GARC Regenerative radio night.

GARC Regenerative radio night. A follow on project from the highly successful Crystal set night. Aim. It is hoped that we can continue on from the very high level of interest and participation that occurred on the Crystal set night. Although these radios appear very simple, we have all learnt things about Construction, Tuned circuits, and the Detection process. I hope that we can build on this and learn more about Q, Amplifiers, and Feedback. It is also a way to follow the development of radio during the early part of last century. The competition Although the aim is to get as many as possible involved, we will issue certificates along the lines of those issued for the Crystal sets. Most Bizarre Most Authentic Best Beginner Most Innovative All of the above will be judged as a People s Choice award Best Performance Best Overall These will be judged on several performance categories as follow. Sensitivity: Selectivity: Antenna coupling/isolation: Regeneration control: The number of stations that can be heard at useable level. The ability to separate stations, and freedom from overload. The ability of the radio to operate over its tuning range on the antenna provided. (Allowance will be made for special cases like VHF superregenerative sets.) Freedom from hysteresis, Smoothness of control. Obviously some of these criteria are subject to interpretation, especially as we may be judging between broadcast, shortwave and VHF sets. Allowance will be made for radios with amplifiers to drive a speaker, and those designed for headphones. A brief history of receivers used mainly for entertainment. Crystal detectors (Diodes) were used in various forms from the earliest days of radio. After Braun discovered the rectifying characteristics of various junctions in 1874 the principle was applied to detecting RF energy by Jagadiss Bose, late that century and was announced by both Bose and Pickard in 1904 as radio receivers. After the first transmission of voice on 23 December 1900 by Fessenden, and the subsequent development to achieve 12 miles range in June 1906 the first audio broadcast was made on Christmas Eve of that year. This eventually lead to entertainment and commercial broadcasting. Crystal sets formed the first receivers used by the public until around 1920, they have continued to be popular as an entry level radio in the 1920 s and a child s first radio at least into the 1970 s. They also remain as a fascinating hobby interest to the current day. 1

In 1912 a third year undergraduate Edwin Armstrong, who had been playing with radio from an early age took the Audion (Triode) invented by De Forest and explained how it worked. (Something not understood by De Forest) He then went on to introduce feedback to the receiver circuit thus inventing the Regenerative detector and enabling a vast increase in both sensitivity and selectivity. This invention, despite his subsequent invention of the Superheterodyne receiver during WW 1 formed the basis of many of the smaller commercial and home built radios through the 1920 s. In the early 1930 s with the lapsing of the patent on the superhet, and its ease of use it replaced the regen and TRF radios of the previous decade. Despite this the regen set continued as the most common ham receiver for short wave until the late 1930 s An interesting sideline is that the simplicity and low cost of this circuit enabled the DKE 38 (Deutscher kleinempfanger (German small ceceiver)) sometimes referred to as Goebbles Schnauze by the people. And the more expensive VE 301(dyn) Volksempfanger (Peoples receiver) to be produced in Germany at a price that enabled most of the population to own a radio during the depressed economic times in that country from both the 1930 s depression and war retributions imposed by the allies. (They cost around one weeks average salary for the two valve DKE38, and two weeks salary for the three valve VE301(dyn)) By this means Hitler was able to get his propaganda to the people, and also show that he was improving their lifestyle. 2

