Data Recovery Case File · Solid State & Flash · Say It Before They Start

The Repair Is Happening This Afternoon

Her enquiry contained a deadline nobody had noticed. An NVMe drive whose "controller chip has become faulty and is getting very hot. It has stopped working on my laptop. I am having a new board fitted to my laptop today — but would I be able to retrieve the information off the faulty board?" The technical answer is yes, usually. But the important part is the timing, and it is not technical at all. The repair is happening today, and the failed drive is inside the machine. Unless somebody says otherwise, in writing, before it starts, there is a real chance the old drive leaves with the repairer and never comes back.

The decode: the sentence to send now, then the technical picture

Why the old drive may not come back: repairers dispose of replaced parts as a matter of routine, and there are several reasons a failed drive does not simply get handed over. It may be returned to a supplier under warranty or as a core exchange. It may go into a bin of dead components. It may be securely destroyed as a data-protection measure, which responsible shops do precisely because customers' drives should not accumulate on shelves. None of that is bad practice — but all of it ends the recovery, and none of it will be reversed by asking afterwards.

What to send, and it takes a minute: an email or message to the repairer, before the work starts, saying that the original drive must be returned and must not be wiped, destroyed or sent back to a supplier. Ask them to confirm. If the machine is already with them, telephone as well as write. This is an entirely ordinary request that shops handle every week, and the only thing that makes it fail is not making it.

The related warning about the heat: she reports the controller running very hot, and that is not merely a symptom. Heat means the device is drawing far more current than it should — a short, or a controller stuck in a fault loop — and that is actively damaging while powered. Elevated temperature accelerates charge leakage from the flash cells that hold the data, and sustained overcurrent can destroy the controller or the memory outright. So the drive should not be powered again to test it, and it should certainly not be left in a running machine.

Why the outlook is otherwise reasonable: the memory holding her data sits behind the controller and is normally unaffected by the controller's failure, provided the device has not been left running hot for long. The route is vendor-mode access where the controller can be brought up, or reading the memory directly at chip level where it cannot — bypassing the fault rather than repairing it.

The general rule worth carrying: whenever a machine goes anywhere for repair — a shop, a manufacturer, an insurer, an IT department — the storage is the thing to ask about explicitly and in writing, before the work begins. Machines get repaired by replacing parts, and the part holding your data is one of them.

On the bench

Retention of the original drive was secured in writing before the repair proceeded, since a replaced component is routinely returned to a supplier or destroyed and no technical work is possible on a drive that has left the building. The drive was not powered again, given that a controller running hot is drawing excessive current and doing progressive damage while energised. The board was inspected under magnification with the current path assessed to establish whether the excess was a short or a controller cycling, and the memory read at chip level past the failed controller where vendor-mode access could not be restored.

The outcome

The original drive retained before the repair, the current fault identified and the memory read past the failed controller. 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 machine is about to be repaired: say now, in writing, that the original drive must be returned and must not be wiped, destroyed or sent back to a supplier — replaced parts are disposed of as a matter of routine, sometimes securely and deliberately, and asking afterwards is too late; and treat a component running hot as urgent rather than merely faulty, because excessive current draw damages flash while it is powered. The memory behind a failed controller is normally intact.

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.