This is only a preview of the August 1990 issue of Silicon Chip. You can view 57 of the 112 pages in the full issue, including the advertisments. For full access, purchase the issue for $10.00 or subscribe for access to the latest issues. Articles in this series:
Items relevant to "Digital Sine/Square Wave Generator; Pt.2":
Articles in this series:
Articles in this series:
Articles in this series:
|
AMATEUR RADIO
By GARRY CRATT, VK2YBX
The old vs. the new - we compare a vintage
shortwave receiver with today's technology
Ever wondered how a Collins communications
receiver from the 1950s compares with a
modern synthesised counterpart? Read on
the answer might just surprise you.
Recently, we had the opportunity
to compare one of the old-time
favourite receivers, made in the
mid 1950s, with the standard
monitor receiver in our lab, a Yaesu
FRG-7700. The results of this comparison highlight the outstanding
design and construction techniques
used in post-war equipment, and
show that these units can still
represent good value for money
today.
Most amateurs have heard of the
brand name "Collins". In the 1950s,
the name was synonomous with
rugged high-quality communications equipment. And the most
famous of all the Collins equipment
was the "S line" series of
transceivers.
Earlier models included the
"75A" series of amateur band only
receivers, and prior to that, the
"51" series of receivers, which surfaced on the surplus market in the
late 1960s.
Unearthing a relic
The particular receiver we
"discovered" is the 51J4, the last in
a series of five receivers, manufactured in the early 1950s. This prize
was unearthed during a visit to a
connoiseur's shack, where it had
•
I\
The Yaesu FRG-7700 is a modern shortwave receiver featuring phase-locked
loop tuning, digital frequency readout and memory circuits. It has better
sensitivity than the Collins 51J4 but is eclipsed in some areas.
82
SILICON CHIP
been buried for the previous 10
years. We decided that it would be
interesting to test the receiver to
determine what advances, if any,
had been made in receiver design
and construction in the last 40
years.
The first and most obvious observation to be made about the 51J4 is
its weight and size! The unit is
massive, measuring 585mm across
the front panel, 265mm high and
380mm from front to rear. It's total
weight is around 20kg. The unit
contains 19 valves or, as the
manufacturer would call them,
"tubes" and the total power consumption is 85 watts.
By contrast, the FRG-7700 measures 334 x 129 x 225mm (W x H x
D) and weighs just 6kg. So, as far as
size and weight are concerned, considerable improvements have been
made to receivers since 1950!
The specifications provided in
the manufacturer's handbook are
quite respectable, even by today's
standards. One notable specification is frequency stability - the
dial calibration is guaranteed to be
within 300Hz at room temperature
and the unit has an operating
temperature range of - 20°C to
+ 60°C. The sensitivity is quoted as
5µ,V for 10dB S/N but our unit
measured 2µ, V for the same ratio
(0.5µ, V for the FRG-7700).
The complete block diagram for
the Collins receiver is shown .in
Fig.1.
The biggest difference between
the two receivers is the complexity
of the Yaesu PLL generator, frequency counter, and memory circuits. The Yaesu receiver has a
first IF of 48MHz and a second IF of
The Collins 51J4
was built like a
battleship. It covers
from 540kHz to
30.5MHz in 30
1MHz bands,
features
mechanical tuning
and contains 19
valves. The unit
weighs about 20kg
and has a total
power consumption
of about 85 watts.
455kHz. Synthesised local oscillators are used to ensure good frequency stability, although the end
result is only as good as the Collins
unit!
The Collins receiver
Basically, the 51J4 is a superheterodyne unit covering 540kHz to
30.5MHz. This tuning range is
covered in 30 1MHz bands. The
receiver also contains AGC circuits, a noise limiter, and a lO0kHz
crystal calibrator. The first IF is
either 2.5MHz or 3.5MHz and
depending upon which band is being received, single, double, or triple conversion may be used. To
enable each 1MHz band to be tuned, a highly stabilised permeability
tuned oscillator is used to
heterodyne against the first IF. This
oscillator is fed from its own DC
regulator to ensure stability.
The second IF is at 500kHz and
the receiver is equipped with both
.NOT[:
BANDS 4-30
EV EN BANDS-INJ. fREQ=B ANO I-JO +2
ODD BANDS-INJ . l='REQ: BAND NO+~
Fig.1: block diagram of the 51J4. It features permeability tuning,
mechanical filters & either single, double or triple conversion.
mechanical and crystal filters at
that frequency .
Mechanical filters
Mechanical filters are electromechanical bandpass devices
whose dimensions are about one
quarter the size of a cigarette
packet. As shown in Fig.2, a bandpass filter consists of an input
transducer, a resonant mechanical
section consisting of a number of
metal discs, and an output transducer. The frequency characteristics of the mechanical section give
almost rectangular selectivity
curves as shown in Fig. 3.
An electrical signal applied to
the input terminals is converted into a mechanical vibration at the input transducer by means of a process called ''magnetostriction''.
This mechanical vibration travels
through the resonant mechanical
section to the output transducer,
where it is converted again by the
magnetostriction process to an electrical signal, which appears at the
output terminals.
In order to provide the most efficient electromechanical coupling, a
small magnet in the mounting above
each transducer applies a magnetic
bias to the nickel transducer core.
