I’ll make a reservation right away - this anthology in no way claims to be a manual on lamp circuitry. Schemes (including historical ones) were selected by a combination technical solutions, if possible with "zest". And everyone has different tastes, so don’t exact if you didn’t guess right ... In the old schemes, a number of denominations are reduced to standard ones.

Skeptics claim that some schemes cannot sound at all "by definition". Here is one diagram that gives just that impression. But still it worked!

This diagram is taken as starting point. The amplifier is made on then new finger lamps, according to the classical scheme on pentodes without a common OOS. The treble tone control circuit is interestingly solved, but it can really work "on the rise" only with a high-quality output transformer. Since the amplifier was intended for an electric player, they saved on a power transformer. If, apart from the pickup, nothing else is connected to it, electrical safety is observed with some stretch. It's good to live in civilized countries - the sockets are correct. Here is phase, here is neutral, here is zero. And for some reason it is the same in all outlets. And in my apartment, for example, some of the switches were not in the phase wire, but in the zero one. What after that to demand from sockets ...

The pentodes in the first stage were abandoned rather quickly. Two triode cascades coped with this task no worse, and the sound quality increased. Further improvement was brought by ultra-linear circuits of the output stages. In such an inclusion, the screen grid is attached to the outlet primary winding output transformer. The resulting local OOS significantly reduces the output impedance of the cascade and increases its linearity, and the gain does not decrease much. True, the ultra-linear circuit was mainly used in push-pull amplifiers. Below is a diagram of a typical single-ended amplifier with an ultra-linear output stage.

fig.2

The values ​​of the parts in the tone control have been adjusted to meet modern requirements - in the original they only hunched the frequency response at 5 kHz. However, the rise of HF was rarely used at that time. Variants of this scheme blossomed wildly in the era of economic councils, when the party and the government decided to flood the country with cheap radio products. The ultra-linear cascade disappeared, the tone control was simplified, and the power transformer was often abolished altogether or only an incandescent one was installed. Saved on everything, and it is noticeable. Many people remember the sound of players in cardboard suitcases - a good middle, but there is nothing else.

When repeating the circuit, you can abandon the tone control, and with it, eliminate the first stage of the gain. Then in the two-channel version, only one double triode is needed for the driver. It is also possible to introduce a shallow FOS from the amplifier output into the cathode circuit of the first or second stage.

An increase in the depth of the OOS in tube amplifiers is prevented by the phase incursion on coupling capacitors. To eliminate this shortcoming, the interstage communication must be direct. And such a scheme appeared:

fig.3

Since the slope of the lamp decreases at low anode voltage, a pentode had to be used to obtain the necessary gain. Triodes with the necessary characteristics appeared later. Another highlight of the circuit is the inclusion of a bridge tone control in the amplifier's overall OOS circuit. The advantage of this solution is that at the maximum rise in the frequency response, input overload is excluded. If the adjustment is made in the preamplifier, there is a risk of such an overload. Therefore, the inclusion of regulators in the OOS circuit of the power amplifier was used for a long time and in amplifiers based on transistors and microcircuits. The sound quality, by the way, clearly benefits from this.

The direct heir to this scheme is the Gubin amplifier, a constant participant in Hi-End exhibitions. It can work with pentode and triode switching of the lamps of the output stage. For complete happiness, you can also provide an ultra-linear option.

fig.4

However, direct-coupling schemes also have disadvantages. The first is the need to apply anode voltage only after the cathodes have warmed up. Otherwise high voltage on grids can damage the lamps or shorten their life. To do this, you need to use devices for delaying the supply of anode voltage, or to make a rectifier on a kenotron with a large thermal inertia of the cathode. At worst, you can use a separate toggle switch for the anode voltage, but this is not very convenient.

The second disadvantage is the contradiction between efficiency and sound quality. When using automatic bias in the output stage, one has to either reduce the anode voltage of the driver, or put up with an increase in power dissipated by the resistor in the cathode circuit.

An interesting solution to this problem was found at http://www.svetlana.com/. You can apply a signal to the screen grid circuit of the output pentode, the constant voltage on it is usually close to the anode voltage of the driver. The auto-bias resistor can then have a relatively small resistance. True, the slope on the screen grid is much lower, but the linearity is better. At the same time, the first grid is grounded, and the pentode turns into a kind of triode that works with the grid current (A2 mode). But the driver will have to be powered by a cathode follower.

fig.5

By the way, if the first grid of the output pentode is not directly grounded, it can be used to supply a local OOS signal, including a frequency-dependent one. And this is the way to create a bandpass amplifier without a separate crossover.

A similar driver solution is used in another amplifier. He got here because of the parallel connection of the triodes of the output lamp. However, there are many disadvantages, first of all - monstrous extravagance. Of all the power consumed by the amplifier, almost a third is in the bias circuits. It would be much more reasonable to use separate rectifiers for bias, and in the driver - SRPP on a medium power double triode.

Lovers of high-quality sound will appreciate our homemade product today: the Magnifique Evolution tube single-ended amplifier.

Made in ex-USSR from the old soviet components.

This is a story about where the unconscious desire to create something out of nothing can lead, and what this “monster” can turn into.

Jazz-rock and roll improvisations

There is such a kind of pastime - viewing all kinds of equipment on the network: Maranza, Denons, Yamahas, Rotels, Nedas, and so on. A similar remake surrounds from all sides, hangs over your head, "begging" to buy, change and buy again. Which sometimes happens.

But in parallel with this, I was in the process of finalizing my technique. With audiophile methods, and not only, I tried to get a sound that, in my opinion, corresponded to the level of quality I needed. I wanted a dynamic (driving), warm, detailed, transparent sound with an unshakable stage. Somewhere like this ... And it finally happened. I got the sound I needed on the transistor class AB. So, what is next? A dead end... If the sound is warm and detailed, transparent and driving, then in the conditions of a particular room it is impossible and ridiculous to make it even more of what it already is, because all this is already there. But you always want something more ... And there is a way out! Japanese amplifier Luxman 550A. Transistor class A. Total 20W per channel. But what watts! Distortion - 0.005%. Inside, everything is not just right, but perfectly right. Large arrow indicators are pleasing to the eye. Hull color and finish. In short - Hi-End! 5700$!!! Wait…

In my case, it's more interesting, that's for sure, to get the same sound on a different type of equipment. Lamps! Lamps, lamps! It's inspiring.

For every self-respecting connoisseur of music should have in his arsenal and. This inspired me, especially the creations of those brave gentlemen who presented their projects on the net.

In order not to reinvent the "bicycle", it was decided to create an amplifier from ready-made circuit solutions with the introduction of their own ideas about the "correct".

