Data Recovery Case File · Second Fixes & Trade Handoffs · Why Matched Heads Sometimes Are Not

Head Swap Failed but the Platters Are Fine

His enquiry was distressed, detailed and asked for something better than sympathy. A Toshiba drive that failed; he unplugged it immediately, which was right; he sent it to another laboratory; he paid separately for a donor drive to be sourced from overseas for its head assembly. Their finding: they "can't get the drive to read service areas on platters", and the swap failed once the donor heads were fitted. They report the platters have minimal damage, and have not been touched. "The drive was only new. I just don't know what to think." Here is what to think, and it is more hopeful than case histories where the surfaces are wrecked: undamaged platters plus an unreadable service area is a very different picture from a head crash — and there is a specific, well-understood reason a correctly-intentioned head swap can still fail.

The decode: what the service area is, and why matched heads sometimes are not

What the service area is: a reserved region of the platters, invisible to any operating system, holding the drive's firmware modules, its defect maps, and the adaptive parameters that describe how this specific drive's heads behave — their individual sensitivity, the corrections needed to read each surface correctly. Every drive is calibrated at the factory and that calibration lives there. Until the service area is read successfully, a drive cannot address its own media, which is why it never becomes usable.

Why a head swap can fail with good platters: the crux of his case, and the reason the other laboratory's result is not necessarily the end. Donor heads must match far more than the model number on the label. They must match the drive's firmware revision, its head map, and its physical generation — manufacturers change internal designs within the same model, sometimes several times, and two drives bearing identical labels can be internally incompatible. Beyond that, replacement heads have their own characteristics, and they will not read the original's surfaces correctly unless the drive's adaptive parameters are handled appropriately — which on some families means the adaptives must be read from the original heads or reconstructed. A swap that is mechanically perfect can therefore still produce a drive that cannot read its own service area. That is not necessarily incompetence; it is one of the genuinely difficult parts of this work, and it is a different problem from damaged platters.

Why "minimal damage, platters untouched" matters so much: it is the most important sentence in his enquiry. The absolute limit in this trade is scored magnetic media — where heads have contacted the surface and removed the recording layer, nothing at any price restores it. His report says that has largely not happened. The data is still physically present. What stands between him and it is an access problem rather than a destruction problem, and access problems have routes.

What an independent look can legitimately try: stated without overpromising. Verifying the installed donor's compatibility against what this drive actually requires, rather than against its label. Attempting service-area access in the manufacturer's technological modes with the adaptive question handled explicitly. Assessing whether the service area itself is physically damaged — which would be a genuine wall — or simply unreadable with the heads currently fitted, which would not. Those are real avenues. They are not guarantees, and each further intervention on a twice-opened drive carries risk, which he should be told before, not after.

And on the money already spent: he paid for a donor drive on top of the work. The right response is a free independent assessment and a written verdict, and if it says no, to say so and charge nothing. He has had enough of paying for hope.

On the bench

The assessment was free and its purpose was to answer his question rather than to find work. Under the laminar flow bench the surfaces were inspected independently — confirming for himself, not on report, how much of the media was actually intact — and the installed donor heads were examined and their compatibility checked against what this drive's firmware revision and head map genuinely require, rather than against the model number. Service-area access was then attempted through the PC-3000 Express in technological mode with the adaptive question handled explicitly. The findings went to him in writing: what was intact, what the obstacle actually was, and whether a route existed — with nothing proposed beyond what the evidence supported.

The outcome

An independent assessment given free, the obstacle identified specifically rather than accepted in summary, and the position stated in writing. Free assessment, one fixed written figure including VAT if a route exists, 50% of parts and labour upfront with the balance only on successful recovery. The decode, for anyone told a head swap failed: undamaged platters with an unreadable service area is a materially better position than a head crash, because the data is present and the problem is access rather than destruction; donor heads must match firmware revision, head map and internal generation rather than the model number, and the drive's adaptive parameters must be handled correctly, so a mechanically perfect swap can still fail to read the service area; and that distinction is worth having established independently, in writing, at no cost, before you conclude anything.

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.