Free diagnosis, fixed written quote, and on most work you only pay if we recover your data. Start online or call our freephone.