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.
| Media | A 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 reported | Western 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 class | Electronic. 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 bench | Both 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. |
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.
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.
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.
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.
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.
A drive that dies during use, with nothing having touched it, is usually an electrical failure
A socket torn off its pads is a repair to the way in, not to anything that holds a file
Finding a deleted file and returning one that opens are two different pieces of work
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.