is
automatkally brought into
use
when a pair
of
high-resistance headphones are plugged into the
CRYSTAL CHECK jack socket; with the aid
of
the
headphones, the difference frequency between
the basic oscillator and the harmonic multiples
of
the calibrator's
333·
3 kc/s can be monitored aurally.
Because the outputs on the four higher-frequency
bands are all derived directly from the multiplier
chain, their frequencies have an exact integral
relationship to the frequency
of
the basic oscillator.
It
follows, therefore, that setting the COARSE TUNE
controi to bring the basic-oscillator frequency to
that
of
a crystal harmonic will also bring the fre-
quency
of
the outputs from the multiplier chain to
a known relationship with the crystal harmonic,
and allow the frequency dial
to
be standardized
with a high degree
of
accuracy.
Outputs on the lowest-frequency band are not
derived directly from the multiplier chain; their
generation involves a heterodyne action between
the
27-
to 55-Mc/s multiplier and a 30-Mc/s fixed
oscillator which
is
not locked to the basic oscillator.
For
this reason, although use
is
made
of
the crystal
calibrator when setting up for
1,5-
to 13'5-Mc/s
outputs, the accuracy
of
standardization
is
of
a
lower order than
that
obtained
on
the four higher-
frequency bands.
(b) Check-Point Frequencies
The calibrator provides a total
of
56
check pOlnts
between
13·5
and 220 Mc/s; these occur as
follows:-
Band
2,
13·5
to 27·5 Mc/s:
at
all multiples
of
l Mc/s from
14
to
27
Mc/s inclusive.
Band
3,
27
to
55
Mc/s:
at
all multiples
of
2 Mc/s
from
28
to
54
Mc/s inclusive.
Band
4,
54
to
110
Mc/s: at all multiples
of
4 Mc/s
from
56
to
108
Mc/s inclusive.
Band
5,
108
to
220
Mc/s: at all multiples
of
8 Mc/s from
112
to
216
Mc/s
inc1usive.
(c) Standardization Procedure
As shown above, the calibrator allows the fre-
quency scale
to
be
checked at
14
different points
on each
of
the above bands, and the adjustable
cursor can
be
set
to
correspond exactly with any
one
of
these points.
When the Signal Generator
is
to
be
used above
13·5
Mc/s
to
provide
an
output at a single spot
frequency,
or
over a narrow band
of
frequencies,
the cursor should
be
set up at the nearest crystal
check point.
When the Signal Generator
is
to
be
used over a
wide range
of
frequencies, and it
is
inconvenient to
OM 99SA/S
SECTION
2
reset the cursor for each material frequency change,
or, altematively, when using the 1·5· to 13'5-Mc/s
band, the procedure
is
varied to reduce the mean
error to a minimum. The method
of
standardizing
the frequency scale for subsequent general use
is
as
follows:-
(l)
Set the INC. FREQ. controis to zero and the FINE
TUNE
controI to mid-position.
(2)
Set the RANGE switch to 13'5-27 Mc/s.
(3)
Using the headphones plugged into the CRYSTAL
CHECK
jack, tune the main dial to a crystal
check point near the centre
of
the band; e.g.
20
Mc/s.
When using the calibrator, the
MOD. SELECTOR
must
be
set to a position other than INT.
MOD.-
F.M.
or
EXT.
MOD.-F.M.
This ensures that the
variable oscillator
is
not being frequency modulated
-a
condition which prevents precise setting
of
the
COARSE TUNE controi for the lowest-frequency beat
note in the headphones, since it gives rise to a
fluctuating tone.
After using the calibrator, the Signal Generator
can,
of
course, be set up for f.m. without invali-
dating this frequency standardization.
(4)
Adjust the milled boss in the centre
of
the dial
to bring the cursor exactly in line with the
calibration mark corresponding to the crystal
check point.
If
the Signal Generator has been out
of
use for
some time,
it
may be necessary to use a coin in the
slot provided in order
to
rotate the milled boss.
(5)
Check the calibration accuracy at several crystal
check points both above and below the check
point at which the cursor was set in
(4)
above.
(6) Readjust the cursor setting to equalize the errors
over the band; e.g. it might be found that, with
the frequency scale indication correct at
20
Mc/s, the indication was high
at
both
15
and
25
Mc/s-in
such a case, the errors would
be
equalized by making the indication a little low
at
20
Mc/s, and thus not
so
high at
15
and
25
Mc/s.
It
will
be noted that, in the above procedure, the
frequency scale
is
standardized on the 13'5- to
27-Mc/s band. This band
is
specified since its
corresponding scale calibrations occupy the longest
arc on the dial. The dial can therefore
be
read with
a high degree
of
diserimination on this band and
the correct cursor setting most easily determined.
Once the frequency scale has been standardized
on
the
13·5-
to 27-Mc/s band, the cursor
is
correctIy
set to give the minimum mean error on the other
three direct-multiple bands.
It
is
also correctly set
for the
1,5-
to 13'5-Mc/s band.
11
1-1/61