Data Recovery Case File · Cameras, Drones & Cards · Interrupted at the Worst Moment
The Card That Failed Mid-Copy
Her enquiry included the one detail that most narrows this fault: when it happened. A professional CompactFlash card reporting "card 1 cannot be accessed" — "the card was working yesterday, but I got the error message this morning while trying to put the images on my laptop." Email contact requested, since her phone is always on silent. The failure did not appear overnight in a drawer; it appeared during a transfer, while the card was being read continuously. That timing is not a coincidence, and it points at something considerably better than a card that has simply died.
| Media | Professional CompactFlash card — functioning normally the previous day; access error reported by the host during a transfer of images to a computer |
| Reported situation | Card in normal use until the day of failure · error arising during an image transfer rather than at rest · access denied since · photographic content required · email contact preferred and honoured |
| Fault class | Failure surfacing under sustained read load — marginal regions or a struggling controller exposed by continuous access; content typically intact |
| Equipment used | Transfer attempts halted · card imaged write-blocked under per-sector timeouts (DeepSpar USB Stabilizer 10Gb) · weak regions revisited on later passes · structures rebuilt on the image · images validated by rendering |
The decode: why the timing matters
Why a transfer is when faults appear: most of the time a card is barely used — it sits in a camera receiving occasional short bursts of writing. A transfer is completely different: it reads every occupied region, continuously, at speed, for as long as the copy takes. That is the most demanding thing a card ever does, and it is therefore when marginal areas that had gone unnoticed are first asked to perform. A card that has been quietly developing weak regions will very often reveal them at exactly this moment.
Why that is hopeful: because it describes a card that was working right up to the point of heavy load — reading, responding, delivering images — rather than one that has failed outright. The problem is likely to be localised regions that cannot be read on demand at speed, rather than a dead controller or a lost filesystem. That is the most recoverable category there is, because the fix is patience rather than repair.
What the error message means: a host that cannot access a card is reporting the outcome of a failed read, not diagnosing the card. It is the same message whether one region is unreadable or the whole device has stopped answering, which is why the message alone tells her very little and the timing tells her a great deal.
Why retrying the copy is the wrong move: the instinct is to try the transfer again, on another machine or another reader. But a copy that fails partway leaves the operating system retrying the same failing regions repeatedly, and each attempt is another sustained load on the part of the card least able to take it. Some of these cards give one or two good reads of a marginal area before it stops responding at all — and spending those on a failed copy attempt is the worst possible use of them.
What actually works: imaging under per-sector timeouts, which is the same intention as her copy done properly. Each failing region gets a strict budget and is deferred rather than retried indefinitely; the healthy expanse is captured first at full speed while the card is at its strongest; and the difficult areas are revisited on later passes, where repeated patient reads frequently recover what a hurried copy could not.
The one instruction: stop transferring, take the card out, and do not let anything offer to repair or format it.
On the bench
Transfer attempts stopped first, since a failed copy leaves a host retrying the weakest regions and spends the card's remaining margin in the least efficient way possible. The card was imaged write-blocked behind the DeepSpar USB Stabilizer 10Gb with per-sector timeouts enforced — healthy ground banked at speed, failing regions deferred to later patient passes where repeated reads returned areas that had defeated the copy. The filing structures were then rebuilt on the image rather than the card, and every photograph validated by rendering at full resolution before delivery.
The outcome
The card imaged under timeout control, the weak regions re-read and the images validated and delivered. 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 whose card failed while copying: the timing is the useful part, because a transfer reads every occupied region continuously at speed and is the most demanding thing a card ever does — so it is when marginal areas are first exposed; that means the card was working until heavy load, which points at localised unreadable regions rather than a dead device; and retrying the copy spends your best remaining reads on the weakest areas. Stop, and have it imaged patiently instead.
Card that failed while you were copying from it
Don't try the copy again — not on another machine and not with a different reader. A transfer is the hardest work a card ever does, because it reads every occupied region continuously at speed, which is exactly why faults show up then rather than while the card sits in a camera. That timing is good news: your card was reading and delivering images right up to the point of heavy load, so the likely problem is a few regions that can't be read on demand rather than a failed device. But a copy that stops partway leaves your computer hammering those same weak regions over and over, and some cards give only one or two good reads of a marginal area before it stops answering entirely. Take the card out, keep it somewhere safe, and refuse any offer to repair or format it.
Don't retry it — call or email Glasgow Data Recovery on 0141 404 0294; imaged under per-sector timeouts, weak regions revisited patiently, every image checked by eye.
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.