This is only a preview of the February 2003 issue of Silicon Chip. You can view 27 of the 96 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 "The PortaPAL Public Address System":
Items relevant to "The SC480 50W RMS Amplifier Module; Pt.2":
Items relevant to "Using Linux To Share An Optus Cable Modem; Pt.4":
Items relevant to "A Windows-Based EPROM Programmer; Pt.3":
Purchase a printed copy of this issue for $10.00. |
VINTAGE RADIO
By RODNEY CHAMPNESS, VK3UG
Aligning the front-ends of TRF &
superheterodyne receivers
In the last two issues, we’ve looked closely at
the IF stage in superhet receivers. This month,
we take a look at the RF (radio frequency)
circuits in TRF receivers, the local oscillator
stage in superhets and at receiver alignment.
TRF (tuned radio frequency) sets
were the most common receivers in
the 1920s but their popularity rapidly decreased as the superhet became
dominant in the mid 1930s. However,
some manufacturers still produced
cheap and simple TRF sets that could
be used in high-signal strength areas.
A typical set of this type is the Astor
“Football” from the 1940s.
TRF radios had from one to four
tuned stages. Initially, it wasn’t nec-
essary to have the stages track one
another as individual variable capacitors were used to peak each station.
However, this was quite a chore and
the radio manufacturers soon decided
to mechanically couple all the variable
capacitors together in the more elaborate receivers – eg, by using brass
straps and dial drums.
This made tuning much easier but
it also meant that the sets had to be
designed so that each stage tuned to
This photo shows two
different twin-gang
tuning capacitors. The
one on the right is a
padderless type.
80 Silicon Chip
the same frequency across the dial –
ie, the stages had to track each other.
However, in many cases, no attempt
was made to accurately align each
stage for best performance.
Later on, a trimmer capacitor was
placed across each tuning capacitor.
This meant that all tuned circuits
could be set to the same frequency
at the high-frequency end of the dial,
thereby ensuring the best performance
possible.
By contrast, accurate tracking at the
low-frequency end of the tuning range
relied heavily on the accuracy of the
winding of the tuning coils. This was
helped by the fact that distributed and
stray capacitance around the tuned
circuit was less of a problem than at
the high-frequency end of the tuning
range. And as time progressed, manufacturers were able to wind the coils
with quite high accuracy, which meant
that it wasn’t really necessary to later
adjust the inductances.
However, I have found that almost
all air-wound RF and antenna coils
give improved performance at the low
frequency end of the dial if an adjustable slug is incorporated in the coil
design. Some older coils suffer from
moisture ingress over the years and
this causes the inductance to alter and
the quality factor (“Q”) to decrease,
thereby lowering the gain of the stage.
In order to assess alignment accuracy, it is best to measure the effect
of any adjustments on the receiver. If
the set has AGC, a digital multimeter
(DMM) connected across the diode
load or AGC line will indicate peak
performance, as described in the previous two articles on IF amplifiers.
However, many sets do not have
diode detectors or AGC, so measuring
www.siliconchip.com.au
the audio level across the speaker terminals is one of the few options left.
Note that this can only be done if a
signal generator with tone modulation
is used. If the alignment is done only
on the apparent aural difference, the
accuracy will be limited, so it’s best to
measure the output if possible.
It’s also important to realise that a
receiver front-end that’s been aligned
with a signal generator connected to
it will no longer be correctly aligned
when connected to an antenna. To
overcome this problem, I adjust the
signal generator output to a high level and wrap the test lead around the
antenna lead. That way, the signal
generator does not detune the antenna
tuned circuit to any extent and the
signal level into the set is relatively
low.
PEAKING AIR-CORED
TUNING COILS IN TRF SETS
These coils are usually wound
solenoid fashion on coil formers
ranging from 25-75mm in diameter.
Some of these coils are in aluminium
cans while others are left unshielded.
Invariably, they don’t have iron dust
cores to adjust their inductance, to
ensure accurate tracking across the
band.
