CW is for Continuous Wave, which determines the radio link where the Morse Code is being transmitted by cutting the oscillator work in the transmitter.
SSB and CW signals are impossible to accomplish with the receivers that use the ordinary diode - type detector (earlier described AM receivers). The receipt can be done only by bringing another signal into the detector, the HF signal from the oscillator, known as the BFO (Beat Frequency Oscillator). Simpler solutions, however, do exist. These are the reaction - type receivers, i.e. receivers with positive feedback.

You have been able to see one of them in the previous project (3.29-a), and here we’ll take a look at another one, which works so nice that we were sometimes having the impression it beats up much more sophisticated, modern supereterodyne receivers. Its electrical diagram is shown on Pic.3.29-b.
The coil L and capacitors C and C1 form a parallel oscillatory circuit whose role is to separate and amplify the signal of the tuned station, and to suppress all others. It doesn’t entirely succeed in that, however. The reason for this is small Q- factor of the oscillatory circuit, being such because of big energy losses in the circuitry. There are many kinds (reasons) of these losses, but we can imagine in first approximation that there’s a resistor RG in the circuit which represents these losses, its resistance being such that the oscillating current transforms itself into heat dissipation energy on it, its amount being the sum of all the (actual) losses in the circuit. We could, furthermore, solve the problem of these losses if connecting serially to RG a resistor RG’, whose resistance would be negative and equal to the value of RG by its absolute value. The overall resistance would then be zero, there would be no energy losses and the Q- factor would become infinite. The oscillatory circuit would, together with the components that create this negative resistance, become an oscillator capable of receiving SSB and CW signals.
We don’t really need an infinite Q- factor while receiving usual (conventional) AM RG by its absolute value. The resistances would not cancel each other completely, but the losses would be made very small, the Q- factor therefore becoming very big therefor increasing both the selectivity and sensitivity of the oscillatory circuit.
Transistors T1 and T2 constitute, together with resistor R3, a two-stage amplifier with strong positive feedback that has a negative dynamic input resistance. This negative resistance is connected between the leg No.3 on the coil and the ground, therefore superimposing itself with the resistance representing losses of the circuit. The quantity of this negative resistance depends on the amount of the DC current flowing through the transistors, which is being regulated by altering the DC voltage on the right end of the R3 resistor (by moving the slider of the P1 potentiometer).
The red LED D and the resistor R2 comprise a simple voltage stabilizer, obtaining 1.8 V of stabilized voltage on the P1. That means that the voltage on the right end of R3 shifts between 0 and 1.8 V while moving the slider of P1. The current flowing through the transistors thereat also changes, causing the voltage on the left end of R3 to vary between 0 and 0.6 V.
The signal of the station is being led from the leg No.3 of the coil into the collector-type detector made of T3, R3, R4 and C4. That is an AM signal detector that performs both signal detection and its amplification. Its name is the Audion. The LF signal is then, from the collector of T3, over the coupling capacitor C5, being led onto the sound volume potentiometer P2 and the audio amplifier. For the latter any of the earlier described devices can be used.
Tuning this receiver on the desired station requires both some knowledge and patience (that’s what finally “buried” this kind of receivers). Put the slider P1 in the upmost position. If strong whistling is heard that means the oscillating began. Move the slider carefully downwards until the oscillating stops. Then start slowly turning the rotor of the capacitor C until you come upon some station. If the whistling re-appears, move the slider of the potentiometer very little downwards, the whistling should stop and you should be able to hear the radio - station programme from the loudspeaker (loud and clear). For the next station tune yourself with C, then move the slider P1 upwards until the whistling appears, then put the slider back until it stops etc. All this may seem rather complex at first, but with a little practice and with two hands all will go quick and smooth.
The abovementioned method is for the signal reception of ordinary, broadcast stations. If you wish to receive the SSB and CW signals you should move the P1 slider upwards until the oscillating is achieved, so that articulate speech (SSB) or Morse code signs (CW) can be heard from the loudspeaker.
* The coil L is being made on the cylindrically - shaped body 6 mm in diameter, about 25 mm long. The plastic - made body taken from an old device is the best, like the one shown on Pic.5.14-b. The screw-shaped core allows the setup of the inductance, adjusting therewith the reception bandwidth of the device. If you cannot find such coil body, any plastic- or carton- made cylinder can be used instead. If you don’t have even that, then make yourself one. Cut the paper band to be 25 mm wide and about 150 mm long and reel it around the flat part of the 4 mm drill, adding every now and then some glue (UHU or similar). When the glue gets dry, remove the coil body off the drill.

The coil L has the total of 20 quirks of the lacquer - isolated copper wire, having 0.3 to 0.5 mm in diameter. A leg should be made on every fifth quirk. Latching of the wire ends (with small holes made in the coil body), as well as leg making (by making wire loops) can be done acc. to the instructions given with Pic.3.6. It can also be accomplished differently, as shown on Pic.3.29-b. First, 4 separate coils, each one made of 5 wire rings, are made side-by-side on the coil body. The starts and ends are fixed with scotch tape. The isolation is then removed from all coil ends, about 5 mm in length, after which they are tinned. On the PCB the legs are being soldered in pairs, the end of one coil with the beginning of the next (they are put together in the same hole on the PCB). For example, the end of the second and the beginning of the third coil should be connected on the same line where contact for the left end of C3 capacitor is, thus creating the leg No.3 of the coil. Putting two wires through one hole is not a very professional solution. The “real thing” are separate junctions, one for each wire, as shown on Pic.3.29-d-c.


* The feedback may happen to be not big enough, causing that there’s no oscillating even when the P1 slider is in the rightmost position. In that case, leg No.2 of the coil should be used instead of No.3. Switching between the legs can be done in many ways, the nicest (?) one given on Pic.3.29-d, made with factory-made contact pins and jumpers. On Pic.3.29-d-c you can see a detail of the PCB for the receiver shown on Pic.3.29-b. In the contacts marked as x, y and z (distance between them is 1/10 inch) the contact pins are soldered. The jumper is in position marked with dashed line, therefore making contacts x and y short-circuited. When it is moved in vertical position, the x and z contacts are in junction. In former case the coil leg No.3 is used, and in the latter it is No.2. In factory-made devices, these jumpers and contacts are used, together with appropriate connectors, to connect the PCB to the loudspeaker, power supply, variable capacitors, various switches etc.

* Setting the collector-type detector circuit to optimum operation is done by changing the R3 resistance, until voltage on the collector of BC549C is 1.2 V.


* The antenna can be a piece of copper wire no longer than 50 cm, but with longer (few metres), external antenna, the results will be much better.

* This receiver is scheduled for the reception of SW stations from 6 MHz till 9 MHz, which is accomplished with C1 value of about 400 pF. The exact value for C1 is being determined experimentally and can be significantly different. Going down to the amateur range (about 3.75 MHz) is performed with bigger C1 capacitance.