The transistor receiver era effectively started in Australia
with the importation of a number of radios from Japan in the mid-1950s. These
early sets were not brilliant performers as I soon found out when I bought a
small Sony pocket receiver. It proved to be extremely "hissy", even when tuned
to a strong local station. In fact, the signal had to be quite strong for the
set to even receive it.
This is the fully-restored set, shown here with its carrying handle raised. Its styling resembled the earlier valve mantel receivers.
Despite these shortcomings, I became hooked on this new
technology – a technology I didn’t understand at the time but wanted to learn
about. I kept the little pocket set for a quite few years but its limitations
meant that it had very little use and I mostly listened to my old faithful valve
Australian manufacturers started making transistor receivers in
the late 1950s. Initially, they assembled the sets in the same way as their
valve sets, with point-to-point wiring, and in some radios, the transistors were
even mounted in special sockets – just like valves. This method of manufacture
was expensive, particularly when Australian manufacturers had to compete with
the Japanese manufacturing techniques of PC board construction and cheap
However, Australian manufacturers quickly realised that
transistors should be treated as just another component. Mounting most of the
receiver on a PC board would also be cheaper, with fewer wiring mistakes. These
techniques coupled with tariff barriers helped Australian manufacturers stay
competitive until the barriers were reduced in the early 1970s.
The transitional Kriesler
Kriesler Australia was one of many firms making both valve and
transistor sets in the 1960s. Like most manufacturers at that time, they built
their transistor receivers in a style that suited valve technology.
Just why they did this isn’t clear. It’s hard to be sure
whether the manufacturers were ultra-conservative when it came to designing
their radios (cabinets in particular) or whether they felt that customers would
not accept the styling changes that were possible with transistors (including
portability). In reality, it was probably a combination of both scenarios.
Fig.1: the Kriesler Model 41-29 employed a conventional superhet circuit based on seven PNP germanium transistors.
The 41-29 is one of these transitional receivers, being called
a "Trans-Mantel". It is in a case similar to the Kriesler 11-90 valve mantel
receiver, although the dial mechanism is quite different. It could be used
either as a conventional portable (although the cabinet may not be all that
rugged) or as a cordless mantel receiver.
Unlike some other sets, the battery life was quite good, as a
286 battery was used to provide power. The 286 was in reality two 276 type
batteries in the one case. In fact, a life of up to 1000 hours has been quoted
for this battery in some of the Kriesler receivers.
The Kriesler 41-29 is housed in an attractive plastic case and
its size suited both portable or mantel-piece operation. It has a large
slide-rule dial and is tuned by a relatively large knob at the righthand end of
Apart from that, there are just two other controls: a combined
on-off/tone control and a volume control at bottom right. It all adds up to a
rather neat and functional layout.
The rear of the set is held in place by two screws, which serve
dual functions. When viewed from the back, the lefthand screw is also the
antenna connection, while the righthand screw serves as the earth
It is necessary to remove these two screws and the back to
replace the battery. In fact, the instructions for this are on a small piece of
paper attached to the underside of the receiver. Another set of instructions,
this time inside the back, describe how the rest of the receiver can be
This same piece of paper also has a layout diagram of the major
parts on the PC board, plus a rather small copy of the circuit. This circuit is
hard to read because of its small size but it’s much better than having no
information at all about the circuitry.
Dismantling and cleaning
The two control knobs fitted to the unit that was given to me
to restore were certainly not the originals. They were both black and much
bigger than the originals that came with the set, so they certainly looked out
Fortunately, I had some old Kriesler knobs stashed away and I
found two which looked similar to the original knobs. The two black control
knobs and the tuning knob were then removed, followed by the four screws that
secured the chassis to the front of the cabinet. That done, the front of the
cabinet came away in three separate pieces (that’s how it was made).
This rear view shows the fully restored set, just before the back was refitted.
