Case record · SSD & flash · SDR-2025-0939
The USB Stick Every Computer Ignores.
The description was short because the testing behind it was not: the client can't get at anything on the memory stick
, and it isn't picked up by any of our machines
— four laptops, a desktop and a powered hub between them. The same silence every time. The survey photographs on it had become hypothetical.
Seeing the same thing yourself?
0800 6890668
The translation.
Being rejected everywhere is diagnostically generous: it clears every machine tried and points squarely at the stick's own doorway. Two suspects lurk on the far side of it — the USB connector with its solder joints, which take a knock each time it goes in, or the controller chip standing between the memory and the host. Neither fault touches the NAND, which is why chip-level work so often rescues sticks that look stone dead.
Kit used on this job.
What happens in a case →| Platform | Its role in this case | Why we use it |
|---|---|---|
| PC-3000 Flash | Identified the NAND and read it without the dead controller | Pulls data straight from memory chips, with a current library that identifies them automatically |
| Rusolut Visual NAND Reconstructor | Corrected the errors and reordered pages to rebuild the logical map | Rebuilds files from a raw NAND image: ECC, XOR, page transforms, reassembly |
| IR preheater & BGA rework station | Getting the package off and reballed so the NAND could be read | Lifts memory chips off and reballs them without burning what they hold |
In the lab.
Rule out the cheap fault first, then the dear one
The connector's solder joints went under the microscope first, since a cracked joint is a far smaller job than anything involving the memory. They proved sound, leaving the controller as the only candidate — and confirming that all that testing across machines had aimed at the right part.
Read the NAND without asking the controller
The memory was reached directly and dumped raw. This is what makes flash recovery possible at all: a controller can be stone dead while the NAND behind it still holds everything ever written to it, laid out in whatever order the controller fancied.
Unpick the controller's layout, then rebuild the files
The raw dump was run through ECC, descrambling and page reordering until it resolved into the logical map the controller had once presented. A filesystem appeared only at that point, bringing the client's reports and photographs with it.
The result.
Recovered and delivered on fresh media, bar three photographs from the last site visit, which came back damaged and were flagged as such.
Similar cases on the index.
Also from SSD & flash.
Does this sound like your own drive?
The rule holds as in every case above: switch it off, and let a free diagnosis come before any decision.