A Samsung EVO went from working to invisible overnight, taking a student's nearly finished dissertation with it. When an SSD's controller fails, the flash inside is usually fine — nothing can simply reach it. We accessed the drive at a lower level and rebuilt the map that turns flash back into files.
The laptop was shut down on a Friday with a term's work on its Samsung EVO; by Monday the drive wasn't detected at all. On it were the student's dissertation, close to submission, and a folder of research, with no backup. An SSD that dies suddenly like this has almost always suffered a controller failure rather than lost its data, so the drive was set aside rather than being repeatedly plugged in, which achieves nothing on a dead controller.
An SSD stores data in NAND flash, managed by a controller that maps the file system's logical view onto the physical chips and constantly moves data around for wear. When that controller hangs or dies, the flash underneath is typically intact but unreachable through the normal interface — the drive appears dead. It's worth noting that this is a failure, not a deletion: on a healthy SSD, deleting files triggers TRIM, which really does erase them and makes deleted-file recovery unlikely. Here nothing was deleted; the data was all present, simply locked behind a failed controller.
The drive was brought up in the controller's manufacturer technical mode — a diagnostic state that allows the flash to be reached even when the drive won't present normally. Where a controller is too far gone for that, the NAND can be read directly instead. Either way, what comes back is raw flash: scrambled, error-coded and physically out of order, not yet files.
Making files out of raw flash meant reconstructing the translator — the lookup table that maps logical blocks to physical flash pages — along with removing the controller's scrambling and applying the correct error-correction to fix read-time bit errors. With the translator rebuilt, the logical drive fell back into order, the file system reappeared, and the dissertation, its references and the research folder were extracted.
The dissertation was opened and checked in full, and files across the research folder confirmed to be whole, before the recovered data was returned — with days to spare before the deadline. About 97% came back. We made the familiar point to a relieved student: an SSD gives no warning when it dies, so a single copy of something as important as a dissertation is a risk that a free cloud folder or a second drive removes entirely.
Controller technical-mode access and direct NAND reading · de-scrambling, ECC correction and translator (FTL) reconstruction · file-system rebuild. All work carried out in-house in Belfast.
Send it to us for a free, no-obligation diagnostic. We’ll tell you what can be recovered and put a fixed price in writing before any work starts — and on most jobs, if we can’t get your data back, there’s nothing to pay. Post your device in, or drop it to us by appointment.
Often, yes. Sudden-death SSDs have usually suffered a controller failure while the flash — and your data — stays intact. We reach the flash at a lower level and rebuild the mapping to recover the files. A free diagnostic will confirm what's possible.
Usually not, because of TRIM: on a working SSD, deleting a file tells the drive to erase it, so it's genuinely gone. Failures are different — nothing was deleted, so the data is still there to recover.
Typically several days, depending on the controller and capacity. You'll get a written quote after the free diagnostic, with no fix, no fee on most jobs.