Data Recovery Case File · Desktop Externals & Aging Drives · Don't Put It in a Working PC

Old Internal Hard Drive: Getting Access Safely

His enquiry was one line: "I've got a single old internal Seagate hard drive I was hoping to get access to." No symptoms, no history, no fault described — which usually means the drive has not been tried yet, and that is genuinely the best moment to receive advice. Because the obvious next step, and the one nearly everybody takes, is to open a working computer, connect the old drive as a second disk, and boot up. That works often enough that it is the standard advice online. It is also the single riskiest way to approach a drive whose condition is unknown, for three reasons that have nothing to do with the drive being old and everything to do with what a running operating system does to storage it does not recognise.

The decode: three reasons not to put it in your PC

One — a running system writes to disks it meets. This is the risk people never anticipate. Attach an unknown drive to a booted Windows machine and the operating system does not simply look at it: it may mount the volume and update timestamps and journals, it may offer to initialise a disk whose partition table it cannot parse, it may propose scanning and repairing a filesystem it finds inconsistent, and system services may begin indexing it. Every one of those writes to a drive that should be read-only until its condition is known — and if the drive turns out to hold deleted material worth recovering, those writes land in exactly the free space that matters.

Two — a failing drive can take the host down with it. If the old drive has weak regions, the machine will stall waiting for reads that never complete, and a stalled storage subsystem hangs or crashes the whole computer. That means an unplanned power cut delivered to a drive mid-operation, which is the fastest way to convert a marginal fault into a serious one — and if the machine is his working PC, he now has two problems.

Three — dormancy makes the first power-up the one that counts. A drive that has been sitting has stiffened bearings and heads that may have adhered to the platters during storage. The first attempt to spin it is the most demanding thing that will ever be asked of it, and it deserves to happen on equipment that can watch what the drive does and stop if it goes wrong — not inside a desktop case where the only feedback is whether Windows shows a drive letter.

What to do instead: if the drive is worth reading at all, it should be connected write-blocked — through a hardware write blocker or a dock in read-only mode — so that nothing can be written to it regardless of what any operating system decides. It should be connected to a machine that is not doing anything else, so a stall costs nothing. And its interface generation should be established first, because a drive of unknown age may be IDE rather than SATA and needs the right connection rather than a cheap adapter. Where the drive is healthy, that turns the whole matter into a simple copy; where it is not, nothing has been spent to find out.

On the bench

The drive's generation and interface were identified before anything was connected, and it was brought up write-blocked on native equipment rather than inside a working machine — so that whatever it turned out to be, nothing could be written to it and nothing else depended on it behaving. The Atola Insight Forensic read its power-on behaviour and health directly, which on a dormant drive is the moment that decides everything: whether it spins freely, whether the heads unpark, and whether the surfaces read cleanly. Where the condition warranted it, a complete image was taken on the PC-3000 Express under Data Extractor before anything else was asked of it, and the contents were read from the copy, verified by opening, and delivered on fresh media.

The outcome

The drive assessed before connection, read write-blocked, and its contents verified and delivered. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode, for anyone holding a salvaged drive: connecting it inside a working computer is the standard advice and the riskiest option, because a running system writes to disks it meets — initialising, repairing, indexing — and a failing drive can hang the host and deliver itself an unplanned power cut in the process; a dormant drive's first spin-up is the most demanding moment it will ever face and deserves to be watched; and the safe version costs nothing extra: identify the interface, connect write-blocked, and look before anything is allowed to write.

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.