Two-stage mains powered amplifier


Two-stage mains powered amplifier

Radioamator i Krótkofalowiec 1961/11. Author: K.W.
(A corner for beginner radio amateurs)

   The simple one- and two-stage battery amplifiers described in the previous issues of the magazine helped us to get acquainted with the basic circuits of this type of amplifiers. We must say, however, that battery power, apart from its specific advantages, has a major disadvantage: it is uneconomical. Therefore, wherever possible, radio equipment is powered from AC power circuits.

   AC powered amplifiers, popularly known as "mains" amplifiers, differ from battery amplifiers in that, apart from the actual amplifying circuit, they are equipped with a power module, usually composed of a mains transformer, a rectifier tube and a rectified voltage smoothing filter. Some details about the layout and operation of the power supply were given in the previous issue when discussing the design of the power supply, intended for cooperation with a two-stage low-frequency amplifier. The power supply is usually constructed as one unit with the amplifier or receiver system (e.g. radio receivers), and only in special cases it constitutes a separate element. The latter solution is used, for example, in the popular tourist receiver "Szarotka".

   There is also a second, fundamental difference between a mains and a battery amplifier: the use of other types of tubes. This issue requires further discussion due to its crucial importance.

   As we remember from the short explanation of the principle of operation of the electron tube ("Radioamator" No. 5/61), the cathode is the source of electron emission inside it. In the case of battery-operated tubes, it is simply a thin filament, heated to an appropriate temperature. The design of the cathode of the vacuum tube adapted to AC power is more complex.

   Figure 1 shows us in cross-section the cathode of such a modern vacuum tube. It is an "indirectly heated" cathode. As we can see, it consists of two basic elements: an electric heater made in the form of a spiral of resistance wire and the actual cathode. The latter, usually made in the form of a ceramic tube, is covered on the outside with a suitable substance which, when heated to an appropriate temperature, emits electrons. As you can see, the filament circuit does not directly participate in the work of the amplification circuit.


Fig. 1. Cathode of an indirectly heated electron tube (cross-section)

Indeed, the fragment of the amplifier diagram with the tube in question shown in Fig. 2 has a filament circuit completely independent of the rest of the circuit.


Fig. 2. Part of the schematic diagram of an amplifier with an indirectly heated electron tube

   Now we can present our readers a schematic diagram of one of the mains amplifiers. As shown in Figure 3, it is a simple and economical system, as it uses only one modern ECL82 electron tube.


Fig. 3. Schematic diagram of the amplifier

Despite its simplicity, the circuit presents full-fledged amplification equipment of quite good quality. Probably some readers have already encountered the term "Hi-Fi" (abbreviation of the English term "High Fidelity") in relation to electroacoustic devices. As you know, this is the name used to describe very high-class equipment, which gives excellent results due to the high quality with which it reproduces, for example, reproductions from discs. The high quality of amplification equipment depends on many different factors, among which one of the most important is the small amount of distortion of the reproduced sounds caused by the system. We say that a circuit does not distort if the amplified signal at its "output" has the same shape as the input signal. This is illustrated in Figure 4a. If, on the other hand, the amplifier causes distortions, it means that the output signal will have a slightly different shape than the signal fed to the "input" of the circuit. This is the case in Figure 4b.


Fig. 4. The voltage shape at the "input" and "output" of the amplifier
a - without distortions, b - with distortions

   One of the most common ways to reduce the distortion caused by an amplifier circuit is to use so-called "negative feedback". The phenomenon behind this new term for us is of great importance for the amplification technique and therefore it is worth getting to know them better. We will do it on the example of our amplifier, the schematic diagram of which is shown in Figure 3.

   The input signal is fed to a potentiometer R1 used to adjust the volume in a conventional circuit. From the potentiometer slider, part of the signal is sent to the control grid of the first stage of the amplifier, in which the triode part of the ECL82 tube works. From the anode resistor of the triode (220kΩ), the amplified voltages are fed through a 20,000 pF capacitor to the control grid of the vacuum tube operating in the output stage of the amplifier. The pentode part of the same ECL82 tube works in this stage. In the anode circuit of the power stage, we see the already known loudspeaker transformer, which adjusts the low resistance of the loudspeaker electrically to the much higher internal resistance of the tube. As you can see, the circuit of our power amplifier does not differ much from the previously known two-tube battery amplifier.

   Now, however, let us turn our attention for a moment. Our amplifier has one more connection, absolutely incomprehensible at first glance, namely: the voltages from the secondary winding of the loudspeaker transformer are fed through a 33kΩ resistor to the cathode of the first stage of the amplifier. This is the circuit of the previously mentioned "negative feedback". The operation of this interesting circuit will be easier to understand if we consider it in a simplified diagram.

