Electrical damage · job log · PLY-2025-0628
The Supply Was New. The Lead Was Anything But.
One press of the button gave him a bang, and then the smell of electronics burning
. A new graphics card had asked for more current, so a wedding filmmaker in Plympton fitted a larger supply to his edit machine and kept the drive lead already in there — cable-tied inside that case since the day I built it
. Nothing about the plug looked any different. Three films remained undelivered, and the WD Blue 4TB with that summer's work on it has not made a sound since.
Sounds like yours? Talk it through.
0800 6890668
Why it happens.
Two plugs can be indistinguishable by eye and wired to completely different schemes. On the drive side the pin order is settled by specification, so that half of any lead is safe wherever it goes. The far end belongs to the manufacturer: it varies between ranges, it varies again once a range is superseded, and nothing on the outside tells you which is in your hand. The lead went in, and pins expecting five volts got twelve. The board then behaved exactly as it was designed to. Its TVS diode shorted itself, dragged the rail to earth, burnt through doing it, and let nothing past into the drive.
The kit this job called for.
How the job runs →| The kit | What it did here | Why we keep it |
|---|---|---|
| ACE Lab PC-3000 Express | Established that the damage stopped at the board before any power was applied | Drops the disk into its maker's service mode, out of reach over plain ATA |
| PC-3000 Data Extractor | Took every sector the moment it reported itself ready | The PC-3000 runs the copy through the drive's own firmware |
| R-Studio Technician | Lifted the NTFS tree from the image and set it against his booking diary | Reads nearly every file system, and rebuilds RAID sets from images |
How it ran.
How far the spike actually travelled
Ten minutes with a microscope and a current-limited supply behind the board told us the whole of it: a regulator burnt out, a shorted protection diode, scorching around both. What would have finished this job is the preamplifier, which lives within the sealed body on the head stack itself, and nothing electrical had come near it.
A matching board, and the one chip that has to move
A board of matching revision was on the shelf, and its own eight-pin ROM was left in place. Off the burnt board came the chip, and onto the donor it went. It is that chip which marries a board to one specific drive: the values it holds were taken from this particular head stack and these particular platters when the unit was built. Put a board on without it and the drive spins up, sounds thoroughly healthy and returns nonsense.
Take the copy while it will still give one
With three deliveries already in the diary, the Data Extractor was wired in and standing ready before the rebuilt board went on. None of it had been felt by the mechanics. Read speed stayed level through the night, no new defects turned up, and the image was complete by morning.
How it ended up.
All three edits came off the copy undamaged, and each couple received their film on the date they had been given. Every lead belonging to the old supply travelled back in the same parcel, taped up as one labelled bundle, so that not one of them can find its way into another machine.
Pages people open after this.
Other power-fault cases.
Seeing the same thing?
Leave it switched off and post it in. Nothing happens until the diagnosis, which lists what can still be read and what cannot.