The CPRC-26 is a small battery operated transceiver, operating on 6 crystal controlled channels in the VHF frequency range of 47 to 55.4 mhz. It has a power output of 300mW with an FM (Frequency Modulation) deviation of plus and minus 15khz. It has a range of about 1 mile (1.6km), is fully sealed and weighs about 10 pounds (5kg). Power was provided by dry batteries and had a useful life of 20 hours. It was designed in Canada and used by many armed services, including the Canadian Army and Navy, the US Army, NATO, and the Australian Army. There were about 4500 built by Philips and Canadian Rogers.

The circuit is very interesting, as it uses only one crystal per channel, used for both transmit and receive. The design is based on valve circuitry, at the time when pencil valves were taking over from miniature valves. The radio contains 11 pencil valves and 2 miniature valves. The receiver is a single conversion superheterodyne, using a crystal controlled oscillator. There is one RF (Radio Frequency) amplifier and the IF (Intermediate Frequency) is at 4.3 mhz, with 4 stages plus one Limiter. There is an FM discriminator and audio amplifier for driving headphones. The transmitter has a master oscillator directly driving the power output valve. The modulator is driven from the microphone and is  also  used to control the transmit frequency with an AFC (Automatic Frequency Control) circuit.


The British built a near copy of the CPRC-26 and called it the Station Radio A40. It looks almost identical, with the most obvious differences being the different knobs, no cover for the second handset socket, a small bar to hold the plug in place, a corner screwed case, and battery attachment with a strap rather than snail cams. The external accessories are interchangeable. Internally they are very similar, with exchangeable modules. Inside the CPRC-26 the battery connector is wired to the radio, whereas the A40 has an extra connector on the radio chassis. The previous PRC-10 radio has some similar modules. The following design, the PRC-510 also has some common modules. Philips built a transistorised version and called it the PRC-261. It looks almost identical from the top view, except that it has an extra switch to enable a second bank of 6 channels, bringing the total number to 12 channels. The radio is not as tall as the CPRC-26. This is because the battery compartment is smaller, even though the radio is taller. The photo shows CPRC-26 (left), A40 (centre), PRC-261 (right).

The following photograph shows a side view of the CPRC-26 (left), A40, (centre), and PRC-261 (right).


The CPRC-26 was normally used by a radio operator, in a mobile role. It was mounted on the operators back in its own canvas backpack, or placed in any other larger pack. The A40 had a special belt arrangement. The 1/4 wave whip antenna would normally be used vertically, when the operator was standing upright. The counterpoise wire can be attached to the radio earth pin and hangs down, thus making a 1/2 wave vertical antenna and improving the range. The counterpoise wire had a little attachment so that it could be fixed to the operators ankle. The whip has a flexible base so that it can be bent to the vertical when the operator is lying down, or the radio is horizontal. If the operator wishes to be concealed, the whip can be removed, and the counterpoise wire can be plugged into the antenna socket and used as a trailing antenna, but giving reduced range. Alternatively, the homing antenna from a PRC-10 can be used to determine the direction of another transmitter. The CPRC-26 could also be used as a portable station with the larger 10 foot ground mounted whip, placed up to 44 feet away from the radio. There are two sockets on the front panel for a handset. The handset was used for normal operation, and when an officer used this, the radio operator could plug in a single earphone for monitoring. There is a PRESSEL switch on the handset. The A40 had a Larkspur type headphones and microphone. There is a NORMAL and WHISPER mode, which changes the audio input and output levels. There is an audio extension lead to allow the handset to be used up to 60 feet away from the radio. There is a battery extension lead for cold weather, to enable the battery to be removed from the radio, and kept warm. There were no facilities for rebroadcasting or relaying. There was no other power than a dry battery.


In Australia the CPRC-26 replaced the post WW2 backpack PRC-10, and was  eventually replaced by the PRC-25 which was introduced during the Vietnam war (about 1970).  In Britain the A40 replaced the Wireless Set Number 88 in the late 1950s, and in the mid 1970s was eventually replaced by the Clansman PRC-350. The CPRC-26 was used by NATO until the 1980s. In Canada it was declared obsolete in 1969.


