Hybrid audio amplifier. Hybrid umzch without oos Relay types are shown in the diagrams
Hello everybody.
I will continue about the terminal cascade of Alexander Pavlovich Deria.
At the beginning of 2017, I published the circuit of the completed Alexander Pavlovich amplifier on this site, and in parallel, to discuss this circuit, I published it on the AP and on diyaudio.ru
During the discussion at the AP, many questions were raised, and these discussions were not in vain.
There's a lot of mannerisms and vomit on DIY, like give an amplifier with a transformer ass
or oh, it's a pity now I'm standing in the hospital in line. And then I would take a picture with a glass So take a picture. It's not necessary to drink. Although it's a pity... In general, moderation on this forum "ordered to live."
Yes, the sad and vile is also present, and happens, in some forums.
This is a classic ITUN with all the consequences. If resistances of 0.5 ... 1 Ohm are included in the emitters of the output transistors (and the corresponding resistors in series with the bias diodes), the distortion will decrease significantly. Yes, and the thermal stability of the quiescent current will be much better.
Alexander Pavlovich drew conclusions and decided to experiment with complementary pairs at the output and field-effect transistors at the input.
The main idea belongs to Alexander Pavlovich. and if you characterize it briefly - “then you don’t have to be afraid of a large output resistance”
We all love numbers, and this is also very necessary and good. As they say, a fact is a fact!
But the fact must not be disguised. It happens that everything is in order with the numbers on the amplifier, but there is no sound
And recent measurements have shown that the amplifier is linear from 20Hz to 20kHz and even higher. By -3 dB 75 kHz!!!
Personally, I was glad that you can shoot from 10 parts, and up to an undistorted sine of 1000Hz 65 watts in the hybrid version.
The lamps were used 6Zh11P 6Zh43P in the triode and 6F4P in the standard inclusion ..
6P9, 6P15, 6E5P, 6E6P and IL861 and El861 were also tested
(I want to note that the glow of the IL861 lamp is -20 volts)
The only thing that can be considered a "fly in the ointment" is a large output impedance from 6Om to -20 Om from the prototype of Alexander Pavlovich, and from 30 to 50 Om for my hybrid version, depending on the lamps used. The output impedance of the amplifier depends on the choice of driver.
Many people think “and know” that the high output impedance of the amplifier has a bad effect on the damping of the acoustics, but part of the small population still believes that the acoustics, moving mechanically in the opposite direction, creates a field that also affects the amplifier no less than the amplifier affects the acoustics and, accordingly, the sound generally!
Some literature says that with an output impedance of 18 Om, acoustic damping is already a fact.
But the majority will not agree with this statement, since the closer to "zero" the output impedance of the amplifier, the more correct.
There is another opinion - that the output resistance in the range of 10-20 Om has a beneficial effect on the final picture as a whole. The sound is not clamped, “off the ground”, widening the panorama, ease of perception, no fatigue even after several hours of listening.
Triode and pentode amplifiers also have different output impedances, but both are entitled to sound, and have their own pros and cons. How many ears, so many opinions.
The following photos provide a rectangle at 1000Hz at 10kHz and at 20kHz. Load 5Om. From them it is clear that the amplifier is in perfect order. These are measurements of a purely transistorized amplifier assembled by Alexander Pavlovich Deriy.
Amplifier Chuyka 1.5v
Power supply + - 24 volt transformer - overall power of only 80 watts (from the amplifier Radio engineering -101)
29 watts of undistorted sine!
0. dB - 20Hz - 20KHz
The bottom of -3db could not be measured, the top of -3db -75kHz
Output impedance 20 ohm.
Looking ahead, the tube hybrid amplifier with the same circuitry produces 65 watts at 0.75v chuyki when powered + - 38 volts
20Hz -0.25dB 20KHz +1dB 45KHz-3dB
The output stage of the amplifier is provided in the following figure.
It can be arranged both with common emitters and with common collectors. In the latest versions, we settled on the version with common manifolds.
It is very convenient to mount transistors on a radiator without mica plates.
Below are two driver versions of 1988 and 2018
The field-effect transistor KP901 can be replaced with a conventional composite transistor KT972, this does not affect the sound quality, this transistor acts as a repeater. Resistors R11 and R12 can and should be replaced by 0.6 Ohm., The stability of the output stage will increase and distortion will decrease. It is desirable to put a zobel chain to the output and put 56 ohms in parallel with the speaker, while the output impedance will decrease by 10-15%.
The quiescent current of the transistors and the zero level are set by resistors R7 and R10 with a decrease in ratings, the currents decrease, and increase with an increase. The quiescent current is set from 100 to 200 mA, it all depends on the grandiosity of your radiators. For example, in the hybrid version, I generally set 280 mA, and this is not the limit.
IMPORTANT! Be sure to install a matched complimentary pair, if this is not done, then the modes can “float away”.
When properly assembled, the amplifier works immediately
Below is a hybrid version of the amplifier. Power supply + - 38 volts. Anode 200 volts. EL861 driver lamps.
Ktr transformer 12.5/1/1 Primary winding is wound with wire 0.25-0.33 3000 turns Secondary 2X240.
I wound on OCM 0.063. Winding was carried out in the following way.
900 turns first. — 120 turns sec. - 1200 turns first. — 120 turns sec. -900 turns perv.
The secondary wire is wound with a double wire from 0.33 to 0.51. Each layer was laid with graph paper.
The transformer is not phase inverted. The role of the phase inverter is performed by the output stage. This is a big plus in this circuitry. Plus, I also think that the transistor collectors are screwed directly to the radiator without mica gaskets.
The amplifier is assembled in a 6mm plywood case. Plywood dampens hums from transformers well, vibration is not transmitted to lamp grids. At 65 watts output, hum is minimal. At 100 dB of acoustics, it is barely audible if you stick your head into the speaker.
Metal top and bottom.
I will provide a photo and video report additionally when I “comb” the installation.
Sincerely, Evgeny Vilgauk Chelyabinsk
I welcome all visitors to the site and present the design of the UMZCH, which in my opinion (ear) is the embodiment of all the best that we can take from modern transistors and vintage lamps.
Power: 140W
Sensitivity: 1.2V
The circuit contains a small number of parts, is easy to set up, does not contain scarce and expensive components, and is very thermally stable.
Briefly about the scheme.The source follower is implemented on complementary MOSFET transistors IRFP140, IRFP9140 and has no special features. Transistor VT1 does not affect the sound, it is needed to stabilize the current when the temperature of the output transistors changes and is installed in close proximity to them on a cooling radiator. It is desirable to have a massive radiator with a large cooling area, to install transistors close to each other on a heat-conducting paste, through a mica gasket. Capacitor C4 provides a "soft" start source follower.
Now about the driver. I had to tinker with the driver, because. the input capacitance of one transistor is 1700pF. Different types of lamps and different switching schemes were tested. We had to abandon low-current lamps, because. the blockage in the HF began already in the audio range. The result of the search was SRPP on 6N6P. With a current of each triode - 30 mA, the frequency response of the amplifier extends from a few hertz to 100 kHz, a smooth decline begins in the region of 70 kHz. The 6N6P lamp is very linear, in addition, the 6N6P driver has a huge overload capacity. Triode modes 6N6P - 150V, 30mA. According to the datasheet Rmax.-4.8W, we have 4.5, almost at the limit. Who cares about 6N6P, you can ease the regime by increasing the values of the resistors R3 and R4, say, to 120 ohms. And yet, despite the fact that the 6N6P lamp has a small gain, it turned out to be prone to self-excitation, maybe the whole thing is in the copies I have, but, nevertheless, measures were taken to stifle this undesirable phenomenon. A standard aluminum screen was put on the lamp, the ninth leg was soldered to the ground, a small coil was installed in the grid - 15 turns of PEV 0.3 wire, wound around a 150 kOhm - 1W resistor. If a smooth frequency response at HF is not the main thing for you, you can try in the 6N8S or 6N23P driver, in SRPP, of course.
