Having completed the installation of the structure, the novice radio amateur cannot make it work because he cannot set the radio tube mode
The term "lamp mode" is commonly understood as the totality of all constant voltages on the electrodes and currents in the lamp circuits in a particular operating circuit. On rice. 1 shows a diagram of a resistive low-frequency voltage amplification stage assembled on a pentode. To the points marked on the diagram Un, the filament winding of the power transformer is connected. The filament voltage can be measured with an AC voltmeter by connecting it between the points 1 And 2 . Current in the heating circuit In measured with an alternating current ammeter, which can be included in the open circuit at the point 2 .
The power supply of the anode and the shielding grid is connected between the points marked +ea And -ea. Power supply voltage Ea measured with a DC voltmeter connected between points 3 (the positive wire of the voltmeter is connected here) and 1 (negative wire). It is customary to determine all voltages on the lamp electrodes (except for the filament) with respect to the lamp cathode. Therefore, the voltage at the anode of the lamp Ua measured between points 4 And 5 , and the voltage on the screening grid Ue- between points 6 And 5 .
Rice. 1
If we break the chain at the point 3 and in the gap we turn on the DC milliammeter plus to the clamp +ea, minus to the output of the anode load resistor Ra, then the device will show the anode current of the lamp Ia. The same current will be shown by the device when it is turned on in the open circuit at the point 4 . It is better, however, to measure the anode current at the point 3 , since the operation of alternating current circuits, which we do not consider here, is less disturbed. Similarly at points 7 or 6 screening grid current is measured Ie. Both of these currents Ia And Ie, add up to the total cathode current of the lamp Ik.
The current in the control grid circuit when determining the cathode current can be neglected, since in most cases it is equal to zero (except for generator circuits). The cathode current of the lamp can be measured at the point 5 . In this case, the positive wire of the milliammeter is connected to the cathode, the negative wire is connected to the output of the resistor Rk.
What device is used to measure lamp modes?
The voltmeter, which measures the voltage in the circuits of amplifiers or receivers, must be high-resistance. This means that its internal resistance must be significant. It is usually determined in terms of one volt. Good high-resistance voltmeters have an internal resistance of about 20,000 ohms per volt. For example, a voltmeter with a scale of up to 300 volts has an internal resistance of 20,000 x 300 = 6 MΩ. Therefore, to the points between which the voltage is measured, an additional resistance of 6 MΩ is connected in parallel. Whether this is acceptable must be decided based on the data of the scheme.
For example, if the resistance of the resistor Re (rice. 1 ) is equal to 300 kOhm and a current of 0.5 mA flows through it, creating a voltage between the points 6 And 7 - 150 V, and the voltage Ea equal to 250 V, then the voltage on the screening grid will be:
250 - 150 = 100 V
The voltage drop across the resistor Rk due to its smallness, we neglect it. When connecting a voltmeter between points 6-1 total resistance of the section screening mesh - point 1 will change. If before it was:
Ue / Ie \u003d 100 / 0.5 \u003d 200 kOhm
then when a voltmeter is connected, it will become equal to:
(6 MΩ x 0.2 MΩ) / (6 MΩ + 0.2 MΩ) = 193 kΩ
This means that the total resistance of the screening grid circuit will be:
300 + 193 kΩ = 493 kΩ
and the current passing through the resistance Re will be equal to the voltage of the power supply divided by the resistance of 493 kOhm, that is:
250 / 493 = 0.508 mA
This current will create on the resistance Re voltage drop:
0.508 x 300 = 152.4 V
and the voltage on the screening grid will no longer be 100 V, but
250 - 152.4 = 97.6V
Therefore, the device will show a voltage less than the true one by 2.4%. You can still come to terms with this. If we use a voltmeter with an internal resistance of 1000 ohms per volt, then the error will be even greater, and the error may become unacceptable. Therefore, it is recommended to use only high-resistance voltmeters to measure the lamp mode and ensure that the internal resistance of the device, connected to the appropriate scale, is 20-30 times greater than the resistances of the resistors in the circuits under test.
How to measure offset on a grid?
The cathode current, which is the sum of the currents of the anode and the shielding grid, flows through the resistor Rk. In this case, a voltage appears on the resistor, plus which is applied to the cathode (point 5 ), and minus - to the common wire (point 1 ). Lamp control grid through a resistor Rc connected to a common wire. Since the current through the resistor Rc does not flow, then there is no voltage drop on it and the potential of both its ends is the same. Therefore, a voltage is applied between the control grid and the cathode, taken from the resistor Rk. It is the bias voltage on the control grid, since it shifts the operating point to the right place in the lamp characteristic. How to measure it?