It is fairly easy to determine whether
the tuned circuits do track accurately.
To do this, first tune to the high-frequency end of the tuning range and
adjust the trimmer capacitors for best
reception, with the trimmers nominally at half capacitance. It doesn’t
matter whether the stations are tuned
at the correct spot on the dial for this
test.
Now tune to a low-frequency station. Make a note of where each trimmer is set, then vary the trimmer capacitance up and down in each tuned
circuit and note the position where
peak performance occurs. Return each
trimmer to its start position before
adjusting the next trimmer.
Note whether more or less capacitance is needed to peak the performance in each case (sometimes, you
may need more capacitance, in other
cases less capacitance). If the trimmer
has to be screwed further out (less
capacitance) for peak perfor
mance,
it indicates that the tuned circuit has
too much inductance and a turn or
two of wire may need to be removed
to improve the tracking. Conversely,
if the trimmer has to be screwed in
www.siliconchip.com.au
Two broadcast-band oscillator coils (left)
and one shortwave band oscillator coil are
shown here.
further, it indicates that the inductance
is insufficient and a turn or two of wire
will need to be added.
As a matter of fact, I had a 1931
Operatic TRF with this problem. In
that case, removing two turns from one
coil allowed the two tuned circuits to
track each other.
Another method of checking the
tracking is to get a small length of ferrite rod and with the set tuned to the
low-frequency end of the dial, insert
the rod into each coil and observe
wheth
er the performance improves
or deteriorates. If it improves, more
turns are needed and if it gets worse,
fewer turns are needed. If there is no
change in the performance, it is accurately tuned.
Some later receivers with air-cored
RF and antenna coils can be converted
to slug tuning for improved sensitivity.
I’ve done this to several receivers by
placing a small-diameter slugged coil
former inside the air-cored coil. This
is held in place by soldering the coil
former lugs (no winding on the form
er) to the larger coil lugs, so that the
former sits quite securely inside the
air-cored unit.
This works well and will noticeably
lift the performance of some sets.
ALIGNING TRF RECEIVERS
TRF receivers are relatively easy to
align, as we shall see.
First, if the coils have no iron-dust
cores, no alignment adjustments can
be done at the low-frequency end of
the dial – unless you are prepared to
Fig.1: circuit of “front end” of Kriesler 11-90. Note the use of a padderless
tuning gang (VC1a & VC1b).
February 2003 81
set that has all the stations displaced
by a similar amount is likely to be
reasonably well aligned. However, it
will still have to be peaked for best
performance.
Starting at the low frequency end
of the dial, adjust the coil cores until
the sought after station (about 600kHz)
appears at the correct spot on the dial.
That done, tune to a high-frequency
station (about 1400kHz to 1500kHz)
and adjust the trimmers for best performance. Note that some adjustment
may be necessary to get the station to
appear at the correct spot.
If you find that you have wound the
iron cores right into or out of the coils
and the trimmers right in or out and the
performance is still poor, reset them
all to half-way and try again. However,
it’s also possible that one or other of
the tuned circuits has a fault and no
amount of adjustment will overcome
the lack of alignment.
Alignment of TRF receivers on
shortwave bands involves the same
principles as those used on the broadcast band. However, there aren’t many
shortwave TRF sets about and those
that do exist tend to have fewer tuned
circuits (often only one in a regenerative detector circuit).
THE SUPERHET RECEIVER
This photo shows the dual-wave coil assembly from a wrecked receiver. The
two coils can be seen at top-right of the metal sub-chassis.
try the techniques previously men
tioned to match the inductances in
the various tuned circuits.
If the inductances do prove to be
matched, it’s a matter of adjusting the
dial scale so that the tuned station
appears at the right spot on the dial.
This is done by loosening the dial
scale pointer or grub screws holding
the dial drum to the tuning capacitor
and adjusting its position, while still
listening to the station. It’s then just
a matter of retightening the screws.