Setting up the amplifier is simple - set R5 to the middle position, and R8 to the lower position according to the diagram and turn on the amplifier. We warm up for 3 minutes, turn R5 - set “0” at the output, then carefully turn R8 - set the quiescent current of the output transistors. The current is controlled by measuring the voltage drop, on any of R15, R16 it should be - 110mV, which corresponds to a current through the output transistors of 330mA. The quiescent current is up to you - it all depends on the heatsinks and fans you have at your disposal. Amplifier setup is complete - enjoy the sound.
I don’t bring the power supply, because. everyone can develop it himself. But I want to warn you that saving on a power supply is the last thing. Put large transformers, huge capacities and you will be rewarded. Don't forget to put fuses everywhere.
Details. The details are the most common, OMLT resistors, JAMICON capacitors, resistors R15, R16 are made up of three parallel-connected OMLT-2 - 1 Ohm, R8 is a wire, ALPS input potentiometer. The use of audiophile components is welcome, especially for power supply capacitors. Separately, it must be said about C3, C4, C5, the sound of the amplifier depends on them, so you better choose the type of capacitors to your taste. I have imported red-brown filmmakers of an unknown manufacturer, I suspect they are made in China. If you do not need the frequency response of the amplifier to be linear from 2 Hz, then the capacitances of capacitors C3 and C5 can be reduced. It is desirable to select the output transistors in pairs according to the parameters.
When the amplifier is turned on, an alternating current background is heard for several tens of seconds, then it disappears. This phenomenon is due to the fact that the source follower has a large input resistance and while the cathodes of the triodes are warming up, the follower input is “suspended” and “receives” the electromagnetic fields surrounding it with the frequency of the industrial mains. There is no need to deal with this phenomenon - you need to implement a delay in turning on the speakers.
Amplifier power - 140W, with Uin.eff. - 1.2V. There is nothing to measure the coefficient of non-linear distortion, but I do not think that this amplifier has a horse, judging by the sound.
Now about the sound itself. The sound of this amplifier is similar to the sound of a triode push-pull, but the bass register is much "meater", the bass is fast, clear and solid. The middle is transparent and detailed, the tops are without the “sand” inherent in transistors.
The amplifier eats everything, shakes any acoustics. The amplifier was conceived for outdoor use - a single-cycle tube at home, but now I'm not sure that it will not be the main one. Let's listen again.
And yet, when building an amplifier, it is desirable to equip it with a system of all kinds of protection, this will improve its performance and protect your speakers from emergency situations.
List of radio elements
| Designation | Type | Denomination | Quantity | Note | Shop | My notepad |
|---|---|---|---|---|---|---|
| VT1 | bipolar transistor | KT602BM | 1 | To notepad | ||
| VT2 | MOSFET transistor | IRFP140 | 1 | To notepad | ||
| VT3 | MOSFET transistor | IRFP9140 | 1 | To notepad | ||
| Diode | KD521A | 2 | To notepad | |||
| zener diode | 12 - 15V | 2 | To notepad | |||
| Lamp | 6N6P | 2 | To notepad | |||
| C1 | electrolytic capacitor | 10000uF x 50V | 1 | To notepad | ||
| C2 | Capacitor | 0.1uF x 63V | 1 | Film | To notepad | |
| C3-C5 | Capacitor | 6.8uF x 63V | 3 | Film | To notepad | |
| R1 | Variable resistor | 50 kOhm | 1 | To notepad | ||
| R2 | Resistor | 220 kOhm | 1 | 1W | To notepad | |
| R3, R4 | Resistor | 100 ohm | 2 | 2W | To notepad | |
| R5 | Trimmer resistor | 33 kOhm | 1 | To notepad | ||
| R6 | Resistor | 86 kOhm | 1 | 1W | To notepad | |
| R7 | Resistor | 56 kOhm | 1 | 1W | To notepad | |
| R8 | Trimmer resistor | 15 kOhm | 1 |
We hope that your home audio system has been replenished with quality from our latest publications. Now it's time to think about the power amplifier. Today we offer you a description of the design of one very interesting hybrid amplifier. author wim de han called his creation "MuGen". In Japanese, this means infinity, but from a technical point of view, the amplifier combines a voltage amplifier - Mu and a current amplifier - Gen, which is reflected in the name.
Today tube amplifiers are undergoing a rebirth - a fairly large number of both commercial and home-made designs have appeared. Unfortunately, their most worthy samples are distinguished by a very immodest price, which is due, in particular, to the need for high voltage for the operation of the amplifier and the presence output transformer. The rather high internal resistance of the lamps does not allow you to connect acoustic systems directly to them. And a cheap output transformer of mediocre quality will negate all the efforts to assemble the amplifier, no matter how expensive and high-quality the other components are, no matter how well the circuit is worked out.
In hybrid amplifiers, the output transformer is replaced transistor cascade, which has a low output impedance, which allows you to connect a load to the output of the amplifier without any tricks. Modern electronic devices at the same time allow you to get very high performance and low distortion.
MuGen Amplifier Parameters and Schematic:
- Input sensitivity: 825 mV (8 ohms) and 770 mV (4 ohms)
- Input impedance: 300 kOhm
- Gain: 29 dB (23 dB with total negative feedback)
- Output power (at 1% THD):
- 70 W into 8 ohm load,
- 110 watts into 4 ohms
- Harmonic distortion (THD) + noise:
- at output power 1 W / 8 ohm:<0,1%
- at 10 W / 8 ohm output power:<0,15%
- Damping factor: 20 (into 8 ohm load)
The amplifier circuit is shown in the figure:
Zoom on click
input stage.
To obtain a given output power, the input stage must provide amplification of the input signal to an amplitude of 25V. In addition, due to the absence of a common negative feedback, this stage should have minimal distortion when driven into a load of 10 kΩ (the input impedance of the output driver).
Based on my experience with lamps, the author chose a differential stage for the input of the amplifier, which, among other things, allows it to be used as a phase inverter and it is enough to simply introduce general negative feedback into the amplifier if such a need arises or a desire to experiment. In this case, the LLCS signal is fed separately from the input signal to the grid of the right triode.
Since the cathodes of the lamps of the first cascade are connected in series by alternating current, this generates local feedback with a depth of about 6 dB, which reduces the distortion of the cascade, but also reduces its gain. Therefore, a lamp with a high gain is needed here. The author chose the ECC83 lamp (similar to 6N2P).
The current source in the cathode circuit is made active, on transistors, which also significantly improves the parameters of the cascade and allows simple methods to implement differential current adjustment. cascade. The final gain of the first stage is 29 dB.
To turn on the general feedback in the amplifier, it is necessary to close the jumper JP1. This will reduce the overall gain to 23 dB, but this is still enough to obtain a given output power.
Let me remind you that a deep general feedback improves the parameters of the amplifier, but, as tests show, worsens its subjective sound. The -6dB feedback depth is a good compromise in this case.
The disadvantage of using ECC83 tubes in the input stage is their high output impedance - about 50 kOhm. Coordination with the low-resistance transistor part is provided by a cathode follower on an ECC89 lamp (analogous to 6N23P) with an output impedance of about 500 ohms.