Turn on the voltmeter between the control grid (point 8 ) and cathode (point 5 ). In this case, parallel to the resistor Rk, on which there is voltage, a chain of two resistances will turn on - the internal resistance of the voltmeter and the resistance of the resistor Rc. They are connected in series and form a voltage divider to which the control grid is connected. If the resistance of the voltmeter is less than the resistance Rc or commensurate with it, then the voltage shown by the voltmeter will be significantly less than the true bias on the grid.
In order for the measurement error to be small, it is also necessary to use a voltmeter with a high internal resistance here, 20-30 times more than the resistance of the resistor Rc. And since the latter is usually equal to 0.5-1.0 MΩ, it is necessary to use voltmeters with a resistance of the order of 10-20 MΩ. The measured voltage here is usually a few volts; therefore, a voltmeter with a resistance of at least 1-2 MΩ per volt is needed. A simple pointer device of a magnetoelectric type is no longer suitable here. Therefore, to measure the displacement at points 5 And 8 apply lamp DC voltmeters with an input resistance of the order of 20-50 MΩ (on any scale).
It is much more convenient to measure the bias voltage in the rice. 1 circuit not directly on the grid of the lamp, but at the points of its occurrence - at the ends of the resistor Rk. Since the resistance of this resistor is small, only a few hundred ohms, in this case almost any, even relatively low-resistance, voltmeter can be used by connecting it to the points 5 And 1 . This method of measurement is only suitable when the bias on the control grid is supplied from the cathode resistance. In other cases, the measurement method will be different.
Often, for simplicity, the anode voltage and the voltage on the shielding grid are measured not with respect to the cathode, but with respect to the chassis connected to the common wire. The resulting inaccuracy in the determination of Ua And Ue is a few percent (does not take into account the voltage drop across the cathode resistor Rk).
When checking faulty radio equipment, it is recommended to measure not only the voltages on the electrodes of the lamps, but also the voltage drops across the resistors Ra, Re. If it is equal to zero, then this means that there is no current in this circuit (for example, a lamp has failed).
People who love good music probably know about the Hi-End tube amplifier. You can make it yourself if you know how to use a soldering iron and have some knowledge of working with radio engineering.
Unique device
Hi-End tube amplifiers are a special class of household appliances. What is it connected with? Firstly, they have a rather interesting design and architecture. In this model, a person can see everything that he needs. This makes the device truly unique. Secondly, the performance of the Hi-End tube amplifier differs from alternative models that use the Hi-End difference in that a minimum number of parts are used during installation. Also, when evaluating the sound of this unit, people trust their ears more than the harmonic distortion measurements and the oscilloscope.
Selection of circuits for assembly
The preamplifier is fairly easy to assemble. For it, you can choose any suitable scheme and start assembling. Another case is the output stage, that is, the power amplifier. As a rule, a lot of different questions arise with him. The output stage has several assembly types and operating modes.
The first type is the single-ended model, which is considered a standard cascade. When operating in the "A" mode, it has a small non-linear distortion, but, unfortunately, has a rather poor efficiency. Also note the average power output. If you need to fully sound a fairly large room, you will need to use a push-pull power amplifier. This model can work in the "AB" mode.
In a single-cycle circuit, only two parts are enough for a good operation of the device: a power amplifier and a preamplifier. The push-pull model already uses a phase-inverted amplifier or driver.
Of course, for two types of output stage, in order to work comfortably with, it is necessary to match the high interelectrode resistance and the low resistance of the device itself. This can be done with a transformer.
If you are a connoisseur of "tube" sound, then you should understand that you need to use a rectifier, which is produced on a kenotron, to achieve such a sound. In this case, semiconductor parts must not be used.
When developing a Hi-End tube amplifier, you can not use complex circuits. If you need to sound a fairly small room, then you can use a simple single-cycle design that is easier to make and set up.
DIY Hi-End tube amplifier
Before starting installation, you need to understand some rules for assembling such devices. We will need to apply the basic principle of mounting lamp fixtures - minimizing fixtures. What does it mean? You will need to discard the mounting wires. Of course, this cannot be done everywhere, but their number must be minimized.