   Figure 5 shows our amplifier with the omission of all irrelevant elements at the moment. Here we can clearly see that the output voltages (from the secondary winding of the loudspeaker transformer) are fed through the coupling resistor Rs (33 kΩ) to the cathode of the first stage of the amplifier. In this situation, a certain part of the output voltages is supplied to the cathode of this vacuum tube, with values depending on the ratio of the resistance R8 to Rk. It is obvious that the lower the value of the Rs resistor, the greater part of the output voltages will reach the discussed cathode. In an extreme case, if the secondary winding of the transformer were to be connected directly to the cathode of the vacuum tube (Rs = 0), then the full output voltage would be applied to it.


Fig. 5. A simplified diagram of an amplifier with feedback

   "Negative feedback" is called "feedback" because the output voltage returns almost to the input of the amplifier, while the expression "negative" is not that clear. The term is used here because the feedback voltages are fed back towards the input of the amplifier in such a way that they are subtracted from the input voltages in the first tube circuit. Feedback voltages are said to be "opposite in phase" to the input voltages.

   Figure 6 shows us this situation visually. The signal fed to the amplifier input is marked with the letter A; after amplification, the B signal is larger, but obviously has the same shape as the input signal, because we assume that the amplifier circuit does not cause distortion. Some part of of the output signal from the transformer secondary winding (C in the opposite phase) is fed by negative feedback towards the input of the amplifier. As a result, the first tube is influenced by the difference in the value of both voltages: input A and feedback C. We see both signals superimposed on each other (D), and the resultant signal (signal difference A and C) is presented as signal E. It is like it is not difficult to conclude, correspondingly smaller than the input signal A. Therefore, after amplification, the output signal F will be smaller than the output signal B. As we can see, the introduction of negative feedback into the amplifier circuit practically reduced its overall gain.


Fig. 6. Illustrative representation of voltage relationships in the amplifier circuit with negative feedback

   As can be seen from the above description, it cannot be said that the use of negative feedback has brought us any specific benefit, because it cannot be said that it is beneficial to reduce the gain of the circuit. However, let us now consider another case, namely the one in which the amplifier circuit causes distortions. Figure 7 will help us in this. We can see the input signal A and a correspondingly larger amplified signal B. This signal is distorted (it has a "spike" at the top of the curve). This is an example of the distortion introduced by the amplifier. Part of the output voltage of the same shape (C) is fed to the input of the amplifier. Both superimposed signals (D) give the resultant of the voltage (E). As you can see, this voltage is distorted (concave at the top of the curve). However, this signal, amplified in the amplifier circuit, results in an undistorted voltage (F). In this way, the amplifier, which normally introduces considerable distortion to the signal, works by employing negative feedback with very little distortion. It is not difficult to say that the negative feedback distorts the input signal in such a way that these distortions cancel out with the distortions introduced by the amplifier.


Fig. 7. Illustrative representation of voltage dependencies in the amplifier circuit with negative feedback, which reduce distortions

   The schematic diagram presented now should be completely understandable to us. One can only add that a small 100 pF capacitance, parallel to the 33kΩ feedback resistor, bypasses it for large (and practically short for very high) frequencies of acoustic voltages. This ensures stable operation of the amplifier (very strong feedback) and prevents unwanted oscillations at supersonic frequencies. A similar role is played by the 2000 pF capacitor, shunting the primary winding of the output transformer (it also eliminates the voltage of higher acoustic frequencies). The value of this capacity determines the "color of the tone" of our amplifier, therefore it should be selected individually, which will be discussed at the end of this article.

   To assemble the amplifier, we will need the following components:

  • Electron tubetype ECL82 - 1 pc.
  • Socket for the "Noval" vacuum tube - 1 pc.
  • Logarithmic potentiometer with a power switch W (from the "Figaro" receiver) - 1 pc.KCapacitors:
    • 20000 pF/250V - styroflex - 2 pcs
    • 4 μF/250V - electrolytic - 1 pc.
    • 25 μF/25V - electrolytic - 1 pc.
    • 2 ×50μF/350V - electrolytic - 1 pc.
    • 2000 pF/350V - styroflex - 1 pc.
    • 100 pF - ceramic - 1 pc.
  • Resistors:
    • 2,2 kΩ/0,5W - 1 pc.
    • 220 kΩ/0,5W - 1 pc.
    • 33 kΩ.0,25W - 2 pcs
    • 820 kΩ/0,25W - 1 pc.
    • 300 Ω/1W - 1 pc.
    • 2 kΩ/1W - 1 pc.
  • Mains transformer (according to the text) - 1 pc.
  • Selenium rectifier from the "Figaro" receiver - 1 pc.
  • Loudspeaker transformer type "Figaro" - 1 item.
  • Radio sockets with nuts - 4 pcs.
  • Speaker type GD 18-13 / 2 - 1 pc..
  • Power cord with a plug, aluminum sheet on the chassis and small assembly elements.

