In the wake of great interest in tube technology, I want to describe the design of a tube preamplifier "for the little ones." Or for those who are not the smallest, but who do not have time for a serious deepening into the tube circuitry, but who want to try the "tube sound" and look at the pleasant warm glow of the lamps in the dark. Definitely - the characteristics of this design are more than modest, but at the same time it is very functional and, most importantly, does not require special skills for assembly and does not contain expensive and rare elements.

The design is based on a common Soviet radio tube 6Zh1P- "high-frequency short-response pentode". Its detailed characteristics and application features are easy to find on the Internet, in particular, on the site that I myself use - The Magic of Lamps. Its main feature, thanks to which we choose it, is the ability to work with low voltage. Yes, if you are interested in tube designs, you should certainly know that the anode voltage in most of them is hundreds of volts, which means you need an anode transformer, expensive high-voltage capacitors, an output (in fact, a step-down) transformer and, in the end, assembly precautions and skills. The second - no less important - is the unique cheapness and availability. All other parts are standard passive elements. You will have to order separately, perhaps only a linear stabilizer for 6V LM7806 (about it - separately), but - even then - it can be replaced with an adjustable stabilizer LM317 or, in general, with a design with a transistor and a zener diode.

So, in order.

This device is considered a pre-amplifier very conditionally due to its rather low (unity) gain, which depends on the supply voltage. The main function of the device is matching the level and output impedance of the signal source with the load, and, of course, introducing into the signal a small level of specific distortions inherent in tube technology.

Source stereo signal for it can be a player, a digital-to-analog converter (possibly as part of a sound card) or an electronic musical instrument (including those with a high output impedance). The output from the device is fed directly to the power amplifier, or any device with a line input.

As the most successful application for this device, I would highlight the following solutions:

As a matching device between the DAC and the power amplifier. So, many DACs do not have an output buffer and are "capricious" to the input impedance of the subsequent device. The preamplifier compensates for this due to the rather high input impedance of the tube stages with a signal applied to the grid. Well and - where without it - some smoothing of "digital artifacts" + typical "warm tube" distortions.

For sound recording of an electronic musical instrument, incl. with a high output impedance or after a digital special effects device (guitar processor). The preamplifier will help you to set the desired signal level and - of course - the "tube character of the sound".

Scheme

It is really possible to assemble this device with all the parts at hand in one evening, taking into account the casework (even such as drilling large holes for lamp sockets). By the way, I strongly recommend taking a metal case. The electronics work will take hardly an hour.

Indeed, for one cascade ( there are two of them in the design - on the right and left channel) there is only a lamp (V1 / V2), a resistor in the anode circuit (R3 / R5) and a blocking capacitor at the output (C3 / C4). In addition, a potentiometer (R2 / R4) for adjusting the input signal level (I recommend a linear potentiometer with a resistance of approximately 50kOhm - 100kOhm), a blocking capacitor at the input - at will (I personally did not install it).

The rest of the circuit is the power circuit. C1, R1 and C2 - power filter and linear stabilizer DA1. It's worth dwelling on the DA1 chip. It is needed so that no more than the required 6.3V is supplied to the heating of the radio tubes. In this design, I used the closest in voltage LM7806 output 6V. As I wrote above, you can replace it with other solutions ( about them, if there is a need, I will tell you separately). It was also possible, of course, to make a separate heating power supply and a separate anode power supply. This would give us a few more options, but - at the same time - would greatly complicate the design... But with this inclusion, the entire circuit can be powered from a standard 12-18V adapter.

Now a few very important words about the power supply. As I wrote above, the gain of the circuit and the dynamic range are the higher, the higher the supply voltage... However, there are limitations here. We will not take into account the maximum anode voltage of the lamps - it is quite high, we will focus on the weak link of the circuit - the stabilizer. The maximum voltage that can be applied to its input is 35V, maximum current - 1A. The filaments of the two lamps together consume about 300mA... It would seem that the stock is pretty decent. However, in practice - the greater the current consumption and the input voltage - the more heat the stabilizer generates... See datasheets for exact thermal specifications and tolerances. Therefore, the maximum allowable supply voltage will be partly determined by the heat sink (radiator) on which the stabilizer will be installed.

