Re: [LAU] Ground lifts, TRS, etc. to PA? USB/other signal crossover to audio?

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
To: <linux-audio-user@...>
Date: Saturday, May 10, 2014 - 1:51 am

On May 9, 2014 12:44:00 AM Jörn Nettingsmeier wrote:

Although it might not relate directly to the OP's question,
since the safety topic came up I would like to relate a story about
one reason you should never remove power ground pins and make
sure your outlets are grounded:

Some relevant info:

Grounding is important not only for day-to-day usage, but in case of
unexpected catastrophic failure.

So this high-power powered PA speaker decides to SMOKE in the
middle of a practice.

Diagnosis, and a theory of what caused it:

A component on the Switch Mode Power Supply board's +15V rail short-circuited.
The short still exists, haven't tried to find the component yet,

It appears that, at the exact moment of the short, the SPMS wanted to dump
a lot of current into the chassis ground while things went haywire
for a moment and the driver was still running and things shut down
and caps discharged and so on.

But here, the Input Board, which is fed by the +/- 15V power rails,
was DESTROYED. The tops of all ten or so 8-pin op-amp ICs were
completely BURNT and CHARRED open.

More tellingly, in this model there's a 10-ohm surface-mount ground RESISTOR
on the Input Board, connecting INPUT signal ground to CHASSIS ground.
The resistor, and the circuit board area, were completely BURNT and CHARRED.
(A mixer was connected to this PA, on the same AC circuit)

I could only conclude there was no CHASSIS (AC) ground, or it was breached
by a bad or loose cord.

If AC ground had been intact, the catastrophic current would have been dumped
into the AC ground and there would have been little or NO potential difference
across the 10-ohm ground resistor to cause THAT amount of damage to it.

The resulting wackiness in voltage swings could be what destroyed all those
Input Board op-amp ICs.

So the current wanted to flow into the AC ground but instead took the
easiest path it could - through the surface-mount 10-ohm ground resistor,
back through the INPUT cable and into the MIXER ground.

(I checked: The mixer has no such resistor. Signal ground = chassis ground.
The mixer still works. Still, I advised the owner to change his input cable.)


Linux-audio-user mailing list

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
[LAU] Ground lifts, TRS, etc. to PA? USB/other signal crosso..., Jonathan E Brickman, (Thu May 8, 7:31 pm)
Re: [LAU] Ground lifts, TRS, etc. to PA? USB/other signal cr..., Jörn Nettingsmeier, (Thu May 8, 10:44 pm)
Re: [LAU] Ground lifts, TRS, etc. to PA? USB/other signal cr..., Tim E. Real, (Sat May 10, 1:51 am)