There are several antenna arrangements for the CPRC-26. There is a four foot whip, a four foot trailing wire, a four foot wire which acts as a counterpoise, and a homing loop. The antenna socket is dual threaded, so that any other antenna may be used. There is also a 10 foot long ground mounted whip. There is a BNC socket used for the homing loop. The antenna is connected to the main tuning coil, which serves as both receiver tuned RF coil and the transmitter tank coil. Different trimmer capacitors are switched in for channel selection. There is one RF amplifier (V7) which is capacitively coupled into the grid of the mixer, the plate circuit has a tuned circuit with a different trimmer for each channel. The mixer (V8) has the oscillator signal injected via the filament. The 4.3 mhz difference frequency is selected in the plate tuned circuit, and connected to the first IF amplifier (V9). Each of the four IF amplifiers have two tuned circuits at 4.3mhz. The output of the last IF amplifier is connected to a limiter. The limiter is identical to the IF amplifier, except for the input coil, which has greater coupling, which contributes to overdriving the grid of the limiter (V10) resulting in amplitude limiting. The plate voltage is higher at 90 volts, the IF only having 45 volts. The output from the limiter is connected to the discriminator, which acts like a balanced bridge. Any deviation from the IF frequency due to FM modulation, unbalances the bridge and causes an output at the rate of deviation (which is the modulation component). This is capacitively coupled to the audio output valve (V6), and then transformer coupled to the headphones. The channel frequency determining component is a crystal oscillator (V4) which has a separate crystal for each channel. The valve oscillates on the third harmonic of the crystal, and is 4.3 mhz below the channel frequency.  The transmitter uses a free running Colpitts master oscillator V2, with  a different trimmer for each channel. The output of the oscillator is capacitively coupled to the power output valve V1 running in class B. The output tank circuit is the same coil and switched capacitors, used by the receiver, and coupled to the antenna. The receiver RF amplifier V7 is not powered, but there is sufficient capacitive coupling within the valve to inject the transmit frequency into the mixer V8. The output from the mixer is the difference between the transmit frequency and the frequency of the crystal oscillator V4. This 4.3mhz signal is capacitively coupled to the AFC valve V5. The output of this is connected to a discriminator which produces a bias for the modulator valve V3. The modulator adjusts the master oscillator frequency, until the transmit frequency is the same as the receive frequency. It has a range of plus and minus 250khz. The handset has a carbon microphone, which is transformer coupled to the modulator grid, to provide FM modulation at voice frequencies.  The changeover from receive to transmit is done simply by turning off the filament supply to some valves , and energising other valves. The receive valves are RF amplifier V7, and the four IF amplifiers V9, V9, V9, V9, and the limiter V10. The transmit valves are the AFC amplifier V5, modulator V3, master oscillator V2 and power output valve V1. Valves common to both receive and transmit are always powered up, and these are the oscillator V4, mixer V8, and audio amplifier V6. The two miniature valves used are type 3B4 (CV2240), in the transmitter power output V1 and master oscillator V2. The pencil valves used, are four type 1AD4 (CV2237) for the RF amplifier V7, crystal oscillator V4, AFC control V5,  and modulator V3, six type 5678 (CV2254) for the mixer V8, four IF amplifiers V9, and limiter V10, and one type 5672 (CV2238) as the audio amplifier V6.