Scheme

It is located on a separate board and contains a double electronic choke with an anode voltage delay according to Chugunov's design. On the same board there is a bias voltage source and a constant voltage regulator for the filament of the driver lamp. The food spectrum is very clean.

The whole circuit is powered by two transformers: the main one - TAN-43 and an additional one, 10V, to illuminate the indicators and to heat the driver lamp. As the second, you can use the TN-30.

The amplifier itself is made in a separate block, according to the scheme of Oleg Chernyshev, with a fixed offset of the output lamps. Capacitors C4 provide treble correction. For a linear characteristic, 100 nF is sufficient. I have 200 nF - 1db boost.

Type-setting volume control, based on the Soviet instrument paired switch for ten positions. I disassembled it, cleaned it, lubricated it with thick silicone grease and installed a softer spring. The entire structure is surrounded by a screen made of copper sheet. The total resistance of the regulator directly affects the level of the lowest frequencies and the noise level. I recommend its resistance in the range of 15 - 24 kOhm.

Output transformers TV-2Sh (TVZ-1-9). They were selected from seven pieces according to the maximum signal level and frequency range.

The input selector on the toggle switch for two sources. Of course, there is an opportunity and a place to build a relay system if you want in the future.

The wires from the input jacks to the selector are wrapped in a thick silver-plated copper screen from some kind of military radio cable.

The body wires from all blocks and screens and from the front panel converge in the center of the power supply at the minus of the capacitors.

The mutual arrangement of all nodes, blocks and connectors was determined by the minimum noise level and controlled by a spectrometer with subsequent comparative measurements of RMAA.

Instead of Oleg Chernyshov’s block, if you don’t like his sound or for experiment, you can put an amplifier block according to such a common scheme http://cxem.net/sound/amps/amp46.php or any other single cycle of the corresponding current consumption.

The PSU components mounted on heatsinks are located upwards with a heat sink. When mounting these components vertically with others , the wiring must be changed .
R24 - adjustment of the filament voltage L1 6.3 V.
R17 - adjusting the output voltage of the electronic throttle 300V.
C14 - determines the time for setting the operating mode of the anode voltage.
R11 - setting the current of the pentodes. At the anode - 300V. On R10 48mV.
R12 - setting the level of the indicators to 0db (2W) 2.85V with a load at the output of 4 ohms.

Frame

Chernyshev called his amplifier Pokemon - a small pocket monster. In my case, due to almost complete ignorance of what I do and what I want, I got an amplifier for a full-format case. Or the pocket should be larger ... I didn’t have the goal of compacting all the details for compactness. Moreover, if you want to change something, then the case will not be a hindrance.

The body was assembled from fiberglass. It didn't work out otherwise. But it turned out to be a fairly stable design. Chassis - 6mm. Back wall - 4mm. Top cover (2mm). Its color and texture, in my opinion, are acceptable and do not require coloring. The front panel is made of foil fiberglass (2mm), sheathed in front with aluminum (2mm). In addition, various aluminum corners were used for reinforcement and decoration. The chassis, of course, will be much more convenient if it is made with windows for boards, as they did in the old days. To do this, you can safely use thick plywood, and not expensive fiberglass. Ventilation grill - soap dish. Support legs - bearings (Primare is resting).

***

Actually, this whole project is a process of creation for the sake of the process itself, and not with the goal of getting some very best amplifier with standard sound. The most simple parts were used, especially for output transformers and transition capacitors (few people recommended K73 for this role).

What is the result?

What do we see and hear? Despite the lack of fundamental perfectionism, and somewhere, a formal approach, it turned out to be a very beautiful and stylish, as for me, product that demonstrated a fairly powerful and beautiful sound, even on low-sensitivity (85db) speakers. Perhaps, of course, this is not Luxman-550A, but one can note the high detail and transparency of the sound, warmth and dynamism, as well as the “indecent” absence of even a hint of noise or background. In general, according to individual feelings, after listening, this is a very good result.

Instrument measurements

Electricity consumption: 66 VA, 46W, 0.3A.

Output power limited by visible distortion of the sine wave, with an input of 1.3V: 2.3W.

Maximum output power: 3.6W.

Frequency range at 1.5db linearity: 30Hz - 18kHz.

Your article will be here if you send it to us :) [email protected]

The author of the scheme of this amplifier has been designing high-quality sound-reproducing equipment since 1963. In my opinion, he succeeded a lot in this. Its designs have an excellent sound, are easily repeatable and have a well-deserved success even with beginners. I will only (with the permission of the author) outline the features of his work.

Readers are offered a simple original power amplifier circuit in two versions. The first is budgetary, with automatic output lamp shift. The second is with a fixed bias from a separate winding of the power transformer.

According to the author of the scheme, the fixed offset version has a deeper and more beautiful sound, although the automatic offset version will not disappoint you, allowing everyone who repeats it not to recognize the sound of their favorite records.

Fig. 1 A variant of A. Manakov's circuit with auto-shifting of the output lamp. Output transformer of the company "Audioinstrument"

The amplifier circuit in the version with auto-bias of the output lamp is shown in Fig. 1. The input signal after the volume control is fed to the control grid of the 6N2P double triode. This lamp has a high gain and high internal resistance, which in this case is not very good. I will not go into the details of this, since you can read about this in any radio engineering literature.

The main feature of switching on the preliminary stage lamp is the parallel connection of two triodes located inside one 6N2P lamp bulb. This achieves a decrease in the internal resistance of the lamp, which entails an improvement in the load capacity and the signal-to-noise ratio. The load resistance was not chosen by chance, while compensating the coefficient of non-linear distortion of the output stage and high signal dynamics. A 470 microfarad capacitor shunting the cathode resistor eliminates the effect of feedback, which reduces the gain of the first stage.

The 0.22 microfarad capacitor is separating and the sound of the amplifier as a whole depends very much on its quality. You can use FT, K71, K78, if you want to get a more “warm” sound K40U-2, K40U-9, K42U-2. BM, MBM are not recommended due to their leakage. It is undesirable to use K73 because of their less natural sound. Another. When using the TVZ 1-9 output transformer, the capacitance of this capacitor should be reduced to 0.047-0.068 microfarads. The fact is that a single-cycle tube with external simplicity is a complex design, for example, the capacitance of this capacitor is included in the calculation of the amplitude-frequency characteristic of the output stage.

Now about the output stage. Lamp 6P43P was not chosen by chance. After listening to many instances of lamps 6P14P, 6P18P, 6P43P, it was the latter that was preferred. The design of the lamp is characterized by the correct geometry of the internal parts, which in itself speaks of the high class of this pentode. Install this lamp. You will be rewarded with a juicy and bright sound, excellent sound detail and its shades.