In Hi-End, mounting petals and strips are used. They are used as additional points. Such an assembly is called hinged. You will also need to solder the resistors and capacitors that are on the lamp panels. It is strongly discouraged to use printed circuit boards and assemble conductors in such a way that parallel lines are obtained. Thus, the assembly will look chaotic.
Interference elimination
Later, you need to eliminate the low-frequency background, if, of course, it is present. Also important is the choice of grounding point. In this case, you can apply one of the options:
- The type of connection is a star, in which all "earth" conductors are connected to one point.
- The second way is laying a thick copper bus. It is necessary to solder the corresponding elements on it.
In general, it is better to find a grounding point yourself. This can be done by determining the level of low-frequency background by ear. To do this, you need to gradually close all the grids of lamps that are located on the ground. If, when the subsequent contact is closed, the low-frequency background level decreases, then you have found a suitable lamp. To achieve the desired result, it is necessary to experimentally eliminate unwanted frequencies. You also need to apply the following measures to improve the quality of your assembly:
- To make the filament circuits of radio tubes, you need to use twisted wire.
- The tubes that are used in the preamplifier must be covered with earthed caps.
- It is also necessary to ground cases with variable resistors.
If you want to power the preamp tubes, you can use direct current. Unfortunately, this requires the connection of an additional unit. The rectifier will violate the standards of a Hi-End tube amplifier, as it is a solid state device, which we will not use.
transformers
Another important point is the use of various transformers. As a rule, power and output are used, which must be connected perpendicularly. In this way, you can reduce the level of low-frequency background. Transformers should be located in earthed casings. It must be remembered that the cores of each of the transformers should also be grounded. It is not necessary to apply when you install devices, so that additional problems do not appear. Of course, these are not all the features associated with installation. There are quite a lot of them, and it will not be possible to consider them all. When installing a Hi-End (tube amplifier), you cannot use new element bases. They are now used to connect transistors and integrated circuits. But in our case, they do not fit.
Resistors
A high-quality Hi-End tube amplifier is a retro device. Of course, the details for its assembly must be appropriate. Instead of a resistor, a carbon and wire element may be suitable. If you spare no expense to develop this device, you should use precision resistors, which are quite expensive. Otherwise, MLT models are applicable. This is a pretty good item, as evidenced by the reviews.
Hi-end tube amplifiers are also applicable with BC resistors. They were made about 65 years ago. Finding such an element is quite simple, just take a walk around the radio market. If you use a resistor with a power of more than 4 watts, you need to choose enameled wire elements.
Capacitors
When installing a tube amplifier, you should use different types of capacitors for the system itself and the power supply. They are usually used for tone control. If you want to get high-quality and natural sound, you should use a decoupling capacitor. In this case, a small leakage current appears, which allows you to change the operating point of the lamp.
This type of capacitor is connected to the anode circuit, through which a large voltage flows. In this case, it is necessary to connect a capacitor that supports a voltage of more than 350 volts. If you want to use quality elements, you need to use Jensen parts. They differ from analogues in that their price exceeds 3,000 rubles, and the price of the highest quality radio elements reaches 10,000 rubles. If you use domestic elements, it is better to choose between the K73-16 and K40U-9 models.
Single Ended Amplifier
If you want to apply a one-cycle model, you must first consider its circuit. It includes several components:
- power unit;
- final cascade;
- a pre-amplifier in which you can adjust the tone.
Assembly
Let's start with the pre-amplifier. Its installation takes place according to a fairly simple scheme. It is also necessary to provide for power control and a separator for tone control. It must be tuned to low and high frequencies. To increase the shelf life, you need to apply a multi-band equalizer.
In the laughter of the pre-amplifier, one can see similarities with the common 6N3P double triode. The element we need can be assembled in a similar way, but using the final cascade. This is also repeated in stereo. Remember that the design must be assembled on a circuit board. First it needs to be debugged, and then it can be installed on the chassis. If you have installed everything correctly, the device should turn on immediately. The next step is to move on to the settings. The value of the anode voltage for different types of lamps will be different, so you will need to select it yourself.
Components
If you do not want to use a high-quality capacitor, then you can use K73-16. It is suitable if the operating voltage is more than 350 volts. But the sound quality will be noticeably worse. Electrolytic capacitors are also suitable for this voltage. You need to connect an S1-65 oscilloscope to the amplifier and apply a signal that will pass from the audio frequency generator. At the initial connection, you need to set the input signal to about 10 mV. If you need to know the gain, you will need to use the output voltage. To find the average ratio between low and high frequencies, it is necessary to choose the capacitance of the capacitor.
