An SSD that appears in the firmware setup with a strange model string and a capacity of zero is not empty. The controller has started in a service mode after failing to load its normal firmware, and it reports a placeholder identity instead of the real one. The NAND still holds the data; what is missing is the table that maps logical addresses to it.
A solid state drive keeps a translation table in NAND and loads it at power on. If that table cannot be read, or the firmware image fails a checksum, the controller falls back to a minimal mode and identifies itself with a firmware or service string rather than the product name.
The most common symptom is a model name such as SATAFIRM followed by a revision, with a capacity of zero bytes. The drive may also show up as an unknown device that cannot be initialised, or may disappear entirely after a few seconds.
This is a controller level failure rather than a worn out NAND. The chips usually still respond, which is why recovery labs can read them directly when the controller will not cooperate.
Enter the firmware setup and read the storage page. A drive listed with a service string and no capacity needs a different approach from a drive that is simply missing from the list.
Open Disk Management and check whether the disk appears as unknown or not initialised. Windows may offer to initialise it; decline that prompt, because initialising writes a new partition table over the area a recovery tool needs.
Note whether the drive appears for a few seconds and then vanishes. A disappearing drive suggests a power or controller fault that a firmware level repair will not fix.
A vendor firmware update run on a drive in this state can rebuild the translation table with empty entries. The drive becomes usable and the previous content becomes unreachable by any software tool.
Initialising the disk in Windows has the same effect at a smaller scale: it writes a new partition table and can trigger a trim pass that clears the pages where the old data lived.
If the drive still reports a capacity for a short window, use that window to copy data rather than to run a repair. Once the content is safe, a reflash is a reasonable way to return the drive to service.
Qiling Deep Data Recovery can scan the drive while it still exposes a readable capacity, even when the partition table is gone. It works at the level below the file system, so a volume that Windows shows as unallocated can still yield files.
Use Format Scan when the disk shows as unallocated or uninitialised, and Fast Scan when it still has a readable volume with missing files.
Pick the disk by capacity and position in the list. Do not select the disk you intend to save files to, and confirm the serial number when two identical drives are fitted.
Let the deep scan finish even when it is slow. Sort the results by type and preview files, because a scan of a damaged file system often returns more entries than the user expects.
Save the results to another drive and stop using the SSD. Keep the copy until you have opened a sample of each file type and confirmed that the content is complete.
Usually yes. The NAND stores the data and the translation table together; a controller that cannot load the table does not erase the pages.
It often makes the drive usable again, but it can also clear the mapping that points at the old data. Recover first, reflash second.
Because the controller is running a minimal service image and reports that string in place of the product name. It is a symptom, not the real hardware identity.
When the drive no longer enumerates, or when it reports a capacity but returns nothing after a scan. Both point to work below the level that software on Windows can reach.
For a drive that vanished after a failed firmware update, see SSD not detected after a firmware update.