An old IBM drive out of a parallel-port enclosure, holding code its owner had written as a child. No light on the adapter, no spin, and a distinct burnt smell — which narrows the diagnosis considerably before anything is opened, and narrows it in a good direction.
The client had kept an old IBM drive from a parallel-port external case, and on it was code they had written as a child. Not valuable in any commercial sense and entirely irreplaceable — a combination we see often, and one that makes a do-it-yourself attempt a poor gamble.
They connected it through an IDE-to-USB adapter and got nothing: no LED activity, no spin, and a burnt smell from the casing. They had the presence of mind to test other drives on the same adapter first, including ones with a higher power draw, which worked normally and ruled out the adapter and the supply.
No power draw at all, plus that smell, points squarely at the circuit board rather than the mechanism — and the board is designed to be the thing that fails.
Drives carry transient voltage suppression on their power rails: components whose entire purpose is to conduct a surge to ground rather than allow it downstream. They fail closed by design. When one does, it short-circuits the rail, the drive draws nothing, and everything past it — the motor driver, the preamplifier, the head assembly — is spared.
That is exactly what had happened. Visible scorching around the power section, a shorted protection device, and collateral damage to the regulation components beside it. Nothing had reached the motor or the heads.
From a recovery point of view this is among the better diagnoses available: the platters and the data entirely untouched behind a circuit that stopped conducting.
The obvious repair is to buy an identical drive and swap the board over. On drives made before roughly the early 2000s that frequently worked, which is why the advice still circulates. On anything since, it does not, and the reason is specific.
Modern drives are calibrated individually at the factory. The characteristics of that particular head stack and those particular platters — preamplifier bias, per-head channel settings, fly-height compensation, and the list of defects mapped during manufacture — are measured on that one unit and written into a small serial flash chip on its own board.
Those adaptive values belong to that drive and no other. Fit a board carrying a different drive’s calibration and one of two things happens: the drive fails to initialise, or — worse — it initialises and drives the heads using values meant for different hardware, reading unreliably or making contact with the surface.
So a board swap does not fail safely. It can convert an electrical fault into a mechanical one.
The board was repaired rather than replaced. A matching donor board was sourced for the family and revision, and the original ROM chip — the one carrying this drive’s own adaptives — was desoldered from the burnt board under a microscope and transferred onto the donor.
That transfer is the whole job. Everything else on the board is generic; the ROM is the part that makes it this drive’s board. On some families the adaptives live inside the main controller rather than in a separate chip, which makes the same repair considerably harder and occasionally impossible — another reason the correct board revision matters as much as the correct model.
Damaged components in the power section were replaced individually, and the rebuilt board was checked for shorts and correct rail voltages on the bench before any power was applied to the drive itself. Powering a drive through a board that has not been verified is how a second fault gets created.
It spun up and identified correctly on the first attempt.
It was then imaged read-only immediately, in one planned pass rather than tested repeatedly. A drive that has been through a surge has an uncertain amount of life left in whatever else the event touched — the motor windings, the preamplifier inside the sealed chamber — and none of that can be assessed from outside. The safe assumption is one good read.
Everything recovered, including the code, returned on new media. Six days, most of it sourcing the correct board revision rather than the repair or the imaging.
Worth noting what made it straightforward: the client stopped after the first failed attempt and did not keep trying adapters and supplies. Repeatedly powering a drive through a shorted board is how a contained electrical fault reaches the motor — each attempt pushes current through a path that is no longer behaving as designed. Single drives are from £300 +VAT after a free 48-hour diagnostic.
Send the device over with a short account of what happened to it. The diagnostic costs nothing and commits you to nothing, and the figure that follows is fixed in writing before anything is opened.
Nothing can begin until the device is on the bench, which makes this the only step that needs anything from you. Pack it properly, put your details in with it, and send it over. What follows is a free diagnostic and a written figure, in that order.
Posting it? Use something tracked and insured — whatever is on the drive is worth considerably more than the postage. Bringing it in? Weekdays, 9am to 5:30pm, and it still wants packing as above for the journey.
Tell us what the device is, what it is doing, and anything already tried — an engineer will come back with a realistic view of the odds and a price band, before you commit to posting anything.
We’ll be in touch shortly. For anything urgent, call 0161 871 0788.
Usually not. A burnt smell with no power draw points at the circuit board, and the board is designed to fail first — most drives include a protection diode that shorts deliberately so a surge never reaches the mechanism. The platters are generally untouched.
Not on anything made in the last twenty years. Each drive is calibrated individually at the factory and those adaptive values live in a ROM chip on its own board. A donor board carrying different calibration either will not initialise or drives the heads with the wrong values.
Moving the serial flash chip holding a drive’s unique calibration from its failed board onto a working donor board. It is what makes board replacement viable at all, and it is why board-level work is a laboratory job rather than a screwdriver one.
No. Once a drive has shown a burnt smell or drawn no power, repeated attempts risk pushing a fault that was contained on the board through into the motor or heads. Stop after the first failure.
Yes, and they arrive regularly — frequently holding material nobody thought to move forward. The recovery is well understood; sourcing a correct-revision donor board for an unusual model affects the timescale rather than the feasibility.
From £300 plus VAT for a single drive, fixed in writing after a free 48-hour diagnostic, with most jobs no fix, no fee.
Start with an instant online quote, or call and talk it through with us first. You'll have a clear, fixed price before any work begins.