Camera cards · bench notes · RDG-2025-2061
The Break Never Touched the Memory.
Twelve months of checking forecasts, repaid inside seven days: two clear nights, no moon, stars down to the horizon
. None of it was copied anywhere. Then his reader clamped onto the card and refused to give it back. He is an astrophotographer, in Caversham, and he went at the slot with a screwdriver — at which point it was the card that came off worse: a piece of the casing broke away right beside the gold contacts
. That chunk vanished under a desk and stayed there. Ever since: not one reader will accept that the slot has anything in it
.
Same fault on yours? Tell us what it is doing.
0800 6890668
What actually caused it.
Very little of an SD card is actually card. Almost all of what you hold is moulded plastic wrapped around one block where controller and memory sit sealed together in resin. A few fine traces connect that block to the gold pads, and they run down precisely the corner he had been levering at. Break the corner and it is the wiring you break, never the storage. Think of a shop with a jammed shutter: the stock is all still inside, but nobody is getting in at the front. Losing the fragment therefore cost us nothing. Magnification shows how far a fracture has travelled, and this one had stopped well short of the die.
What this one needed on the bench.
How every job runs →| Bench tools | Its part in this job | Why it is here |
|---|---|---|
| Rusolut Visual NAND Reconstructor | Error correction applied, then the dumped pages returned to their true order | Error correction, XOR mask, page order, reassembly — a raw NAND dump turned back into files |
| Infrared preheater & BGA rework station | Controlled heat, gentle and held only long enough to take the coating off the pads | Warm enough to release the chip and reball its pads, and never hot enough to put the data at risk |
| PC-3000 Flash | Matched the monolith against our ID library and mapped its pinout, with no power applied | Reads the NAND direct, past the controller, against a maker-ID library kept current |
What the bench did.
The break told its own story
With no fragment to inspect, the card itself had to answer the question, which took about a minute under the stereo microscope. Part of the contact block had gone, traces ended where it had been, and the crack ran out well before the die. Once you have seen a card in that state it stops looking finished and starts looking like ordinary chip-level work.
The route the factory left open
On the underside of a monolith is a row of service pads, which is how the factory tests it. No maker publishes the layout and no two agree on one, so probing continued until the map was clear. Wire thinner than a hair then goes on to the few pads that matter, and the memory replied down a path the fracture had never gone near.
Raw pages first; pictures a long way further on
A chip gives up its pages in the order its controller chose to store them, and in no other order at all. Error correction runs before anything else can. The XOR mask then comes off, the pages are put back into their real sequence, and only after a filesystem has been built over the whole lot does a folder name appear. Closing that gap is the work, and no reader sold anywhere will close it for you.
What went back.
Both nights came through intact: each frame opened, each one checked, calibration files as well. The card is in a bin somewhere and nobody mourns it.
If this sounds like yours, start here.
Other card jobs in the log.
If that sounds like yours, stop there.
Post it in as it is, switched off. Looking costs nothing, and you get one fixed figure in writing before any chargeable work begins — £250 + VAT for a card or stick, £300 + VAT for a single drive, from £500 + VAT for RAID and NAS. On most jobs, nothing is charged unless your data comes back.