The radio is in a sealed cast aluminium box. The battery is in a removable box with a seal around the edge. The radio contains a H style aluminium chassis, with many plug in modules on one side. These are all the same shape and height and contain all the active electronics. They are packed in rows 3 across and 6 along. All are colour coded for their special function. They plug in so that they can be changed easily, and are held in place by a cover. Each module is a (usually) square tall metal can, with a 7 pin valve type base. Each one may contains a pencil valve, tuned circuits, capacitors and resistors. They are sealed at the base with solder and contain dry air. There is one vacant position. This normally contains a desiccant when the radio is sealed in the case. When removed from the case, a cable can  be plugged into the vacant socket, to measure aspects of the radios performance. The other side of the chassis has a cast aluminium cover, which holds the switch mechanism and all the trimmer capacitors. This cover is made of magnesium for lightness. It is removable so that the bases of the 7 pin sockets can be accessed There is a bank of 6 crystals on the same side at the rear. On the rear of the chassis is the battery connector. At the front of the chassis, is the wiring to the front panel switches and plugs. The whole of the radio insides is closely packed. The four IF amplifier modules are interchangeable. The two discriminators are interchangeable. The two 3B4 valves are interchangeable. The normal dry battery is a BA-289/U which provides 4 different voltages, from one battery block. On receive the requirements are: 90v@3ma, 45v@12ma, 1.5v@550ma, 3v. On transmit the requirements are: 90v@30ma, 45v@8ma, 1.5v@850ma, 3v.


There are three main pieces of test equipment for the CPRC-26. There is a small Condition  Indicator which shows the transmitter output. There is also a Battery Tester which checks all battery voltages on a meter. Those are mainly field testing units.   The main testing device is intended for bench repairs and is called the CTS-3/PRC Test Set. This a test meter in a bakelite case that has several functions. It is connected between the battery and the radio. The switch on the front then enables the  battery voltages to be checked, and the radio current consumption to be monitored. It is actually a VTVM (Vacuum Tube Volt Meter) so it also has a self test position. When the radio is opened, there is a flying lead with a probe, that can be plugged into the vacant position on the radio chassis. This allows the monitoring of several conditions within the radio to enable alignment. It has switch positions for, AFC TEST, TUNE RF, TUNE MO, TUNE PA. There is also a 7 pin socket to enable the condition of the valves in the radio to be checked for emission, which include the 3B4, and the modules AF, MODulator, AFC, XLO, MIXer, RF AMPlifier, IF amplifier and LIMiter. The modules would normally be checked before tuning and alignment of the  radio.


Several items are needed to align the radio. A power supply that provides 1.5 volts DC at about 1 amp (for filaments), 90 volts DC at 50 milliamps (for the transmitter), 45 volts DC at 25 milliamps (for the receiver), and 3 volts DC at about 1 milliamp (for bias). A shorting link or handset to enable the receiver. A shorting link or clamp to enable the transmitter. A dummy load, or field strength meter for the transmitter. The CTS-3 provides most tuning functions. If no CTS-3 is available, a VHF signal generator with FM modulation, and an audio output meter is necessary. For tuning the IF amplifiers, a soldering iron capable of opening the modules is required, and a sweep generator is useful. There is little space behind the front panel, and if the wiring loom is not placed correctly, it can foul the ON-WHISPER- NORMAL rotary switch, and as the panel is tightened up, the switch will jam or break. The tool used for tuning the trimmers needs to be non conductive, as some trimmers have high tension on them. 


