Home / Case Studies / He Brought the Board. It Needed the Chip
Power, Boards and Liquid · case file

He Brought the Board. It Needed the Chip

A haulage firm at Narborough lost a Western Digital disk that had been sitting in an office machine holding tachograph downloads, delivery paperwork and the photographs their drivers take at collection. It stopped dead. The transport manager, who is practical, matched the model number online, bought a replacement circuit board and then read enough to suspect that fitting it might not be the end of the story. One of their drivers runs down the A14 twice a week, so the disk and the new board were handed in over the reception desk at the lab on a Tuesday morning rather than posted.

Hard DriveElectrical / PCBDead / No Power
// the job in four lines
MediaA Western Digital hard disk that stopped without warning, handed in together with a replacement circuit board of the same model bought online by its owner.
What was reportedWestern Digital disk, dead with no warning, no sign of life on any machine, a matching replacement board supplied with it by the owner, tachograph records, delivery paperwork and collection photographs wanted back.
Fault classElectronic. A component on the disk's own board had failed, taking the supply down with it and leaving the drive with no power path at all. The mechanism and the recorded material were untouched.
On the benchBoth boards examined under a stereo microscope and their revisions compared; the failed part found by tracing the current draw on a current-limited bench supply; the calibration chip on the original board identified, read out and transferred to the replacement board, with the failed component repaired on the original as a fallback; the drive brought up under supervision behind PC-3000 UDMA; imaged on DeepSpar; the volume parsed in R-Studio and the paperwork opened before it went to fresh media.
// reading the fault

Reading the fault

The instinct was right and the missing step is the one nobody mentions. Buying a board of the same model is a sensible reaction to a dead disk and it is how the fault used to be fixed. What changed is that boards began carrying a small chip of settings measured at the factory for one specific mechanism: the strength of each head, the surface map, the tuning that makes that particular disk readable. Those settings are not the same on two disks that came off the same line on the same day.

A bare swap does not simply fail, it fails misleadingly. The drive will usually power up and spin, which convinces the owner the repair has worked, and then read nothing or read nonsense. In a proportion of cases it goes further and writes an updated set of tables using the wrong calibration, at which point a straightforward board fault has been converted into something considerably longer. The disk arriving unswapped was worth more than the board he had bought.

Moving the chip is fiddly and it is the whole job. The calibration lives either in a separate eight-pin chip or inside the main controller, depending on the generation, and on this one it was separate. It is read out, written into its counterpart on the donor board, and the assembly is then the original disk's board in every way that matters. Where the failure has damaged that chip as well, the alternative is repairing the original board instead, which is often the surer route and was kept in reserve here.

None of this touches what is stored. A board is the small computer that drives the mechanism. When it dies it stops the mechanism being used; it does not reach through and alter the surfaces. Everything the firm needed had been sitting undisturbed since the afternoon the disk stopped.

// bench work

Bench work

The two boards were compared under the microscope first, since a same-model board of a different revision is a common and wasted purchase. His was the correct revision, which is better luck than most. On a current-limited supply the original board pulled hard and warmed at one point, and the failed part was identified from where the current went.

The calibration chip was read off the original and written to the replacement, and the drive was brought up under supervision behind PC-3000 UDMA. It identified correctly, reported its own capacity and read at full speed. Imaging on DeepSpar took one pass, R-Studio parsed the volume, and a month of tachograph downloads and a run of collection photographs were opened before anything was written out.

// what went back

What went back

The lot came back, including the driver paperwork the firm had been anxious about for audit reasons. The repaired original board and the now-calibrated replacement both went back with the disk, on the basis that the firm had paid for one of them and might as well keep both.

One disk is £300 + VAT, and supplying the board did not alter it, because the board was never the expensive part. Board work is electronic and the published exclusion names it, so 50% fell due before the work started and the balance once the paperwork had been opened.

Before you fit a replacement board

Do not. Not because the idea is silly, but because the settings that make a board work with its own disk stay on the failed board, and fitting a substitute without them frequently makes the situation worse rather than merely failing to help. If you have already bought one, keep it and send it in with the disk, since the correct revision is genuinely useful and saves sourcing one. Do not keep powering the disk up to see whether it has changed its mind, because a board that has failed short can take a power supply with it. Leave it disconnected, put it in an antistatic bag if you have one and a plastic one if you have not, and send the disk on its own if it came out of a machine. Boards fail cleanly and predictably; it is what gets done to them afterwards that turns a routine job into a long one.

Is this your drive as well? Every file on this page began with somebody describing a noise or a message, and the route in is the same each time. Nothing is charged for finding out what has happened. That examination finishes two working days after the parcel is logged in — never sooner, because rushing it is how evidence gets destroyed — and it ends in one number, written down, which does not move afterwards. Where a single drive is recoverable it is generally back with you two to four working days later. How you pay turns on which kind of fault it is, and the wording does not vary across this network: no fix, no fee all jobs except electronic and mechanical failures, chip level work, DVR and Forensic jobs. So a deletion, a broken file system, a lost partition, a firmware or service-area repair all keep the guarantee — nothing comes back, nothing is owed. Opening a drive under laminar flow, rebuilding a board, lifting a memory chip, or anything that arrived wet or burnt does not: half the agreed figure is taken before the work, and the balance only if your files come home. The published figures run: £250 + VAT for a memory card, a USB stick or a pen drive; £300 + VAT for any one hard drive or SSD, NVMe included; £400 + VAT for a recorder disk or a BitLocker volume; from £500 + VAT for an array, a NAS or a server; and £800 + VAT for a forensic investigation that ends in a written report. Five bands, all of them public, and the only one outside the guarantee is the forensic work. On packing: if the storage is screwed inside a laptop, a tower, a Mac or a server, take it out and send the bare unit, because dismantling equipment is not work this bench does. Three things go the other way and travel complete — an external drive in its own case, a NAS, and a WD My Passport or My Book with its cable, since on those the key sits on the bridge rather than the disk. A Fusion Mac needs both of its drives, labelled. A RAID or a server sends its member disks only, numbered by bay, with the front of the chassis photographed first. Flash soldered onto a mainboard, as on Apple Silicon Macs and many thin laptops, cannot be taken in at all, and neither can phones or tablets. Post it tracked and insured to Cambridge Data Recovery, Compass House, Vision Park, Chivers Way, Cambridge CB24 9AD, or put your own courier on it; the Cambridge reception takes devices across the counter too, Mon–Fri 9:00am–5:30pm. Nothing is collected, and there is no shopfront in the county to call at. Freephone 0800 689 0668. Packing notes and the shipping form live here.
Start a free diagnostic

Each of these is a real enquiry, taken from a household or a business somewhere in Leicester or the surrounding county, with the name of the customer, the firm and anything else that would identify them taken out. What is left is the reasoning: what turned up, what had actually failed inside it, the method that fault required and the equipment it was worked on. One of the thirty-six is a job that was refused, and it belongs here every bit as much as the successes do.

// nearby on the shelf

Three more files along the same lines

See every case file →

Is this your drive as well?

Examining it is free and takes 2 working days from arrival, the figure goes to you in writing and does not move afterwards, and a logical fault carries no fix, no fee. Start online or ring the freephone.