You can see a photo of the Hi-End tube amplifier below. For this model, 2 lamps with an octal base were used. A double triode is connected to the input, which is connected in parallel. The final stage for this model is assembled on a 6P13S beam tetrode. A triode is mounted in this element, which allows you to get a good sound.
To set up and check the performance of the assembled device, you must use a multimeter. If you want to get more accurate values, then you should use a sound generator with an oscilloscope. When you have taken the appropriate devices, you can proceed to the setting. On the cathode L1 we indicate a voltage of about 1.4 volts, this can be done if you use the resistor R3. The output lamp current must be specified as 60 mA. To make a resistor R8, you need to install a pair of MLT-2 resistors in parallel. Other resistors can be of different types. It should be noted a rather important component - a decoupling capacitor C3. It was not in vain mentioned, since this capacitor has a strong influence on the sound of the device. Therefore, it is better to use a proprietary radio element. The other elements of C5 and C6 are film capacitors. They allow you to increase the quality of transmission of various frequencies.
A power supply built on a 5Ts3S kenotron is worth finding. It complies with all the rules for constructing the device. A homemade Hi-End tube power amplifier will have high-quality sound if you find this item. Of course, otherwise it is worth looking for an alternative. In this case you can use 2 diodes.
For a Hi-End tube amplifier, you can use the appropriate transformer, which was used in old tube technology.
Conclusion
To make a Hi-End tube amplifier with your own hands, you must perform all the steps consistently and accurately. First, connect the power supply with an amplifier. If you set up these devices correctly, you can mount a preamplifier. Also, using the appropriate technique, you can check all the elements in order to prevent breakage. After assembling all the elements together, you can begin to design the device. Plywood might work well for the body. To create a standard model, it is necessary to place radio tubes and transformers on top, and regulators can already be mounted on the front wall. With them, you can amplify the tone and see the power indicator.
GENNADY SEMENOVICH GENDIN, "HIGH-QUALITY TUBE AUTO AMPLIFIERS"
The level of intermodulation distortion and the minimum achievable level of own background and interference, and, consequently, the real dynamic range of the entire amplifier, which is one of its most important parameters, depend on the correctness and literacy of the installation.
In order to firmly understand the principle of competent installation common to any amplifiers, we will carefully consider the drawings illustrating the connection of the grid circuit of the lamp to the input connector, which is located at some distance from the lamp.
Examples of incorrect installation
We emphasize once again that this principle is common for connecting any two sections of the circuit, one of which is a signal source, and the other is a receiver. This can be a microphone and a lamp of an amplifier of a microphone stage, an input jack of a tape recorder and a switch of the type of work, or, as in our case, the first two stages of an ultrasonic frequency converter and a tone control unit.
Correct installation
In this case, special attention should be paid to the fact that in this case the signal source is the anode of the lamp of the first stage, and the signal receiver is the grid of the lamp of the second stage and, therefore, no grounding within this section is unacceptable. In other words, inside the tightly grounded metal case of the tone control unit, no part should be grounded directly to the chassis or shielding casing, but only to a special bus that is well isolated from the case. The above is illustrated in the figure.
Now about the shielded wires themselves. None of the industrially produced types of wires in a "pure" form suits us. All shielded wires will have to be done by yourself. This is done easily. If you look closely at the figure, you will see that two wires are placed inside the shielding braid: one is indicated by thin lines, the other is thick. Such a conditional division corresponds to the actual one. Indeed, all the shielded wires in our amplifier are made according to the principle of a doll-matryoshka: two wires of different diameters are placed inside a conventional metal shielding braid - one thin (signal) colored stranded wire in PVC or PTFE insulation with a cross section of 0.2 ... 0.35 mm, the other also stranded, but with a cross section of at least 0.5 mm - zero, i.e. grounded. Both of these wires, together with the shielding braid, are placed in an insulating PVC stocking.