After much experimentation, the author chose the mode that provides the least distortion and makes it possible to match both tube stages directly, without an isolation capacitor. In addition, it provides a smooth increase in voltage (from 0 to 194 V) across the cathode resistor R7 when the amplifier is turned on, due to which the capacitors C2 and C3 are smoothly charged, which eliminates clicks and negative effects on the transistor part.
Separating capacitors.
The voltage amplification stage (tube part) and the current amplification stage (transistor part) are interconnected via coupling capacitors. You can't do without this in the circuit, because the voltage at the cathode of the ECC88 lamp is about 194 V. Unfortunately, these capacitors significantly affect the sound of the amplifier.
After conducting tests on listening to this amplifier, the author opted for capacitors ClarityCap SA series, which have a very good price/quality ratio. Due to its high operating voltage (600 V), the SA series is very well suited for use in lamp circuits.
The topology of the printed circuit board allows the use of high-quality capacitors from other manufacturers in the design, including Wima And solen. The value of 3.3 µF is chosen to provide a frequency response drop below 10 Hz. The isolation capacitor together with the input resistance of the transistor stage constitute a filter, the cutoff frequency of which can be determined by the formula:
1 / (2π* 3.3µF * 10kΩ)
The operating voltage of isolation capacitors must be at least 400V.
output stage.
The output stage of the amplifier is built on bipolar transistors. Of course, one could also use MOSFETs like BUZ900P or 2SK1058, but the author intentionally weeded them out. The selected transistors are quite often used in audio amplifiers and, with very good performance for audio applications, they have a very modest price and high reliability.
The output stage is quasi-complementary, i.e. built on transistors of the same conductivity in both arms. This configuration was widespread in the 70s and 80s due to the lack of available p-n-p complementary transistors. And, in general ... deserved a bad reputation. But! The author believes that there are no completely complementary transistors in principle, and therefore, using the same type of transistors, you can achieve a greater real shoulder symmetry cascade. The well-known company Naim uses only such an output stage configuration in its amplifiers.
The value of the supply voltage is 38 V, what is optimal for this output stage and allows for a 4-ohm or 8-ohm load to operate the amplifier without problems.
Learn more about schema elements.
Resistor R1 is the grid resistor of lamp V1a. Its value is not critical, but its presence is mandatory! Resistor R2, together with the input capacitance of the lamp, forms a low-pass filter to protect the amplifier input from interference. A similar role is played by the resistor R5 for the cathode follower.
The values of the resistors R3 and R4 are chosen to obtain a voltage slightly more than 190V on the anodes of the lamps. In this case, the current through each lamp is 0.8 mA. Current source for dif. The cascade is built on transistors Q6, Q7 to increase its internal resistance. The LED sets the reference voltage, and the P1 trimmer can conveniently and with high accuracy set the required source current. To power the current generator, a stabilizer on the LM337 chip is used.
If desired, a general negative feedback can be introduced into the circuit. Its depth depends on the values of the resistors R6 and R8. With the values indicated in the diagram, the depth of the LLC is 6 dB. To increase stability, a small capacitor (56pF) can be connected in parallel with R8. If you do not like experiments or are an ardent opponent of negative feedback, then the elements R6, R8, JP1, Cfb can be omitted. Even without overall feedback, this amplifier has very low distortion.
The quiescent current of the cathode follower lamp is chosen to be about 9 mA. To reduce distortion and the output impedance of the cascade, it is desirable to set this one more, but this may adversely affect the lamp life. The author made a compromise.
Transistor Q1 sets quiescent current transistor output stage. To ensure thermal stabilization, it must be fixed as close as possible to the output transistors on a common heatsink. Resistor P2 must be multi-turn and with a reliable contact of the engine.
Resistors R11, R16, P3 determine the input resistance of the transistor part of the amplifier (at the indicated ratings, it is about 10 kOhm). Using field effect transistors The values of these resistors can be significantly increased. Trim P3 is used to adjust the "0" output of the amplifier. The author deliberately did not use the integrator for these purposes, as he believes that it negatively affects the sound.
Elements R12 / C4 and R20 / C8 are additional power filters, and it is highly recommended not to exclude them from the circuit. The capacitances of capacitors C4 and C8 can be in the range of 220uF-330uF.
Transistors Q2 and Q4 form a classic composite Darlington transistor, which gives the required current gain. Transistors Q3 and Q5 form a composite Shiklai transistor, simulating a complementary PNP transistor. Since Q4 and Q5 are of the same type, according to the author, complementarity is achieved here more completely. To reduce the distortion of the Shiklai stage, a Baxandall diode is usually added to it. The author replaced it with a diode-connected transistor (marked Qbax in the diagram), which made it possible to further reduce the distortion of the output stage. The measured distortion at 1 W of output power with the diode was 0.22%, and with the 2SC1815 transistor turned on by the diode, only 0.08%. At higher output power levels, the difference between diode and transistor decreases. The printed circuit board allows you to install transistors of the 2SC1815 or 2SC2073 types or just a 1N4007 diode.
Due to the presence of local negative feedback, the output stage has low distortion and good thermal stability. Resistors R21 and R22 must be non-inductive and possibly smaller.
Elements R23 and C7 form a Zobel circuit to ensure the stability of the amplifier at frequencies above 100 kHz. Base resistors R13, R17, R14, and R18 also prevent possible excitation at high frequencies. With a capacitive load of this amplifier, to increase its stability, you can connect an inductance in series with the output (as is often done). The coil contains 16 turns of copper wire with a diameter of 0.75 mm, wound on a mandrel with a diameter of 6.3 mm or on a 15 ohm 2 W resistor.
Device diagram protection and turn-on delay speaker systems shown in the figure:
Zoom on click
It provides a delay in connecting speakers 30 seconds after turning on the amplifier and turning them off when a dangerous DC voltage appears at the output. To minimize the impact on the sound, the relay for this unit must be selected with reliable and high-quality contacts.
Power Supply
The high-voltage part of the circuit is powered by a stabilizer built on the TL783 chip. The input voltage should be about 360V. The chip is mounted on a small heatsink and is securely isolated from the case. The output voltage of 315V is set by the divider resistors R39/R40. Resistor R41 serves to discharge the capacitors after the amplifier is turned off.
R42/C27 and R43/C28 are additional filters for the left and right channels. After them, the output voltage of the power supply is 310V.
If you can't find a capacitor for C23 Wima FKP1(see specification) it is better to exclude it from the circuit!
Zoom on click
The secondary winding of transformer T1 with a voltage of 30V is used to power the AC protection device (not stabilized).
The filament voltage is connected to a common wire (to reduce the background) through capacitor. It cannot be directly connected to the “ground”, since the voltage at the cathode of the ECC88 lamp is 194V, which is more than the maximum allowable voltage of the cathode-grid. A capacitor easily solves this problem. Resistor R36 is selected experimentally so that the filament voltage is ~ 6.3V.
The output stage of the amplifier is powered by an unstabilized voltage of 38V. All transformers in the author's design are toroidal.
Design.
All amplifier blocks are assembled on printed circuit boards. Each channel of the amplifier is assembled on a separate board, so for the stereo version you will need two of them.
The author guarantees that you will get the best results if you use exactly the elements that are listed (see below). Meanwhile, nothing prevents replacing them with other similar ones - available or in the plan of the experiment.
Zoom on click
Amplifier printed circuit boards are designed to mount transistors on radiators or the amplifier base (which will serve as a radiator):
Zoom on click
All connecting wires must be of the appropriate cross section and as short as possible.