The capacitance of the capacitor in the auto-bias circuit can be increased to 1000 microfarads (compare the sound), and the resistor connected in parallel with this capacitor sets the cathode current of the output lamp within 50 mA (in the auto-bias version).

The author used the TVZ 1-9 output transformer from a tube TV, sorted and “welded” in paraffin anew, replacing the paper in the gap with drawing tracing paper, but I used the TW6SE transformer from the Moscow company Audioinstrument.

In my opinion, different, for example, from the opinion of Simulkin, whose amplifier circuit is given in the magazine "Radiohobby" No. 2 for 2003 (p. 57), no other mode than the triode should be used. Stanislav's reasoning on page 58 about the inclusion of a pentode output lamp for rock music, ultra-linear for chanson and reggae, and a triode for classical music seems disputable to me. You can do eclecticism, but it has nothing to do with sound. The fundamentals of building high quality amplifiers have been the same for decades. It:

1. The shortest, with the least loss, signal path.

2. High quality components.

3. Triode mode of the output stage.

Flicking a switch, and even in the anode circuit, is illogical and impractical. With this to the audiologist.

Rice. 2 PSU circuit for amplifier A. Manakov on 6P43P with auto-bias

The power supply option is shown in Figure 2. The power supply circuit does not differ from those described many times and does not need comments. It is not necessary to feed the glow with direct current, this will lead to a deterioration in microdynamics.

Rice. 3 A variant of A. Manakov's circuit with a fixed offset of the output tube.

For the variant of the amplifier with a fixed offset of the output tube, the circuit of which is shown in fig. Z, an additional bias voltage source is added to the power supply, the circuit of which is given in Fig. 4. The trimmer resistor R2 sets the voltage to 0.04-0.05 volts checkpoint K.T. on the amplifier circuit Fig.3.

Rice. 4 PSU scheme for the fixed-bias variant.

In conclusion, I present the parameters of the amplifier at a fixed bias, measured by A. Manakov.

P out \u003d 2.5 W with SOI \u003d 2-3% at a frequency of 1000 Hz. At Pout=2.2 W SOI=0.8-1% When using TVZ 1-9, the frequency range is from 35-40 Hz to 18-19 kHz with an unevenness of 1.5-2.0 dB. (Depends on the quality of performance of TVZ 1-9). When using the TW6SE from Audioinstrument, the frequency range is even wider. More details about the products of this company can be found at the link on the website of my good friend Mikhail Toropkin www.metaleater.narod.ru

Don't let the low output power scare you - complete with acoustics, sensitivity from 90 dB, 2-3 W is enough.

In the future, it is planned to acquaint readers with many of A. Manakov's schemes, which are distinguished by simplicity and originality, as well as excellent sound.

29 comments: High quality single ended Manakov power amplifier

— the majority of connoisseurs of quality music, who know how to handle soldering equipment and those who have some experience in repairing radio equipment can try to assemble a high-end tube amplifier, which is usually called Hi-End, on their own. Tube devices of this type belong in all respects to a special class of household radio-electronic equipment. Basically, they have an attractive design, while nothing is covered by a casing - everything is in sight.

After all, it is clear the more you can see mounted on the chassis electronic components, the more authority the device has. Naturally, the parametric values ​​of the tube amplifier are significantly superior to models made on integral or transistor elements. In addition to this, when analyzing the sound of a tube device, all attention is given to a personal assessment of the sound, rather than the image on the oscilloscope screen. In addition, it differs in a small set of used parts.

How to choose a tube amplifier circuit

In the case of choosing a pre-amplifier circuit, there are no particular problems, then when choosing a suitable final stage circuit, difficulties may arise. Tube Audio Power Amplifier may have several options. For example, there are single-stroke and two-stroke type devices, and they also have various modes operation of the output path, in particular "A" or "AB". The output stage of the single-ended amplification is by and large a model, because it is in the "A" mode.

This mode of operation is characterized by the lowest values ​​of non-linear distortion, but its efficiency is not high. Also, the output power of such a cascade is not very large. Therefore, if necessary, voicing inner space medium-sized, you will need a push-pull amplifier, with the "AB" mode of operation. But when a single-cycle device can be made with only two stages, one of which is preliminary and the other is amplifying, then a driver is needed for a push-pull circuit and its correct operation.

But if single tube audio power amplifier may consist of only two stages - a pre-amplifier and a power amplifier, then a push-pull circuit for normal operation requires a driver or a stage that generates two voltages of identical amplitude, shifted in phase by 180. The output stages, regardless of whether it is single-ended or push-pull, assume the presence in the circuit output transformer. Which acts as a matching device for the interelectrode resistance of a radio tube with low acoustic resistance.

True fans of "tube" sound argue that the amplifier circuit should not have any semiconductor devices. Therefore, the power supply rectifier must be implemented on a vacuum diode, which is specially designed for high-voltage rectifiers. If you intend to repeat a working, proven tube amplifier circuit, then you do not need to immediately assemble a difficult push-pull device. To sound a small room and get the perfect sound picture, a single-ended tube amplifier is fully enough. In addition, it is easier to manufacture and configure.

The principle of assembly of tube amplifiers

There are certain rules for the installation of radio-electronic structures, in our case it is tube audio power amplifier. Therefore, before starting the manufacture of the device, it would be desirable to thoroughly study the paramount principles for assembling such systems. The main rule when assembling structures on vacuum radio tubes is the wiring of connecting conductors along the shortest possible path. The most effective method is to refrain from using wires in places where you can do without them. Fixed resistors and capacitors must be installed directly on the lamp sockets. At the same time, special “petals” should be used as auxiliary points. This method of assembling a radio-electronic device is called "hinged mounting".

In practice, when creating tube amplifiers, printed circuit boards are not used. Also, one of the rules says - avoid laying conductors parallel to each other. However, such, at first glance, chaotic wiring is considered the norm and is fully justified. In many cases, when the amplifier is already assembled, a low-frequency background is heard in the speakers, it must be removed. The primary task is the correct choice of the "ground" point. There are two ways to organize grounding:

• The connection of all wires going to the "ground" at one point - called "asterisk"
• Installing an energy-efficient electrical copper bus around the perimeter of the board, and solder the conductors to it.

