Table of Contents
Advertisement
JBUUREI
MODEL
7110
peak-to-average
ratio
of
3
dB
—
^to
the
human
voice, with
a
peak-to-average
ratio
of
around 10
dB
(depending
highly
on
the
particular
voice)—
to
the
plucked
string
instruments
and
electronic synthesizers,
which
may
have
peak-to-average
ratios of
20
dB
and
higher.
Depending on what
we
are
trying to
achieve
with
our compressor,
we may
need an
average
responding
type,
peak
type
or both.
In
the
7110
we
do have
both.
If
we
are using
a compressor
to
match
the perceived
levels
of different
sounds
in
an
overall
mix,
we
will
probably
be
primarily
concerned
with
the
average
level,
because
that
approximately
matches
the
response
of
the
human
ear.
Peak
signals
of
short duration
may
be
literally
too
fast to
be
heard.
But,
when
we
get
these
peak
signals
into
an
audio system,
they
may
give
us problems.
The
first is
overload, the
second
is
noise.
Overload occurs
because each
piece
of
electronic
equipment
in
our audio chain
is
powered
by
a
DC
voltage.
If
our audio
signal
attempts
to
exceed
that
DC
voltage,
it
cannot.
If
this
occurs
for
an
extended
period
of
time,
we
will
definitely
hear
it.
But
we may
hear
it
after
only
a
very
short
time,
depending on
the
design
of
the
equipment,
and what
additional
problems
the overload
causes.
These
may
range from
mild
to
severe.
Severe examples
include:
severe
distortion;
latch-up;
overmodulation
of
radio transmitters with resulting interference to adjoining
frequencies,
severe
distortion
or signal drop-out;
excess
modulation
of
record
cutting
styli
causing groove
cut-through;
massive
overload
of
digital
systems,
etc.
The
first,
and most
obvious,
remedy
to
overload
is
to
reduce
the signal
level.
This
rriay
work
in
some
cases,
and
not
in
others.
The
major problem
with
reducing the
signal
level
is
that
it
degrades
the signal-to-noise
ratio of
the
system.
If
we
only
need
to
turn
it
down
by 3
dB,
that
may
be
acceptable. But
a 20
dB
decrease
will
likely
not
be-especially
if
the offending high
amplitude
signal
is
present
only occasionally.
We
could
try
to
coritrol
this
by hand,
but,
if
the
situation
occurs
with
regularity,
some
type
of
automatic
control
will
provide
more
predictable
and
consistent
results.
As
we
discussed
earlier,
the difference
between peak and average
is
time.
Now we
can
begin
to
see where
we
want an average
responding
compressor
versus
a peak
responding
type.
If
our
program
material
is
a
vocalist,
and
the
vocalist
is
weaving and bobbing around
the
microphone,
there
may
be
some
major
variations
in
the
electrical
output
of
the
microphone.
These
variations
may
be so
great
that
the
sound
of
the
vocalist
is
lost
beneath
some
accompaniment
at
one moment, and
completely
drowns
out the
accompaniment
at
the
next
moment.
This
is
not
artistically
satisfactory.
We
would
like
a
better
balance
between
the
vocal
and
the music.
What
we
have
here
is
a problem
with the
average
level of
the
vocalist-
a
job
for
the
average
limiter.
Or
take another case:
we
have a saxophone
that
we
can't turn
up
loud
enough
to
balance
in
our
mix,
because
the
peak-to-average
ratio
of
the
sax
sound
is
so
great.
If
only
we
could
cut
down some
of
those
peaks
that
the ear doesn't
hear...
And now
we
have a
job
for
the
peak
limiter.
The
front
panel
control labelled
Detector
is
the
means
by which
we
choose
predominantly
average
or
peak
compression,
and
in
what
balance.
Think
of
it
as a
second
threshold
control.
With
the
control set
full
CCW
to
the
Average
mark, the threshold
for
peak
limiting
is
20
dB
higher than
that
for
average
limiting.
This serves
as a
sort of
"safety net" allowing small
amounts
of
average compression
(which
typically
sounds
better
than
large
amounts)
to
be
used
while
still
retaining protection
against high amplitude,
short duration signals
As
the
control
is
rotated
clockwise toward the
Peak
setting
the threshold
for
peak
signals
is
progressively
reduced.
You
therefore
have
very
good
control
over
the actual peak-to-
average
ratio
of
the
program
material.
Page
12
Advertisement
Table of Contents
