XP Falcon Car Radios

These radios were made by AWA for Ford and were fitted to 1965 XP Falcons. I am not an expert on early to mid 60's Fords, but it seems they may have been also fitted to other Fords of the era, e.g. XL, XM.
Two versions of this radio came in for repair. One is the low cost version with a single transistor output stage, and the other more expensive version has push-pull output. Both have push button tuners. They are fully solid state using germanium transistors.
No model number is printed on the case for either type, but it was easy to narrow down the possibilities of what model they were, or had been based on. AWA car sets of this period did not vary greatly in terms of circuit design, although many different types case and dial were used. Many Australian radios were given nick names based on their shape or some other feature; for example the STC "Jelly mould", the Healing "Scales" or the Kriesler "Beehive". As far as car sets go, "Diamond Dot" refers to a series of Astor models. The Falcon radio being described here seems to be known as a "5 Star" - note the stars on the right side of the dial.



Single Output Transistor type.


Set under test with new repro speaker.

This set came to me for repair from an owner of a 1965 XP Falcon. It had been installed in the dash of the car, but not connected. No speaker was present. In fact, it seemed the car had not been provided with a radio when manufactured, since the paint was intact around the speaker screw holes at the top of the dash. The absence of proper mounting brackets also suggested the radio was a recent addition, with the back of the radio held up by a piece of coat hanger wire. However, it was the correct radio for that model of car.

I was familiar with this model of radio, having first played around with one back in 1981. Back then I had no idea what it was from, but I knew it was made by AWA. Unusually, there is no model number printed on the radio, but I knew from experience that I'd have a circuit for something that was the same. I trawled through my AWA manuals and circuits and found the 991-A to match. These are listed as Falcon radios in the AWA manuals, and are also listed as XM prefixed  models, suggesting XM Falcons.

The Circuit.


The set uses six transistors and is, as standard by then, permeability tuned. VT1 functions as a tuned RF amplifier. C1 and the aerial capacitance resonate TR1 to the input frequency. C1 is made adjustable since different aerials may have different capacitances. It is peaked at around 1400-1500Kc/s.
Transistor VT2 is an autodyne frequency converter. That is, it is a self oscillating mixer. The oscillator coil is L3, and tuned by L4 and C9 in series with C11. Feedback from the emitter to cause oscillation is connected to the mid point of these two capacitors.
VT3 is the one and only IF amplifier stage. Generally transistor sets use two IF stages, so this is a little unusual, especially in a car set where utmost sensitivity is required. AVC voltage is developed from rectifying the voltage across the 2nd IF transformer primary (TR3). C16 shunt feeds diode MR1, and is filtered by R6 and C3. Only the RF amplifier is AVC controlled.
Detection is by means of diode MR2 in the conventional way. R17 would apply a small bias to MR2 to improve sensitivity.
The volume control has a loudness tap for improved bass response at low volume. In addition, a tone control is also provided, shunting more or less of the high frequency component to earth via C24. C25 is a 4uF electrolytic to couple the audio into the first audio amplifier, VT4. It should be replaced as a matter of course in these radios, and is the most common cause of low volume.
The audio amplifier is a simple direct coupled class A design, using transistors VT4, VT5 and VT6. DC feedback is provided for stability. If the collector current of VT6 increases, the voltage across R25 increases, increasing the emitter voltage of VT4, causing VT5 to conduct less, and thus reducing the base current for VT6. Note that VT6 is an NPN type. If C27 loses capacitance, volume will suffer, because of the excessive negative feedback at audio frequencies than can now occur.
Further temperature stabilisation is provided by thermistor TH1. As temperature increases, its resistance reduces, reducing bias current. Collector load is a simple choke. The speaker is connected directly across it. The choke prevents excess DC flowing through the speaker, which would be problematic at the 500mA collector current. The 220R presumably provides some protection against excess voltage breaking down the output transistor, if the set is operated without a speaker.
Note that as most of the transistors are PNP types, the set is operating with a negative supply rail. In the case of a radio designed for negative earth, it means the set is constructed upside down in the electrical sense.

Despite being all solid state, no PCB is used. Construction is as per a valve set, with point to point wiring and tagstrips. The transistors are mounted in rubber grommets in the chassis. It's a method used for some of the first Australian transistor portable radios in the late 1950's. I was surprised to see it still being used in 1964 when this radio was built.

