INSTRUCTION MANUAL
I
Direct
analysis
of
sound signals
Used solely
as
an
amplifier
the
NAG
RA
IV-SJ
is
a
precision sound level meter,
which
operates
with
any calibrated
microphone
cartridge. The sound
pressure measurements must
be
taken in relation
to
the 0 dB reference level,
which
corresponds
to
a
pressure
of
0.0002
Ilbar, i.e.
201lN/m
2
or
20
IlPa.
This level coincides in practice
with
the
audibility'
threshold
of
the human ear
at
1
kHz;
at
this fre-
quency, the sound level can
be
expressed in phones,
the phones value being equal
to
the
dB
value read
in relation
to
the
0.0002
Ilbar
reference.
The frequency response
of
the human ear
is
not
linear and depends very much on the sound level; a
sound level meter
must
therefore include weighting
filters
which
modify
the
values measured in accor-
dance
with
certain criteria,
so
that
they
can
be
interpreted in relation
to
the subjective impression
felt
by
the human ear. There are three standard
weighting curves,
A,
Band
C.
Curve A
is
used
for
low
and
medium
sound levels,
which
are the sub-
ject
of
most analyses; curves
Band
C relate
to
higher levels. Curve D relates
to
contour-lines
of·
perceived noisiness, in
particular
with
noise pro-
duced
by
aviation:
this
weighting
is
used
for
the
measurement
of
sounds
which
cause
annoyance
in general. Curves A,
B,
C and D are shown
after
the specifications.
The sound level measured
is
shown on a dual
galvanometer,
with
one needle per channel. Its
measuring
circuit
determines the average value
of
the signal on
RMS
and its peak value on
PEAK,
with
different
dynamic
characteristics in
relation
to
the
integration
time.
On
RMS
FAST
integration
time
is
200
ms: a signal at 1
kHz
lasting
200
ms
gives a reading 1 dB
lower
than
that
which
would
correspond
to
the steady signal. On
RMS
SLOW, integration
time
is
500
ms
and
a signal lasting
500
ms
gives a reading 4 dB
below
that
which
would
correspond
to
the steady signal.
These
two
integration times are in accordance
with
Measurements
the
CE
I standard
179
for
precIsion sound level
meters.
Naturally,
any
signal
which
is
shorter than
the
integration
time
will
be
shown
below
its real
value.
For
analysis
of
pulse signals
or
strong
transients a more useful reading
is
obtained on
PEAK,
i.e. peak value
with
an
integration
time
of
only
5 ms. This very fast reading
cannot
be
used
because
it
causes
visual fatigue: in
order
to
over-
come
this
disadvantage the measuring
circuit
holds
the signal
for
about
one second,
thereby
increasing
the
fall
time
of
the needle.
Recording and
playback
of
the
sound signals
In all
cases
where
direct
measurement
is
not
sufficient
for
studying
the
signals picked
up
by
the
microphone,
these signals can
be
recorded on mag-
netic
tape and analyzed later in
the
laboratory.
While the
two
tracks used
for
direct
recording are
in
use,
the
third
track
can record FM signals
for
synchronization,
a
commentary,
or
a measuring'-
signal
from
D.C. up
to
4
kHz.
The three tracks
thus store signals in
fully-synchronized
form,
which
is
an
important
factor
for
analysis.
However,
this
storage
is
restricted
to
some
extent
due
to
the
limitations
of
present magnetic tapes.
These
limitations
concern the
following
charac-
teristics in
particular:
the response
cu
rve
Attenuation
at
the
upper and
lower
extremes
of
the spectrum depends on the speed used.
Very
low
frequency
signals,
which
are
difficult
to
play
back
with
a small reproduce head, should
be
analyzed
by
frequency transposition.
distortion
This increases very
rapidly
as
soon
as
the
maximum
recording level
is
exceeded, and tape saturation
occurs, generating
harmonic
frequencies (in par-
ticular
2nd and
3rd
harmonics),
which
falsify
the
analysis
of
the signals.