where a transformer is desirable even if the input is
electronically balanced. For example, where there is a
signiftcant amount of electrostatic or electromagneti-
cally induced noise, particularly high-frequency high-
energy noise (the spikes from SCR dimmers, for ex-
ample), the common mode rejection ratio (CMRR) of an
electronically balanced input may be insufficient to
cancel the noise induced in the cable. In such cases,
input transformers can be useful. Also, there is incom-
plete ground isolation with an electronically balanced
input. For the ultimate in safety, there are instances
when a transformer will isolate the console ground from
the external source. Consider what happens, for ex-
ample, when a performer is touching a mic and also
touches an electrically “hot” item such as a guitar which
is electrically “live” due to a fault in the guitar amp; if
the mic is grounded, current will flow. The performer
can be subjected to very high currents, and to conse-
quently severe AC shock. If the mic is isolated from
ground, via a transformer, then that low-resistance
return path for the AC current is eliminated, and the
performer has a better chance of surviving the shock.
(In reality, the transducer capsule in a microphone is
generally isolated and insulated from the mic case, so
an electronically balanced input still would not permit a
current to flow through the mic... assuming everything
is wired correctly in the microphone.) If a transformer is
used in this way, primarily for ground isolation and to
obtain the benefits of a balanced line, it is said to be an
“isolation” transformer.
If the transformer is also used to prevent a low
impedance input from overloading a high impedance
output, it is known as a “bridging” transformer (not to
be confused with the “bridged” connections of a stereo
power amp output in mono mode).
In general, the PM4000 has no need for input
transformers since it already has electronically bal-
anced inputs. In the occasional instances where abso-
lute isolation of the grounds between the console and
the other equipment must be obtained, as cited above,
there is no viable substitute for a transformer, and an
optional input transformer kit (Model IT3000) can be
installed in individual input modules. Similarly,
PM4000 outputs can be transformer isolated by pur-
chasing one or more optional output transformer sets.
The Model OT3000 output transformer set contains
8 transformers, with XLR connectors, in a compact
19-inch rack mountable box that is external to the
PM4000. In this way, those inputs or outputs which
require a transformer can be so equipped, and it is not
necessary to pay the price, carry the weight or incur the
slight performance penalty that comes with the trans-
formers.
NOTE: There are other ways to achieve isolation. The
most common means is with a wireless radio mic. One
can digitize the audio signal and transmit it by means of
modulated light in fiber optics, but this is much more
expensive than using a transformer, with no great
performance advantage. One can use the audio signal to
modulate a light, and pick up the light with an LDR
(light dependent resistor), thus achieving isolation at the
expense of increased noise and distortion. Some systems,
such as those for hearing impaired theatre goers, even do
this over 10 to 100 foot distances using infra-red LEDs
for transmitters and infra-red sensing photo sensors for
receivers. The guitarist who places a microphone in front
of the guitar amp speaker, rather than plugging a line
output from the guitar amp into the console, has
achieved electric isolation between the guitar and console
by means of an acoustic link.
4.4.5 Noise And Losses In Low and HighImpedance LinesThe length and type of cable can affect system
frequency response and susceptibility to noise. The
impedance of the line has a major influence here, too.
Signal cables from high impedance sources (actual
output impedance of 5000 ohms and up), should not be
any longer than 25 feet, even if low capacitance cable is
used. The higher the source impedance, the shorter the
maximum recommended cable length.
For low impedance sources (output impedances of
600 ohms or less), cable lengths of 100 feet or more are
acceptable. For very low impedance sources of 50-ohms
or less, cable lengths of up to 1000 feet are possible with
minimal loss.
In all cases, the frequency response of the source, the
desired frequency response of the system, and the
amount of capacitance and resistance in the cable
together affect actual high frequency losses. Thus, the
cable lengths cited here are merely suggestions and
should not be considered “absolute” rules.
Susceptibility to noise is another factor which affects
cable length. All other factors being equal (which they
seldom are), if a given noise voltage is induced in both a
high impedance and a low impedance circuit, the noise
will have a greater impact on the high impedance
circuit. Consider that the noise energy getting into the
cable is more-or-less constant in both instances. The low
impedance input is being driven primarily by current,
whereas the high impedance input is being driven
primarily by voltage. The induced noise energy must do
more work when it drives a lower impedance, and
because the noise does not have much power, less noise
is amplified by the input circuit. In contrast, the in-
duced noise energy is not loaded by a high impedance
input, so it is amplified to a greater degree.
Page 4-14