As is typical with Ford, the dial lamp is fed separately from the dash light circuit in the car. To those unaware, the red and black wires emanating from the radio would infer positive 12V (red) and earth (black). This is not so! Black is the positive 12V supply, and red is for the dial lamp. If one wanted to, the red and black wires can be both connected to 12V and the dial lamp will always come on with the radio, rather than only when the park or headlights are turned on. A fuse is not provided with the set, this being taken car of by a specific radio fuse in the car's fuse block. If one of these radios is used elsewhere, it is important to see the supply is fused. 2A to 5A would suffice.


Very spacious layout makes servicing easy.


Dead parts replaced.

Repair.
First thing evident was a burnt up emitter resistor (R25) for the output transistor (VT6). This usually indicates the transistor has shorted, and this was confirmed. The resistor is one ohm and from appearance looks like a small RF choke, being wound on the body of a 1W resistor.
I replaced it with a 5W resistor. The output transistor was replaced with an AD149 since I have a stock of these. AD149 is essentially the European version of the 2N301.
Powering up the set showed normal current consumption, so presumably the amplifier stage was operating, and those were the only two damaged components. Connecting a speaker indicated all was indeed good here, but with a lack of volume. Experience with mid 60's AWA car radios has taught me that the coupling capacitor in series with the volume control wiper (C25, 4uF) would be the culprit. These low value blue Philips electrolytics are always suspect. Replacing it brought forth good volume.
C29, the 640uF across the 12V supply, had leaked electrolyte so it was replaced. All the other electrolytics were tested and found to be OK. These AWA sets use an AEE "Microcap" type of paper capacitor in other parts of the circuit. As is well known, paper capacitors are notoriously unreliable in valve circuits because the paper absorbs moisture, and also being organic, is unstable. In the case of low voltage and low impedance transistor circuits, small amounts of leakage is generally not a problem. For this reason, the paper capacitors were left in situ, except for one which had suffered the not uncommon splitting of the plastic encapsulating. A polystyrene capacitor mounted adjacent to the output emitter resistor had suffered damage from the burn up and was also replaced.

With the set electrically working, the tone and volume control were given a clean, and the push button tuning mechanism cleaned and lubricated.

For such a simple set, the performance was surprising. Early afternoon in the Blue Mountains brought in reception from 2GN (Goulburn) and 2AY (Albury).



Deluxe version - Push Pull Output Type.


Second set as received for repair. Note two output transistors.

This set came from an early Falcon enthusiast referred to me by the owner of the previous set. For the purpose of servicing, I used the circuit for the 975-A. The complaint was that it received some stations but with low volume. Immediately I thought of the audio coupling capacitor (and subsequently confirmed this to be faulty). However, as this set was from a lot further away and with the owner not personally known to me, I considered the best thing would be to give it a full restoration. This was agreed to,  given that it will have the best reliability possible.
There are some interesting differences between this set and the previously described version. At the time, the extra power transistor, and transformer to drive it, would have increased the purchase cost. Also, this model has the power switch as part of the push button tuner (AWA called it "Pressmatic"). The left most push button switches the set off, and to turn it on any of the other buttons can be pressed. Power output is nearly twice that of the cheaper set. This time a PCB is used for all the small signal circuitry. I noticed the same PCB has been used in several different models of AWA car radio. The audio output and power supply circuit is all point to point using tagstrips.


Component side of the radio before parts were replaced.

The Circuit.


Circuit of the AWA 975-A.

As can be seen, the circuit is more elaborate than that of the first set. As before VT1 is the RF amplifier. However, now we have a separate oscillator (VT2) and mixer (VT3). There are also now two IF amplifiers, VT4 and VT5. AVC controls the RF amplifier as before, but also the first IF amplifier. Furthermore, there is an AVC overload diode, MR1. This shunts the RF input with very strong signals. To see how this works, note that the AVC is of the forward type; that is, collector current of the controlled transistors increases, rather than decreases with signal strength. This often provides better control characteristics than reverse AVC in transistor circuits.
Turning now to VT1, we can see that as signal increases, and thus AVC, the base current increases, and then so does the emitter current. Thus, the voltage across R6, 680R increases. Now, see how MR1 and R3 are connected. DC-wise, they are connected across the emitter resistor. The more current that flows through the emitter resistor, the more MR1 is biassed on, and the signal is shunted by it and the 220R series resistor.
Unlike the six transistor set, AVC comes from the detector diode (MR2). Again, a the volume control circuit has a loudness and tone control. Note the presence of SW1 across the volume control wiper. It mutes the audio amplifier when the push button unit is actuated, so that one does not hear a rapid succession of stations as the tuner goes from one end of the band to the other.
Transistor VT6 is the audio driver which produces the push pull signal via transformer TR5. The output transistors are again 2N301's and drive a tapped choke as their collector load. Note that the speaker is not connected across the full choke. Temperature stabilisation is again provided by a thermistor (TH1). Being push pull, this stage operates in class B. This saves on battery current since current consumption of the set is more dependent on the volume level than is the case with class A.