The photo shows the mounting option for the output transistors and the thermal stabilization transistor:
Zoom on click
Please note that all transistors are isolated from the case/heatsink. For best results, the author advises to first fix the transistors to the heatsinks, then bend their leads at a right angle, then insert the leads into the holes of the board and fix it. Soldering the leads should be the very last thing, when the transistors and the board are finally positioned relative to each other and fixed.
In the author's design, two large radiators are used as the side walls of the amplifier case, on which the printed circuit boards of each channel are fixed. In the central part there are toroidal power transformers, a power supply board and an AC protection board:
Zoom on click
To save space, the power supply board is fixed above the transformers:
Zoom on click
To reduce the level of background and interference, all "common" wires must be connected at one point, as shown in the diagram:
Zoom on click
Setting up the amplifier.
Before turning on, make sure that the transistors are securely isolated from the heatsink / case and from each other, the polarity of the electrolytic capacitors is not reversed, and the lamps are in place (they are not interchangeable!)
As noted above, the amplifier has three controls:
- P1 sets the operating current of the ECC83 lamp.
- P2 controls the quiescent current of the output transistors.
- P3 adjusts the DC voltage level at the output of the amplifier.
Before switching on, the P2 engine must be placed in the upper position according to the diagram(close to collector Q1). This will ensure the minimum quiescent current of the transistors after switching on.
Trimmer P1 must be set to approximately 800 ohms (set before soldering to the board).
After turning on the amplifier without applying an input signal and without connecting a load, adjust the voltage at the control point TP3 with the trimmer P1, which should be 1.6V. In this case, the voltage at the cathode V2a should be 195 V (± 5%). These tensions are interrelated. If any voltage is very different from those indicated, one of the lamps will have to be replaced.
Then trim P3 set zero voltage at the output of the amplifier. It can range from -50mV to +50mV. This is fine. After that, use the P2 trimmer to set the quiescent current of the amplifier in the region of 100-150 mA. To do this, you can control the voltage across the resistors R21 or R22, which should be in the range of 22 mV-33 mV.
After warming up the amplifier for half an hour, check the set values and, if necessary, correct them.
The amplifier uses a high operating voltage. Remember the safety precautions when working with electricity!!!
Conclusion.
Despite the absence of general negative feedback, the amplifier provides low signal distortion at low power levels and good damping factor, which is usually a problem for amplifiers without general feedback.
The amplifier has a great sound with good dynamics and high detail. He is especially careful with micro-details (low-level signals). At the same time, there is no pronounced lamp color in the sound.
MuGen embodied the best of both worlds - transistor dynamics and tube warmth of sound (within reason, without transistor harshness).
It should be noted that this amplifier is operated by the author already since 2007 and so far no other amp has surpassed it in terms of musicality!
Zoom on click
List of elements.
Amplifier and power supply
(For the stereo version, all details must be taken in double quantity)
Resistors
(1% metal film, 0.5W unless otherwise noted)
R1 = 392 kOhm
R2, R5, R12, R20, R32 = 1 kΩ
R3,R4 = 150kΩ 2W (BC PR02 series)
R6, R15, R19, R45 = 100 Ohm
R7 = 22kΩ 3W (BCPR03 series)
R8 = 2.43 kOhm
R9 = 274 Ohm
R10 = 560 Ohm
R11 = 18 kOhm
R13, R17 = 392 Ohm
R14,R18 = 2.2 ohm
R16 = 20 kOhm
R21,R22 = 0.22 Ohm 4W (Intertechnik MOX)
R23 = 10 ohm 2W
R24,R26 = 182 Ohm
R25 = 1.5 kOhm
R27 = 3.3 kOhm
R28,R29 = 1 MΩ
R30 = 330 kOhm
R31 = 10 MΩ
R33, R34, R35 = 100 kΩ
R36 = selectable (approximately 0.22 ohm)
R37,R38 = 100 Ohm 1W
R39 = 330 Ohm
R40 = 82 kOhm 3W
R41 = 150 kOhm 3W
R42,R43 = 1kΩ 1W
R44 = 4.7 ohm
P1 = 2 kΩ, multi-turn
P2,P3 = 5 kΩ, multi-turn
Capacitors:
C1=100nF 400VDC
C2,C3 = 3.3uF 400VDC (ClarityCap SA 630V audiophile quality)
C4,C6,C8,C10 = 270uF 50V (Panasonic FC)
C5,C9,C12,C14,C22 = 100nF 50V
C7 = 100nF (Vishay MKP-1834)
C11,C16,C17 = 10uF 50V
C13 = 47uF 50V
C15 = 1uF 250V (Wima type)
C18 = 22uF 63V
C19,C20 = 47uF 25V
C21 = 220uF 50V
C23 = 2n2 (Wima FKP-1/700 VAC)
C29,C30,C31,C35 = 2n2 (Wima FKP-1/700 VAC)
C24 = 150uF 450V
C25 = 100n 450 VDC
C26 = 10uF 400V
C27,C28 = 22uF 400V
C32,C33,C34,C36,C37,C38 = 4700uF63V (BC056, 30x40mm, Conrad Electronics)
C39 = 10uF 25V
Cfb = 56pF (optional)
Active elements:
D2,D3 = UF4007 (if not available, you can supply - 1N4007)
D4,D5 = 1N4001
D6,D7,D8 = 1N4148
D9,D10,D11,D12=BY228
D13=1N4007
LED1 = LED, 5mm, red LED
Z1 = Zener diode 110V 1.3W
Q1=BD139
Q2=2SC2073
Q3=2SA940
Q4,Q5 = 2SC5200
Q6,Q7=BC550B
Q8=BS170
Q9,Q10=BC547B
Qbax = 2SC1815BL
U1=LM337
U2=LM317
U3=TL783
Lamps:
V1 = ECC83 (pref. JJ Electronics), 6N2P
V2 = ECC88 (pref. JJ Electronics), 6N23P
Miscellaneous:
B1 = bridge rectifier 600 V, 1A (DF06M)
B2,B3 = bridge rectifier 400V, 35A
T1 = transformer with secondary voltages: 30V + 250V +6.3V (Amplimo type 3N604)
T2 = transformer with secondary voltages: 2×28 VAC, 300VA (Amplimo type 78057)
RLY1 = 24V relay (e.g. Amplimo type LR)
Radiators U3 Fischer SK104 25,4 STC-220 14K/W
Radiators U1 and U2, FischerFK137 SA 220, 21K/W
Heatsinks for Q4 and Q5, with a thermal resistance of 0.7K/W or better.
9-pin panel for lamps - 2 pcs.
PCB drawings (original in pdf format) download .(rar-archive, 186 kb)
Latest version of PCB drawings in Sprint layout format from our readers (the editors of "Radio Gazeta" HAVE NOT CHECKED!) download (rar-archive 117 kb).
The article was prepared based on the materials of the Elector magazine.
Free translation - editor-in-chief of Radio Gazeta.
Successful creativity!
, which use lamps and field-effect transistors, have a number of advantages over purely microcircuit or transistor ULF. Here, due to the use of a radio tube, excellent signal amplification is achieved, which can be immediately applied to the buildup of powerful output field-effect transistors. Thus, we have only 2 stages of amplification, while if we take a preamplifier with an opamp, then the sound inside the microcircuit will go through a dozen stages, filled with albeit small, but distortions. So let me introduce you, dear visitors of the site "", the project of a new hybrid UMZCH.Amplifier
Schematic diagram of a hybrid amplifier
The first stage of the amplifier is built on a double triode according to the schemeSRPPin order to reduce its own non-linearity and increase the load capacity. The lower half of the lamp according to the scheme is engaged in signal amplification, and the upper one plays the role of a dynamic load. The positive features of this inclusion are the high gain and low output impedance of the cascade. The anode voltage is chosen to be 150-180 V.