Now tune to the high-frequency end
of the dial and note where a particular
station appears. If it tunes with the
gang further out of mesh than it should
be, adjust the trimmers for less capacitance until the station appears in the
correct position. Conversely, if the
gang is further in mesh than it should
be, increase the trimmer capacitance.
Note that it is necessary to check
the tuning at the low-frequency end
of the dial following any adjustments
82 Silicon Chip
at the high-frequency end, and vice
versa, until the stations appear in the
correct positions.
Aligning a TRF with slug-tuned RF
and antenna coils is straight forward.
In this case, it isn’t necessary to muck
about with adding or subtracting turns
on the coils as occurs with some older
sets.
Generally, the frequencies at the
ends of the dial travel are marked in
some way. If not, just assume that later receivers tune from 540-1620kHz,
while older sets tune from about 5501500kHz.
First, tune the receiver across the
band and note where stations appear
on the dial. Alternatively, if you have
access to a signal generator, check
the entire tuning range. If stations are
consistently displaced (eg, by 10mm)
from the correct positions on the dial,
it is necessary to adjust the dial pointer
by a similar amount in the opposite
direction to correct this inaccuracy. A
The RF and antenna circuits of a
superhet receiver are identical with
those used in TRF sets of the same era.
Instead, the big difference between
the two types of sets is in the variable
tuned circuits, due to the addition of
a local oscillator in the superhet. This
local oscillator is tuned to a frequency
that’s offset from the antenna and RF
circuits by the intermediate frequency
(IF).
In most sets, the IF is 455kHz and
the local oscillator must accurately
maintain this offset right across the
band. This is not an easy task. Receivers using tuning gangs with identical
capacitance ranges (plate shape) will
usually only have the local oscillator
displaced exactly 455kHz from the
signal frequency circuits at three spots
across the tuning range.
This means that the tuning of a superhet receiver can vary sufficiently
for differences in sensitivity to be
apparent across the band. However,
this is mainly a problem in earlier
sets –later receivers use automatic gain
(volume) control (AGC) and therefore
these inaccuracies are hardly noticed.
www.siliconchip.com.au
Another view of the dual-wave coil assembly from the wrecked receiver.
On some “broadcast band only”
receivers, it became reasonably common to use a twin or triple section
tuning gang which had one section
specifically for the oscillator tuning.
This section usually had fewer plates
and they were shaped differently to
achieve accurate tracking. These are
called “padderless tuning gangs” as
no padder was required.
The miniature tuning gangs used in
broadcast-band transistor sets are of
this type too. One such tuning gang can
be seen in one of the photos, where one
section is noticeably smaller than the
other. The smaller section is used for
the oscillator. The maximum capacitance for each section is 210pF for the
signal frequency section and 90pF for
the oscillator section.
By the way, I have several transistor
sets (different brands) that use the
3-gang version of this tuning capacitor
(MSP). Unfortunately, I have found
that the oscillator and the antenna
and RF tuned circuits do not track,
not matter what I try do to overcome
the problem. Whether this was just a
faulty batch or is due to some other
problem, I don’t know. On the other
hand, my Kriesler 11-90 and 11-99 sets
use another brand of padderless gang
and they track perfectly. As a result,
their performance is very good.
www.siliconchip.com.au
Padderless tuning gangs are only
suitable to use on the broadcast band
and in sets with a 455kHz IF. They can
be designed for other bands and IFs but
I’ve not seen any. Where multi-band
operation is required, all sections of
the tuning gang are identical.
ALIGNING THE FRONT-END OF A
SUPERHET WITH A PADDER
Early superhet receivers used aircored coils in both the RF/antenna
circuits and the local oscillator. This
meant that it was necessary to have
some means of adjusting the oscillator
at both the low and high frequency
ends of the dial, so that the stations
were at their correct locations on the
dial.
This was achieved by having an
adjustable padder (mounted on the
chassis) to align the circuit at the
low-frequency end of the dial, plus a
trimmer capacitor to align the high-frequency end. There was only one adjustment for the antenna or RF coil and
that was done at the high-frequency
end of the dial.
