Re: [LAU] ground loop hell

From: jim <jim@email-addr-hidden>
Date: Sun Feb 08 2015 - 22:20:30 EET

( hoping this helps someone(s) )

     Best, or at least good, practice suggests ensuring you use a
single earth ground. Practically this means using a single
receptacle with a power strip with enough receptacles for all
your equipment. This model forces separate ground paths for
each device to a single, common grounding point. (15 amps
at 110VAC is usually enough power for a band and PA and
recording equipment; think of the entire setup as a single
device comprising amps, mics, outboard equipment, mixers,
and so forth: all one frankenstein device.)

     Given one single common receptacle for all power, then
the earth ground wire properly connects to the frames of
the devices so that the cases act as a Faraday cage, capturing
EMF (e.g. from fluorescent lamps) and shunting that kind of
noise to the earth ground point on the ground (green or bare
copper) wires not shared by the devices' power supplies.

     Note that both balanced and unbalanced signal cables
connect to device frames, and therefore to earth ground.
     In the case of balanced cables, the frame connector (the
sleeve for phone plugs and pin 1 for Cannon XLR three-pin
cables) extends the Faraday shield of the frame to the sheath
conductor over the two signal wires.
     In the case of unbalanced cables, there is only one signal
wire (tip of a phone or RCA plug); the sheath conductor
provides a current path for both the signal return and the
earth ground.
     Balanced is better, if possible.

     Regardless of balanced or unbalanced, there is a nuance
to keep in mind: ground loops.
     A ground loop is a current path to two different ground
connections. Ground loops can be the reason noise enters
the signal paths; ground loops may also be the reason for
annoying shocks to humans.
     For electrical power connections, using two different
receptacles risks having two different ground points (even
if ultimately the grounding wires connect to the same earth
grounding lug). If there is a difference in length or the
number of connections in the two ground wire paths, then
it is likely that some impedance difference is presented to
each of the receptacles, and therefore a voltage difference
between the two receptacles' ground lugs.
    For signal connections, consider the case of routing a
signal from some source to some device and from that
device to another device (chaining devices). Note that the
frames of all devices are connected through the signal
wires' sheath conductyors. If each device has a separate
power connection, there may be an impedance (and
therefore, voltage) difference between the ground lugs of
the two devices, and therefore between the two devices'
frames. If so, there may be electrical current between the
two frames. This current may be that of low voltage but
high frequency noise that may bleed into the power supply
and thence to the signal circuitry, or it may be high,
out-of-phase power voltage current (which likely will trip
the breakers or fuses--if not, this presents a truly
dangerous shock hazard to humans).
     If you're using unbalanced cables, then your only fix is
to lift the ground of the power cables for all but one device
so that all the other devices must make their connections
toground through the sheath conductor of the signal cables.
     If you're using balanced cables, then you can either use
the above approach or you can add an insulating device to
the signal cables' sheath ground such that the signal cable
connects all three pins (or tip-ring-sleeve) to one device but
only the two signal pins (or tip-ring) to the other device.
This lets each device extend its frame grounding over its
inbound cable (or outbound cable, either one, not both),
and the ground connector at that device's output cable
connection has insulation to disconnect that device's frame
ground from the sheathing of its outbound cable (which in
turn gets connected to the next device in the chain and its
frame). Thus any differences in frame ground voltages
between devices have no way to flow to adjacent devices
(e.g. microphones).

On 02/08/2015 10:57 AM, Ralf Mardorf wrote:
> On Sun, 08 Feb 2015 12:46:20 -0600, Jan Depner wrote:
>> I've been doing that for the last 5 or 6 years as well. It also keeps
>> you from getting the crap shocked out of you when you're playing guitar
>> and singing ;-)
> I've got ambivalent feelings with ground on the guitar strings. I kept
> the ground since 32 years on my electric guitar's strings. Touching the
> guitar strings and at the same time touching a mixing console's or
> synth's metal case or kissing a microphone, several times was
> unpleasant.
> _______________________________________________
> Linux-audio-user mailing list
> Linux-audio-user@email-addr-hidden
> http://lists.linuxaudio.org/listinfo/linux-audio-user
>

_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@email-addr-hidden
http://lists.linuxaudio.org/listinfo/linux-audio-user
Received on Mon Feb 9 00:15:03 2015

This archive was generated by hypermail 2.1.8 : Mon Feb 09 2015 - 00:15:03 EET