Call us — 0113 322 3083
Mon–Fri · 9am–5:30pm · No fix, no fee
Start a free diagnostic →

Data Recovery Case File · Solid State & Flash · The Twin Is a Warning

One Stick Failed and Its Twin Still Works

Her enquiry contained a detail she mentioned only in passing, and it is the most important thing in it. A 32GB memory stick holding photographs and videos now says it needs to be formatted — and she knows what that means: "I'm aware this means all the stuff on there will be erased and I don't have any other copy." Then the detail: "I've got two of the exact same memory sticks purchased at the same time. The other one, with all of my university work, is fine." That twin matters twice over. It is genuinely useful to the recovery of the failed one, in a specific technical way. And it is a warning, because two devices from the same production batch, bought together and used over the same period, tend to reach the end of their lives at around the same time — which means her university work is sitting on a stick that is now on notice.

Media32GB USB flash drive holding photographs and video — presenting a format prompt, contents inaccessible; an identical stick from the same purchase remains functional
Reported situationFormat prompt on the affected stick, correctly declined · no other copy of the contents · identical twin device still working and holding university coursework
Fault classFilesystem or controller failure on one of a matched pair — twin available as an identification reference; batch-correlated risk to the surviving device
Equipment usedWrite-blocked imaging where the device enumerates (DeepSpar USB Stabilizer 10Gb) · chip-level read where it does not, with the twin used to confirm controller family and layout (PC-3000 Flash, Rusolut VNR) · filesystem reconstruction and validation

The decode: the format prompt, the useful twin, and the warning

The prompt, briefly: her instinct is right and worth confirming. "This memory stick needs to be formatted" means the computer could not read the volume's filing structures and is offering to write fresh empty ones over them. It is not a report that the contents are gone — the photographs and videos sit intact behind the unreadable filing — and accepting would replace exactly what a reconstruction rebuilds from. Declining every time, as she has, keeps this straightforward.

Why the twin is genuinely useful: this archive has explained elsewhere that flash recovery, unlike hard drive work, does not need a donor — there is no part to transplant. But an identical device from the same purchase is valuable in a different way: as a reference. If the failed stick has to be read at chip level, the work begins with identifying the controller family, its pin layout, the scrambling pattern it applies and the error-correction scheme it uses — and an identical working device can be examined non-destructively to establish all of those quickly, rather than determining them from the damaged unit alone. Two sticks bought together from the same batch will almost certainly share the same internals, even where a manufacturer has changed components between production runs. So bringing the twin along genuinely helps, and it is returned unharmed.

Why the twin is also the warning: and this is the part worth acting on today. Flash devices wear as they are written to, and two sticks bought at the same time, from the same batch, and used over the same period have accumulated comparable wear on comparable silicon. One failing is not a coincidence that leaves the other safe; it is the first of a pair reaching the end of its usable life. Her university work is on the survivor, and the correct response is not to wait and see: copy it off today, to a computer and to something else besides. That advice costs nothing and is worth more than the recovery it accompanies.

The route for the failed one: established at assessment. If it still enumerates, it is imaged write-blocked and the filing structures rebuilt on the copy. If it is electrically silent, the memory is read directly at chip level past the controller and reconstructed in software. Which applies is determined free of charge before any figure exists.

On the bench

The failed stick's response was established first, since it decides the route. Where it enumerated, it was imaged write-blocked behind the DeepSpar USB Stabilizer 10Gb and its structures rebuilt on the copy until the folder tree returned. Where it did not, the memory was read at chip level on the PC-3000 Flash — and the twin earned its journey here, examined non-destructively to confirm the controller family and layout against the Rusolut VNR database before a probe touched the failed unit. Her photographs and videos were validated by rendering and playback, and the twin was returned working, alongside a written note recommending that its contents be copied off without waiting.

The outcome

The failed stick recovered and its contents validated, with the twin used as a reference and returned intact. 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. Three things worth taking from a matched pair: the format prompt is always the offer to refuse, and refusing it is what keeps a recovery simple; an identical device from the same purchase is not a donor — flash does not need one — but it is a valuable reference for identifying the controller and its layout, so bring it; and above all, treat the survivor as a warning rather than a reassurance, because two sticks from one batch, used over the same period, wear out at around the same time. Copy the working one off today.

One of a matched pair of drives has failed

Copy the surviving one off today — that is the action item, and people almost never take it. Devices bought together come from the same production batch, contain the same components, and have usually done comparable work over the same period, so the first failure is a timing signal rather than bad luck striking one of two. Whatever is on the working device should be somewhere else by this evening. For the failed one: keep declining the format prompt, since accepting writes empty structures over the damaged ones a recovery rebuilds from, and stop reinserting it to check. And bring the healthy twin with the failed one when you send it — it isn't needed as a donor, because flash recovery has nothing to transplant, but an identical working device lets the controller family and layout be confirmed quickly, and it comes back to you unharmed.

One stick failed and you have its twin?
Copy the twin off today, then call Leeds Data Recovery on 0113 322 3083; failed unit imaged or read at chip level, the twin used only as a reference and returned intact.
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.

0113 322 3083