In my design, for example, the metal body of the device is used as a scattering surface - the microcircuit is screwed to the wall through thermal paste. By the way, the insulating pad not required if you, as in most classic solutions, combine case with minus power supply(in our design, the power supply is unipolar and "minus" will be "mass" and, accordingly, shield the circuit). The case does not dissipate heat very well (not much, but it heats up noticeably in an hour of operation), so I limited the supply voltage to 12V. If you install the stabilizer on a sufficiently massive radiator ( just please don't overdo it! the basic idea of ​​the design is compactness!!! ), then the voltage can be increased to 18-20V. Achieve limit value 35V I categorically do not advise, since they significantly reduce the life of the element and soon it may fail from overheating!

Well, a few words about the design and a couple of assembly tips.
The green numbers on the diagram next to the lamp leads are the electrode numbers. The location of the electrodes on a standard 7-pin panel is shown below.

Just in case, here is the purpose of the contacts for the linear stabilizer.
And, finally, the design itself.

Any metal case the size of a pack of cigarettes will do. In my case, it was once the D-Link Media Converter. Using a cone drill, I made two large holes with a diameter of 22mm in the socket. It was decided to mount the installation hinged. For this design, a PCB is completely redundant. With so many radio elements, only two contact blocks of 10 contacts were enough, and they were not fully involved.

Don't forget about star connection- all taps going according to the scheme to "ground" must be connected at one point with power supply and housing. True, again, for such a simple circuit with a low anode voltage, this principle is not critical, although you should accustom yourself to observe it everywhere. Experienced electronics engineers will surely point out to me that the wires inside are not laid out the way they do in complex and expensive amplifiers. Of course, it is worth striving for this, but it is not for nothing that I wrote in the headline - "... in one evening." With such conditions, there is no time for perfectionism, but - on the other hand - I think this is a good demonstration that even the most novice radio amateur can cope with the assembly of the device.

That's all. A properly assembled structure works immediately. Personally, I am quite satisfied with the sound - at least it corresponds to the level. You can feed from an ordinary adapter, as mentioned above, with a voltage of 12-18V, but - preferably - stabilized. In this case, the likelihood of power pickups will be reduced. I listened through Soundtech Series A on Quested S6, the signal was sent from E-mu Tracker.

Digging into the box with lamps, I found a dozen 6Zh1P lamps. Where can they be applied?
This high frequency pentode is designed to amplify high frequency voltage in radio engineering devices.
Remarkably, it can operate with a low voltage of about 40 volts.
If you have high-impedance headphones, you can assemble a tube amplifier.

In an hour, a mock-up of the amplifier was assembled, without a power choke, of course, without it phonit. The model was assembled in order to estimate the power that these lamps at 40 volts can give out. See the photo of the 6Zh1P headphone amplifier layout below.

After listening to this amplifier on 32 ohm headphones, it seemed to me that the signal strength was insufficient. When connected to the output of a transformer, there is already enough power. Taking into account the fact that the necessary parts can be easily found in the collection of any radio amateur, and the absence of output transformers, I recommend assembling.

Addition.
When unsoldering the breadboard and assembling the amplifier on the chassis, an error was found in the circuit. 2 conductors were not drawn, from the 6th leg of the lamp to the +40 volt power supply. I promise to fix it on the diagram when I finish another design with smaller dimensions and a separate external power supply.
At this stage, the final version of the 6Zh1P headphone amplifier has been completed.

Below is a diagram of the UMZCH for reproducing G. Krylov's recordings. Its output power is 6 W with a nonlinear distortion factor of 3%; at an output power of 4 W, the harmonic distortion factor is 1%. Frequency response unevenness in the range from 25 Hz to 16 kHz - 1 dB. Input sensitivity - 170 mV. Background level -55 dB. A feature of the amplifier (Fig. 13), which consists of a pre-amplification stage, a push-pull output stage and a rectifier, is a kind of excitation circuit of the final stage without using a phase inverter.