It is necessary to verify the place for the ground point by experiment, listening for the presence of a background. To determine where the low-frequency background comes from, you need to do this: You need to use the method of sequential experiment, starting with the double triode of the preamplifier, to short the lamp grids to ground. In the case of a noticeable decrease in the background, it will become clear which circuit of which lamp is “phoning”. And then, also empirically, you need to try to eliminate this problem. There are auxiliary methods that are mandatory for use:

Pre stage lamps

• Vacuum lamps of the preliminary stage must be closed with caps, and they, in turn, must be grounded
• Cases of trimmer resistors are also subject to grounding
• Lamp wires need to be twisted

Tube Audio Power Amplifier, or rather, the filament circuit of the pre-amplifier lamp can be powered with direct current. But in this case, you will have to add another rectifier assembled on diodes to the power supply. And the use of rectifier diodes in itself is undesirable, as it breaks the constructive principle of manufacturing a tube Hi-End amplifier without the use of semiconductors.

By pairing the output and mains transformers in a lamp device, it is enough important point. These components must be installed strictly vertically, thereby reducing the background level from the network. One of them effective ways installation of transformers is their placement in a casing made of metal and grounded. The magnetic circuits of transformers must also be grounded.

retro components

Radio tubes are devices from distant times, but again come into fashion. Therefore, it is necessary to complete tube audio power amplifier the same retro elements that were installed in the original lamp designs. If it concerns fixed resistors, then carbon resistors with high parameter stability or wire ones can be used. However, these elements have a large spread - up to 10%. Therefore, for a tube amplifier, the best choice would be to use small-sized precision resistors with a metal-dielectric conductive layer - C2-14 or C2-29. But the price of such elements is significantly high, then MLTs are quite suitable for them.

Particularly zealous adherents of the retro style get the "audiophile's dream" for their projects. These are BC carbon resistors, developed in the Soviet Union specifically for use in tube amplifiers. If desired, they can be found in tube radios of the 50-60s. If, according to the scheme, the resistor should have a power of more than 5 W, then PEV wire resistors coated with vitreous heat-resistant enamel are suitable.

Capacitors used in tube amplifiers are basically not critical to one or another dielectric, as well as to the element design itself. Any type of capacitor can be used in the tone control paths. Also in the rectifier circuits of the power supply, you can install any type of capacitors as a filter. When designing high-quality low-frequency amplifiers, isolation capacitors installed in the circuit are of great importance.

It is they who have a special influence on the reproduction of a natural, not distorted sound signal. Actually thanks to them we get an exceptional "tube sound". When choosing isolation capacitors to be installed in tube audio power amplifier, special care must be taken to keep the leakage current as low as possible. Because the correct operation of the lamp, in particular its operating point, directly depends on this parameter.

In addition, we must not forget that the decoupling capacitor is connected to the anode circuit of the lamp, hence it follows that it is under great tension. So, such capacitors must have an operating voltage of at least 400v. One of the best capacitors working as a transition capacitor are JENSEN capacitors. It is these capacitances that are used in top-end HI-END class amplifiers. But their price is very high, reaching up to 7500 rubles for one capacitor. If you use domestic components, then the most suitable would be, for example: K73-16 or K40U-9, however, they are significantly inferior in quality to branded ones.

Single Ended Tube Audio Power Amplifier

The presented tube amplifier circuit has three separate modules:

• Pre-amplifier with tone control
• The output stage, that is, the power amplifier itself
• Source of power

The preamplifier is made according to a simple scheme with the ability to adjust the signal gain. It also has a pair of separate bass and treble tone controls. To increase the efficiency of the device, an equalizer for several bands can be introduced into the design of the pre-amplifier.

Preamp Electronics

The preamplifier circuit presented here is made on one half of a 6N3P double triode. Structurally, the preamplifier can be made on a common frame with an output stage. In the case of the stereo version, then naturally two identical channels are formed, therefore, the triode will be fully involved. Practice shows that when starting to create any design, it is best to first use the circuit board. And after adjusting, already assemble in the main body. Provided that the preamplifier is assembled correctly, it starts to work synchronously with the supply voltage without any problems. However, at the setup stage, you need to set the voltage of the anode of the radio tube.

The capacitor in the output circuit C7 can be used K73-16 with a rated voltage of 400v, but preferably from JENSEN, which will provide the best sound quality. Tube Audio Power Amplifier not particularly critical to electrolytic capacitors, so any type can be used, but with a voltage margin. At the stage of tuning work, we connect a low-frequency generator to the input circuit of the pre-amplifier and apply a signal. An oscilloscope must be connected to the output.

Initially, the input signal range is set within 10 mv. Then we determine the value of the voltage at the output and calculate the amplifying factor. Sound signal in the range of 20 Hz - 20000 Hz at the input can be calculated throughput amplifying path and depict its frequency response. By selecting the capacitive value of the capacitors, it is possible to determine an acceptable proportion of high and low frequencies.

Setting up a tube amp

Tube Audio Power Amplifier implemented on two octal tubes. In the input circuit, a double triode with separate cathodes 6N9S is installed, connected in parallel, and the final stage is made on a rather powerful output beam tetrode 6P13S, connected as a triode. Actually, it is the triode installed in the final path that creates the exceptional sound quality.

To perform a simple adjustment of the amplifier, an ordinary multimeter will suffice, and in order to perform accurate and correct adjustment, you must have an oscilloscope and an audio frequency generator. You need to start by setting the voltage on the cathodes of the 6H9C double triode, which should be within 1.3v - 1.5v. This voltage is set by selecting a constant resistor R3. The current at the output of the beam tetrode 6P13S should be in the range from 60 to 65 mA. If a powerful constant resistor 500 Ohm - 4 W (R8) is not available, then it can be assembled from a pair of two-watt MLTs with a nominal value of 1 kOhm and connected in parallel. All other resistors indicated in the diagram can be installed of any type, but preference is still C2-14 is given.

Just like in the preamplifier, an important component is the decoupling capacitor C3. As mentioned above, the ideal option would be to install this element from JENSEN. Again, if there are none at hand, then Soviet film capacitors K73-16 or K40U-9 can also be used, although they are worse than overseas ones. For the correct operation of the circuit, these components are selected with the lowest leakage current. If it is impossible to perform such a selection, it is still advisable to buy elements from foreign manufacturers.

Amplifier power supply

The power supply is assembled using a 5Ts3S direct-heated kenotron, which provides rectification alternating current, which fully complies with the design standards for HI-END class tube power amplifiers. If it is not possible to purchase such a kenotron, then two rectifier diodes can be installed instead.

