What Was Happening
The machine had powered on normally for years. One morning it reached the HP splash screen, then either dropped to an automatic repair loop or showed the unhelpful “no bootable device found” message. The customer had tried the usual self-help steps — power-cycle, unplug for ten minutes, try a different mains socket — none of which made any difference.
They were worried about the data more than the hardware: family photos, scanned documents, several years of email archives, none of it backed up.
Our Diagnosis
A non-booting machine has three broad failure classes, and each gets a different first response:
- OS corruption with healthy hardware. Fixable in place — repair the boot partition, run system file checks, often boot from a recovery USB. Customer’s data stays put.
- Failed storage with the controller intact. Recoverable on the bench using a drive duplicator and SMART-aware tools. We image the drive bit-for-bit to a healthy disk and recover from the image.
- Failed storage with controller damage. Not safe to attempt on the bench. Every power cycle and every read attempt risks pushing the drive into a state that even specialist recovery can’t reverse.
The diagnostic order matters because step three is irreversible. We always start by minimising power-on time to the drive.
For this HP:
- BIOS showed the SSD on one cold boot, missing on the next two, present again on the fourth. Intermittent BIOS detection is a textbook controller symptom.
- Connected the drive to our diagnostic bench via a USB bridge (no risk of installing or writing). The drive enumerated for about 30 seconds then dropped off the bus.
- SMART read attempts failed — the controller wasn’t responding to ATA identify commands reliably.
- No clicking, no spin-up issues (it’s an SSD — no moving parts to listen to), but the controller’s behaviour matched the textbook pattern of NAND flash partial failure with the controller getting confused trying to map around bad blocks.
At that point we stopped. Continuing would have meant repeatedly power-cycling a drive that was already failing, and the data on it was the customer’s primary concern.
How We Fixed It
Honest conversation with the customer, in writing as well as on the phone:
- The drive had failed in a way that on-bench tools could not safely recover from.
- The longer the drive sat powered on or being probed, the worse the odds got.
- For data this important, the right route was a specialist cleanroom recovery house — they can extract the NAND chips directly and rebuild the data from the raw flash if needed.
- That route is materially more expensive than a typical on-bench recovery, but it is the only one with realistic odds for a drive in this state.
We powered the drive down, sealed it in an anti-static bag, and gave the customer the contact details for the specialist we work with regularly. We also offered to handle the courier and act as the intermediary so the customer wasn’t navigating it alone.
In parallel we prepared the rest of the work that was always going to be needed:
- Replacement SSD sourced (correct form factor for this HP All-in-One — 2.5" SATA in this chassis, not M.2)
- Clean Windows installation prepared, with current driver bundle for this HP model staged on a USB ready to apply post-install
- Migration plan documented so when the recovered data comes back from the specialist, it lands cleanly back into the user’s profile
The Result
Customer made an informed decision based on the real trade-offs rather than a generic “your data is gone” or a misleadingly optimistic on-bench attempt. The drive is currently with the cleanroom specialist. The HP is back together with a healthy replacement SSD and a fresh OS, ready to receive whatever the specialist can recover.
Why This Happens
Consumer SSDs fail in two ways. Cell wear is the slow one — over years, the NAND flash that stores the data wears out a little with every write. Most modern SSDs handle this gracefully, slowing down and eventually going read-only when they hit their limit.
Controller failure is the fast one. The controller is the small chip on the drive that manages everything: which cell holds what data, error correction, the mapping table that turns logical addresses into physical locations on the flash. When it fails — usually from a power event, sometimes from firmware corruption — the drive can become unreadable in seconds, even with perfectly healthy NAND underneath.
The intermittent detection pattern on the HP above is the most common controller-failure signature we see. It is often misinterpreted as a “loose cable” and the user goes round in circles re-seating SATA connectors that are fine. The real fault is the controller giving up on reliably responding to enumeration.
How to protect yourself from this
There is only one reliable answer to SSD failure: backups. The drive will fail. The only variable is when.
- Set up an automatic backup that runs without you thinking about it. Windows File History to an external drive, or a cloud service that syncs your Documents and Photos folders. The good ones cost less than ten pounds a month.
- Don’t keep the backup plugged in permanently. A backup drive that lives in the same machine as the original drive will be hit by the same power surge, the same ransomware, the same accidental deletion. Rotate it.
- Replace SSDs proactively at five to seven years on heavily-used machines. A drive that has been running daily for that long is on borrowed time. Migrate to a new drive while the old one is still healthy enough to copy from.
- Watch for early warning signs. Sudden slowdowns, the OS reporting “scanning and repairing drive” on boot, files that won’t open with no obvious reason — all worth taking seriously rather than waiting for the drive to die outright.
Local Help in Fulham SW6
Data recovery is more about what you don’t do to the drive than what you do.
We image first, recover from the image, and tell you honestly when specialist cleanroom services are the right call rather than risking the data ourselves.
Drop in to our Putney workshop (SW15), call 020 7610 0500, or contact us urgently if the drive is showing warning signs.