 The alignment of the radio involves adjusting three trimmers for each channel. The antenna circuit is tuned with the PA trimmer, and the RF amplifier is tuned with the RF trimmer, for maximum output indicated on the CTS-3 meter, with the switch set to TUNE RF. That is all that is tunable for the receiver.  For the transmitter, set the CTS-3 switch to TUNE MO, and adjust the MO trimmer to set the meter to the red line on the scale, after tuning through a peak. Switch the CTS-3 to TUNE PA and the CPRC-26 to transmit. Readjust the PA trimmer for maximum meter reading and dummy load lamp brightness. Leave the CPRC-26 on transmit, switch the CTS-3 to TUNE MO, then readjust the MO trimmer for no deflection, as you switch the CPRC-26 from receive to transmit. Repeat for each of the 6 channels.  I didn' t like this procedure much. Tuning for a bright lamp is reasonable, but it doesn' t take into account what antenna you are using, so I prefer to tune the PA trimmer using a field strength meter. I looked at the transmitter output with a spectrum analyser, and noticed that if the AFC circuit was maladjusted, the transmit frequency would lock in after a second or two. I could see the carrier move and lock in on the screen. With the spectrum analyser it was easy to adjust the MO trimmer so that it was correct immediately. Modern equipment can make things easier. I used a VHF FM signal generator and output meter, to tune the RF and PA trimmers for maximum, and seemed to get better sensitivity. However, there seemed more than one peak (receive frequency) and they all seemed a little off what they should be, according to the crystal. The receiver alignment assumes that the IF is correct and this appeared not to be the case, so I used a sweep generator to check them. The first graph shows that the IF was not exactly tuned to 4.3 mhz and showed a non uniform passband. Tuning the CPRC-26 with this characteristic could produce any of three different receive frequencies, and a poorer sensitivity than possible. I could hand select IF modules for the correct frequency. That would be fine if I had spares.   I decided to open the IF and limiter modules. The coils had dobs of glue on them which chipped off with a sharp point and gentle pressure. I freed up the slugs which are a small ferrite cylinder that rotates around the outside of the coil. Since each IF has two slugs, and they interact, I had no success tuning the ten slugs, either by ear, by output meter, by oscilloscope or by sweep generator. I made up a little jig that plugs into the CTS-3 module test socket, which enabled me to tune each module to 4.3 mhz individually, while powered up. I then plugged them all into the radio, and tuned them for one single peak on 4.3 mhz. This was measured after the input coil of the limiter. I then monitored the discriminator input and tuned the last coil in the limiter module to give a double peak of the same height, centered on 4.3 mhz. The result is shown as the middle graph. The lower curve shows the discriminator characteristic, which is responsible for the double peak in the output. The slugs are brittle and easily broken, and I have yet to find a supplier of them. The receiver alignment  was redone and better output was obtained, and on the correct frequency. There is so much gain in the IF and limiter strip and I was operating them without the shielding covers, that I had a lot of instability. I had to reduce both the 90 volts and 45 volts supply to about half, to do the alignment. The modules have to be dried in a dessicator before reassembly.


The CPRC-26 is extremely small and light compared to its predecessors. It has six crystal locked channels that ensure stability over normal temperature ranges. Each channel can be set with only one crystal. Repairs can be done extremely quickly by changing internal modules. The coloured disc on the side, indicates which set of channels are fitted. It was common practice to have several different groups of channels, for different battle roles. All frequency groups had one common channel for liaison. The frequencies commonly fitted are: 50.0, 50.2, 50.4, 50.6, 50.8, 51.0 mhz. There are few controls so that operator error is reduced and unskilled operation is possible. Only one of the handset sockets has transmit capability, so accidental transmission is avoided, and battery life is extended. Battery supply is through the handset, so if no handset is connected, the batteries cannot go flat. The whip is short so obstacles can be avoided. The flexible base allows any radio orientation. The radio is sealed with dry air, so a long life in tropical conditions can be achieved.


The battery is small, so operation time is limited. Transmit power is only 300mw, so the range is limited. The transmit function is provided by energising the output valve filaments, and even though they are quick heating, the operators first few words may be lost. Operators needed to be trained to be aware of this. On some channels, the strong FM band radio stations can enter and beat with the ninth harmonic of the receive crystal, to generate interference.


CPRC-26 Circuit Diagram (92k)
CPRC-26 Parts list(65k)
CPRC-26 Battery Top view(33k)
CPRC-26 Battery Terminals (9k)
CPRC-26 repairs (3k)


RADIO SET CPRC-26, EMEI (Aust) F542, March 1957
TEST SET, RADIO CTS-3/PRC, Canadian Army Operators Handbook, January 1955
WIRELESS SET TYPE A40, Working Instructions, EKCO
Wireless for the Warrior, (http://home.hccnet.nl/l.meulstee/larkspur/larkspur1.html)
Wireless Set A40, Murray McCabe, June 1999

Copyright: Ray Robinson
Back to Museum