3.8 The Audion - Direct Receiver with Drain Detector
During the experiments with this receiver, the Author had decided to name this chapter “The BEST Direct (TRF) Receiver”, which he gave up on later, having in mind the old Latin saying: DE GUSTIBUS NON DISPUTANDUM EST (Tastes should not be discussed). It is, however, very hard to make something better with so little components. Anyway, the Author leaves to the readers of this chapter to name their own Best Receiver candidates, picking one of these described in this book, having in mind their own criteria for concepts of the beautiful, simple, cheap and useful. You can mail your voices to me on the address: ETŠ "Nikola Tesla" (Praktièna ELEKTRONIKA), Narodnog Fronta 31, Beograd, or by E-mail: tesla@drenik.net, Subject: Pe9.
On Pic.3.25 you can see the electrical diagram of this, however, anyway, nevertheless... Device. The signal of the tuned-in station is lead to the Gate of the BF256 transistor. Please notify that the signal is being taken from the upper end of the coil, and not from its leg, as it has been in previous projects. This is possible due to big input resistance of the FET (bigger than MegaOhm), compared to the one of the bipolar transistor (couple of kiloOhms). Why is this so important? Pic.3.26 shows the voltage - frequency curve of the parallel oscillatory circuit that is made of the coil L and capacitor C, while being tuned to the station whose frequency is fs2. In case the circuit is not loaded (the next stage of the device is not connected to it), this curve is shown in solid line and, as previously explained, the voltage (measured at the ends of the osc. circuit) of the station with carrier frequency fs2 is significantly bigger than the voltages of the stations with frequencies fs1 and fs3, although all of them have the same size in the antenna. However, when the next stage, containing the bipolar transistor as, for example, the one on the Pic.3.12, is connected, is small input resistance is damping the circuit and the bandpass curve has
he shape shown with the dashed line, marked as Q2.When the FET is connected to the oscillatory circuit (as on Pic.3.25) there is practically no damping, and the bandpass curve remains as shown in solid line. This is, clearly, much better, since all the other station voltages are more suppressed (reduced), comparing to the voltage of the tuned station. Considering the curve marked as Q3, we will be discussing it more in context with the Pic.3.29-a.
It has been earlier noticed that the most important characteristic of a parallel oscillatory circuit being used in a receiver is its resonance frequency
Its second most important feature is the Goodness Factor of the receiver, which is most often being marked with the letter Q, and is therefore also known as the Q-Factor. A loaded circuit has smaller Q-factor than the non-loaded one, as shown on Pic.3.26, consequently being Q1>Q2.For example, the goodness factor of the circuit from Pic.3.6 is Q=95. Since it is not loaded, the oscillatory circuit on Pic.3.25 has smaller (narrower) bandpass and therefore better selectivity. Additionally, since the whole signal from the circuit is led in the next stage (instead partly, when getting the signal from the coil’s leg), the receiver has got bigger sensitivity (it’s capable of receiving weaker signals).
The FET, together with R1, R2, C2, C3 and C4 forms the so-called Drain Detector (its analogous circuit with bipolar transistors is the Collector Detector, and with the electronic tubes - the Anode Detector. The popular name for the anode detector was - the Audion). The LF signal being detected is received on the drain (D). It has the same shape as the LF signal obtained on the output of the diode detector, but is significantly bigger than it, since the drain detector also amplifies the signal. The LF signal is then led on the volume regulation potentiometer, over the filter that is used to suppress the remains of the HF signal carrier (R3 and C7). After that it goes to the audio - receiver.
* The R7 resistor can be omitted. If you do that you should switch off the battery (over the switch S) every time you are removing the coil from the circuitry (during experiments). If you fail to do so, the oscillation in the circuitry will occur, and loud hum will be heard from the loudspeaker.
* Since FETs have very different characteristics compared to each other, it might be necessary to change the value of R1 resistor. The simplest way to do it is to place a 50 kOhm variable resistor instead, tune the receiver to some radio station and then achieve the best possible reception by moving the slider. The resistor is then removed from the circuit, its resistance measured, and the (fixed value) resistor of similar resistivity soldered in the circuit. The same goes for R2.
