DAVE CAUDREY RE-VISITS A VINTAGE RECEIVER THAT BEGAN THE RADIO CONTROL HOBBY HERE IN THE UK.
A RESTORATION PROJECT.
These two photos show the beginnings of the restoration of the IVY superegenerative carrier wave receiver dated around 1950, transistors were not yet available at the time and these early designs used glass vacuum valves or gas filled types.
Workshop equipment used included a 50MHz oscilloscope, a variable voltage supply up to 67.5 volts for supply to the receiver.
In the early days of radio
controlled model aircraft the IVY receiver was the inspiration for receivers
from other manufacturers. Both the ECC951 Telecommander
and ED Boomerang receivers of the 1950s
and possibly others were based upon the IVY. (Compare the circuit schematics on
pages 17 and 19) The IVY is a super-regenerative receiver employing a single
‘hard’ valve. (An explanation of hard and soft valve super-regenerative
receivers is given on Page 17).
Some years back a friend provided me with a non-functioning example of the IVY receiver and the temptation of trying to get it to work has been hard to resist. All that was needed was a valve and a replacement for a missing r.f. coil. Some research and calculation indicated that the latter should have two separate windings each of 11 turns. An example was constructed and fitted together with a new 3S4 valve and the circuit was powered up using 1.5 V supply for the filament and the specified 67.5V supply for the anode. The circuit was found to operate immediately having languished for 60 years with an impressive 1.6mA anode current drop with a ‘good’ signal. One thing which was puzzling was that the relay could not be adjusted to drop out with such an impressive current drop.
(Little more than half a milliamp was often typical for some old receivers). Eventually the penny dropped, the relay is polarised and it had been wired in the wrong sense. Swapping the relay coil connections produced a useable receiver.
I had built model aircraft when I was at school and following National Service, in which I received some training in radio, I felt ready to have a go at radio control. In the late 1950s I successfully made both hard and soft valve receivers with virtually no test gear other than a homemade multi meter and a signal generator that I bought on hire purchase for £6.50. (The generator, made by Radio and TV Components Acton Ltd., is still in my possession although it has been modified superficially and provided with reduction gearing for the tuning capacitor and a logging Vernier). Its schematic is shown on page 9 pt.2 of this site.
When the designs of early super-regenerative receivers were first published waveform information did not appear, possibly because few had access to suitable oscilloscopes and because it is virtually impossible to connect anything to a super-regen. circuit without upsetting it. To counter this omission, I decided to display some waveforms if possible. Because of the touchiness of the circuit waveforms, with the exception of the quench waveform at terminal 4 of the schematic on page 17, were produced via a two turn pick-up loop slipped over the r.f. coil of the receiver and thus have no direct voltage reference. However they do indicate what goes on for different signal levels.
Having obtained a working receiver the question is what to do with it? A demonstration assembly using an available Elmic Commander rubber powered escapement and simulated rudder is a possibility.
The totally restored IVY valve receiver is shown in this picture. With its required power supply of 67.5 volts for HT and a 1.5 volt battery for the valve heater it will give an HT current change of 1.6 milliamps for the polarised relay. It just shows a receiver that began the journey of many pioneers of radio control and it can still be made to work.
Would we trust this receiver in a model today? Mmm…Well many fliers did 65 years ago !
A Retro Receiver or a trip back in time:
I started building radio control equipment around 1957. In those days all receivers used valves in super-regenerative circuits as described in pages 17 and 19 of this site and both hard and soft valve versions were made with minimal test equipment. It is imagined that these circuits evolved from their designers’ wartime experience.
I experienced little difficulty constructing established designs and the main problems were of reliability resulting from the number of batteries required and their connections. Each receiver needed a 1.5 volt battery to supply the valve filament, a 22, 45V or 67 volt battery to supply the valve anode circuit and a 4.5 volt battery to drive the actuator. The batteries were of necessity zinc–carbon types with limited life and the need for battery monitoring was ever present.
Transistorised super-regenerative circuits would later operate from a single battery but transistors available before 1960 were mainly audio types and ones capable of operating even at medium wave radio frequencies were very expensive. Transistors capable of operating at 27Mhz. simply didn’t exist.
The hard valve circuits I built and those of the commercially available receivers available at the time appear to have been developments of the IVY receiver above, in which a sensitive and delicate electro-mechanical relay is partially de-energised by reduction of valve anode current on receipt of a carrier wave signal. The valves used were miniature valves used in the output stages of portable radios e.g. the DL92 as used in the IVY. Sub-miniature valves such as the DL68, used in hearing aids, were available but a relay capable of operating from a change of anode current for these devices would be a very special one. In fact the Frog/Triang organisation did market a simple receiver which employed the sub-miniature DL68 valve but I doubt that it was very reliable because it soon disappeared from the market. (The schematic for this receiver is shown on P19 with comment).