We strongly recommend that you make it a rule to assign certain wire colors to the various circuits of the amplifier. The choice of colors themselves can be arbitrary. It, as a rule, depends on the actual range of wires available from the radio amateur, but it is still better to adhere to some rules. So, it is better to make all zero wires to be grounded black and thick (section 0.5 ... 0.75 mm), the positive wires of the rectifier are red, and if there are several rectifiers, then red, pink and orange. All signal wires of one of the stereo channels are green, and the other is blue (or cyan). Lamp filament circuits - white or gray. For circuits of auxiliary devices and systems, brown, yellow, lilac or thin black can be distinguished. This order will greatly simplify the installation process and eliminate confusion when wiring dual volume and tone controls (which of the wires is from the left channel, which one is from the right).
For self-manufacturing of shielded connecting cables, you need to either take a “clean” metal braid, or remove it from an industrial single shielded wire, then thread two insulated wires into the braid (one thin - signal, the other thick - zero) and stretch all this together with the braid inside PVC stocking of the appropriate diameter. This can be done in two ways: to make each individual specific wire of a predetermined length, or to immediately prepare 10 ... 15 m of wire, and then cut off pieces of the desired length as needed. From practice, we can say that the second method saves time significantly.
For mounting incandescent circuits and network wires, both wires are placed inside one braid (you can use the same color) and the braid is also insulated with a vinyl chloride stocking.
Now about the "zero" bus mentioned above inside the shielded blocks. If the block contains a printed circuit board with radio elements, then one of the printed tracks can play the role of a bus. It should be as wide as possible. To reduce its resistance, the track must be tinned and soldered on top of it along the entire length of a piece of bare copper, and even better, a silver-plated wire. If the installation inside the block is not printed, but mounted (for example, on a switching switch), then the same piece of bare wire, fixed with its ends on the “idle” switch terminals or on special insulating reference points, can play the role of a busbar.
Please note that all signal interstage and input circuits of tube amplifiers have input and output resistances an order of magnitude greater than transistor ones and are measured in hundreds of kilo-ohms and mega-ohms. In this regard, the intrinsic capacitances of the shielded wires begin to have a significant impact on the frequency response of the ultrasonic frequency response. It is known that this capacitance is directly proportional to the length of the shielded wire and inversely proportional to the distance from the inner wire (core) to the braid. Therefore, do not strive to use modern thin and ultra-thin (3, 2 and even 1.5 mm in diameter) branded shielded wires and, if possible, make shielded connections shorter.
I built with output stages on beam tetrodes and pentodes. For some reason, output triodes did not fit into my designs. Perhaps the common opinion that the triode sound is good only for classics, vocals and jazz (mostly I listen to rock) played a role, or maybe because I have a larger selection of tetrodes and pentodes. In general, be that as it may, I decided to fill this gap and try the triode. For eminent and rather expensive 300V, 2A3, 2S4S, etc. there was no point in grabbing without experience, so I stopped my choice on indirectly heated powerful stabilizing triodes 6N13S (6N5S is their complete analogue). For a single-ended amplifier, they are not suitable because of the "congenital curvature", but for a push-pull fit. The Uym de Jaeger scheme, which is the classic Williamson, was chosen as the initial one. You can get acquainted with all its advantages and disadvantages, here I will only give the original scheme.
With Huim de Jaeger, I immediately disagreed on the issue of biasing the output tubes. With a selected quiescent current of each triode of 70 mA, four auto-bias resistors turn into a hellish stove, so it was decided to convert the bias of the output stage lamps to a fixed one. In addition, this approach also improves the sound of the amplifier. There were no other disagreements with the author of the scheme, so the rest of it remained unchanged. As an input lamp, 6N2P was originally planned, in the second stage 6N1P, and the place in the output stage, as I said, was staked out by 6N13S. I redrawn the scheme in my own way and offer it to your attention.
Well, now is the time to think about nutrition. In my slightly modified amplifier circuit, I had to separate the anode power of the first two stages and the output, since the 6H13C is a relatively low-voltage lamp. Therefore, in the power supply, I will need two anode rectifiers - 180 V for the output stage and 270 V for the first two. You will also need two "underground" sources: -50 V for the drive stages and -100 V for biasing the output stage lamps. In order to extend the life of the lamps, which is far from infinite, it does not hurt to build a simple soft start with a delay and a smooth supply of anode voltage. Based on these considerations, a terrible at first glance power supply circuit appeared.