Restoration.


Audio and power supply capacitors replaced.

This time I replaced all the paper capacitors with polyester and ceramic types. All electrolytics were also replaced. It is not a difficult job as service access is good, but just time consuming. It was noted the 12V supply choke L11 had been overheated. Although it hadn't actually gone open circuit, the insulation had burned off the winding. I rewound it with slightly heavier gauge enamelled copper wire. The radio functioned normally once this work had been done, but I had noticed an anomaly with the output transistor bias circuit. The circuit shows the bias current comes from R41, 195R. In actual fact this 195R resistor is two parallel 390R's. Current consumption with the volume all the way down seemed higher than it should be for a set with a class B output stage, and what caught my attention was that a 330R had been soldered across the two 390R's. It was not a factory job given how the soldering had been done. Why it had been done is not clear, because with it removed the audio quality was normal with no crossover distortion evident. With a class B stage, if initial bias is insufficient, crossover distortion makes listening unpleasant. With the foreign resistor removed, current consumption was much more satisfactory.


Not clearly visible, but all the paper and electrolytic capacitors have been replaced.

Next point of attention was the Pressmatic unit. Some of the buttons were loose on their shafts, flying off when the buttons popped out. Some were jammed and wouldn't disengage when pressed in. This also resulted in not being able to manually tune the radio, because the clutch was disengaged to the tuning knob. It transpired that the jammed buttons were so because they had in the past come off the shaft and been incorrectly replaced. The buttons have several slots into which the shaft may be inserted. If anything but the central slot is used, the button is tight against its neighbour. Once this was sorted out and the mechanism lubricated, it worked very well. The final job was to fix the loose buttons. They are a push fit on the shafts which have a slight toothed profile, but if the plastic wears they fail to grip. One could use contact cement to secure them, but I found the insulation off telephone wire to be perfect for packing the gap between the slot in the knob and the shaft. The buttons are a tight fit, but could be removed if need be.
The final job was to replace one of the rivets securing the aerial socket. It had been replaced with a self tapping screw which did not provide the right grip. An ordinary pop rivet works well for this.



Another Radio.


This set was fitted with a LED dial lamp. It's the brightest car radio dial I've yet seen.

This third set was sent up from Melbourne for restoration; the six transistor 991-A. The complaint was the usual one of it being very weak. The usual cause is C25 open circuit, and indeed this turned out to be the main fault. Since this set was going back interstate, I replaced the other likely capacitors to get the best reliability. They are not expensive, and especially for the simpler tagstrip version, not difficult to replace. The 991-A would have the best serviceability of any transistor car radio that I've worked on.


Capacitors replaced.

While replacing the capacitors, it's a good opportunity to clean out some of the grime on the chassis.


Spring at left is foreign.

The remaining job was the tuning mechanism. Not surprisingly, the clutch was slipping. This happens because the clutch material between the tuning control and the tuning mechanism becomes hard and brittle. In this instance, someone had at some time added a spring to provide extra pressure against the clutch. While that helped, the extra pressure bears against the push buttons to the extent they have difficulty returning to their positions.
I replaced the material on the tuning control side of the clutch with cork gasket material. Needless to say the foreign spring was removed. This, and lubricating the whole mechanism restored correct operation.

As expected the radio worked normally at power up. The dial light was a bright white. Someone had replaced the original incandescent bulb with an LED E10 based bulb.
Performance was typical, with all the Sydney stations easly received. By late afternoon, 4ZR from Roma in QLD was coming in.

The owner had supplied a short "rubber ducky" style aerial. Out of curiousity I tested it, and just as well I did, because it was vastly inferior to the standard telescopic type that I used for testing. It may give passable results in a metropolitan area, but at my location 60km from the transmitters, it is not usable.


Not recommended.

The aerial is only 32cm long which does not provide enough capacitive loading for the input tuned circuit of the radio. The pickup was so bad that I wondered if the conductor inside the rubber actually extends the full length. No doubt this aerial would work better for VHF FM, and it's possible this is what it was designed for, to the exclusion of AM.
It should be pointed out however, that modern car radios are not critical with loading capacitance, and have in fact dispensed with the trimmer adjustment. This aerial may be more suited to them.
 




Home