Raimund wrote:
>=C2=A0> I mean that the microphone is drawing current in an unbalanced way=
.
and David wrote:
>=C2=A0but if the mic is unbalanced, or has one leg referenced to ground, o=
r has=C2=A0
> different + and - impedances, then phantom power noise will no longer can=
cel=C2=A0
In what follows, I'm not trying to start an argument, but rather have a dis=
cussion on what's causing the observed problem.
I think that we are all speaking of microphones which have unbalanced signa=
l outputs but the output impedance is intended to be balanced. =C2=A0There =
are numerous examples such as the Oktava MK-012 and the DPA 4011. =C2=A0The=
circuitry of each of these drives only pin 2 of the XLR but pin 3 is termi=
nated by a resistor in series with a capacitor to mimic the output impedanc=
e of the active side. =C2=A0Power for the microphone is derived by equal va=
lue resistors from pin 2 and pin 3 to a filter circuit to reduce the noise,=
thus determining the DC part of the output impedance. =C2=A0These two exam=
ples should draw equal current from each half of the phantom power circuit.
Not only that, but in every phantom power implementation that I've seen, th=
ere is a single 48 V supply which then feeds two 6k8 resistors attached to =
pins 2 and 3. =C2=A0Drawing more or less current from the phantom supply mi=
ght result in a change in the amount of noise in the phantom supply, but it=
should still be the same in each leg.
So the only means that I can see by which microphone imbalance could cause =
an increase in observed noise is if there is an imbalance in the AC compone=
nt of the differential output impedance of the microphone.
From: " [naturerecordists]" <=
ps.com>
To:
Sent: Tuesday, May 13, 2014 4:05 PM
Subject: [Nature Recordists] Re: Fostex FR-2LE ?
=C2=A0
> I mean that the microphone is drawing current in an unbalanced way. As l=
ong as it draws exactly the same current from both legs, a potential noise =
component on the phantom power voltage will be rejected successfully by the=
(differential) preamplifier.
Eric,
Mics can be simple 2-wire unbalanced, floating 3-wire where the mic circuit=
is not connected to anything but the input + and - signals, or fully
balanced where the mic output is symmetrical about ground as with a centre=
tapped transformer or the equivalent circuit.
Passive moving coil mics are often fully floating, but anything needing
power also needs a power negative connection.
Phantom current power is supplied from a fairly clean medium-Z source
equally applied to + and - signal connections, fully balanced both ends, bu=
t
- here is the problem - the power return has to be through the grounded
screen. The mic capsule can't now be fully floating, but has to be balanced=
to eliminate power supply noise.
Normally any noise from the phantom power cancels out on the + and - inputs=
,
but if the mic is unbalanced, or has one leg referenced to ground, or has
different + and - impedances, then phantom power noise will no longer cance=
l
out.
Any mic which doesn't have fully balanced outputs will pick up any
(unbalanced) noise on the phantom powering. The noise requirement for
unbalanced power is of the same order of mic noise down near -120dB.
David Brinicombe
|