The signal from the volume control R1 is fed to the control grid of the 6Zh1P type lamp, amplified by it and fed to the control grid of the L2 output lamp of the 6P15P type. The signal voltage from the cathode of the L2 lamp is fed further to the cathode of the L3 lamp.

The signal voltage U supplied to the L3 lamp can be determined from the formula: U = (I1 - I2) (R7 + R8), where I1 and I2 are the alternating components of the L2 and LZ currents.

It is not possible to increase this voltage, since for good use of the L3 lamp, the current I1 must be close to I2, and it is impossible to increase the resistance of the resistor R8 due to a decrease in the anode voltage. Therefore, this circuit is of interest only when using lamps with a high slope, operating at a low excitation voltage. Of the common lamps, the 6P15P pentode meets this requirement.

To reduce harmonic distortion and reduce the output impedance, the amplifier is covered by negative feedback with a depth of 14 dB. The feedback voltage is removed from the secondary winding of the output transformer and is fed through a resistor to the cathode of the L1 lamp.

The power transformer is assembled on a core made of Sh32 plates, the thickness of the set is 32 mm, the window is 16x48 mm. The mains winding contains 880, and the anode 890 turns of PEL 0.33 wire, the filament winding consists of 28 turns of PEL 0.8 wire.

The output transformer (Fig. 14) is made on a core made of Sh26 plates, the thickness of the set is 26 mm, the window is 13X39 mm. The primary winding contains 1200X 2 turns of wire PEV-2 0.19, the secondary - 88 x 3 turns of wire PEV-2 0.47. It is necessary to strictly maintain the equality of the number of turns of the sections of the secondary winding and connect the sections in parallel.

The amplifier is mounted on a 1.5 mm aluminum chassis with dimensions 240x92X53 mm. The first stage should be as far as possible from the power and output transformers. The body of the potentiometer R1 should be connected to the chassis.

The distance between power and output transformers must be at least 15 mm. The axes of their coils must be mutually perpendicular.

Adjusting the amplifier is reduced to adjusting the amount of feedback by changing the resistance of the resistor R10. If the amplifier is energized, then the terminals of the secondary winding of the output transformer must be reversed. To avoid self-excitation of the amplifier at ultrasonic frequencies, the feedback depth should not be more than 15 dB.

The bridge rectifier on diodes D209 can be replaced with a selenium rectifier ABC - 120-270. It is advisable to replace capacitors C5, C6 with one 150 μF capacitor for a voltage of 300 V. The loudspeakers of the acoustic unit must have an impedance of 8-10 ohms. The author used two 5GD10 loudspeakers connected in series.

I haven't written an article for a long time - the summer cottage season has begun. There was a lot of work on the site and other troubles. Nevertheless, sometimes there was time for a hobby. For a long time I began to be interested in radio tubes, and to be precise, since 2013. Although I have a large park of tube radios at home, but beyond repair I did not delve into this wonderful world. Nevertheless, I really wanted to try something tube. Having outlined something interesting for myself, I began to read forums, download interesting schemes. Slowly put aside details and think about future cases. But as I already wrote in the blog, I had a difficult period associated with the move, and I had to put homemade products on the back burner. Little by little, I came to the conclusion that I want not only the classic ULF SE 6n2p + 6n14p, but the legend 6p3s, and I also wanted to get a tube VHF receiver, or rather only the FM detector unit, since I see no reason to fence the HF part on the lamps. Further more - I wanted to assemble a tone block, a super-regenerator receiver for HF and ULF for headphones. The latter will be discussed. I realized that there are few schemes without trances and they have quite a few problems. There are also problems with a rather high anode voltage. In this regard, I decided not to bother for headphones with ULF, but to collect SRPP for 6n1p / 6n23p / 6n2p. However, leafing through the vastness of the Internet, I came across a simple 6zh1p circuit with a power supply of only 12 volts. Pentode 6zh1p is connected by a triode. Diagram of a stereo amplifier for 6zh1p for headphones (on the left are the pins of the 6zh1p lamp):

Everything in this circuit is terrible - both the low supply voltage, and the absence of a transformer, even the very option of turning on the headphones, and that implies the flow of anode current through the speaker coil. Nevertheless, I remembered my first designs such as a single-transistor ULF on a kt-315 or mp-41 and thought - why not?

