Data Recovery Case File · Solid State & Flash · Heat Means Stop
A Stick That Runs Hot: Unplug It Now
His enquiry contained one detail that changes the advice from stop trying to stop immediately. A dual-connector SanDisk USB stick holding 70–80GB of a teacher's material: "for some reason it isn't being picked up by my MacBook or my Surface. When I have it connected to any device, the stick seems to be quite hot to the touch." Most symptoms in this archive are things a device is failing to do, and the cost of retrying is gradual. Heat is different. A stick that becomes hot while doing nothing is drawing far more current than it should, and that is actively destructive while it is plugged in — so unlike almost every other case here, the instruction is not merely to stop testing but to disconnect now and not reconnect at all.
| Media | Dual-connector SanDisk USB flash drive holding approximately 70–80GB of teaching material — not enumerated by two separate hosts; becoming noticeably hot when connected |
| Reported situation | Not detected on either of two computers · device heating significantly on connection despite no data transfer · no physical damage reported · professional material held |
| Fault class | Excessive current draw indicating a short or a controller in a fault loop — progressive damage while powered; memory typically intact |
| Equipment used | No further connection · board inspected under magnification with the shorted path identified · current draw measured on controlled supply · chip-level read past the failed controller (PC-3000 Flash, Rusolut VNR) · files verified by opening |
The decode: what heat means, and why it is urgent
What a healthy stick does: almost nothing, thermally. Flash devices draw modest current, and a stick sitting idle in a port should be barely warm even after hours. Some heat during sustained transfer is normal. Heat while the device is not being read at all is not.
What heat therefore indicates: current going somewhere it should not. Two usual causes. A short circuit — a damaged component or a bridged path allowing current to flow directly to ground, which produces heat continuously and immediately. Or a controller in a fault loop, repeatedly attempting to initialise, failing, and restarting, each cycle drawing power without ever completing. Both produce a device that gets hot and never appears, which is precisely what he describes.
Why it is actively damaging: this is the part that makes it urgent rather than merely diagnostic. Heat is one of the few things that harms flash memory directly — elevated temperature accelerates charge leakage from the cells, which is what stores the data, so a stick left plugged in and cooking is degrading its own contents. And sustained overcurrent can destroy the controller outright, or damage the memory packages, converting a recoverable fault into an unrecoverable one. So every minute connected has a cost, and the cost is not gradual in the way that repeated insertion of a merely dead stick is.
Why his testing is already complete: two independent computers failing to see it eliminates ports, cables, drivers and operating systems in one step. There is nothing further to establish, which means there is no reason to connect it again — and given the heat, a strong reason not to.
Why the outlook is still reasonable: the memory holding his material sits behind the controller and is normally unaffected by the controller's failure, provided the device has not been left powered for long. It is read directly at chip level, bypassing whatever is shorting or looping. The urgency is entirely about limiting how much heat the memory is exposed to before that happens.
The practice note: a teacher's working material on a single stick is the arrangement this archive sees most often. Once it is back, it wants a home that is not carried between buildings in a pocket.
On the bench
The stick was not connected to a computer again. Its current draw was measured on a controlled supply rather than a host port, which establishes whether the excess is a short or a controller cycling without risking a machine, and the board was inspected under magnification to identify the shorted path. The memory was then read at chip level on the PC-3000 Flash with the pinout confirmed against the Rusolut VNR database, bypassing the failed controller entirely, and the raw data descrambled, error-corrected and rebuilt through its translation layer until the filesystem stood up. His material was verified by opening and delivered on fresh media.
The outcome
The overcurrent identified on a controlled supply, the memory read at chip level past the fault, and the contents verified and delivered. Free assessment, one fixed written figure including VAT, 50% of parts and labour upfront with the balance only on successful recovery. The decode, for anyone whose stick gets hot: a device drawing enough current to warm up while doing nothing has a short or a controller stuck in a fault loop, and unlike most symptoms this one is actively destructive — heat accelerates charge leakage from the flash cells that hold your data, and sustained overcurrent can destroy the controller or the memory outright; so disconnect it now and do not reconnect it, even to test; two computers failing to see it is already a complete elimination; and the memory behind that controller is normally intact if it has not been left cooking.
Device that becomes hot when connected
Unplug it now and don't plug it back in — this is the one symptom where continuing to test causes real, immediate damage rather than gradual wear. A stick or drive that warms up while doing nothing is drawing far more current than it should, which means either a short circuit or a controller stuck repeatedly trying and failing to start. Both are destructive while powered: heat accelerates the charge leakage that erases flash memory from the inside, and sustained overcurrent can kill the controller or the memory packages outright, turning a recoverable fault into a permanent one. If two computers have already failed to see it, your testing is complete anyway and there's nothing to gain from another attempt. Don't try a powered hub, don't leave it connected hoping it settles, and don't run diagnostic software at it.
Unplug it right now — call Glasgow Data Recovery on 0141 404 0294; current draw measured on a controlled supply, shorted path identified, memory read at chip level past the fault.
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.