The chassis was dusty but a few minutes work with an old
paintbrush fixed that. The dial pulleys were then oiled, as were other moving
In this set, the volume and tone control shafts are split down
the centre and require a knob with a metal insert (this insert anchors to the
sides of the knob). The Kriesler knobs that I’d dug up (originally scrounged
from a Kriesler TV set) were suitable but to accommodate the metal inserts, I
had to increase the width of the slots (the original knobs obviously had a much
narrower insert). This was done by carefully filing them with a needle-nosed
That done, the inner retaining screw for the tuning knob was
cleaned and repainted with gold spray paint, as was one stud underneath the
cabinet front. In addition, the dial pointer was resprayed with white paint, as
it too was looking a bit grubby.
The speaker had some fluff trapped at the front of the cone,
which meant it had to be removed. This involves first removing the PC board
(more on this later), after which it’s a matter of undoing the nuts and bolts
that hold the speaker in place.
In the end, it was necessary to only partially remove the
speaker, after which the fluff was easily brushed out. The speaker was then
correctly refitted into position.
Next, the cabinet and tuning knob were washed using soapy
water, a nailbrush and a toothbrush. However, I did take care to ensure that the
paper stickers didn’t get wet. Any shallow scuff marks in the case were then
removed using automotive cut and polish but there were also some marks that were
too deep to get out – at least not without cutting well into the plastic case.
However, the remaining scratches are not particularly obvious.
Removing the PC board
The PC board wasn’t as easy to remove as it could have been due
to the location of the "earth", which is also the mounting plate for a screw
holding the back of the set against the metal chassis.
This rear view shows how the PC board could have been hinged for easier service access.
First, the single-strand wire from the oscillator section of
the tuning gang to the PC board has to be desoldered, after which the four
screws holding the board in place are removed. The "earth" plate is then sprung
outwards so that the board can be removed. In practice, the board can then be
turned over (so that the parts face upwards) and a piece of cardboard or cloth
placed underneath it to prevent shorts to the frame during testing.
This procedure could have been simplified by mounting the board
in a slightly different position, so that it could be directly removed without
the "earth" being in the way. In fact, with a little more thought, the board
could have been hinged on the edge near the centre of the receiver, which would
have made it a dream to service.
The old AWA P1 portable TV set had a hinged board and it made
the set very easy to service. In fact, AWA did such a good job of making the P1
accessible that the picture tube could be replaced in 15 minutes.
The circuit configuration is similar to many other transistor
receivers of the era – see Fig.1. It’s a standard 7-transistor superhet circuit,
with the front-end using an OC170 transistor as an autodyne converter.
Kriesler economised on the windings on the loopstick antenna by
connecting the low-impedance section of L2 to both the base of TR1 (via R1, C1
& R3) and to the antenna via a broadly-tuned coil (L1). Coils used in the L1
position are usually tuned below the broadcast band when used with a 7-10m long
antenna. This coil boosts the performance at the low-frequency end of the dial,
as the performance here is usually inferior to that at the high-frequency
Despite its age, no PC board parts required replacement. The alignment required adjustment though, to get the set to tune correctly.
The output from TR1 is fed via 455kHz IF (intermediate
frequency) transformers IFT1A and IFT1B to the base of the first IF amplifier
(TR2, OC169). It then goes to IF amplifier stage TR3 (OC169) and from there to
detector stage D2 (OA79) which also provides the AGC voltage. Resistors R17 and
R18 forward bias D2 almost to the point of conduction, thereby greatly
increasing its sensitivity and reducing distortion.
As the signal strength increases, D2 applies an increasingly
positive voltage to R10 and hence to the base of PNP transistor TR2. This AGC
voltage in turn causes TR2 to draw less current as it cuts off. As a result, the
voltage at the junction of IFT2 and R12 becomes more negative (note: the circuit
is positive earth).
Now let’s consider the action of diode D1 (OA90), which is
effectively wired between the collector circuit of TR1 and the collector circuit
of TR2. Normally, D1 does not conduct as TR1’s collector is at -7.1V and TR2’s
is -5.1V (ie, D1 is reverse biased). However, as TR2 shuts down due to AGC
action, its collector voltage progressively becomes more negative. When this
voltage goes below about -7.3V, D1 becomes forward biased and conducts, thus
causing the signal from TR1 to largely bypass TR2 and go straight to TR3 (via
As a result, the gain of the set is reduced on strong signals.