Another role for the regen set was in Clandestine radios used by all sides during WW 2. The Paraset transceiver and Sweetheart short wave receiver were examples of this technology used by the allies. 31 1 Sweetheart Receiver with battery box and accessories 31 1 Schematic After the Second World War the superior performance and abundance of war surplus superhets pushed the regenerative set into the status of The beginner s second radio and a limited number of very basic commercial offerings, and home built/kitset radio s for both broadcast and short wave. Effects of regeneration on a tuned circuit. The ability of a tuned circuit to select a desired signal and reject an undesired one is determined by the Quality factor Q of the circuit. This is related to losses in the tuned circuit and any energy being removed by other circuitry that is connected to it. A simple approach is to consider all losses and loads in terms of resistance that takes energy out of the circuit. This takes many forms; some is resistance in the wire in the coil and insulation around the capacitor. Also the dielectric of the capacitor itself has loss along with eddy current s set up around the coil. Another major issue is that to be useful, we need to couple energy in to and take energy out of the circuit. Just as an antenna can capture energy and feed it to our receiver, it can also take energy away as we desire with a transmitter. It can obviously be considered as a load and we say it Loads our tuned circuit. In the same way, whatever is coupled to the circuit in the way of a detector or amplifier also takes energy out of the system. Anything that loads or removes energy from the tuned circuit reduces the Loaded Q of the tuned circuit. If resistance reduces the signal amplitude, and Broadens the frequency response of the tuned circuit, it would appear that we actually want Negative resistance to achieve the selectivity and signal strength we desire. Negative resistance is at first a strange concept, however if we consider that resistance takes energy out of a circuit it is simply a case of putting energy back in to the circuit In Phase. This is exactly what we do in a regenerative receiver. The circuit is the same as that of an oscillator with the addition of some form of control on the feedback. For a regenerative receiver to work properly we need to carefully control the feedback to just below the point of oscillation to receive AM. For CW (Or SSB) we need to have the unit just oscillating. It is interesting to tune in a strong broadcast station and change the regeneration setting whilst listening to music. The effect on bandwidth can easily be heard on the recovered audio. Another effect that is noticeable is called Ringing This is a characteristic of very narrow filters; you sometimes hear it when you use a CW filter on your transceiver. You often set the regeneration control back some way from the point of oscillation to get nicer sound. 3

Methods of producing and controlling regeneration. One very important aspect of regenerative radios is the ability to Feed Back a controlled amount of RF that is in phase to the tuned circuit in the regenerative stage(s). There are many ways to achieve this; however some are easier to control than others. The most common method is to use a separate coil that is referred to as a Tickler to couple energy from the output back to the input. This requires some means of adjusting the amount of feedback (Energy). If a variable capacitor is used this is sometimes called the Throttle capacitor. Another consideration is that the antenna can Suck out sufficient energy to prevent the receiver from going in to regeneration on some frequencies. This is usually when the antenna approaches a quarter wavelength, or multiple that has a low impedance. The AM broadcast band is around 180 to 550m, so if the antenna is approaching 45m long it can start to cause problems. (The antenna at the GARC would fit in to this category!) Another issue is that if the antenna swings in the wind it can cause some sets to go in and out of oscillation. This all sounds like things are All too hard, the reality is that these sets usually just work. We do need to consider how long the antenna is when we decide how to couple it to our set. Armstrong s first patent was for a TPTG(Tune plate, tune grid)design. This relies on capacitive feedback from plate to grid within the triode valve. This would have been a truly cantankerous beast of a receiver that would have needed three hands to tune! Refer to the image on page two. This 1920 s Receiver belonged to Lional Learmonth OA3PL later VK3PL in Hamilton. It shows the adjustable coupling coil where feedback is controlled by moving the tickler coil closer or further away from the tuning coil. This was satisfactory for use on LW and MW, however the adjustment of regeneration would also effect the tuning. This was tolerable at low frequencies, but became unmanageable on SW. The third coil is most likely the antenna coupling, it is adjustable so you can obtain sufficient coupling whilst not preventing the set from regenerating. A very popular way of providing and controlling regeneration is to connect a tickler coil from the Plate, Collector, or Drain of the active device and using a throttle capacitor to vary the amount of energy fed back to the input coil. This method is more satisfactory than the adjustable coils, and was used from at least the 1920 s on, with it being the main technique for short wave receivers in the 1920 s and 30 s. Depending on the design, this capacitor is often at a high voltage above ground. Refer to the DKE38 and 31 1 schematics on page two and three. Another popular method that was used from at least the early 1930 s was to place a pot, or rheostat across the tickler coil. This has the advantage of being cheaper than the variable capacitor. When transistors came along in the mid 1950 s the pot was even more desirable, as it suited the low impedance of transistors, and generally allowed more compact construction. This example is from Radio TV and Hobbies, July 1962. I am using this as the basis of the GARC Regen set. The new design will use a NPN Silicon transistor. 4