The first step is to shift the dial
pointer (if necessary) so that it has
equal overlap at either end of the
dial scale. Then, with the receiver
operating, check that the stations are
received at the correct dial locations
KALEX
PCB Makers!
• High Speed PCB Drills • 3M Scotchmark
Laser Labels • PCB Material – Negative
or Positive Acting • Light Boxes – Single
or Double Sided; Large or Small • Etching
Tanks – Bubble • Electronic Components
and Equipment for TAFEs, Colleges and
Schools • Prompt Delivery
We now stock Hawera Carbide Tool Bits
718 High Street Rd, Glen Waverley 3150
Ph (03) 9802 0788 FAX (03) 9802 0700
ALL MAJOR CREDIT CARDS ACCEPTED
February 2003 83
(1) Do the same as (1) above (ie, by
overwinding the coil), except the turns
must now be in series opposing; or
(2) Adjust the padder to higher values of capacitance so that the tuning
capacitor plates are further out of mesh
for a given oscillator frequency. That
done, continue to adjust the circuits
in the same way as mentioned above,
except that the padder is continually
increased in value. Do this until the
performance is at its best.
ALIGNING FRONT-ENDS WITH
IRON DUST/FERRITE CORED COILS
This assortment includes one 13mm ferrite rod
and one 9.5mm rod (both without windings),
plus an antenna assembly from a wrecked set.
(they probably won’t be at this stage).
Next, at the low frequency end
(around 600kHz), adjust the padder
so that a known station appears at
the correct dial location. That done,
go to the high-frequency end (around
1400-1500kHz) and adjust the oscillator trimmer so that a known station
appears at the correct spot. For best
results, double check these two adjustments.
Next, adjust the antenna/RF circuit
trimmer(s) at around 1500kHz for best
performance. Now if all is well, the set
is correctly aligned.
In some cases, however, the alignment may not be correct for the
antenna and RF circuits at lower frequencies. To check this, slide a small
ferrite rod into these coils and observe
any differences in performance, as
described earlier for TRF receivers.
If the performance improves, this
indicates that the anten
na/RF coils
have insufficient inductance. In that
case, there are three ways in which the
alignment can be improved:
(1) Overwind a few turns (experiment with the number) of enamelled
copper wire onto the coil. That done,
connect one end in series with the
grid end of the coil and the other end
to the point in the circuit where the
grid connection had originally been
made.
(2) Install a small amount of ferrite
material inside the coil.
(3) Reduce the “apparent” induct84 Silicon Chip
ance of the oscillator coil by reducing
the padder value. This means that
the tuning capacitor plates have to be
more in mesh for a given oscillator
frequency.
Method three is the easiest to implement. First, tune to a weak station
at the low-frequency end of the dial.
Now adjust the padder so that the
station appears closer to the end of
the dial and note any improvement
in signal strength. Keep doing this
until no further improvement can be
obtained.
It is then necessary to alter the dial
pointer so that it points to the appropriate frequency or marking on the dial
scale. That done, readjust the oscillator
alignment at the high frequency end
of the dial and peak the RF/antenna
circuit(s).
However, in some cases, it many no
longer be possible to receive all the
stations that should be received. The
receiver may now only tune from (say)
600-1600kHz instead of 550-1600kHz.
If this proves to be the case, then this
particular method of obtaining the best
performance at the low-frequency end
of the band isn’t appropriate. Instead,
one of the other methods must be used
– or you can just forget about getting
the best performance at the low-frequency end of the dial.
If the antenna/RF coils have too
much inductance, there are only two
ways of improving the low-frequency
alignment:
It is much easier to adjust the frontend of a receiver if all the coils have
iron-dust or ferrite cores, plus trimmer
capacitors. However, you must use a
non-metallic alignment tool to make
the adjustments if the core is inside
the coil.
The first step is to adjust the dial
pointer so that it travels from about
the 520kHz mark (ie, maximum
capacitance) to about the 1620kHz
mark (the exact frequencies at either
end of the band will vary from model
to model). That done, tune to a frequency around 600kHz and adjust the
oscillator so that the station appears
at the correct spot on the dial, then
tune to around 1400kHz or 1500kHz
and do the same by adjusting the
oscillator trimmer.