The single-cycle output stage is built on a field-effect transistor according to the scheme of a powerful source follower, loaded on a stable current generator on a transistor of the same structure. The input signal from the lamp through the interstage capacitor is fed into the input circuit of the amplifying transistor, providing good performance, especially considering that the amplifier is not covered by feedback.
The integrator is assembled on an operational amplifierOPA134(can be appliedk140ud6), which ensures automatic holding of zero potential at the output of the amplifier. In addition, the integrator has an equivalent cutoff frequency of 3 Hz at infra-low frequencies, which favorably affects the damping of acoustic systems.
Power Supply
PSU circuit of a hybrid amplifier
Power supply - 300 W toroidal transformer removed from a halogen chandelier, with two windings of 12 V 6.5 A each, to which 4 turns of the same wire are wound, an anode 140 V 200 mA is wound and an incandescent winding 6.36 V 0.7 A I selected the number of turns experimentally, having wound 10 turns, I measured the voltage by counting how many turns per Volt are needed. As an interlayer insulation, I used a FUM fluoroplastic tape, fixed with adhesive tape. Diode bridges are taken with a margin for charging rectifier capacitors. Hanging installation. Everything fit in the case of a computer PSU.
An electronic filter with capacitance multiplication is located on the ULF main board. The capacity in the base is multiplied byh21ecomposite transistor. Due to the fact that the transistors are bipolar, it was necessary to fence a multi-link filtering RC chain into the base. The 12 kΩ resistor shunting the last capacitor of this chain sets the voltage drop across the transistor, preventing it from saturating. When power is applied, the voltage gradually increases as the base capacitors charge, thereby ensuring a smooth exit to the entire ULF mode. Transistors can be replaced with compositeTIP142 And TIP147which already have a diode on board, this will simplify the circuit a little.
Speaker protection
AC amplifier protection circuit
Turn-on delay and DC current protection AC - Universal speaker protection against DC voltage, clicks and surges when power is turned on, infra-low frequencies. When a constant voltage of any polarity of more than 1.5 V appears at the ULF output, the corresponding key opens, which closes the field-effect transistor and the relay opens the AC circuit. The protection provides a delay in connecting the speakers when the power is turned on (for a period of 5 s), thereby preventing the penetration of noise into the speaker system caused by transients in the amplifier.
Amplifier setting
The setting is reduced to setting the zero potential with the integrator correction trimmer with the lamps removed and setting the required maximum current of 2.5 A in the current-setting circuit with the least resistance attenuator position.
The attenuator is designed on a biscuit switch10P4N, consists of 4 biscuits for 11 positions with a wide blade of the rotor for seamless switching of the signal. One of the important parameters of volume controls is the matching of the right and left channel sections, since this phenomenon determines the spatial characteristics of the entire amplifier. The volume control consists of a multi-link divider of eleven resistors.
Resistors were selected by selecting the same resistance from a dozen for each rating. The other two switch jacks are used to change the current of the output transistors in relation to the output amplitude of the volume attenuator. The divider is wound with nichrome with a diameter of 0.5 mm, twisted in two on a mandrel of 6 mm. and soldered to the contacts of the biscuit every third turn so that between adjacent contacts the resistance is 0.15 ohms.
Technological features
Electronic filter, amplifier, acoustics protection - made on the board of one-sided textolite by laser-ironing method. The board is made in separate blocks for setting up and testing each circuit individually. Field-effect transistors are soldered from the side of the tracks and pressed through mica spacers with an aluminum plate to the body of 4 m2 radiators, at full power they heat up to 60-75 0s.
When power is applied, the lamp, warming up through the interstage, affects the field-effect transistor, thereby shifting zero, the protection circuit does not connect the speakers until it enters the lamp mode for 30-40 seconds. If the attenuator knob is turned sharply, the action will cause the appearance of infra-low frequencies and the protection will work for 5 seconds. With a significant increase in current with a trimmer resistor, distortion may occur at high frequencies, it is better to reduce the current resistor to 0.25 - 0.28 Ohm. The interstage capacitor must be chosen according to the best so as not to spoil the whole sound picture.
ULF turned out great, I've been listening to it for more than a month. The output stage exactly repeats the amplified input signal from 10 Hz to almost 1 MHz, it was checked by a generator without lamps with signals - meander, saw, sine wave, complex noise. The lamp gives the sound a characteristic coloration, which in turn is well perceived by the listener. The classic meander roll-off at 50,000 Hz is a characteristic feature of lamp-based cascades. This is a soft and imperceptible limitation of the amplitude of the sound signal on dynamic bursts, a kind of improver.
P.S.Anyone who wants accurate signal transmission can remove the lamp, and use the integrator as a preamplifier by adding two resistors to the circuit, but then you need high-quality op-amps, and they are expensive.All the necessary files and datasheets for the used radioelements are in the general archive.
There is also an idea to try a current generator on an opamp by tying it with a dual resistor with a volume control in order to smoothly regulate the current depending on the power. In general, the project has room for development, O. Senenko was with you.
The published material is intended for a wide range of radio amateurs who do not have a special technical education, complex metalwork tools and experience in building such structures, therefore, some issues may be covered in too much detail in someone's opinion.
It should be immediately noted for criticism that in this article the author expresses only his own vision of solving this issue and, therefore, the material presented does not claim originality and indisputability both in judgments and circuit solutions, and in the practical implementation of the designs of the amplifiers themselves, and their individual components. .
The main objectives of this publication:
- obtaining a universal amplifier design that allows it to be assembled by a radio amateur who does not have much experience in building such devices and is not highly qualified;
- to enable radio amateurs, without serious alterations in the design, to experiment with individual nodes, to use (be able to replace) the most common domestic medium-power generator lamps in the amplifier circuit;
- the use in the design of the power amplifier of the maximum number of widely available factory-made widely used parts;
- the possibility of using in the manufacture of the amplifier a minimum of complex metalwork and turning equipment, as well as service equipment and measuring instruments when setting it up.
The amplifiers described in the article were operated with various types of transceivers: UW3DI-2; RA3AO; Ether-M; The wave, UA1FA (transmitter attachment), used ONWA and LINCOLN to drive on CB and 10m. In all cases, the quality of the output signal was uniquely determined by the signal quality of the transceiver used.
MAIN TECHNICAL CHARACTERISTICS OF AMPLIFIERS
The amplifier circuits used lamps GI-7B, GI-7BT, GI-6B (2 pcs.), GU-72 (2 pcs.), GMI-11, GU-74B, 6P45S, 6P42S, 6P36S (4 pcs.), GU-50 (3 and 4 pcs.), G-807 (4 pcs.), GK-71. The amplifiers operate in class AB1 (in SSB mode) and class C (in CW mode).
Operating frequency range ………………………………………………..1.8 - 28.7 MHz
Type of radiation …………………………………………………………………..SSB, CW, RTTY
Power supplied to the anode circuit for a long time in the mode
"clicks" ………………………………………………………..…………650 W max.
