Home / Case Studies / Solid State & Flash
Solid State & Flash · case file

The SSD Times Out on Every Read

He tested methodically and described the fault cleanly. His 2TB Samsung 870 EVO "stopped being recognised by BIOS"; Windows "detects the drive but can't read it and all drive related actions time out causing explorer to become laggy". He moved the SSD to a USB enclosure to rule out the PC, and "behaviour is the same". An SSD that is detected but times out on every read, identically across machines, has a controller or firmware fault — the drive is answering enough to be seen but can no longer reliably reach its own memory, which is a laboratory-level SSD recovery rather than anything a different port can fix.

SSD / NVMeFirmware / SMARTNot Detected
// case at a glance
MediaSamsung 870 EVO 2TB SATA SSD — dropped from BIOS detection, detected by Windows but unreadable with all actions timing out; identical behaviour in a USB enclosure.
Reported situationSSD no longer recognised by BIOS · Windows detecting the drive but unable to read it · all drive actions timing out, making Explorer laggy · moved to a USB enclosure, behaviour unchanged · data recovery required.
Fault classSSD controller or firmware fault — the drive part-responding but unable to reliably access its NAND, causing read timeouts; the memory likely intact behind the failing controller.
Equipment usedPower and thermal behaviour profiled on a current-monitored bench supply before extended host attempts · the controller addressed on PC-3000 SSD in technological mode, the drive's firmware and translator structures assessed and rebuilt · safe-mode reads established where the controller part-responds, with tight timeouts · chip-off reading of the NAND where the controller is beyond revival, subject to Samsung's internal encryption · full image at first stable access, hash-verified.
// the decode

The decode

Detected-but-times-out is a precise symptom, and it names the controller. That BIOS lost the drive while Windows still detects it, yet every read times out, describes a controller that is partly alive — enough to announce itself intermittently — but unable to reliably serve data from the NAND behind it. The timeouts are the drive trying and failing to reach its own memory. This is a controller or firmware fault, the most common way SSDs fail, and distinct from worn-out memory.

His cross-machine test already ruled out the computer, which is the right diagnostic. Moving the SSD to a USB enclosure and seeing identical behaviour eliminates the motherboard, the SATA port, and the PC's configuration as causes. When a drive misbehaves the same way everywhere, the fault is in the drive — specifically, given the timeouts, in the part that manages access to the memory. That test saves a step and points the work in the right direction.

The laggy Explorer is a consequence, not a separate problem. Windows becoming laggy when the drive is connected is the operating system waiting on read requests that never complete — each timeout stalls the system briefly. It confirms the drive is present enough to be queried but never answers, which is exactly the controller-fault picture. It is a symptom of the timeouts, not an additional fault.

The memory is very likely intact, which frames the outlook. A controller or firmware fault threatens access to the NAND, not the data stored in it. If the fault is the controller, as the symptoms strongly indicate, his data sits intact in memory that the controller can no longer reliably present — and recovery turns on restoring that access through the controller's service modes on professional equipment, or reading the memory directly where the controller cannot be revived.

Continued attempts risk the firmware state, so controlled access matters. Repeated power cycles and read attempts on a part-failed controller can further disturb the firmware structures that map the memory, and timeouts under load stress the drive. Access is re-established under laboratory control with tight timeouts rather than by the host's generous retries, spending the drive's stability where it counts.

The honest fork is controller versus memory, with a Samsung-specific note. If the controller or firmware failed and the NAND is intact — the common, hopeful case — the data comes back. Samsung SSDs commonly encrypt the NAND internally, which makes reviving the controller the preferred route, since reading the memory around a dead controller yields ciphertext. Which case applies is established at assessment, and the prognosis stated before any charge.

// on the bench

On the bench

Power and thermal behaviour were profiled on a current-monitored bench supply before extended host attempts. The controller was addressed on PC-3000 SSD in technological mode, the drive's firmware and translator structures assessed and rebuilt, and safe-mode reads established with tight timeouts where the controller part-responded. Where the controller was beyond revival, the NAND was read at chip level subject to Samsung's internal encryption. A full image was taken at first stable access and hash-verified.

// the outcome

The outcome

Controller addressed in technological mode, firmware rebuilt, and a full image taken at first stable access. The diagnostic costs nothing and completes within two working days of arrival, and the quote is one fixed written figure with VAT already in it; if the data can't be brought back, no charge is made. The decode: detected but timing out, identically everywhere, is a controller losing its grip on the memory — not a dead PC and not worn-out NAND. Your data is very likely intact behind the failing controller, and reviving that access is what reaches it.

An SSD that's detected but times out on every read

Stop power-cycling it and running read attempts — a part-failed controller can have its firmware state disturbed further by repeated tries, and software can't push past timeouts that never complete. Your cross-enclosure test was exactly right: identical behaviour everywhere means the fault is the drive, not the PC. This is a controller or firmware problem with the memory intact behind it, needing laboratory SSD tools. Keep it cool and unpowered until it can be assessed.

Sending this in from Swansea? Every case opens with the free diagnostic — completed within 2 working days of your media arriving — and one fixed written quote before any work: logical jobs run no fix, no fee; invasive drive-opening work takes 50% of the quote upfront; forensic-classed work is payable upfront in full. If the data is inside a laptop, PC, Mac or server, remove the hard drive or SSD and send us just the drive; we don’t provide an internal drive-removal service, and we don’t recover storage soldered to a motherboard (e.g. Apple Silicon Macs) — only drives that can be removed and sent to us. Post or courier tracked and insured to Bristol Data Recovery, Castlemead, Lower Castle Street, Bristol, BS1 3AG — full sending instructions and the shipping form are here.
Start a free diagnostic

Our case files are written up from genuine enquiries our lab has handled for customers across Swansea and South Wales, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery approach our engineers apply to that fault, using the equipment listed.

// related case files

More cases like this one

Browse all case studies →

Got a device with a story like this?

Free diagnostic, fixed quote, no fix no fee — start now or call the freephone.