Repeat this procedure until the stations appear in the correct positions
on the dial (or as near as practical).
Now do exactly the same thing for
the antenna and RF coils but use a
frequency nearer 1600kHz for the high
frequency adjustment. Note that these
adjustments should be carried out with
the set coupled to the antenna. In each
case, you adjust the coils and trimmers
for a signal peak by measuring the
AGC or detector output voltage, as
mentioned earlier in the article.
When ferrite rod loop antennas are
used, the coil (or a small auxiliary
coil) is simply slid along the rod for
best performance at the low-frequency end of the dial. It’s then locked
in position with some bees wax (or
similar).
The nominal value of the padder
capacitor for the broadcast band and
a 455kHz IF is 425pF. For a 175 kHz
IF, it is around 550pF or higher. On
the shortwave bands, the padder value can vary from upwards of 2200pF
to 4500pF. In some cases, the manu
facturers didn’t worry about tracking
www.siliconchip.com.au
Photo Gallery: National
Panasonic R-100 4-Band
9-Transistor Radio
Purchased at a charity bazaar a few years ago for the
princely sum of $10, this National Panasonic R-100 4-band
9-transistor radio was the top-of-the line model from Matsushita in the late 1960s.
Covering the range from 525kHz to 26.1MHz, the R100
has an IF of 470kHz. Most of the transistors are on a tightly-packed PC board but the oscillator and antenna coils
are chassis-mounted, as are the band switch and trimmers.
Weight without batteries is 4.2kg. This unit has been
modified to run from a 9V DC plugpack and still performs
very well. (Note: if anyone has a manual, please contact
Leo Simpson).
on the shortwave band and dispensed
with the padder altogether.
RECEIVER ALIGNMENT ON
SHORTWAVE BANDS
On the shortwave bands, the local
oscillator operates above the received
frequency – with a few exceptions.
The alignment procedure is the same
as for the broadcast band. However,
the problem of image breakthrough
is quite evident, as discussed in the
earlier articles on IF alignment.
Assuming that the IF is 455kHz, a
station on 17MHz will be heard on
the receiver dial at both 17MHz (or
close to it) and at 16.09MHz (the image
frequency). For this reason, make sure
that the antenna/RF coil(s) are peaked
on the 17MHz frequency.
Unfortunately, shortwave stations
are often difficult to identify. Several frequencies often carry the same
www.siliconchip.com.au
program and fading is common, all
of which makes alignment rather
difficult.
It’s best to use a signal generator
for this job, as this eliminates any
ambiguities in the results. Once again,
tune the receiver to about 17MHz, then
adjust the signal generator to 17MHz.
Adjust the receiver for a response on
or near 17MHz, then shift the signal
generator frequency until the receiver
responds to another frequency.
It should respond when the generator
is shifted to 17.910MHz if the receiver
is reasonably well aligned. However,
if it is badly out of alignment, the response will be at 16.09MHz instead.
The correct alignment frequency
is the lower generated signal. If the
response is at 16.09MHz, the oscillator
trimmer will have to be reduced in
value so that the receiver can respond
at 17MHz, as marked on the dial
scale. The RF/antenna coil trimmers
are adjusted for peak performance on
this frequency, if the tuning range is
6-18MHz.
Shortwave bands are very approximately aligned, with few receivers
having accurate dial calibrations. In
fact, shortwave bands on many sets
were purely a selling point, with poor
perfor
mance, poor dial calibrations
and inadequate alignment facilities.
SUMMARY
So that’s it – a comprehensive alignment procedure for most TRF and
superhet receivers. In summary, the
general procedure is to adjust the coil
cores (or padders) at the low-frequency
end of each band and the trimmers at
the high-frequency end for peak performance. That said, you should be
guided by the alignment procedure for
SC
each model if this is available.
February 2003 85
|