* The filter (R3 and C7) used to suppress the remains of the signal carrier affects the colour of the tone of the LF signal. If you wish more bass tones you should increase the C7 capacitance. Similarly, if you wish more high pitch tones, C7 should then be decreased.
* The receiver will not start operating the very same moment the switch S is engaged. That is due to a fact that the FET doesn’t work under small supply voltages. Its supply voltage is the one on the C5 capacitor and the detector won’t work until C5 doesn’t fill. This is being achieved through R4. Since this resistance is quite big and so is the capacitance of C5, the filling time is rather long. If, however, you just need the receiver that will have a “late start”, you should be increasing the capacitance and the resistance until reaching the desired delay time.
* This receiver works well also with the ferrite antenna. On Pic.3.27-a you can see the symbol for it, and on Pic.3.27-b its shape and dimensions are given. The simplest thing to do isusing the antenna from an old pocket radio, probably the same one you took the variable capacitor from. During the dismount, you should by no means cut the coil ends or shorten them later. Instead, you should carefully un-solder and unhook them from the PCB (the coil is made of the “litz wire”, consisting of a dozen very thin lacquer - isolated copper wires, wrapped together with the thread. If you cut this cable, you will find very hard to re-solder it, since it is difficult to remove the lacquer from all the wires without damaging (some of) them). Such an antenna, as seen on picture, has four ends. We shall be using the coil L, therefore only significant ends for us are those marked as 1 and 2. The end No.1 is easy to identify, it is the single one, which is not the case with the end No.2. To detect it in the group of three you will need an ohm-meter of some other conductivity tester, which you should connect to the end No.1, and then search for No.2 by touching those 3 remaining ends.
In the lower right corner of Pic.3.27-b you can see means of attachment of such an antenna to the PCB. The envelope being shown is made of carton. It is wrapped around the ferrite core, and then glued to it. The screw is mounted through the hole and the antenna is then attached to the PCB. Instead of screw and nut, the antenna can also be glued onto the board. If the ferrite rod is longer than 6 cm, the envelopes should be mounted on both rod ends.
* C6 is the block-type capacitor that (together with R4) prevents the receiver from working in an unstable regime. If such a thing occurs, its capacitance should be increased. If that doesn’t help either, or if the detector is working improperly, you should try increasing R1 and R4.
* With this receiver the reception of the SW band stations can also be achieved. All that is to be done is making a new coil. For these purposes we utilized a piece of carton cylinder already used for building our coil L (described in previous projects). On it, we bended tightly 6 bends of 0.6 mm copper wire. The wire diameter isn’t critical, practically any can be used. This coil is shown on Pic.3.28-a, together with attachment plates. This coil should replace the coil L, as seen on Pic.3.25. With 6 m long antenna and the antenna capacitor C1=12 pF, the reception bandwidth should be app. from fd=7 MHz till fg=10 MHz. This can be changed by changing the number of bends on the coil and /or C1 capacitance. I such a way you can “take a peek” what’s happening in the civil area, what are the radio - amateurs doing, some professional links etc. You can even make a multiple - legged coil, such as on Pic.3.28-b (number of bends isn’t critical, it may be useful trying out some other values, too), and to be able to choose SW1, SW2, SW3 with a selector switch. Please do have in mind that the reception quality of the SW stations isn’t the same during the day. It is good in the afternoon hours, during the night and in the morning, the weakest reception quality is around noon. But, that isn’t all. It also depends on the season, solar activity etc. Anyway, you should see it for yourself.
There’s also a possibility of receiving professional stations working on considerably higher frequencies. The coil, which is then also an antenna, is given on Pic.3.28-c. It is made of stronger, thicker wire or an metal band, being circularly shaped and then attached to the wooden plate with two screws. With same screws cables connecting the antenna with the variable capacitor C are affixed. The antenna diameter varies from couple of cm till few dozens cm, the real value being found by experiments. It is a directive antenna, which means that the amount of voltage being induced in it depends also upon the direction where the waves are coming from. This gives you the opportunity of achieving the optimum reception of the desired station and simultaneous suppression of others by rotating the antenna. Similar antennas are used in radio - location (searching for whereabouts of an unknown radio transmitter).