At the same time, the selectivity is also reduced, as IFT1 is bypassed and only
IFT2 and IFT3 are effectively in circuit. This loss of selectivity probably
doesn’t matter a great deal and actually has the benefit of improving the audio
quality on strong signals.
Following the detector, the audio is fed to volume control R21
and then to the first audio amplifier (TR4, OC71). This is then followed by
audio stage TR5 (OC75) which in turn drives two class B audio transistors TR6
and TR7 (OC74s) via transformer T1. TR6 and TR7 operate in push-pull
configuration and drive the loudspeaker via transformer T2. Negative feedback is
applied from the loudspeaker voice coil to the emitter of TR5.
The output stage quiescent current is regulated by R37, a
negative temperature coefficient (NTC) thermistor. Germanium transistors are
particularly prone to thermal runaway as their temperature increases. In fact,
it can be so bad in circuits like this that the transistors can
To overcome this, the NTC thermistor decreases its resistance
with increased temperature, thereby reducing the forward bias on the bases of
transistors TR6 and TR7. This compensates for the tendency of the TR6 and TR7 to
conduct more heavily with increasing temperature.
In this circuit, the thermistor senses the ambient temperature
inside the case but some later transistor equipment had the thermistor
physically connected to the output transistor heatsinks to improve the response
time and provide more effective control.
Initially, this set was completely dead but this was quickly
traced to a faulty on/off switch. Not having a replacement on hand, I decided to
simply bypass the switch until a suitable control became available. The set then
showed signs of working.
Despite its apparent simplicity, the Kriesler Model 41-29 Trans-Mantel is not all that easy to disassemble for service.
Next, I sat my Leader LSG11 signal generator on the other side
of the workshop, so that it could provide a weak test signal for the set. This
was tuned to 455kHz and I could immediately hear a beat tone from the radio, so
it was quite a sensitive set.
The cores of the IF amplifier transformers had all been sealed
but their alignment appeared to be pretty much spot on. However, I found that
the set would only tune down to 600kHz instead of the more normal 525kHz.
This was easily fixed. All I had to do was adjust the
oscillator coil so that it tuned down to 525kHz with the gang shut, then adjust
the oscillator trimmer so that it tuned to 1620kHz with the gang fully open.
There seemed to be little interaction between these two adjustments so I then
aligned the antenna circuit by adjusting the position of the loopstick coil for
best performance at about 600kHz and the antenna trimmer for best performance at
The set was now working rather well and that was with no
external antenna. In fact, it worked so well that the addition of an external
antenna almost caused overload.
At this stage, I ran into a problem. The dial scale has a
"start point" marked on it but I couldn’t get it right. I tuned to either end of
the dial and the rest point for the pointer changed which wasn’t good.
I then found that the plastic gear shaft on the dial reduction
drive had expanded, so that when tuned to the end of the dial, you could keep
turning it even though the tuning gang was now stationary. This was easily fixed
by drilling through the plastic gear and the brass tuning capacitor shaft and
then locking the two together with a wire pin.
Kriesler made many interesting receivers over the years and the
41-29 Trans-Mantel was an excellent little set with long battery life. The dial
drive system is a little more complicated than needed and the PC board could
have been mounted in such a way that it could have been hinged for easy service.
However, these are comparatively minor quibbles.
Finally, note that the circuit indicates that the on-off switch
is part of the volume control. In practice, it’s part of the tone control
instead, so someone didn’t check the circuit too well.
In Vintage Radio for June 2004, I stated that the suppressor
grid on a valve caught the electrons that bounced off the plate. This isn’t
strictly correct as the electrons that hit the plate dislodge other electrons,
which are collected by the suppressor. This is called "secondary emission".
It has also been pointed out that the screen grid reduces the capacitance
between the grid and the plate. This should have added to my statement that the
screen grid screens the grid from the plate.