In all single valve super-regenerative radio control receiver circuits the valve has to be capable of oscillating at two frequencies simultaneously i.e. at 27 MHz and at a frequency of a few tens of kilohertz, called the quench frequency. The latter alternatively suppresses and enables the 27MHz. oscillations.
During the enable periods of the quench waveform 27MHz oscillations build up from circuit noise to a very low level before being suppressed by the quench period of the waveform. However the presence of a received 27MHz. results in a rapid build-up of oscillation even to saturation level. The drop in the mean anode current of the oscillator valve as a result of this strong oscillation causes a relay in the anode circuit to drop out and thereby to operate an actuator via its contacts. A tricky balance of requirements but capable of producing an extremely sensitive receiver with the minimum of components.
With the availability of transistors it soon became apparent that if the transmitter output could be keyed off and on at an audio frequency (tone control) the oscillator wave form would replicate the modulation and that the transistor circuits could be used to detect the audio content and to drive an actuator without the use of a relay, a much more reliable system. (Transistors were also used in oscillator circuits to effect the necessary keying of the transmitter but, in some applications an electric motor drove a contact cam). This was about the stage where I gave up radio control and moved on to other things.
Now that I have reached what is euphemistically termed as a good age and radio control has become beyond the capability of the amateur, I can only look back and wonder whether there was the possibility of doing things better in the old days. Thinking about this, it occurred to me that one possibility would have been to eliminate the valve filament and anode supply batteries by the use of a transistor dc.to dc. Converter circuit powered from the actuator battery. Furthermore this converter could operate at the quench frequency of a few tens of kilohertz and thereby to provide the quench waveform.
Because transistors were expensive it would have probably been preferable to use a single transistor in a ‘sine wave’ oscillator, subject to the availability of a suitable transformer, and I realised that a miniature output transformer from a pocket transistor radio would have been ideal for this purpose. (Miniature output transformers were available for home constructors at the time and still are). This transformer would provide the quench waveform from its secondary or speaker winding and it would produce a ‘high’ ac. voltage suitable for rectification to produce an anode supply from the ‘high impedance’ primary winding. The one thing missing would be a supply for the valve filament. Having bought a suitable transformer from the late lamented Maplin store and rigged up a suitable test oscillator I was able to determine that a 40 turn overwind would produce around 1.5V at 25mA for the XFY10 valve heater after rectification. (Installing this winding necessitates disassembling and reassembling the transformer core which is not difficult but a bit tedious). With this winding in place, all was thus ready to breadboard and test the elements of a receiver prototype separately; for which the overall schematic is shown on page 19 and which is repeated as Figure 1 below. This worked remarkably well and I wondered about the possibility of making a retro receiver unit suitable for installation in a model aircraft or boat. However I decided not to proceed with the project due to the lack of period components. Subsequently, at a loss for something else to try, I decided that it would be interesting to install the complete circuit on a board no larger than that for the IVY receiver above and to use slightly more modern components where necessary.
The completed ‘retro receiver’ is shown in Figure 2. Unfortunately during testing I inadvertently managed to burn out the filament of the probably irreplaceable XFY 10 valve and I was forced to substitute a Raytheon JAN 5678 valve. This valve operates satisfactorily in the circuit but the filament requires twice the power of that for the XYF10, which is beyond the capability of the 40 turn overwind on the converter transformer. Consequently it is necessary to operate the JAN 5678 filament via a resistive dropper from the 4.8V common supply. As this is not within the spirit of original aim of the exercise, I have decided to discontinue further development unless, perchance, an XFY10 valve becomes available. Some wave forms from the retro receiver with the JAN5678 valve installed are shown in the figures below.
Figure 1 (from page 19)
Figure 2 (Complete receiver assembly)
Figure 3 (Quench waveform with incipient 27Mhz. oscillation apparent)
Figure 4 (Oscillator burst during enable period with a very weak signal)
Figure 5 (Oscillator burst during enable period with a stronger signal)
Figure 6 (Oscillator bursts with 277Hz. keying of the signal generator)
Figure 7 (Filtered keyed burst prior to rectification and power
amplification for driving the actuator coil) (analogue scope)
Thanks for reading when Radio Control began.