Memory cards · bench notes · HUL-2025-0907
Full-Size Files, and Nothing to Watch.
A videographer in Scunthorpe posted us a card that had gone against him overnight, after a marquee wedding on a farm out on the Isle of Axholme: the last three files from the evening will not import at all
. His next sentence framed the whole job — they still show full size on the card, so the video has to be sitting there
— and the couple’s flight left in under a fortnight.
Sounds like yours? Ring us.
0800 6890668
What causes it.
His reading of it was correct. A camera records in two parts. The frames go down on the card as they are shot. The index — the list that says where each frame starts and how long it lasts — is written last, at the moment the file closes properly. Pull the card early, flatten the battery, or fill the card, and the frames stay put while the table never arrives. An editor navigates a clip by that table and nothing else. So a file stuffed with video but carrying no map looks, on screen, no different from a blank one.
The tools this one needed.
The steps we follow →| The gear | What this one did | Why we have it |
|---|---|---|
| Klennet Carver | Rebuilt the frames that survived into files his editing software would open | It knows the internal shape of a video file; cruder tools return scraps that play nowhere |
| PC-3000 Flash | Took the full sector-by-sector copy of the card before anything else happened | Goes round the controller and reads the NAND itself, matched to a maker-ID library kept current |
| UFS Explorer Professional Recovery | Pulled the good clips and the folder structure back out of that copy | Opens the volume formats most tools cannot: APFS, ReFS, XFS, ZFS, Btrfs |
The stages.
The copy comes before the diagnosis
Nobody said a word about the fault until there was a copy. A delivery date already given to someone else will alter the pace at which a job is run. It does not buy anyone a free hand with the one original in the room. From there on, every step was taken against the image. The card itself was bagged in antistatic and put aside, kept in case a further read was ever wanted.
Bank the clips that still open, first of all
Most of the night came off the copy without argument and was written somewhere safe before anything else was attempted. Only then were the three difficult files opened up and read through. He had called it right: frame data on all three, a half-written index on two, no index at all on the third. That is what set the order of work.
Take the pattern from the clips that survived
Video is seldom laid down in one unbroken run, and that scattering is where one-click software stops being useful — what it hands back will not open anywhere. Undamaged clips from earlier on, off the same body at identical settings, gave a pattern to work from. Each damaged file was built again around the frames it still held, then tested the only way worth anything: played from the first frame to the last.
How it finished.
Two of the three came back complete, and both were played end to end on the bench before they were sent. The third runs cleanly to within a few seconds of its end; those seconds were still in the camera’s buffer when the card was pulled. The edit reached the couple in time for the flight.
What people read after this.
More cases under Memory cards.
Is yours doing the same?
Turn it off. Send it to us. The diagnosis comes first. It says which files are still readable and which are not.