(depends on the excitation power and is limited by the power of the anode voltage source);
load power * in the frequency band 1.8 - 28.7 MHz …………….…….300-350 W
(depending on the efficiency of the output circuit in this range);
input (output) impedance of the amplifier ………………………… 75 (50) Ohm;
power consumed by the amplifier from the network in the "pressing" mode ... .. 700 W max;
in the "silence" mode ……………………………………………….…….130 W;
in receive mode ………………………………………………………………60 W.
overall dimensions of the amplifier (without legs) mm ……………………..352 ´ 153 ´ 350;
Amplifier mass ……………………………………………………… about 25(13)** .kg.
*- This refers to the guaranteed output power, i.e. power obtained at the nominal values of currents and voltages of transformers and 70% duty cycle, the output power of individual instances reaches 500 watts.
**- For circuits with transformerless power supply.
POWER AMPLIFIER DVA-300
Power amplifier uses one GU-74B, GMI-11 tube, two GI-7B (GI-7BT, GI-6B), GU-72 tubes, three or four GU-50, four 6P45S (6P42S, 6P36S), four G- 807 tubes, four G-811 tubes, GK-71. The PA covers 160-10 m (also all WARC). It requires 10-40 watts to produce full power. The PA uses AB class (SSB), C class (CW) grounded cathode circuit. The AC power supply is built-in and can be set for 220/230 VAC
Frequency range …………………………………………..……………….1.8-28 MHz Modes……………………………………………… …………………SSB, CW, RTTY
Input Power……………………………………………..………………650 watts max.
Power Output……………………………………………..……..……..300-350 watts
DriverPower………………..………………………………………………5-40 watts
Efficiency………………………………………………………………..……55-65%
Input/Output impedances……………………………………..……………75(50) Om
Plate Voltage……………………………………………….……………….1300 volts
Harmonics…………………………………………….…………………35 dB typical
Front panel status indicators……………….……..………standby, operate, transmit
Metering………………………………………………………………….……..Ig, Iout
Primary Power……………………………………………………..220/230VAC, 3A
Dimensions………………………………………………. …….350 x 150 x 350 mm
Weight ……………………………………………….…………………… …….25 Kg
In all the diagrams and assembly drawings below, the numbering of elements and parts that perform the same purpose is preserved from diagram (drawing) to diagram (drawing). If the drawing does not contain any next element number or size in the drawing, this means that it was in the previous diagram (drawing) and, accordingly, newly appearing elements have a number that has not been seen before.
one . POWER AMPLIFIER POWER SUPPLY.
Schematic diagram of the power supply (hereinafter PSU) is shown in Fig.1. PSUs for all amplifier options (with the exception of transformerless ones) are assembled according to the anode voltage doubling scheme, which is mainly due to the type of transformers used to obtain the anode voltage (the so-called Latour scheme). The doubling scheme can only work for a capacitive load, the rectified voltage ripple frequency is doubled above the mains frequency. In terms of its energy characteristics, this circuit is not inferior to a bridge circuit operating on capacitance.
The anode voltage rectifier is made on four 210 V KD diodes. In practice, it is customary to use one diode for each arm of the doubling circuit for every thousand volts of rectified voltage, so they are connected in series, two in each arm. This type of diodes allows their use in series connection without shunting resistors. When using older types of diodes, they must be connected in parallel with resistors to evenly distribute the reverse voltage (based on 750-1000 kOhm per 1000 V voltage) and shunt them with capacitors with a capacity of 0.01-0.05 μF to protect against the possibility of electrical (not thermal) breakdown by short-term pulses that occur in circuits for various reasons.
As shown by the three-year practice of operating amplifiers (according to the above diagrams, several variants of such amplifiers were made on various lamps), it is absolutely safe to use a rectifier with voltage doubling and electrolytic capacitors as a capacitive load in amplifiers, and the signal quality practically depends only on the signal quality of the transmitter used . The overall power of the power transformer can be only 10-15% more than the power supplied to the final stage. In addition, at the same time, its secondary winding has half the number of turns, and the wire cross section, on the contrary, increases, which facilitates the winding of the transformer.
The value of the anode voltage was chosen based not only on the type of transformers used, but also taking into account the obtaining of a larger value of the equivalent resistance of the anode load (Re = Ua / 2Ia), since with a small Re the lamps operate with large anode currents (little Ua), as a result of which due to the increase in the required drive power, both the efficiency of the cascade and the service life of the lamps are reduced.
Considering the use in amplifiers for turning on lamps according to common-cathode circuits, the source also provides a full set of other voltages: necessary for the operation of the amplifier: screen and control grid voltages, filament voltage and service voltages necessary to power the automation circuits and signal circuits.
Insignificant differences exist only in the power supply circuit of filament circuits, it is performed depending on the filament voltage of a particular lamp, while using various filament transformers. The PSU uses only industrial-made transformers that have passed state tests in extreme operating conditions and provide the possibility of continuous round-the-clock operation at rated voltages and currents in harsh climatic conditions, thereby increasing the reliability of the amplifier operation. And given that the average power of the amplifier when operating in SSB mode is about 30% of the peak power, and the duration of the full power peaks is quite short, it is possible to get a large output power from the amplifier.
It should be noted that if you are going to use the amplifier to work with digital modes or FM (i.e. during operation, constant carrier radiation is assumed), then in this case, firstly, the anode voltage may drop directly to the value (effective value) of the voltage the output winding of the transformer, which leads to distortion of the output signal, and secondly, to overheating and, accordingly, failure of the output stage lamp itself. Therefore, in such cases, the output power must be reduced. In addition, the network windings of these transformers contain taps that allow the use of transformers with increased or decreased supply voltage, which is especially important for rural areas. And the presence of taps in the secondary windings allows them to vary the value of the anode voltage over a wide range. Options for replacing anode transformers are shown in Table 1. All of the above, in no way excludes your initiative to manufacture transformers on your own in the absence of the possibility of purchasing factory copies. Just in their manufacture, you must consider the following:
Firstly, the high-voltage winding must be reliably isolated from all other windings (it is best to wind it last).
Secondly, the transformer must be reliably impregnated with varnish. Our "shacks" are often not the most ideal place in an apartment (if in an apartment!) And an increase in air humidity is often the cause of breakdown of the windings.
The very procedure for calculating transformers on iron of the PL brand is not given here, since it has been repeatedly described in various literature, for example, see.
The power supply of the amplifier provides the following output parameters:
anode voltage …………………………………………….1330 (1500) V / 500 mA;
stabilized screen grid voltage ………………………300 V / 50 mA;
stabilized control grid voltage …………………100 V / 50 mA;
glow voltage (variable) ……………………..…26 V / 2.1 A (12.6 V / 7.0 A);
relay supply voltage ……………………………………….………..24 V / 700 mA;
signal lamp supply voltage (variable) …………………6.3 V / 700 mA.
NOTE:
1. The capacitors used in the voltage multiplication circuit must have the same leakage voltage.
3. Since when using GI-7B lamps in circuits with a common grid, there is no need for a separate source of bias voltage, the anode voltage can be increased to 1500 volts by using additional windings 15-19 and 21 - 22 of Tr transformers connected in series for this purpose. .1 and Tr.2. In this case, capacitors C1-C8 of type K50-20 must be changed to K50-7 or similar, designed for an operating voltage of 450 V. It is even better to use imported capacitors, for example, Samsung, which do not require selection, although their cost is three times higher .
4. For a symmetrical power supply of the filaments of the lamps, the winding of the transformer Tr.3, from which the filament voltage is taken, if possible, is best done with a midpoint that must be planted on the circuit ground, as shown in Fig. 1A.2 (this applies to all described schemes). If the winding does not have a midpoint, it is easy to obtain it using diodes, as shown in Fig. 1A.3 heating voltage. Almost all modern high-power diodes meet these requirements.
