Electrical damage · bench log · CDF-2025-0628
New Supply. Old Lead. Cooked Board.
A new graphics card wanted more power, so a wedding filmmaker in Adamsdown fitted a new supply to his edit machine. Three couples' films were still to cut. He left the drive lead in place — cable-tied along the inside of the case for years
— and to the eye the plug was identical. The first press of the button brought a bang, and then the smell of cooked electronics
. Since then the WD Blue 4TB with that summer's weddings on it has not made a sound.
Does that match yours? Call us.
0800 6890668
Why that happens.
Only one end of a modular lead is standardised. The end that plugs into the drive itself is set by specification. The plug at the other end is not: each maker wires its own pin order for a given range, and the range that replaces it can be wired differently again, with no warning and no visible difference. That is why a lead slides happily into a socket it was never made for. Here, the pins expecting five volts were handed twelve. What came next is the board doing its job. The TVS diode did what it is there for: it failed short, dropped the rail to ground and burnt out, and the surge never got past it.
What we used on this one.
Every step, in order →| The tools | What it did here | Why it helps |
|---|---|---|
| ACE Lab PC-3000 Express | Confirmed the damage stopped at the board, then brought it up | Opens the maker's service commands, which a plain ATA link will not pass |
| PC-3000 Data Extractor | Imaged the full surface the moment it reached ready | Imaging driven from inside the drive's own firmware, by the PC-3000 |
| R-Studio Technician | Read the NTFS tree from the image, checked against his diary of bookings | Reads nearly every file system, and rebuilds arrays you can rely on |
How it went.
Where the spike stopped
Under a microscope, on a current-limited bench supply, the board gave up its story in ten minutes: a burnt regulator, a protection diode shorted straight through, scorching around both. The part you never want to find damaged is the preamplifier — sealed inside the drive body, mounted on the head stack itself. The surge had not reached anywhere near it.
A matching donor board, and the chip you must move
The donor was an identical revision, but its eight-pin ROM stayed behind — the chip off the burnt board went across instead. That chip is what ties a board to one drive and no other; the figures held in it were measured from this head stack and these platters. Without it the drive still spins up and sounds perfectly well, and every sector it hands back is nonsense.
Image it while it still runs
Three couples were waiting on delivery dates, so the Data Extractor was already wired in when the rebuilt board brought the drive up. The mechanism had felt none of the surge: read speeds stayed level, no new defects turned up overnight, and the full image was finished by morning.
Signing it off.
All three projects came off the image whole, and every film reached its couple on the date they had been given. The leads from the old supply went home in one bundle with the drive, banded and labelled so nobody plugs them into anything again.
Pages people often read next.
More from the power cases.
Is that what yours is doing?
Power it down, get it to us, and hold off on any decision until the diagnosis says what can still be read.