A B&B’s Fujitsu laptop kept shutting down mid-use, ran hot to the touch, and stopped loading files — booking records and accounts trapped on an M.2 NVMe SSD. The drive was overheating: fast NVMe SSDs run hot, and in a cramped laptop they can cook themselves into throttling, crashes and firmware instability. We stabilised it, imaged it while it was cool and cooperating, and recovered the business.
A B&B owner brought in a Fujitsu laptop that had become unreliable: it shut down at random during normal use, the area around the SSD got noticeably hot even after a few minutes, and Windows had started failing to boot with files reported missing. A local repair shop had spotted an SSD problem and pointed them to a recovery specialist. On the drive were guest bookings, reservation history, invoices and tax records — everything the business ran on. The random shutdowns were the key clue: they weren’t a software glitch but a drive protecting itself from heat, and continuing to run it was making things worse.
Modern M.2 NVMe SSDs are fast because they move data at high speed over a direct PCIe connection — and that speed generates heat, concentrated on a tiny stick of a drive. In a slim laptop with limited airflow and often no heatsink over the SSD, that heat has nowhere to go, and the drive can run hot enough to cause problems. When it does, the controller does something deliberate called thermal throttling: it has a built-in temperature sensor, and when it gets too hot it slows itself down — or halts — to avoid damage. That’s what was behind the crashes and freezes. Sustained overheating goes further, though: it can destabilise the controller and corrupt firmware (causing the drive to drop offline or stop responding) and it accelerates NAND wear, so some memory cells degrade faster than they should. The result is a drive that’s intermittently inaccessible and getting less stable over time.
The instinct with an overheating laptop is to boot it up quickly and copy the important files off before it crashes again. On a drive that’s failing from heat, that’s risky: every session lets it overheat again, drives more throttling and instability, and stresses already-degrading NAND — and a controller that keeps dropping offline can leave the file system more corrupt each time. The safer approach is the opposite: stop, let it cool, and capture the whole drive once, in a single controlled pass, while it’s stable — rather than repeatedly pushing a hot drive to the point of another crash.
Recovery started by taking heat out of the equation. The SSD was worked on in a cool, controlled state, with adequate cooling and power-management features that provoke throttling disabled, so the controller stayed stable and responsive instead of overheating and dropping out. In that stable state the drive was imaged sector by sector to healthy storage through a write blocker, capturing the full contents in one clean pass while it was cooperating. Where the overheating had left firmware instability, that was addressed so the drive would present its data reliably for the imaging; had the controller been too unstable to image at all, the fallback would have been to read the NAND flash directly at chip level and rebuild the data from the raw memory. All further work then happened on the image, with the drive left cool and untouched.
From the stable image, the file system was repaired and the B&B’s data extracted — guest bookings and reservation history, financial reports, invoices and tax documents, and staff and admin files — with a few Excel and PDF files that had been caught by the crashes repaired at file level. Everything was checked for integrity and written to a new SSD — not the overheating one — before return, a full recovery. The advice we shared is practical for anyone on an NVMe laptop: keep the machine’s vents clear and airflow good, consider a heatsink if the drive runs hot, don’t keep using a laptop that shuts itself down from heat, and back up — because thermal problems tend to get worse, and a drive that’s throttling and crashing today is warning you it’s under stress.
Thermal stabilisation and throttle-provoking power features disabled · sector imaging through a write blocker in a controlled, cool state · SSD firmware stabilisation · chip-level NAND reading as a fallback for an unstable controller · file-system repair, file-level repair and verification. All work in-house at our Belfast lab.
An SSD that overheats and crashes is under stress and can get worse with use — so stop, let it cool, and send it in for a free, no-obligation diagnostic. We’ll image it in a stable state and recover your files, and put a fixed price in writing before any work starts. On most jobs it’s no fix, no fee. We recover NVMe, M.2 and SATA SSDs for customers across the UK.
It can be. NVMe SSDs run hot, and in a laptop with limited cooling they can overheat and trigger thermal throttling — the controller slowing or halting to protect itself, which shows up as crashes, freezes and shutdowns. Sustained overheating can also destabilise the drive’s firmware and lose access to files. If a laptop shuts itself down from heat, it’s best to stop using it.
Usually yes. The trick is to stabilise the drive — keep it cool and stop it throttling — so its controller stays responsive long enough to image it cleanly, and if the controller is too unstable, to read the NAND chips directly instead. Both routes recover the data; what makes the difference is not repeatedly running a hot drive to the point of another crash.
Keep the laptop’s vents clear and airflow unobstructed, consider fitting a heatsink or thermal pad if the drive runs hot, avoid heavy sustained workloads on a poorly-cooled machine, and don’t keep using a laptop that’s shutting down from heat. And back up regularly — thermal stress tends to worsen over time.
SSD & NVMe recovery · Laptop & PC recovery · SSD not detected