The PSU is turned on by pressing the S1 "ON" button. In this case, power is supplied only to the network winding of the incandescent transformer Tr3. From the same transformer, voltage is obtained for the rectifiers that feed the control grid circuits, signal lights, relays and the fan. The use of a separate transformer makes it possible, firstly, to turn on the anode supply voltage only when the filament voltage is present and the lamps are warming up; turning off the high voltage during long-term operation of the radio station only for reception.
All amplifiers are equipped with fans for blowing lamps. This can come in handy during the hot season, when working in competitions, as well as when working with RTTY, PACKET, etc. The rectifier circuit for powering the fan is assembled on VD15, VD16 and C13, C14. In order for the voltage on the fan under load to be 12 V, the capacitance of capacitors C13, C14 must be 470 uF each.
The cooling fan is switched on either at the same time as the lamp filament voltage is turned on, or independently by pressing the S1 “VENT” button. In the variants of the amplifier circuit on lamps that operate only with forced cooling, a VVF 71M type fan is used, which has relatively small dimensions and sufficient performance - 45 cubic meters. meters of air per hour. The passport for metal-glass and metal-ceramic lamps says that cooling should be applied to the lamps before the filament voltage is turned on and stop no earlier than three minutes after the filament voltage is turned off. Therefore, the fan is turned on automatically when the S2 "NAC" button is turned on, and when the filament voltage is turned off, if necessary, the fan can be left on (for lamps operating with forced cooling) by pressing the S1 "FAN" button. For the convenience of work, those who wish can put a thermal relay (for example, RB 5-2) in parallel with the fan on button, and then the fan will automatically turn on when the temperature reaches 60 degrees. For long-term and noiseless operation, the fan must be serviced periodically: cleaned monthly and lubricated with disassembly once every six months (of course, if lubrication is provided for by the specifications for the fan).
To obtain the anode, screen and bias voltage, two transformers TA262-127 / 220-50 Tr1 and Tr2 are used, the secondary windings of both transformers are connected in series. When the button S3 "ANODE" is pressed, the relay K1 is activated, which, with its contacts, connects the primary windings of the transformers to the network (through the fuses FU1 and FU2).
Resistors R1 and R2 serve to limit the surge in the charge current of capacitors C1 - C8 when the amplifier is turned on, their value is 3 - 10 ohms. In circuits with transformer powered anode, the EMF of self-induction of the secondary windings of anode transformers prevents the current surge when the power is turned on, so the value of R1 and R2 is chosen equal to 3 - 4 ohms. In the case of a power supply for the anode circuits of the amplifier according to a transformerless circuit, the load of the rectifier becomes purely capacitive. At the same time, the starting current increases significantly and with R1 and R2 ratings equal to 3 - 4 Ohms, when the source is turned on, their conductive layer instantly evaporates, while the resistors themselves do not even have time to darken from heating. In this case, the value of the resistors must be increased to 560 - 1200 Ohms, and in order to exclude a voltage drop on them in operating mode, it is necessary to add a starting circuit assembled on R26, C28, K1A, which, after charging C1 - C8, shorts R1 and R2 (in Fig. 1 it is indicated by a dotted line). The value of R26, on which the turn-on time of K1A depends, is selected during setup.
Resistors R3 - R6, on the contrary, serve to discharge C1 - C8 when the anode voltage is turned off. On resistors R9 - R13, a voltage drop occurs to the stabilization voltage of the zener diodes VD11 - VD13 included in the screen grid circuit. The value of the resistors is selected based on the stabilization current VD11 - VD13.
Resistors RSH1 and RSH2 are designed to measure the current of the anode and screen grid, respectively. The resistance of the resistors depends on the type of devices used. So, for devices of the M2001 type with a total deviation current of 1.0 mA, their resistances are 0.28 (0.14) and 2.8 Ohms, respectively, while their scales will correspond to 500 (1.0 A) and 50 mA. In the basic design, measurement screen grid current is not provided, tk. this requires additional switching of the device and the resistor RSH2 is "an amateur".
* - When using transformers of these types, the rectifier circuit is carried out according to the bridge circuit without doubling the voltage, the diodes used in this case must be designed for the corresponding voltage.
** - The resistance is selected until the normal stabilization current VD11-VD13 is obtained.
*** - When using transformers of these types, it is necessary to increase the width of the PSU chassis to 160 mm and adjust the location of the holes for mounting the transformers, adjust the layout and length of the harness wires, and also lengthen the det. 12 - det.13 up to 160 mm. Accordingly, the dimensions of the case change.
The bias voltage source of the control grid of the lamp is also made according to the voltage doubling circuit on the diodes VD5, VD6 and capacitors C10, C11, then the bias voltage is stabilized by the zener diode VD14. Variable resistors R22 and R23 are designed to set the quiescent current of the lamps in SSB and CW mode, respectively. The exact value of the operating point voltage is set by the minimum of out-of-band emissions. This should be remembered when replacing lamps, the value of the quiescent current of the new lamp should be set equal based on the above condition. The choice of the mode is made by the switch S4 "SSB-CW".
Electrolytic capacitors of the K50-20 brand were used to smooth the ripples of the anode voltage. It is often written in the literature that their use is undesirable due to the severe thermal conditions inside the amplifier case, and numerous arguments are given. However, twenty years of personal experience in servicing computers of the Minsk-32, ES-1022 and ES-1045 types, operating around the clock for months without turning off the power, proved that they behave very reliably. The only thing these capacitors do not like is long periods of inactivity without voltage supply. So, if when you first turn on the amplifier or when you turn it on after a long downtime (three months or more), you will be pointed to the "background", do not worry - a couple of days of work on the air and everything will fall into place. In addition, the capacitors are separated by a partition from the installation site of the lamps and practically do not heat up. In general, before installation in the circuit, in order to avoid lumbago, it is best to mold the capacitors, if only because they can be caught in the 80s or even 70s of release. This is done either before inserting them into the circuit using a simple circuit, or directly in the circuit (see Chapter 5).
In the XP2 socket, if necessary, you can turn on the transceiver, or some auxiliary device.
The + 24V (+ 12V) voltage is output to the XP8 connector, which can be used to power the antenna switch or, for example, an electronic key.
In the PSU for an amplifier for 2 GI-7B, two capacitors (C12, C15) are used to obtain the service voltage, this is done in case, for example, you do not get the desired transformer from the TN series, and you come across a transformer that has different current filament windings, for example TN-56. When using it, to obtain the required filament current, it will be necessary to combine windings. To obtain service voltage, you can also easily switch to a doubling circuit using only one 6.3V winding, as shown in Fig. 2A1 (this also applies to other circuits).
2. GENERAL DESCRIPTION OF THE AMPLIFIER CIRCUIT
For the construction of amplifiers, those generator or modulating tubes are best suited, in which the anode terminal is located separately from the other terminals and is located on top. With this design of the lamp, when mounting the amplifier, it is easier to separate the anode, grid and filament circuits from each other, which will reduce the likelihood of their mutual influence, and, accordingly, the tendency of the amplifier to self-excite when turned on.
