Taking work now — the first look is freeSSDs posted in from anywhere in the UK, or handed in at ten drop-off pointsQuicker still, give us a ring:0800 6890668
SDRSSD Data Recovery 0800 6890668 Price my job
SDR / Every kind of SSD / NVMe M.2

M.2 · NVMe · PCIe 3.0, 4.0 and 5.0

NVMe SSD data recovery. The drive that goes without warning, and usually takes nothing with it.

An NVMe SSD talks to the processor directly over PCIe lanes, which is why it is fast and why, when it fails, it simply stops being there. There is nothing to listen for. You shut the machine down as usual; next morning the BIOS has no drive on its list, or lists it at 2MB, or Windows wants to initialise it. Most of the time the flash is fine and the controller or its firmware has fallen over. The work is persuading that controller to hand its map back without erasing anything on the way. The first look costs nothing. A single NVMe drive is £300 + VAT, written down before any chargeable work begins, and it normally takes 3–5 days at the bench after the free look. On most jobs, if nothing comes back, there is nothing to pay.

Free first lookOne fixed figure in writingNo data, no bill on most jobsReturn postage paid

Rather talk it through? An engineer answers the bench line
0800 6890668

NVMe symptoms, and how long each gives you.

Not listed? Describe it on the form →
It is not in the BIOS or in Disk Management at allThe controller did not finish starting. The flash behind it is almost always sound.Do not delay
It shows as 2MB, 0GB, or a controller's name instead of the modelFirmware fallen back to ROM: the drive is up, its map is not.Do not delay
It appears, then vanishes during a large copyA controller crashing or overheating under sustained writes. Each return costs something.Do not delay
Read-only, or writes silently failMany controllers lock themselves read-only when their spare blocks run out. Usually fully readable.Everyday job
A file was deleted, or the drive was formattedTRIM is working through those blocks by the hour. Cut the power to it now.Do not delay
The laptop is dead and the drive probably is notTake the stick out, or have a repairer do it, and post the stick alone.Everyday job
Packing it and posting it: a bare SSD travels in an anti-static bag inside a small padded box, with nothing able to move. Insure it for what the files are worth rather than the price of the drive, and use a tracked service. The posting address is not printed anywhere on this site; it arrives by email in reply to the form, with a booking sheet to print; the sheet inside the parcel is what matches it to your enquiry when it is opened. Or hand the sealed parcel in at the nearest of ten drop-off points, your name on the outside and the sheet inside; say where you are on the form and it comes by email. We pay the postage home either way. The whole of it is written up on the guide to packing and posting.

Brand by brand, and what tends to go wrong.

Samsung 970, 980 and 990In-house controllers on Samsung V-NAND. The 980 Pro had a documented firmware fault that left drives read-only; the 990 Pro a health-reporting fault. Both fixed by firmware.
WD_BLACK and WD Blue SN-seriesMostly WD's own controllers on BiCS flash. The SN550's flash was changed mid-life in 2021 without a change of name.
Crucial P3, P5 Plus, T500 and T700Phison on the P3 and T-series, Micron's own on the P5 Plus. The T700 shut down without a heatsink until firmware changed the throttle curve.
Phison E-series under many badgesCorsair MP-series, Sabrent Rocket, Seagate FireCuda, Kingston KC3000, Gigabyte Aorus, MSI Spatium: one controller family under a dozen badges, and one set of failures.

What the message actually means.

Describe yours to us →
What is going on The usual reason for it Where that leaves you
No bootable device, and nothing in the BIOSThe controller is not enumerating on PCIeFirmware-level work; the flash is usually intact
Capacity shows 2MB, 0GB or 1GBFirmware running from ROM, the translation table not loadedDiagnostic-mode access and a map rebuild
Vanishes mid-copy, back after a rebootA controller crash or a thermal shutdownStop using it; image it before it gets worse
The disk is write-protectedThe drive has locked itself read-onlyOften fully readable on the bench
Health percentage falling fast in the maker's toolReal wear, or a firmware reporting faultBack up now; a warning, not yet a recovery
Boot device not found, straight after a firmware updateA firmware write that stopped half-wayKeep it switched off; bench work

From the box arriving to your files going back.

Work we have closed →
01

Logged the day it lands, and the first look costs nothing Free

A case number goes on it the day the parcel is opened, and an engineer settles what has actually failed before anything else happens. The drive is read on our own equipment, never in a computer that might send it a TRIM command. Back to you come two things together: a straight note of what is liftable and what is not, plus one figure, fixed and written down. Accept it, or decline and owe us nothing.

Nothing to pay for lookingA single figure, put in writingNo TRIM, no writes
02

Powered on our terms

An NVMe drive that works for thirty seconds and then vanishes should not be tried again, and it is not tried here either. It is brought up over PCIe in the controller's diagnostic state, where the bench can talk to the flash directly and the drive's own housekeeping stays switched off.