After the first shock and a closer examination, we see that there is nothing terrible. Three capacitors in the primary circuit of the power transformer T1 form the simplest network noise filter, the resistor R1 is a discharge resistor. Winding 4-5 of the transformer - anode for powering the output triodes. Fast diodes in rectifiers are used to reduce switching noise. The rectified voltage is smoothed by capacitive and electronic filters, and the electronic filter has a separate active element (transistor) for each channel, which is done to isolate stereo channels by power supply. According to a similar scheme, an anode voltage source of +270 V was assembled for the first two stages. Electronic filters are switched on by the electromagnetic relay K1 approximately 45 seconds after the amplifier is connected to the network. That is, at first the incandescence of the lamps warms up without an anode voltage, and then this voltage gradually increases for about 1 minute.
Resistors R10 and R17 discharge the filter capacitors after the amplifier is turned off. From the windings 8-9 and 10-11 of the power transformer, two "underground" rectifiers connected in series are fed. They provide two negative voltages: -50V for the driver stages and -100V for biasing the output stage lamps. The “silovik” has two filament windings - one for each channel. Resistors R2...R5 form the middle point, to which a positive potential is applied from the divider R6R7. This is done in order to get rid of the 50 Hz background that would inevitably occur.
All fixed resistors in the amplifier itself are MLTs of the power indicated on the diagram. It would be better to pick up carbon ULM or BC, but, as they say, we have what we have. Current measuring R19 and R20 with a tolerance of 1%. It is desirable to set the volume control R1 to a high-quality one, a lot depends on it. I still have a Chinese TOMY, the flight is normal. Electrolytic capacitors will be hot, so I had to fork out for 105-degree ones. The requirements for interstage capacitors have long been known to everyone, I used MKR X2, which showed their best performance at a low price. C1, C8 and C9 - film. For the time being, the output transformers will be incandescent TN33, if I can find human ones, I will replace them. In the power supply, the same picture - MLT resistors and 105-degree electrolytes. Moreover, the filter capacitances are shunted with a film, and the capacitances at the outputs of electronic filters are shunted with paper in oil. A hastily assembled layout of one channel inspired some hopes for the success of my enterprise.
My body, as always, starts with anything. This time I came across a piece of duralumin of suitable size, but two rectangular holes were cut in it, with which I had to suffer pretty much, since they were by no means out of place. For a very long time I spun this piece this way and that way, until I still managed to more or less “compose the design” more or less acceptable. After marking, drilling and cutting the necessary holes, I thoroughly walked the sheet with a large sandpaper with kerosene, and this is what happened.
The next step is to mount the lamp panels and a milliammeter, which I previously disassembled and glued two yellow LEDs into it to illuminate the scale.
I mounted the amplifier circuit in a hinged way, on the petals of the lamp panels and a common bus, which is mounted above these panels. Trimmer resistors for adjusting the bias are placed on the far side of this panel-chassis and allow, if necessary, to quickly adjust the quiescent current of the output triodes.
Now it's time to make the body itself. And I decided to make it from beech cutting boards, cutting them into blanks of the right size. The back wall is made of textolite 6 mm thick and pasted over with self-adhesive under beech. It contains input connectors and a socket for a network cable, combined with a fuse box. In the front panel, I drilled mounting holes for self-tapping screws, as well as holes for the volume control and toggle switches - network and switchable OOS. I blew out the wooden parts with a glossy clear varnish from a spray can.
When they were dry, I attached paper-oil capacitors to the sides with tin staples. I installed the volume control, also providing it with a yellow LED and decorating it with two gray and black plastic discs. I collected everything in a heap and saw that I had missed the height of the sides. I had to screw a wooden ruler to them from below to increase their height. Result:
Finally, both parts of the body are screwed together with self-tapping screws. Further assembly will now be in a full-fledged case.
A box with four holes behind the capacitor is a block with trimmers for adjusting the bias. The resistor shafts are slightly recessed into the surface of the block so that accidental contact does not disturb the adjustment. Next, I install the output terminals, output transformers and connect them according to the diagram. I connect a network toggle switch, a milliammeter with a switch. Well, and so on.
Now the power supply. I assembled it on a printed circuit board and fixed it in the basement under one of the output transformers, and the power torus under the other. I cut the wires to the required length so that there are no loops, and tighten them with cable ties.
First turn on! There are no sparks and smoke, the lamps are warming up, the anode voltage is increasing ... Christmas trees! Instead of 270 V anode I see 340, and instead of 180 V - 210! An annoying miss! I measure the lamp modes - in the first two stages, the dissipated power on the anodes does not go beyond the maximum allowable limits, in the output stages it exceeds by 1 W. Well, increasing the supply voltage makes the driver more linear, it's even better. And I will slightly reduce the currents of the output lamps, although this is not necessary. Now you can proceed to the measurements of the parameters of the amplifier. I wonder what kind of animal it turned out.