  All the parts needed to build an amplifier are readily available. Only the power transformer can be a bit of a problem. Here are its technical specifications:

  • primary winding - 220V,
  • secondary winding I - 200 V / 50 mA,
  • secondary winding II - 6,3 V/ 1 A.

The model amplifier uses a commercially available power transformer from the "Tatry" or "Bolero" receivers. This transformer was adapted to our needs by removing 375 turns of the winding. This procedure is not difficult because the transformer core can be easily disassembled (screwed together). After releasing the body, unwind the filament coil on the top (32 coils of thick wire) - carefully and delicately so as not to damage the insulating enamel. Then we unwind five layers (75 turns each) of a thin secondary winding. Cut off the unnecessary wire and lead out the previously removed insulating spacer. Then we wind the filament winding back. we fix them and start assembling the core. The whole operation is relatively simple and should not pose any special difficulties to anyone. After assembling, the core is screwed with screws.

   It is also possible to use a self-made transformer, as long as someone has the appropriate materials. In this case, a core with a diameter of the middle column of approximately 8 cm2 should be used. and make three windings:

  • primary (220V) containing 1100 turns, enamel winding wire ∅ 0.35 mm,
  • secondary containing 1000 turns, enamel wire ∅ 0.15 mm,
  • filament containing 32 turns, enamel wire ∅ 0.6 mm.

  Data for making a loudspeaker transformer on one's own:

  • core with a central column cross-section of approximately 3 cm2,
  • primary winding: 2100 turns, enamel wire ∅ 0.15 mm,
  • secondary winding: 56 turns, enamel wire ∅ 0.5 mm.

  It's best to start building an amplifier with a metal base. The assembly diagram of the system as well as the chassis dimensions are not given in this description. This is right, first of all, because in practice it is not possible to assemble elements of the same dimensions as in the model. In particular, these problems relate to electrolytic capacitors, which come in various designs and sizes. Figure 8, showing the location of the main components in the model amplifier, can be very helpful. In addition, for guidance, the photo shows the exterior of the amplifier.


Fig. 8. The arrangement of components in a model amplifier


Fig. 9. Photograph of a model amplifier

  In order to determine the dimensions of the metal base, one should properly position all the essential components of the amplifier on a sheet of paper and draw lines for cutting and bending sheet metal on it, as well as the location and sizes of holes for the assembly of individual parts. In order to eliminate mistakes, it is advisable to make a cardboard chassis model first, test its suitability, and then transfer its dimensions to a sheet of metal.

  It is best to start assembling the amplifier with mounting all the larger elements, such as: transformers, electrolytes, potentiometer, tube socket, radio sockets and a selenium rectifier. Regarding the latter, it should be remembered that dry rectifying elements of modern construction are adapted to be mounted directly on the "chassis". In this way, the metal casing of the rectifier can dissipate the heat generated inside it to the large mass which is the "chassis" of the amplifier. Improper mounting of the rectifying element, e.g. in the air or on a non-metallic plate (bad heat conductor), can lead to its overheating and destruction..

  After mechanically fixing all the larger elements, we carry out the electrical assembly of the system. As usual, correct and careful soldering is required, which will allow us to avoid many problems during the operation of the amplifier. Cables connecting individual elements should be as short as possible; Small resistors and capacitors, connected to the electrodes of the electron tube, are soldered directly on the tube socket. The grounding conductor is made of thick copper wire with a diameter of at least 1 mm. It should connect one of the input sockets, the left potentiometer lead, the metal pin of the tube base and the electrolyte housing by the shortest route. This wire is connected to the amplifier's base at one point, e.g. with a screw securing the tube socket and a soldering pad. Proper mounting and connection of the electrolytic capacitor is shown in Figure 10. The filament wires - twisted together - are placed directly on the "chassis" sheet. Other connections, particularly wiring in the control grid circuits, should be routed away from the chassis base and other components.


Fig. 10. The way of mounting the electrolytic capacitoro

  Initially, we do not assemble the amplifier completely - without connecting the feedback circuit (33 kΩ, 100 pF). We check the compliance of the connections made with the schematic diagram, especially in the supply part (the polarity of the rectifier is shown in the diagram in Fig. 3, as well as marked on the rectifier element), and then we connect the system to the network - so far without the ECL82 electron tube in its socket. If the moment of switching on "passed" calmly, without any disturbing signs (humming, crackling, smoke, etc.), it means that we did not make any serious mistakes during the installation. We can then - if we have a measuring device - start checking the supply voltages. In the absence of a measuring instrument, one should limit himself to a short observation and stating that none of the elements is heating up, and then turn off the amplifier.