The power supply unit installed in the amplifier does not require any adjustment - it turned on and that's it. The topology of the circuit makes it possible to use any chokes with an inductance of at least 5 Gn. As an option: the use of such devices from outdated TVs. The power transformer can also be borrowed from old Soviet-made lamp equipment. If you have the skills, you can make it yourself. The transformer must consist of two windings with a voltage of 6.3v each, providing power to the radio tubes of the amplifier. Another winding should be with an operating voltage of 5v, which are fed into the kenotron filament circuit and the secondary one, which has a midpoint. This winding guarantees two voltages of 300v and a current of 200mA.

Power amplifier assembly sequence

The assembly procedure for a tube sound amplifier is as follows: first, a power source and the power amplifier itself are made. After the settings are made and the necessary parameters are set, the preamplifier is connected. All parametric measurements measuring instruments you need to do not on a "live" speaker system, but on its equivalent. This is in order to avoid the possibility of removing expensive acoustics from standing. The equivalent load can be made from powerful resistors or from thick nichrome wire.

Next, you need to deal with the case for the tube sound amplifier. The design can be developed independently, or borrowed from someone. The most affordable material for the manufacture of the case is plywood. Lamps for the output and preliminary stages and transformers are installed on the upper part of the housing. On the front panel there are devices for adjusting the timbre, sound and an indicator for supplying voltage. In the end, you may end up with devices like the models shown here.

First of all, let me thank the radio amateurs who sent their feedback in response to the publication of my articles in magazines and the Internet. The vast majority are satisfied with the sound of the amplifiers and almost no one had any particular difficulties in repeating the described designs.

If you remember, in the article "Single-cycle tube ..., returning to the printed" I promised to give descriptions and circuits of amplifiers, in the output stages of which triodes are used. I am happy to keep my promise.

First, a few general points to clarify the choice of amplifier circuitry, which I will talk about, the radio components used in them, etc.

The range of direct-incandescent lamps, moreover, relatively affordable, is limited to several types. These are 300B, 2A3, 6C4C, 6B4G, GM70. The choice of indirectly heated triodes, mainly intended for voltage stabilizers, is also not very large. These are 6S19P, 6S41S, 6S33S, as well as double triodes 6N5S and 6N13S. Although there are a number single-cycle designs on lamps 6N5S, 6N13S, it should be noted that the current-voltage characteristics (CVC) of these lamps are less linear, and the coefficient of nonlinear distortion (THD) is high (reaches 10% at rated power and Ra / Ri = 4), while in 6S19P, 6S41S, 6S33S it does not exceed 3% under similar conditions. Therefore, 6H5S, 6H13S are best used in push-pull cascades.

Each of these lamps has its own unique sound, so it is very difficult to describe it in a nutshell. I will state my perception, and agree with it or not, your right.

GM70 - breadth and scale. On this lamp you can create an amplifier with an output power of more than 20W!!! The voltage at the anode of the lamp can reach up to 1000 volts, the anode current - up to 125 mA, so the output transformers must have high dielectric strength (about 3 kilovolts). The sound is very powerful and, in my opinion, a bit straightforward. Small nuances of a piece of music seem to be suppressed by this power and pressure, but I like a more delicate sound. In general - for an amateur.

2A3, 6S4S - very beautiful, detailed and melodious sound. I would call it "cozy and homely", but at the same time - accurate. The lamps are dual-anode designs with a common jumper and differ in voltage and filament current. In 6C4C, the filaments inside the cylinder are connected in series, and in 2A3, in parallel. As you understand, this affects the background level. In the case of using 2A3, it is possible to power the filament circuit with alternating current, but in the case of using 6C4C, it is better to use direct current.

6B4G - western analogue of 6С4С. It has a slightly more analytical sound. Since 6C4C and 6B4G have the same pinout, you can reveal your preferences by simply replacing one lamp with another. By the way, the Saratov "Reflector" also produces a single-anode version with the same CVC and parameters.

300B - is considered the "queen" of direct filament triodes. In my opinion, the lamp occupies an intermediate position between GM70 on the one hand, and 2A3, 6C4C, 6B4G on the other, combining (to a reasonable degree) the advantages of these two types of lamps. Judge for yourself. The output power of a single-ended amplifier on a 300B tube is 8.0W, versus 2.5-3.0W for 2A3 and 6C4C, with a fairly detailed and full sound.

Unfortunately, the sound of straight filament triodes, especially the 300B tube, is very dependent on the year of manufacture and manufacturer. I was able to listen to several modern amplifiers on this tube. To put it mildly, I was surprised and disappointed. They reproduced classical music without problems, but modern and dynamic, inexpressive and dreary. The reason (from my point of view) is that the 300V tubes were turned on in auto-bias mode, and this tube sounds best fixed. And only one of the amplifiers showed a decent sound. I was not allowed to remove the casing (apparently the developer was afraid of divulging his company secrets), but, according to him, the 300B lamps were imported, made in 1958, and the offset was fixed. The amplifier coped well with any musical material, providing a full-fledged sound.

6S19P - from the family of indirectly heated triodes, the lowest power (Pa = 11W). There are no foreign analogues. Therefore, when using one such tube in an amplifier, you have to be content with three watts of output power. But if you install two lamps, turning them on in parallel, the output power will increase to 6W. The sound is quite beautiful and detailed, so you can safely use these devices in the output stages of amplifiers. Naturally, in this case, it is necessary to select lamps in pairs or take measures to equalize their parameters.

6С41С - also a triode with indirect heating (Pa \u003d 25W), has an approximate foreign analogue EC360, and with an octal base. On the Internet in various forums, I had to meet a variety of assessments of the sound of this lamp, and absolutely opposite. I will not quote the authors of these statements, since, in my opinion, most of them did nothing on this triode, since no one discussed operating modes or switching circuits. My experience of using a 6S41C lamp in the output stage of a single-ended tube amplifier, as well as the experience of A. I. Manakov, D. Andreev, V. A. Starodubtsev, allows us to say that the 6S41C is a great-sounding lamp, and with any type of bias. Excellent, well-articulated bass and very voluminous and detailed sound reproduction are hallmarks of the 6C41C sound. In addition, you will be surprised, the power of a single-cycle cascade on it is about 7 watts! The sound of 6S41S is somewhat similar to 300V with a fixed bias, and not one of the worst specimens. But the 300V lamp loses a little to the 6С41С lamp (this is not only my opinion) in dynamics. The disadvantages of a purely constructive nature can be considered the need to purchase special (not cheap) lamp panels and a high filament current. Some designers also consider a longer time to "enter the mode" (approximately 20-30 minutes) as a disadvantage, compared to direct filament lamps. However, I do not consider this fact a disadvantage, rather a feature, because any tube amplifier starts to sound better after a 20-30 minute warm-up. Such obvious advantages as excellent sound, high output power, no hum problems inherent in direct filament lamps, a simpler output transformer (Ra = 800 ohm is enough) due to the low internal resistance of the lamp (which is also good), etc. - more than compensate for these shortcomings.