I must say in fairness that the first tests were not very encouraging. For some time I played with the lamps of the first and second stages and finally settled on such a “configuration”. In the first cascade there was a pair of successfully purchased 6N2P-V with "VP" stamps. But in the second cascade, the obviously second-hand Novosibirsk 6N1P-E of the 60s, found in the bins, was unexpectedly registered. Interestingly, their electrode system is completely different from that of conventional 6N1P, it looks more like 6N3P. So: these lamps sound just great! I give a picture: on the left is 6N1P-E, on the right is the usual 6N1P.
I now return to the topic of measurements. I conducted them in two modes - with an open OOS and with a closed one at the same output power - 10 watts. In the first case, the sensitivity was 0.2 V, in the second 0.45 V. The test results in the plate:
Recorded frequency response for modes with open and closed feedback, respectively:
Yes, this, of course, is not a fountain, but let's remember for a moment what output transformers are in the amplifier. That's right, TH33. Can we expect miracles from them? Of course not. But even with all this, I am very pleased with the sound of my first triode (namely, triode, and not pseudo-triode, where tetrodes and pentodes are switched on in a triode circuit) amplifier. One feels power, freedom, looseness in sound, excellent bass, purest highs. Accuracy and focus, hence the name of the amplifier - Focus. There is no hint of lethargy and dryness, as seasoned representatives of the "pentode chamber" characterize the triode sound. When the OOS is closed, the sound becomes somewhat clamped, as if compressed. I prefer OOS-free sound, despite its worst parameters. This is exactly the case when the scales tilt in favor of subjective perception, and not the results of measurements.
It remains to close the amplifier from below with a lid with holes for ventilation, screw the legs, put the output transformers in stainless mugs and seal the mounting holes on the front panel with decorative overlays, which was done. Now, it's finally done!
This is the final look of the tube amplifier, which I decided to call "Focus". The author of the project is Gamzan.
Tell in:To check the emission of an electron lamp in fig. and first turn on the filament circuit of the lamp. After 60 ... 120 s, a milliammeter is connected (scale 300 mA or less). No voltage is applied to the other electrodes of the lamp. The more the arrow of the device deviates to the right, the better the emission, and hence the lamp. In double lamps, it makes sense to determine that "half" of the lamp for which the arrow of the device deviates more. To determine the emission of a lamp, you can use an ohmmeter fig b. For a new lamp, the resistance of the gap between the filament and the control grid can be, for example, 900, for a used lamp it is 2000, for those that have lost emission, 4000 ... 4500 Ohms. The measurement readings of the devices are compared with similar measurements for a new lamp and the degree of emission loss of the tested lamp is determined.
A push-pull bass amplifier will only work properly if its arms are symmetrical.
Balancing the lamp stage can be done in the following simple way: voltage of the same phase is applied to the lamp grids of the output stage and by adjusting the variable resistor R1, the minimum signal at the amplifier output is achieved (switch B1 is in the lower position according to the diagram). After that, the switch B1 is set to another position and anti-phase voltages are thereby applied to the grids of the lamp. The output signal in this case should be maximum. Such an adjustment is recommended not only when adjusting the amplifier, but also when changing lamps.
In conclusion, I propose two schemes:
Diagram of a simple two-stage amplifier 
The output stage is made on a 6P14P lamp operating in a typical mode. The pre-amplification stage is made on one of the triodes of the 6N3P lamp. It provides a signal amplification by 27 times, resulting in an amplifier sensitivity of about 0.3 V.
Amplifier circuit with a push-pull output stage, negative feedback and frequency correction.
The output power of the amplifier is about 10 V * A.
The phase-inverted stage is made on one triode of the 6N2P lamp, the second triode plays the role of a preamplifier. Negative feedback covers a part of the amplifier, consisting of cascades: push-pull together with an output transformer, phase-inverted and preliminary on one triode of a 6N2P lamp. The feedback depth is three (1+B K=3).
With the help of R1, the frequency response is adjusted in the region of high sound frequencies, with the help of R2 - in the region of low frequencies.
The coefficient of non-linear distortion of the amplifier is about 2.5%, the sensitivity is about 0.1 V Section.