AWA Radio Australia
AWA AR8

AR8/AT5 Manual

Special Thanks to Ray Robinson for his help getting me going on the filament string,

References to http://www.shlrc.mq.edu.au/~robinson/museum/AR8.html

The AT5/AR8 is a MF/HF transmitter receiver set. Designed originally for the Royal Australian Airforce, it saw extensive use in aircraft, in the ground mobile and ground fixed applications, and marine usage.

It consists of the AR8 receiver, the AT5 transmitter, Aerial Tuning Unit for  the AT5, and a 12, 24, or 240V power supply. There is also a junction box and a remote control unit.

The AR8 receiver consists of two independently tuneable front ends alongside one another, one for HF, the other for MF. There is a MF/HF switch which selects the frequency range.  The receiver can operate from the main antenna fed via the change over relay in the Aerial Coupling Unit, or a free aerial may be used. The RF stages are protected during transmission as the keying circuit disconnects the cathode for the tubes in the RF stages. The intertuning function allows this function to be over-ridden. A winding on the modulation transformer routes sidetone into the receiver during transmission. There is a facility for a loop connection to allow direction finding bearings to be made. 

Oak switches are used to select the frequency range. Two large vernier dials select the receiver frequency, one dial is for MF the other for HF. There is a mode switch to select AM or CW/DF. There is a volume control, a tone control and a BFO note control. There are additional controls for taking DF bearings.

The receiver is a dual conversion superheterodyne type. The aerial input is protected by a 6X5 diode. The audio output stage also doubles as an intercom amplifier. Microphone and headphone positions are connected into the junction box.

The transmitter consists of two seperate master oscillators, one for MF and one for HF. On MF a VFO is used. HF allows the use of either the VFO or a crystal oscillator. In HF mode the 6V6 oscillator drives an 807 buffer amplifer, which drives the two 807 finals. In MF mode the 807 buffer serves as the oscillator, directly driving the finals. A 6V6 serves as a tone oscillator or voice amplifier which modulates the finals on MCW and RT modes. The oscillator uses an overtone system at the higher end of the transmission frequency mode, thus output power is reduced at higher frequencies.

Controls allow the selection of bands for the Master Oscillator, Buffer Amplifier and PA stages. There are separate tuning controls for MF and HF operation. The line switch disconnects the transmitter from the Aerial Coupling Unit and switches a resistor in series with the screens of the finals. This allows the transmitter to be tuned without emitting an on-air signal. The mode switch turns on the transmitter HT supply (normally a motor generator set) and selects CW, MCW or RT. There was an optional external pulse unit which could be used with the transmitter.

The aerial coupling unit connects the transmitter and receiver to the type of aerial likely to be found in an aircraft. The unit actually consists of two independent couplers in the one case. One unit is used for  tuning trailing aerials on MF, and the other tunes a fixed MF aerial. Current transformers are used to give an indication of aerial current. The MF stage used the massive switched coil/variometer arrangement one would expect of an MF tuner. A variable capacitor is also used. The HF tuner uses a roller inductor and a variable capacitor. Capacitance can be switched in as series or parallel. The antenna coupling unit also contains the keying relay and the MF/HF relay.

The Type G power supply is the 24Volt version. It consists of two motor generator sets, one is the receiver machine, the other for the HT. There are various control circuits in the power supply. The 12Volt version is very similar but it uses 12Volt MG sets. The 240V version was used primarily for fixed applications. It is mounted in the same case as the DC power supplies, however the MG sets are replaced with 866 mercury vapour rectifiers and a suitable 240V transformer. This is contained in a cage above the base of the power supply.

The AT5/AR8 is a classic Australian Designed example of a World War 2 vintage transmitter receiver. It saw extensive use in most roles. In the ground role it was used as a beacon transmitter, RADAR relay stations used the AT5. It was used in radio trucks, mounted on the back of jeeps, as emergency radio equipment on RAN ships, and as primary radio equipment on light boats. In the air role it was used in aircraft such as Beauforts and Catalinas.

This set is quite fun for CW or AM operation on the ham bands. I found that the VFO chirps like crazy on CW, I have not used the set on AM. The receiver is quite ordinary, although I have not aligned it, but it isn't really a top line comms receiver. The transmitter is good fun, although you need to use crystals to avoid chirp. I think you have to expect this with most sets of this era. The ATU can be a handful, it depends on the type of antenna you have. Generally it is difficult to tune on 20 metres, although I have had good success on 40 and 80 metres. 

TECHNICAL SPECIFICATIONS

Frequency Range:
    Receiver    140-740Kc, 765Kc-2Mc, 2Mc-20Mc
    Transmitter    140-500Kc, 2-20Mc
Modes:    AM, CW, MCW, Pulse. D/F on receiver.
Power Output:    50W on fundamental frequency

CONSTRUCTION

Construction of the AT5/AR8 is quite interesting. The general layout is a typical WW2 design, but the sections of each unit have been designed in a modular fashion. For example, the HF RF stage of the receiver can be removed as one unit by de-soldering a small amount of connection allows the entire module to be removed an replaced in the event of a fault. 

 

AR8 Restoration.