6S33S (6P18S) - a very powerful triode of indirect heating (Pa = 60W). It has no Western analogues. The tube has been used in amplifiers for a long time, many circuits have been published in various publications and on the Internet. It must be said that this instrument is best used in auto-bias mode due to time and temperature instability and a tendency to self-heat. The sound of a tube in a single-ended amplifier I would describe as somewhat mundane and heavy, with a lack of air, but this is just my opinion, so I leave the choice to you. I emphasize that we are talking about a single-ended tube amplifier with an output transformer. At A. Klyachin's house, I listened to a 6C33C amplifier, made according to the scheme without output transformers (OTL), and so, that amplifier sounded great.

The output power of the amplifier when using 6S33S (6P18S) will be about 12W. The lamp "enters the mode" for an even longer time, compared to 6C41C.

Now let's talk a little about output power in general. For analysis, I will allow myself to introduce the term "comfortable power". This, as a rule, is the power at which the device operates for a long time, its sound does not irritate and allows for the most expressive performance of all the nuances of a piece of music. So, it turned out that for me in a room of 18 square meters, "comfortable power" was about 0.5W per channel. The vast majority of my friends who own single-ended tube amps have confirmed this fact. Someone had 0.4W per channel, someone had 0.7W per channel, in general, the numbers were similar.

Feel what I'm getting at? Considering that the maximum output power per channel of 2.5-3.0 W is more than sufficient for our apartments, as well as the great scarcity and high cost good lamps 300B, the choice fell on the use of direct-heated triodes 6C4C, 2A3 or 6B4G in the output stage. If you need a more powerful amplifier, use indirectly heated triodes 6S19P, 6S41S.

Move on. One of the disadvantages of triodes is considered to be a large buildup voltage. Let's consider this moment in more detail. We open our favorite SE Amp CAD program and model the cascade on the 6B4 lamp. With a supply voltage of about 300 volts and a current of 55 mA, the output power when using a transformer with Ra \u003d 4 kΩ will be 2.44 W at an input voltage of about 40 volts. It would be foolish not to take into account the fact that output voltage of modern CD players with delta-sigma DACs and operational amplifiers on analog outputs is 2.0 volts nominally (my Rotel RCD-02S has an output impedance of 100 ohms and a nominal output voltage of 2.0 volts, respectively, an amplitude of 2.8 volts). Therefore, 40 volts to drive the output triode can be obtained from a simple preliminary stage on resistors, using a lamp with the gain you need. In my case, this condition is fully satisfied by lamps 6S5S, 6S2S or 6N8S.

They are very linear and have a deep opening of the anode characteristics when biased on the grid down to -24 volts. In addition, these types of lamps are perfect for working with direct-heated triodes, mutually compensating for each other's distortions.

If the output voltage of your signal source is small, then you can do the following. Firstly, you can use a lamp with a high gain, for example 6N9S, 6N2P, ECC83, E41CC. Secondly, apply an isolating transformer with a ratio of 1:2. Thirdly, use a pentode (tetrode) as a preliminary stage lamp. To opponents of the use of pentodes, I can say that the best examples of single-ended tube amplifiers of the last century had a pentode in the input stage, and their sound is still considered a reference. A little lower I will give diagrams of preliminary lamp stages on a pentode and a circuit that uses an isolation transformer.

Let's move on to the diagram in Fig.1. We use it as a base, and by applying various lamps and changing their modes of operation, we will try to create an apparatus that meets your specific tastes.

As you can see, the circuit is very simple, and consists of only two stages, preliminary and final. I always adhere to the principle of the minimum possible number of gain stages, since adding extra elements in the signal path will degrade the sound.

The preliminary amplification stage is resistive. Since there are calculations for the cascade on resistors in almost any literature and the Internet, I do not give them. I think that in our case it would be more useful to talk about the sound of the preamplifier tubes. When discussing the amplifier circuit with A.I. Manakov, he proposed the 6S5S lamp as the most linear, having a cylindrical design of the electrode system. In second place - 6S2S. If you open the reference book, you will see that the parameters of these lamps are almost the same, which cannot be said about the internal design. This explains the difference in sound. Despite the individual differences (and they are), both lamps sound very good. I did not notice any shortcomings (I do not consider one triode in a cylinder a disadvantage, rather an advantage). I suggest you try both options and decide which one you like best, especially since you don’t need to redo anything. If you could not find these lamps, use a 6H8S double triode (we connect both halves in parallel). The features of such an inclusion are described in my last article "Single-cycle tube ..., returning to the printed one", so I will not repeat myself. You can also use a 6H8C lamp without connecting the halves in parallel, in which case one lamp will work on both channels (there is a space saving).

I consider it necessary to tell you about one more thing. A 6C2C lamp is not half of a 6H8C lamp (as many "experts" on Internet forums mistakenly believe). Reference data are similar, the design of the electrode system is similar, but there are differences. Due to the larger anode area in 6C2C, the steepness of its characteristic is higher, and the real internal resistance is lower than that of the half of 6H8C. The gain is the same (about 20). Traverses for mounting the electrode system 6S2S and 6N8S are the same, however, in the case of 6S2S, they attach one triode, not two. This explains the almost complete absence of the microphone effect in 6C2C. As you understand, because of this, the difference in sound (though not very big) will be necessary. The same must be said about the 6C41C lamp, which is not half of the 6C33C lamp, as many believe. Look carefully at the passport values ​​​​of the parameters of these lamps, as well as the volt-ampere characteristics. It is clear that the difference in sound will be significant.

In addition, you must remember that the real dynamic gain of the cascade on resistors is always less than the static gain of the particular lamp used. In order not to clutter up the article with formulas, we can assume that it is 25 percent. Thus, when using a 6C5C (6C2C) lamp, the dynamic gain of the real cascade will be 15-16. This moment must always be taken into account when calculating the lamp cascade on resistors.

You can use a choke instead of a resistor in the anode of the input lamp. According to some radio amateurs, the choke stage sounds better. Unfortunately, I cannot agree with them. I understand that everyone has different tastes, but I must express my (and not only) opinion about the sound of such cascades.

If you like to listen to symphonic or jazz music, then a choke-loaded cascade is not the best option. It sounds harsh, I would even say annoying. The overtones of string and wind instruments are strongly emphasized. Reed instruments (saxophone, etc.) sound unnatural, with some unpleasant overtones. If you have the opportunity to listen to both stages (resistive and choke) at the same time (naturally with the same final stage), then put on a good recording of Dizi Gilespie (trumpet) or David Sanborn (saxophone). I think you will immediately hear the difference in sound.