The electrical impulses then add to
or subtract from this magnetic bias,
causing a vibration of the filter
elements which corresponds to the
AUGUST 1990
83
5
I
10
I
15
I
20
I
~
25
dB I
COUPLING
ROO
\
MAGNETOSTRICTIVE - -------1
DRIVING ROD
30
I
35
I
40
I
45
I
COIL
50
I
ELECTRICAL SIGNAL
(INPUT OR OUTPUT)
ELECTRICAL SIGNAL
(INPUT OR OUTPUT)
Fig.2: inside a mechanical filter. The input signal is converted to
mechanical vibration by magnetostriction and travels via a number or
resonant discs to the output transducer.
exciting signal. There is no
mechanical motion except for the
imperceptible vibration of the metal
discs.
Magnetostrictively driven mechanical filters have several advantages over electrical equivalents. In
the region of 100-500kHz, the
mechanical elements are extremely
small and so a mechanical filter
with better selectivity than the best
of conventional IF systems can be
enclosed in a package smaller than
a crystal filter.
In addition , the frequency
response characteristics of the
mechanical filter are permanent no alignment is necessary or possi-
ble. Unfortunately, cost and advances in modern crystal filter performance mean that mechanical
filters are rarely seen in these days.
Another major drawback is that
mechanical filters are limited in
frequency to 500kHz or so, while
quite reasonable crystal filters can
be made for frequencies up to
30MHz or so. However, considering
the age of such devices, their performance is unbelievable.
Phasing control
Another most useful feature in
the 51J4 is the "phasing" control.
This allows control of a variable
notch filter within the IF response
The circuit techniques may be ancient but they still add up to excellent
performance. They sure don't make 'em like this anymore!
84
SILICO N CHIP
kHz
Fig.3: typical response curve of
a mechanical filter. The result
was excellent selectivity.
curve, allowing unwanted heterodyne s from adjacent channel
signals to be notched out. This
feature is found only rarely in
shortwave receivers and is also
seldom used in amateur receivers.
One of the tests we used to determine the ability of a receiver to process strong signals is called the
"blocking test". For this test, we injected a signal on the desired frequency, and then injected a second
interfering signal 3kHz away. The
amount of interfering signal required to degrade the SINAD
reading of the desired frequency by
3dB determines the receiver's ability to withstand a "blocking" signal.
Fig.4 shows our test layout.
Most modern receivers have
rather poor performance in this
area. The "fix" on these receivers,
[including the FRG-7700) is to provide an attenuator which can be adjusted by the operator to minimise
the effect of strong signal overload
(and also receiver sensitivity).
In a well designed receiver like
the Collins 51J4, no attenuator is
necessary due to the excellent
design of the RF, IF and AGC
stages. In fact, our unit exhibited a
30dB improvement over the FRG7700 in this critical area.
There was one area where our
vintage receiver lacked in performance and that was SSB reception.
This is quite important for amateur
service these days but perhaps not
so in the 1950's. Fortunately, the
subject of improving SSB performance was covered by the
American magazine Ham Radio in
their February 1978 issue.
Basically, this issue detailed the
modifications required to change
the BFO to a product detector, and
to change the AGC attack and
decay time constants to allow correct reception of SSB. Some audio
stage modifications are also detailed. After we implemented these
recommended modifications, the
"DESIRED
FREQUENCY"
SIGNALGENERATOR
SINAD
METER
COLLINS 51J4
RECEIVER
SIGNAL
COMBINER
"INTERFERING"
SIGNAL
GENERATOR
Fig.4: the test layout used to determine a receiver's ability to
withstand a "blocking" signal. The 51J4 was considerable better
than the FRG-7700 in this important area.
C206
5pF
TD
------...11--<)DETECTDA
V110A
T106
,-------------7
I !JI
I,{;[
I 1
II
___ ._.....
I
I
I
I
B+
I
BFD
PITCH
I
I
C218
.01
12
L - - - _ _ _ _ _ _ _ _ _ _I
.,.
Fig.5(a): this is the original 51J4 BFO circuit. The tube is turned off by
switch S112 which short circuits the screen voltage to ground.
.05
TD PINS 6,7
V110A
AM
o---
~--
,,
S1128
('
5pF1
TD PIN 3
XV112A
VIA C209
-----
I
I
TD R154
(VOLUME)
I
I
I
I
I
I
n------xo_1..,
" ' ~ _______ ~,:/_ _ __
ovI1-
SJV.,.
01
5
J
1N270
TD PIN 3-""41t--\l\Mr--l-.,_-1---.----------<--- TD R171
V110B
AND XV111A
C205B
1-1.5 +
1M
0.1
l
100VJ
Fig.5(b): the revised circuit substitutes a 6BE6 tube for the original 6BA6 to
provide a product detector. The AGC attack and decay time constants are also
modified as shown to allow SSB reception.
receiver performed equally as well
as the FRG-7700 on SSB signals.
Fig.5 shows some of the necessary circuit changes.
In summary, considering the age
of this receiver and the techniques
used, its performance is amazing.
Although the sensitivity is somewhat worse than modern receivers
[and this can be overcome with a
good HF antenna), it outperforms
modern receivers in many other
respects. It's also worth noting that
the price of a Collins 51J4 on the
secondhand market is similar to the
price of modern pre-loved Japanese
receivers.
If size and digital frequency
readout are not of major importance and a few valves can be kept
as spares, the Collins 51J4 receiver
should continue to provide good
performance for many years to
oome.
~
A UGUST 1990
85
|