  Next, connect the loudspeaker to the appropriate sockets and insert the electron tube into its socket (be careful not to bend the delicate feet). After re-switching on, we should find that the cathode slowly glows orange. At the same time, a slight hum and noise should be heard in the loudspeaker (the ear in the immediate vicinity of the loudspeaker). We check the operation of the volume potentiometer by turning its axis; as we know - the minimum volume is achieved in the extreme left position of the knob.

  Now we can start testing the amplifier: we supply an acoustic signal to the input sockets (eg from an adapter or a detector receiver) and we judge the quality of the reproduction "by ear". If the amplifier has been made correctly and using good quality components, we should immediately obtain the most satisfactory results.

  The last step will be to connect negative feedback. It should be done temporarily, i.e. without connecting to the secondary winding of the output transformer, because we must first of all determine the correct phase of the voltage drawn from this winding. We perform this procedure while the device is in operation, therefore it is necessary to exercise utmost care to avoid touching the elements that are under voltage, which is not very pleasant and dangerous..

  The prepared feedback circuit is connected for a very short time to the unearthed end of the secondary winding of the loudspeaker transformer and we observe the behavior of the amplifier. If you hear a strong hum or whine from the speaker, disconnect the feedback immediately as it may damage the speaker. Such "disturbing" behavior of the apparatus proves incorrect connection of the feedback branch, which - as it is not difficult to guess - is "positive" in this case and leads to the excitation of the system. In such a situation, the terminals of the secondary winding of the loudspeaker transformer should be inverted, i.e. the so far free end should be grounded, and the feedback voltage taken from the terminal currently connected to ground..

  With the correctly selected phase of the voltage taken from the secondary winding of the loudspeaker transformer, when connecting the feedback branch, you should hear only a very slight click in the loudspeaker, while at the same time the humming noise and hum that so far slightly audible should slightly decrease..

  If the amplifier is excited regardless of the phase of the feedback voltage taken from the secondary winding of the loudspeaker transformer, which may occur with incorrect installation of the circuit or incorrect output transformer, the 100 pF capacitor shunting the 33 kΩ resistor in the feedback branchshould be desoldered.

  To make it easier, the external appearance of the loudspeaker transformer used in the manufactured model ("Figaro" type) is given in Fig. 11, with the marking of the terminals being repeated in the schematic diagram (Fig. 3). Mounting the loudspeaker transformer according to these guidelines will allow you to avoid any surprises when starting the amplifier. For curiosity, it is also worth experimenting with the switching on and off of negative feedback when the amplifier is operated with a connected signal source, e.g. a turntable. The operation of this circuit discussed at the beginning of the article can then be practically stated: without the feedback loop, the amplification of our equipment is slightly higher and clearly decreases when it is turned on. The quality of the reproduction is difficult to judge "by hearing", but it is undoubtedly noticeable that with the negative feedback loop turned on, the amplifier works "softer".


Fig. 11. The pinout of the FIGARO loudspeaker transformer.
Designations according to the schematic diagram in Fig. 3

  The quality of the reproduction that we get with our amplification equipment is also dependent on the rest of the set, namely the loudspeaker and the audio signal source. For good results, the speaker should be mounted on a screen of the appropriate size. An example of a solution to this problem is presented in Figure 12. The screen is cut from a plywood 8 ÷ 12 mm thick. It is very interesting to experiment and find out how the screen works in practice..


Fig. 12. Approximate dimensions of the speaker screen and the way of placing it in the corner of the room

To do this, before the final mounting of the loudspeaker, start the apparatus and bring the loudspeaker close to the screen opening several times. The difference in loudspeaker operation and reproduction quality is very clear; it is particularly easy to see that the presence of the screen strongly emphasizes the bass. As the source of the audio signal, we most often use slow-speed turntable records or a simple detector receiver.

  Receiving a local radio broadcast with our set, shown in Figure 13, will be of very good quality, better than normally encountered.


Fig. 13. A simple system for receiving a program transmitted by a local radio station

  Due to the simplicity of the circuit, our amplifier is not equipped with a tone control, which is usually found in more complex equipment. The proper sound of the reproduction can be selected depending on individual taste by changing the capacitance (2000 pF) of the capacitor connected to the primary winding of the output transformer for another one within the range of 500 ÷ 5000 pF. Obviously, a larger capacity eliminates high frequencies more, giving the so-called "dark" color of the sound of the broadcast.

The content of the article for electron tube enthusiasts was provided by Grzegorz 'gsmok' Makarewicz