As you know, the inductor is an inductance, the preliminary stage lamp (driver) has an output capacitance, and the final stage lamp has an input capacitance, respectively. As a result, we have a resonant circuit tuned to a frequency, which is determined by the sum of these capacitances and the inductance of the inductor. F=1/2P multiply by the square root of the product LC. You should be aware that with a large inductance of the inductor, the resonance will move from the ultrasonic region to audio frequencies and, despite the fact that the circuit is shunted by the internal resistance of the driver lamp, and is significantly weakened, it is still present. At the resonance frequency, the rise can reach up to 10 dB.

And one moment. The inductor resistance increases with increasing frequency, as a result, we get an uneven gain of the cascade (it increases with increasing frequency). Naturally, this lengthens the spectral "tail" of the harmonics, which does not have the best effect on the sound.

Since we are talking about preliminary cascades, it should be noted that there are many schemes whose authors use batteries or accumulators to organize the bias. Many people think that electrochemical current sources in bias circuits are preferable to the traditional resistor and capacitor, which adversely affect the sound. It must be said that batteries or accumulators can stand both in the grid circuit and in the cathode circuit.

I have tested seven types of batteries and three types of batteries various manufacturers available in stores. Of the lamps, the following were tested: 6N1P, 6N2P, 6S2S, 6S5S, 6N8S, 6N9S, 6S4P, 6E5P. Accumulators in cathode circuits are preferable as there is no need for recharging (they are charged by lamp current). The only thing, so that there is no overcharging, you need to choose their capacity of at least 20 * I lamps. In my case, I chose the battery capacity in the range of 700-1000mA / h.

The first impression was very good, but as I listened, a small flaw was discovered. In my opinion, the sound acquired some "rigidity" (regardless of the type of electrochemical current source), which was not there when using a resistor and a capacitor. top scores were obtained using NiCd batteries, moreover, standing in the cathode circuit, and not the grid.

Of course, it must be said that I use Black Gate Rubicon electrolytic capacitors in the cathodes. Perhaps a battery or battery stage sounds better than a traditional one, especially when used Chinese capacitors and poor quality resistors taken from computer boards and power supplies. I don’t have such radio elements, so I suggest you listen to both options yourself and choose the one that you like best.

Further, the signal through a separating capacitor is fed to the input of the final stage, made on a 6C4C straight-line triode. I wrote about the types of isolation capacitors many times, so now I will only talk about one nuance. When using lamps with a low gain in the input stage, it is best to use capacitors such as FT-3, K-77, K-78 as a separator, but if a tetrode or pentode is used as a driver, then paper in Jensen, K40U- 9, K42U-2, etc.

The final stage has no features. The lamp is on in auto-bias mode. In previous articles, I described the advantages and disadvantages of fixed and automatic offset types, so it makes no sense to repeat everything again. Choose yourself. Let me just say that when using Black Gate electrolytes (on the C6 and C9 diagrams), there is practically no difference in sound, but there are much fewer disadvantages inherent in a fixed bias.

To avoid problems with the background when using 6C4C, I powered the glow with direct current. In the case of using KD226 diodes, the heating voltage under load is 6 volts. If you use other diodes (necessarily "fast"), it may be necessary to adjust the filament voltage using an additional 0.3-0.5 ohm resistor. And one moment. For a direct-heated triode, the cathode and filament are the same, so the connecting wires of the filament circuits must be of high quality (unlike lamps with indirect filament). If you use a 2A3 lamp, then its incandescence can be powered by a "change", its background level is initially lower (I repeat, due to the parallel connection of the filaments of both triodes inside the cylinder).

It must be said about why I used a transformer with Ra \u003d 4k. The fact is that many in their designs have already used the Audioinstrument transformer TW6SE, and it has Ra \u003d 4k. In order not to spend extra money on buying a new transformer, use the one you already have. Of course, it is better to use a transformer, overall power which is 100W, for example TW10SE, low frequencies in this case they will be played even better, but with TW6SE you will not be disappointed, since the overall power of the output transformer is selected within 20 * Pout or more.

In general, the maximum output power is achieved when Ra=2Ri, where Ra is the AC resistance of the primary winding of the output transformer and Ri is the internal resistance of the lamp. Unfortunately, in this case, the non-linear distortions are too high (about 6%). Therefore, the resistance of the primary winding of the transformer Ra is chosen within 3-5Ri (sometimes up to 7Ri), as a compromise between the magnitude of non-linear distortion and output power. But it must be taken into account that the power of the cascade decreases linearly, and the coefficient of nonlinear distortion (THD) exponentially, with all the ensuing consequences, therefore, there is the concept of reasonable sufficiency. In addition, an excessive increase in the anode load reduces the dynamics of the cascade. In our case, when using 6C4C or 2A3, with internal resistance Ri = 800 ohm, this condition is met.

To illustrate the above, I give data on the output power of the amplifier and the coefficient of the second and third harmonics at various values ​​of Ra (at 40 volts AC voltage at the lamp input, anode current 60mA and 250V anode voltage). I cited these current and voltage values ​​​​as an example not at all by chance. In the textbooks of Tsykin and Voishvillo, it is these modes that are recommended to achieve the best sound quality.

Ra=4.0kΩ, Pout=2.22W, 2nd harmonic 3.1%, 3rd harmonic 0.2% Ra=3.5kΩ, Pout=2.4W, 2nd harmonic 3.4%, 3rd harmonic 0.1% Ra=3.0kΩ, Pout=2.54W, 2nd harmonic 3.8%, 3rd harmonic 0% Ra=2.5kΩ, Pout=2.7W, 2nd harmonic 4.4%, 3rd harmonic 0.1% Ra=2.0kΩ, Pout=2.9W, 2nd harmonic 5.3%, 3rd harmonic 0.3% I hope no comment is needed.

The quiescent current, as always, is controlled by the voltage drop across the cathode resistors. If you use the details indicated in the diagram, then it will be 55-60mA for a 6S4S lamp and 5-6mA for a 6S5S lamp.

Now let's move on to cases where the input voltage of the amplifier is less than two volts, or when a lamp is used in the output stage that requires a large buildup voltage (for example, 6C33C). Figure 2 shows a diagram of a pre-amplifier on a tetrode 6E5P in a triode connection, and in Fig. 3 in a standard tetrode connection.

