On Wed, 27 Jul 2011 18:36:27 +0100, "Gareth Magennis"
wrote:

Hi,

I have a Roland drum machine in for repair. The machine powers up and goes
through the Operating System upgrade, up to the point the Flash is written
to, where it fails and the unit will not boot, showing a "damaged memory"
warning. I suspect the Flash chip is faulty.


When I informed the owner of my findings, he then tells me he thinks it was
damaged by a PAT test. It is powered by a Wall Wart, and he thinks it was
plugged into the unit during the test, as it stopped working immediately
afterwards.

An internal inspection shows around 10 caps around the internal power
supplies are bulging and have leaked guff onto the PCB. The 3.3v rail is at
over 4 volts, so presumable the regulator is toast.



So, if you PAT test a Wall Wart plugged into a unit and turned on, what is
likely to happen? What I have just described perhaps?
Or did someone just use the wrong power supply instead and isn't letting on?

Possibly. In general, wall warts are class II items with no earth pin
so visual inspection is all I would do.
In the (unlikely) event that it was a Class I device with earth pin
and exposed metalwork then a voltage test would be made between line
and the metalwork and a resistance check from metal to earth pin.
I would not test a wall wart with equipment (e.g. printer) connected
as the equipment will be working on low voltage and not be subject to
testing.
I doubt that a short pulse of high voltage would cause caps to bulge
anyway.