On my hands I had the necessary nests, a small box and a free evening (even 2), having assembled a layout on lamp panels - I was disappointed at the beginning. The amplification of the cascade floated in region 1, i.e. the amplifier was pseudo, moreover, the input was 0.3 V and that's it - distortions began. so I decided to listen and compare the sound at night. The difference was noticeable, especially when connecting the tablet. The sound in this ULF became a warm tube and there was a certain rise in the LF. Although it cannot be called pure, there are still distortions. As for the volume, it is quite sufficient for a headset (with a resistor-adjustment and a microphone), in droplets of 32 ohms - a little quieter and no bass. Having estimated that when I am still going to assemble a full-fledged ULF, I decided to assemble it in a case.

I didn't bother - an ordinary plastic box (mounting for wires, inside the terminal block). The terminal blocks that were inside, I removed and cut off the plastic pins so that they did not interfere. I drilled holes for lamp sockets, sockets. Editing was carried out directly on the petals. The wire used MGTF. I tried to comply with the installation rule - the minimum length of the conductors and the correct cultivation of the earth. A capacitor has been added to the circuit - an electrolyte of 100 microfarads parallel to the power socket. The incandescences of the lamps are connected in series. For 6zh1p, these are pins 3-4. The amplifier is powered by direct current, the unit is pulsed 12 V 2 A (Huawei HW-120200E6W). No background is bugged.

However, you need to take into account some peculiarities. For example, lamps must be matched in pairs. Otherwise, there may be a bias in the power supply of the filament or a difference in volume. It is worth noting that I have 1 phonil lamp, if you touch its body with your hand, the most interesting thing is that the amplification of the ULF is such that there is almost no background of 50 Hz if you touch the entrance with your hand, but when you touch the lamp bulb, the background appears. I just changed the lamp to another and that's it. You should also pay attention to the power supply unit - the conversion frequency of the UPS should be significantly higher than the audio range, then 50 khz and higher, otherwise it is possible to listen to the squeak in the headphones. And in general, it is better to power it from a transformer unit, it is possible not to be stabilized, but the voltage should be within 12-13 volts. As you can see in the diagram - originally an EF95 lamp, it was replaced with 6zh1p. I decided to try to pick up analogs of 6zh1p with the same pinout so that I could listen and choose a lamp that would give an even warmer tube sound :-) There were 6zh38p and 6zh5p available, with them the sound is worse. Especially with 6zh38p. Also, a significant disadvantage of 6zh5p can be considered a higher filament current and strong heating of the cylinder. So 6zh1p is the best option for this sound tube generator. Very important! Since the filaments are connected in series, it is impossible to put different lamps together. The spiral of a lamp with a lower incandescence current may burn out. Before connecting the headphones for the first time, you must be extremely careful - after all, there is no protection in the circuit, if it turns out that the lamp has a short circuit between the cathode and the anode or there are errors in the installation, then the headphones may burn out, since an unlimited current of the power supply will flow through them! Also, this circuit cannot act as a tube buffer for most ULFs. The output of this ULF is designed to turn on a relatively low-resistance resistance of the order of 32-600 ohms, and moreover, the load must ensure the flow of the anode current. Of course, the output can be altered by putting a 100-500 ohm resistor instead of a headphone and connecting an additional ULF through a 100 uF isolating capacitor, but that's a completely different story ... that is. scheme.

Appearance of the assembled amplifier:

Conclusion: miracles do not happen, if any radio element has optimal operating modes, then when it goes beyond it, a sharp deterioration in performance is possible, and lamps are no exception. You should not expect high-quality and magical sound from such schemes. They are of interest only from the point of view of self-education (in this case, it is extremely doubtful), or experiment - is it possible for the lamp to work in such an unusual mode of operation. Is it worth collecting this circuit? I don’t know ... I am not planning to assemble and disassemble it, since there is nothing valuable in it for future designs.