You may ask why 6E5P? The fact is that while experimenting with various pentodes (6Zh4, 6Zh52P, etc.), I could not get a sound that would completely satisfy me. In some cases, transparency disappeared, in some cases, dryness appeared, etc. etc. And only 6E5P provided the necessary sound quality. The general impression is that the sound is very similar to the triode, only a little brighter. Deep, well-articulated bass, transparent highs and very detailed mids are hallmarks of the 6E5P's sound. My rating is excellent! In any case, it's up to you to choose and listen, and I will give the parameters of the lamp in triode and regular switching.

Triode connection: Ri=1.2kom; S=30mA/V; Kus=30-35. Tetrode connection: Ri=8kom; S=30mA/V; Cus=200. Well, how impressive? Naturally, having such parameters, the lamp will be able to freely "rock" any triode, be it 300V, 6S41S, 6S33S, GM70, etc.

It should be noted that the broadband tetrodes 6E5P, 6E6P with low internal resistance were "discovered" for audio applications by AI Manakov. They are successfully used by many designers in drivers (triode and tetrode mode) and as output lamps. On the same lamps at the end of 2003 A.I. Manakov also developed a resistive ultra-linear cascade, which also has a very good sound.

Now consider a variant of the circuit using an interstage transformer. The advantages of such inclusion are considered to be:

maximum possible amplification
1. possibility of coordination with any load
2. high efficiency
3. lower stage supply voltage
4. more dynamic sound.

However, not everything is so smooth. The disadvantages of the scheme are:

1. large dimensions and mass
2. the need for shielding
3. high price
4. high price

If these problems do not scare you, then Fig. 4 shows a diagram of a preliminary stage using an interstage transformer with a transfer ratio of 1: 2. The features of such cascades are repeatedly described in various sources, so I do not consider it necessary to consider them in detail.

The article would not be complete if you do not give an amplifier circuit in the output stage of which an indirect filament triode operates. I chose 6S41S, because there are very few circuits using this lamp, unlike 6S33S.

I strongly recommend that you give this design a try. You will be simply surprised by the sound. Compared to a 6C4C or 300V amplifier, I would describe it as more versatile. The amplifier equally well and naturally reproduces both classical music and modern, with a large number of impulse components.

The circuit using a 6E5P lamp in the input stage is shown in Fig. 5. As always, it's fairly simple and repeatable, so you shouldn't have any trouble making this variation. You can try different tubes in the input stage and choose the one that sounds best to you. The 6E5P lamp is turned on by a triode, so the sensitivity of the amplifier will be 1.8-2 volts. If this is not enough, apply the circuit in Fig. 3 or Fig. 4. The sensitivity of the amplifier in these cases will be 0.35-0.4V and 0.8-1.0V, respectively.

I will say a little about the choice of lamp modes 6S41S. The anode-cathode voltage is 165-175 volts, with a current through the lamp of about 93-95mA. This means that the dissipation power will be about 16 W, which is one and a half times less than the passport value (that is, the lamp operates in light mode).

Offset -70 volts. If you also look at the volt-ampere characteristics, you will see that the operating point of the lamp is in a linear region. The total current consumption of one amplifier channel is about 110 mA. Thus, if you are making a stereo amplifier, then it will be enough to use one 5Ts3S (5U4G) kenotron in its power supply. The rated rectified current of this kenotron is 220-230mA (reference value). If you decide to increase the current (which is quite acceptable), then you will need to use two kenotrons connected in parallel in the power supply of the amplifier, or make the amplifier in the form of two monoblocks. Naturally, the primary winding of the output transformer must also be designed for this current.

On forums on the Internet, I once saw a discussion of the power supply of an amplifier using television damper diodes, for example 6D22S. I must warn you that when using these lamps, the sound of the amplifier loses volume and detail, the depth of the stage disappears, it seems that the musicians are on the same line. This sound does not suit me, but you yourself have the right to decide this issue. If there is no desire to make a power supply on kenotrons, it is more expedient to use "fast" semiconductor diodes - "fasts" and "ultrafasts", designed for the corresponding current and voltage, shunting each of them with K78-2 capacitors with a capacity of 0.01-0.022 Mkf, to eliminate switching noise when switching.

The power supply circuit is similar to the circuit shown in Fig.1. Since the incandescence of the 6C41C lamp is powered by alternating current, diodes D1-D8, as well as filter capacitors C12-C15, must be excluded. Remember that the filament current of one lamp is 2.7 amperes, so the filament windings of the power transformer must be designed for it.

The cathode resistor of the 6C41C lamp gets very hot, so its dissipation power must be at least 15-20W.

The output transformer used in this circuit is made by "Audioinstrument" and has the following parameters: Ra=1kom; Ktr=12.5; Pgab=100W; I=150ma. Primary resistance direct current- about 150 ohm.

Even better sound quality was obtained when using output transformers wound on OSM-0.16 cores, made at my request by Dmitry Andreev, for which special thanks to him. The parameters of these transformers are as follows: Ra=1kom; Ktr=10.05; Pgab=160W; I=200ma. The resistance of the primary winding to direct current is about 50 ohms. In both cases, the bias was -70 volts, and the power dissipation of the 6C41C lamp in the second case increased by only 1W. The sound has acquired even greater volume and detail, the band of reproducible frequencies has expanded (up to 70kHz) and the depth of the stage has increased.

The installation of all the amplifiers that I talked about was done in a hinged way, using a Kimber TC series copper stranded cable. I like the neutral sounding character of this connector, as well as the resistance of its Teflon insulation to heat. The cost is about $30 per meter. But by purchasing 1 meter of this cable, you actually get 8 wires of 1 meter each (4 blue and 4 black). Agree that $ 4 per meter of good wire is not so much.

The wiring of the "earth" is made by a "star", in the last article I described this method in detail. AC hum is only audible if you bring your ear close to the speaker system. If this is not the case, you need to tinker with the relative position of the radio elements. In my case, the power supply chokes are in the basement of the chassis, and the power and output transformers are on top.

Well, it seems that's all. In conclusion, I would like to thank my friend A.I. Manakov E-mail: detector(dog)surguttel.ru for constant consultations and assistance in editing this article (all circuits were personally tested by Anatoly Iosifovich long before me), as well as for the 6E5P and 6S41S lamps sent to them.

I must also tell you that the peculiarities of the perception of music are very individual, so you should not get hung up on any individual circuits or lamps. Not only straight-wire triodes provide high-quality sound. Both pentodes and triodes of indirect heating with proper circuit design, right choice operating point and modes, no worse. So learn, try, listen, experiment. Don't Forget Theory electrovacuum devices and building amplifiers on them so that there are no empty "influences" and "revelations from above." Only in this case you will be able to create an apparatus that will fully correspond to your musical tastes.