 

A shot of the front of the AR8 after removing all of the controls and two covers. I removed these primarily to get to some of the wires that run behind this panel and connect to the filament string.  This was one of two receivers I had acquired in my travels. This one had most of the rubber wiring in tact and looked like an easier option to get going. Boy how wrong was I. Anyway after some help from Ray Robinson I managed to get the filament string reconstructed as it had been re-wired for 6V , Why you would do this as standard it would run from 24V or 12V is beyond me but somebody had a go and it looks like they gave up. A number of wires were just missing and others were soldered in the wrong places. After careful study of the other receiver I managed to work out where all of the band switch and other wires went.

Testing my handy work I rewired a plug for power  and yes tubes all lit up.. I ran through the tubes checking voltages as in this type of set with two or more tubes inter connected filament voltages can vary. They all ran @ about .4 of a volt from each other so I was happy with that.

This is a good shot of the AR8 under side before major surgery. As you can see a few dark brown wires and heat shrink to replace the sections cut out of the original loom. Also centre you can see one of the two 21 Ohm resistors I needed to replace as they were missing from the filament string.

Here we are looking at the rear of the receiver and it is positioned upside down. This was the best way I found to work in it. As you can see all of the original capacitors were in place. I had hoped to keep them so the set would look original so I attempted to reform a few caps. The net result is that all of the caps in this set made very good 100K resistors but were not much good as capacitors, so they all had to be replaced.

In my initial diagnostics of the set I noticed that there was a significant drain on the high voltage supply. I first went through every tube and checked the plate voltage to see if we had HV on the plate of every tube. As my buddy had mentioned every set he had seen had the mike transformer open on the primary. This set was no acceptation. With the primary open no HV was getting to V3 so this needed to be rectified. A temporary jumper across the BFO switch that is dead centre of the rear of the set fixed this while I went through all of the other issues.

I had decided to get the audio going so at least I could work my way forward. This proved to be a cause of much aggravation. I could inject a signal into the grid of the output tube and got some audio but it was very weak, I decided to start moving along stage by stage injecting signal into the grid of each preceding stage. With the scope attached to the grid of V4 the driver for the audio stage I could see signal as I moved forward until I got to T3. T3 is the last IF can before the Audio driver. By this time it was 3am and as I moved the signal generator around funny things were happening. I left the set for the night and a call to my Buddy Ray resolved my questions. I should have read his article more closly. The IF was 750Khz not 455... HMMMM.

Ok now with my trusty signal generator set to the right IF frequency I started to get results. I could peek the IF cans at the right frequency and worked my way forward until I reached T1. This is the first or two IF cans connected back to back in what I would say is a bandwidth controlling mechanism. As the set has no bandwidth control it appeared to me that this was the AWA engineers way of pre setting the bandwidth. At this point I found a flaky T1. the top adjustment was damaged and the coil was intermittent. AS I had a scrap AR8 I raided a can to replace it until I had time to go back and repair it. You will also notice in the picture that most of the capacitors at this stage are changed. This was a necessary evil to get the set operational. Most of the original resistors were OK so I left them in place. They were well made and coated very well so they don't absorb humidity and change with the weather.

Dead T1.

Once I was happy with the IF strip I moved to removal of the RF decks as they were not functioning at all. You will notice in the above picture 3 capacitors that are connected to V1 centre shot. They too needed to be replace as they were a partial short.

Removal of the RF decks was not as hard as I first thought. After I had returned all of the wiring to original all I needed to do was de solder the terminal strips and two wires that connected the RF decks together. Removal of two side screws and one on the top of the deck and the section comes out for service.

This shot shows the LF deck removed and ready for work.

Ok with the LF deck out I started to meter all of the components to check them. I started with the paper capacitors as all of the others so far had failed. These were no acceptation. You can notice that some of them are under the band switch. These are a particular bugger to get to but not imposable. Cut the legs on the tube socket side and then turn over the deck. After removal of the side plate I noticed that all of these caps were soldered to ground lugs along with wires that go to the band switch. WARNING!! don't just cut all the wires. I found that if you are careful to select the cap wire and just cut tem as close to the grommet hole as possible it is a easy task to solder the new capacitor to the old leg using a pigtail method.

And here we have the finished LF deck after all of the failed components were replaced.

A close view shows some of the pig tails on the old capacitor legs. I have found this a good method when soldering new components to what may be fragile and un-replicable band switches.

This shot shows the HF deck with some very hard to get to caps under the LF MF selector switch With a little care and using the same method described above I managed to remove all the dead capacitors and replace them.

I though there were a few capacitors missing as I worked on the MF side.

And after removing the side plate I found them, See below.

These two are poked through the chassis to connect to chassis ground. The other side connects to the band switch so I cut them just above the trimmer board and pig tailed new capacitors in place. It would have been imposable to get to the band switch as this was the under side of the band switch.

Note the little buggers hidden under the band switch.

By cutting the leg long enough so you can get it with long nose pliers you can get tem out.

This shot shows the mess of the bench during the operation.

The two decks back in place and ready to align. I noticed that after I got the receiver re-assembled that the LL section was not working properly. A quick check showed that the local oscillator on this side was indeed dead. Hmmm but why.  After looking at the band switch I traced the wire from the switch to the pent grid  mixer oscillator and noticed that a pin was missing from the tube.. bugger! I had placed the wrong tube in the socket and thus no oscillator. The correct tube fixed this. I also noticed that on some sets the HF local ocilator is a 6V6. On this set a ^v^ didn't work and the correct tube was a full size 6J5

AHHH finished and working. Well I still need to do the RF alignment but the set operates What a fun job.. I will add more including some of the hand written notes on the filliment string later.

 Home