Vendor diagnostic modeNo hopeful reboots
03

Past the fault, to the map

Working code is loaded into the controller's memory rather than onto the flash, the service area is read, and the translation table is rebuilt from what is there. Where the controller itself is dead, the chips are read one by one and the controller's scrambling and error correction are undone in software.

Loader in RAM, never on the flashMapping tables put back together
04

A full image, then the volume

Nothing is worked on live. The whole drive is imaged once, weak pages retried in their own passes, and the file system is put back together on the image. Encrypted volumes are unlocked there too, with the key you supply, and never on the original.

One careful imageThe volume verified end to end
05

You see the file list before you pay

What was recovered is listed for you first, and only then does a bill exist. Approve the list and it is invoiced; turn it down and it is not — and where nothing has come back, most jobs carry no charge at all. Recovered data travels home on fresh media bought in for your job, with the postage at our end. Your case is not closed until you have opened the files on a machine of your own.

No charge until you accept the figureFresh media, supplied with the job3–5 days at the bench

What arrives most often

  • Do not put a failing NVMe stick in a USB enclosure. The bridge chip in the enclosure has its own ideas about TRIM and power states, and hides the error messages we need. If you want to know whether the BIOS sees it, try it once in an M.2 slot, briefly, and then stop.
  • DRAM-less drives borrow your computer's memory. Budget NVMe drives hold their map in a slice of system RAM (the host memory buffer) rather than on a chip of their own. They can behave in a laptop and misbehave in an adapter, and the fault is real in both.
  • Every controller family lays its flash out differently. Page order, scrambling and error correction have to be worked out family by family before a raw dump means anything. Lifting the chips takes an hour. Reading what comes off them is the part that takes the day.
  • Deleted on flash really does mean deleted. Windows, macOS and Linux all tell an NVMe drive which blocks no longer matter, and the drive erases them in the background within hours. On an SSD you have the day it happened, not the month.

Before you accept a firmware update: flashing rewrites where your data physically sits, and should power drop half-way there is no earlier copy beneath it. The great majority go through without trouble, so take this as no reason to skip updating — just copy your files off beforehand. The one drive that should never be put through it is one already misbehaving.

One job, followed all the way through.

UK · SDR-2026-0417JOB LOGGED ✓

A 1TB WD_BLACK SN770 that went missing after a Windows update

A Lenovo laptop stopped at a black screen after an update, and the BIOS no longer listed a drive. The stick came out and came here on its own. In the controller's diagnostic state it answered with its map intact but two firmware modules damaged; a temporary loader brought it up read-only and a full image took eleven hours. A decade of a photographer's edits opened first time. Three video files were partly damaged, and were reported as such.

99.6% of files recovered4 days here, and back by post
Illustrative example — replace with a genuine case

What helps, and what harms.

Do this much first

  • A full shutdown, please: neither sleep nor hibernate
  • Write the full model and part number down
  • Deleted something? Send the form the same day
  • The stick travels on its own — no caddy, no enclosure

What sets us back

  • Rebooting over and over hoping it turns up
  • Pushing new firmware at a drive already misbehaving
  • Letting Windows initialise it when it asks
  • Running recovery software on a drive that keeps dropping out

Questions answered before you commit.

Can data be recovered from a dead NVMe SSD?

Usually, yes. Most dead NVMe drives have a firmware or controller fault with intact flash behind it. The exception is a drive whose controller has physically failed on a model that encrypts in hardware, where the original controller has to be revived because the flash on its own is unreadable. The free look tells you which yours is.

Is NVMe recovery harder than SATA?

Different, not harder. The routines are controller-specific, and the newest PCIe 5.0 controllers have the least mature support. We say at the free look whether yours is one we can take, and we do not take jobs we cannot finish.

Should I try it in another PC first?

Once, briefly, in an M.2 slot, to see whether the BIOS lists it. If it is missing or shows the wrong size, stop. Every further attempt is another chance for the drive to run its own housekeeping.

Does my BitLocker key matter?

Yes. If the drive was encrypted, the 48-digit recovery key is needed to read the files once the drive is imaged. It is usually in the Microsoft account that set the machine up, or with your employer. Find it before you post.

How long does it take?

3–5 days at the bench after the free look, for one drive. Priority handling is quoted separately when you ask for it.

Begin here if yours is doing the same thing.

M.2 SATA, the one with two notches →Not detected at all →Shows 0GB or the wrong size →Samsung →Western Digital →Crucial →

Nothing gets worse while the power is off.

Looking at it is free. Back comes a list of what opened and what did not, together with a single price to finish, set down in writing while you are still free to say no. On most jobs an invoice only follows the data. Until that list reaches you, leave the drive switched off.

0800 6890668

Price my jobPacking and posting