Data Recovery Case File · Solid State & Flash · The Enclosure Is Eliminated
External SSD Invisible After Changing the Caddy
His enquiry had already done the work most enquiries leave undone. A 2TB M.2 external SSD formatted for macOS: "it stopped working and suddenly doesn't appear in the Finder window or Disk Utility. I have checked the power of the USB ports with similar external SSDs and there is no issue with them. I have also changed the casing and power but still no luck. I would like to understand if the data can be recovered even if it's a physical issue. There was no incident that caused the SSD to become physically damaged." Swapping the casing is the test that matters here and almost nobody performs it — because an M.2 external is two separate products in one object, and substituting one of them settles decisively which has failed.
| Media | 2TB M.2 solid-state drive in an external enclosure, macOS-formatted — absent from Finder and Disk Utility; enclosure, cable and host ports eliminated by substitution |
| Reported situation | Sudden loss of recognition with no incident · host ports verified with other working externals · enclosure and power substituted without change · owner asking about recovery prospects including physical faults |
| Fault class | Drive-side failure with enclosure electronics excluded — controller or NAND fault on the module itself |
| Equipment used | Module read directly on native equipment rather than through any enclosure · vendor technological and safe-mode access (PC-3000 portfolio, SSD support) · controller state and firmware area assessment · imaging on restored access; contents verified by opening |
The decode: two products in one, and what remains after the swap
Why an M.2 external is two things: unlike a sealed portable drive, this is a bare module — the drive itself — inside a separate enclosure containing a bridge chip that converts its native interface to USB, plus a connector and power circuitry. Either can fail independently, and from outside the symptoms are identical: nothing appears. That is why so many of these cases stall — owners try different cables and computers, which tests the host, and never test the half of their own device that is most likely to have failed.
Why his swap is the decisive test: because bridge chips fail more often than the drives they carry. They run hot in small metal enclosures with no airflow, they take the physical stress of every connection, and they are the cheaper component. By putting the module in a different casing and finding the same result, he has eliminated the bridge, the connector, the cable and the enclosure's power circuitry in one move — and separately verified his host ports with other working drives. What remains is the module itself. That is as complete an elimination as home testing allows, and it means no further substitution is worth doing.
Why "no incident" fits: he offers it almost apologetically, and it is entirely consistent. Solid-state drives fail without physical cause and usually without warning — controllers stop abruptly, firmware states become corrupt after an interrupted write, and none of that requires mistreatment. Heat is a genuine contributor in compact enclosures, where an NVMe module and a bridge chip generate real warmth in a sealed metal shell with nowhere to shed it, which is worth knowing for the replacement rather than as blame.
His actual question, answered honestly: he asks whether recovery is possible even if it's a physical issue, and deserves the straight SSD picture. The realistic route runs through the controller — vendor technological and safe modes that can bring up a drive whose normal start-up fails, allowing the firmware area to be assessed and repaired and the drive imaged. That works often. What is not a dependable fallback on modern NVMe is reading the memory packages directly, because the address mapping and often the encryption live inside the controller, so raw flash read without it yields little. That limit is stated in advance rather than discovered.
One thing the macOS formatting does not change: a filesystem is above the layer that has failed. The drive not being seen at all has nothing to do with how it was formatted, and the formatting neither helps nor hinders the recovery.
On the bench
No further enclosure was tried, since his substitution had already settled the question. The module was read directly on native equipment, removing every bridge chip from the path, and access was attempted through the PC-3000 portfolio's SSD support: the controller addressed in its vendor technological modes, its state examined, and the firmware area assessed for the corruption that typically prevents a drive completing start-up. The honest limits had been set out in writing beforehand. Where access was restored, the drive was imaged immediately and completely, and the volume was rebuilt from the copy with his files verified by opening.
The outcome
The module read directly on native equipment with the enclosure removed from the path, accessed at controller level and imaged. Free assessment, one fixed written figure including VAT; where a chip has to be removed, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode, for anyone with an M.2 external: it is two products in one — a bare module and an enclosure with its own bridge chip — and the bridge fails more often than the drive, so swapping the casing is the decisive test that most people never run; doing it, as he did, along with verifying the host ports on other drives, is a complete elimination and no further substitution is worth trying; SSDs fail without incident, so "nothing happened to it" is entirely consistent; and on modern NVMe the realistic route runs through the controller rather than around it.
External SSD that has gone invisible
Swap the enclosure before concluding anything. An M.2 external is two separate products — a bare drive module and a caddy containing a bridge chip, connector and power circuitry — and the bridge fails more often than the drive does, because it runs hot in a sealed metal shell and takes the stress of every connection. Trying other cables and computers only tests your host; putting the module in a different caddy tests the half you actually own. If it stays invisible after that, and your ports work with other drives, your elimination is complete and further substitution is wasted effort. Don't be puzzled by there having been no incident — solid-state drives fail abruptly without physical cause. And set expectations honestly: on modern NVMe the route runs through the controller using vendor modes, because reading the memory chips directly isn't the dependable fallback it is on cards and sticks.
Your testing is complete — call Glasgow Data Recovery on 0141 404 0294; module read directly on native equipment, controller addressed in vendor modes, honest limits stated in writing first.
Request a quote online →
Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.