Another method that suites Tetrode / Pentode valves is to take the signal from the screen to provide the feedback to the tickler. This usually also works in combination with screen voltage adjustment to change the gain of the valve. A variation on this is to control the gain of a dual gate FET using Gate 2. From the 1957 ARRL handbook If you are happy to have a taped coil the use of the Cathode, Emitter, or Source as the means of returning energy back to the input tuned circuit. This example uses a FET with the device gain controlled by varying the operating voltage by means of a pot. From seekic.com This example takes a different approach. The antenna connects to an untuned RF amplifier. This has the advantage of isolating the antenna from the regenerative stage. Although it will contribute little to the gain it will improve the control, and in this example allows the regeneration to be fed back to the cathode without a tickler coil. This enables the use of a standard RF coil in the receiver. This design is slightly different; it actually uses a Q Multiplier followed by a detector. This probably results in less audio distortion. Radio TV and Hobbies, June 1964 Other types of regenerative radios. Most of what we have seen relates to the regenerative detector, it is also possible to have a regenerative RF amplifier. The only difference is that the amplifier is biased for linear operation (Class A). In a straight receiver this would be followed with a diode detector just like in a crystal set. Sometimes this regenerative amplifier is in the IF of a communications receiver, in which case it is called a Q Multiplier. 5

There is another class of receiver called a Super regenerative receiver that is designed to work at VHF. Its principle of operation is different to the conventional Regen set. Safety recommendations. There is little risk from low voltage battery sets. Valve types on the other hand often have high voltages, sometimes as high as 350 V DC. These types are almost always mains powered, so there are at least two lots of potentially lethal voltages present. It is quite possible to make a very satisfactory valve set that works from 9 to 13.8 V on the anode. As the current involved is very low it is also practical to supply the HT (High Tension anode supply) from 2 to 5 9 V batteries. In the 1920 s it was common to run the detector on 22.5 V and only use the HT on the audio amplifiers. Nowadays we can cheat and use a simple IC audio amplifier, or perhaps an old pair of computer speakers. I strongly recommend that you stay away from mains operated sets unless you are experienced in their construction. What to build, Where to find information. A web search will reveal thousands of designs for regen receivers. Old magazines are also a good place to look. Several manufacturers also sell kits; I would question the value of some of these, although that is up to you. Perhaps you made a regen radio in your younger days and wish to revisit the project. Two club members have already found the original articles in Radio TV and Hobbies! If you don t know where to start I have tested a modified version of a project from Radio TV and Hobbies, July 1962 with the option of using a simple IC audio amp, or the traditional headphones. This can be made from modern parts available from Jaycar for around $25. If you want to scrounge some components, or modify your crystal set you can probably build it for nothing or perhaps $7.95 for the Champ audio amp. I will make plans for this version available after the talk for anyone who is interested. (This is also available on the GARC website.) Another design that has been tested, and works well if you want to make a Valve Radio was published in AR May2012 by Peter Parker. I have also made a Battery Valve Radio using a 1T4 with screen grid regeneration that will run from a 1.5V and 4 or 5 X 9V batteries. All of these will be available to play with after the talk. There is a potential issue when using a audio stage on simple regen radios, A class B audio stage will draw more current on audio peaks then with no output, also there is a lot of RF in the output that can cause trouble with some audio stages. Many regen stages are very sensitive to supply variations and uncontrolled feedback, and may Howl when using a common power supply and audio amp. If you are using a 9 V battery, consider using a separate battery for the audio stage. You could also use a 5 V regulator for the receiver, although this is starting to complicate things. I also recommend placing a 1k resistor in series with the output, and a 0.0047uF to ground in parallel with a 10k Volume pot. Additional measures may still be needed. Now it is time to warm up the soldering irons! Images used in this note are from a variety of web sites, and scans from magazines and personal photographs. David Learmonth VK3QM 2012 6