A schematic diagram of the high-frequency part of the power amplifier is shown in Fig.2. The circuit of the anode part of the base amplifier is common for all variants and is made according to the parallel power supply circuit. The anode circuit is a traditional P-circuit, consisting of range coils L4 and L5, an anode capacitor C20, a coupling capacitor with an antenna C21. The only feature of the scheme is the inclusion of the receiver antenna in the “hot” end of the P-loop of the output stage, which gave additional signal selectivity at the reception. With this inclusion, it became possible to configure the circuit for transmission in the “cold” mode. This eliminated the overvoltage mode of the amplifier with a detuned circuit in the tuning mode, since the tuning operation is performed purely in the receive mode without applying high voltage and emitting a signal on the air, and the settings in the receive and transmit modes practically coincide, a slight difference is observed only on the range 10 meters. To reduce the "noise" of the lamp during reception due to the residual current through it (otherwise the zener diodes will not work, because at low currents they simply will not enter the stabilization mode), the negative voltage blocking the lamp is chosen large enough.
Experiments carried out in the design of amplifiers showed that with a normalized input resistance of the receiver (Rin). equal to 75 (50) ohms, the value of the capacitance of the coupling capacitor C19, included in the "hot" end of the circuit, must be at least 15 pF. Otherwise, the signal at the input of the receiving path will have a large attenuation, however, in this case, the capacitance value of the capacitor in the range of 10 meters becomes commensurate with the capacitance value of the anode capacitor C20, which leads to some difference in the settings. In addition, the total capacitance of these capacitors becomes already significant for a range of 10 meters, and therefore it may be difficult to tune the circuit for transmission, since C19 is connected in parallel with capacitor C20 during transmission, so the latter should have as little initial capacitance as possible (with the exception of option E).
The use of broadband transformers (SHPT) at the input of the amplifier, which doubled the input excitation voltage (for circuits with a common cathode), had to be abandoned. Numerous experiments were carried out, rings with a permeability from 1000 to 20 HF were used, the number of turns of the windings and the pitch of the wire were changed, a series-connected circuit was used to compensate for the blockage of the characteristics at the HF, the switching circuit of the SHPT windings was changed, and still approximately the same results were obtained. . Yes, it works great as a resistance transformer over the entire range, but at frequencies above 11 MHz, the signal amplitude began to fall, and at 28 MHz its level was two times lower than the input signal level, and given the decrease in the gain of the circuits themselves with OK with increasing frequency , got the corresponding result. Thus, it turned out that it is impossible to cover the band of almost 28 MHz with one FFS, which was to be expected, but using several FFSs - we get the same input band circuits. But the use of input range circuits at the input of the amplifier immediately significantly complicates and increases the cost of its design. This also leads to the complication of the switching circuit, so in this case it is necessary to use additional relays for switching the input circuits, or they need to be mechanically connected to the output P-circuit switch, which ultimately leads to difficulty in repeating the circuit by unskilled radio amateurs. Although, of course, an unskilled radio amateur with the first category is a paradox, but still. Naturally, if you wish, you can use both options (at the same time and independently check all of the above). Diagrams of possible options for connecting the FFS at the input of the amplifier are shown in Fig. 2C2
If you are still going to put band circuits at the input of the amplifier, or you plan to use transceivers like RA3AO, URAL-84 or similar ones together with the amplifier, which contain low-power broadband amplifiers (up to 5 W) and whose power is not enough to build up a powerful output stage, and it is not possible to build an additional stage due to lack of space in the transceiver case; in this case, band-pass filters can be installed at the amplifier input. It is best to use inductively coupled circuits for this purpose, which, firstly, provide galvanic isolation (which is a prerequisite for transformerless circuits) and, secondly, provide good range filtering. The diagram of the input part of the amplifier with such filters is shown in Fig. 2.22 (for circuits with OK) -2.24 (for circuits with OS), and the drawing of the universal board is in Fig. 13H.
In the basic version of the amplifier circuit, there is no “BYPASS” mode, since the amplifier was not intended for local communications. In addition, firstly, for local QSOs there is a telephone, CB and 144 MHz, secondly, at present, even almost all of our "home made" are equipped with output power controls and, thirdly, if the dynamics of your correspondent's Radio is home does not allow him to listen to you, you can talk to him just sitting on a bench in the yard (saving "at the same time QSO" on electricity for communication with DX).
If you still want to have the “BYPASS” mode in the amplifier, it is necessary to make changes to the RF circuit of the amplifier according to Fig. 2.1 and Fig. 2.2, while the K5 relay is additionally installed on the PSU chassis (Fig. 11), and on the PSU front panel - relay K3 using a bracket poz.106. In this case, holes are additionally drilled in both the front panel of the amplifier and in the false panel for the S6 button - “BYPASS”, corresponding changes are made in the wiring diagram.
If you do not provide for the use of the P-circuit of the amplifier in the receiving path, the antenna is switched according to the recommendations given in Fig. 2.3. In this case, the relay K3 is installed on the front wall of the power supply unit using the bracket poz.106, there is no need for a capacitor C19, and holes for mounting K3 are not drilled on the partition of the RF unit. Accordingly, changes are made in the layout of the harness.
If a transceiver is used to work with an amplifier, in which the antenna is switched from reception to transmission directly in the transceiver itself, it is necessary to make changes to the circuit diagram of the RF part of the amplifier according to Fig. 2.4 and Fig. 2.5. In this case, the XP1 connector on the rear panel of the amplifier (Fig. 4) is not installed and, accordingly, a hole for it is not drilled. Relay K5 is additionally installed on the PSU chassis (Fig. 11) and appropriate changes are made to the wiring diagram. If the P-circuit of the amplifier is not used at the reception, changes are made to the circuit according to Fig. 2.5 and Fig. 2.6, and if the “BYPASS” mode is also required, according to Fig. 2.6 and Fig. 2.2.
When using the amplifier both in conjunction with a transceiver with internal antenna switching, and with a transceiver with separate sockets for receiving and transmitting antennas, the RF part of the amplifier is made according to Fig. 2.7 and Fig. 2.8. Relay K5 is installed on the chassis of the PSU, relay K3 is installed on the front panel of the PSU, a hole is drilled on the rear panel, dia. 8mm to install switch S7 “2 – 3”. Appropriate changes are made to the wiring diagram.
If in this variant the P-circuit of the amplifier is not used at the reception, the HF part of the amplifier is performed according to Fig. 2.8 and Fig. 2.9, if the “BYPASS” mode is also required, then the K6 relay is additionally installed on the PSU chassis, button S6 "BYPASS". Installation in this case is carried out according to Fig. 2.10 and Fig. 2.11.
All amplifiers are equipped with built-in devices that allow monitoring the state of the antenna-feeder economy (SWR meter) during operation, as well as approximately measuring the power at the amplifier output. For this purpose, a ready-made and well-proven scheme by V.A. Skrypnik, given in the book “Instruments for monitoring and adjusting amateur radio equipment”, only unlike the author, for the convenience of using the device, two arrow indicators are used at once. The first of them shows the level of the incident wave, and the second one can immediately evaluate the SWR readings of the antenna-feeder system. The SWR meter is turned on by pressing the S5 button.
Now I would like to separately highlight the issue of using lamps, especially the old years of production. Again, there is an opinion that old lamps that have lain in warehouses for ten or more years cannot be used in powerful cascades operating at high voltages, because. breakdown or discharge inside the lamp is possible due to partial loss of vacuum due to old age. This opinion is especially readily supported by resellers of lamps (for well-known reasons). Indeed, during long-term storage of lamps, their parts and the shell can emit a certain amount of gas. At the same time, the vacuum necessary for stable operation and ensuring stable parameters of the lamps inevitably deteriorates. However, in most cases it is possible to improve the vacuum inside the lamp and make it quite suitable for work by special training of the lamp. Therefore, when the lamp is turned on for the first time after long-term storage, as well as after being inoperative for more than six months, the lamp must be subjected to training, which is commonly called "hardening".
In the presence of a spark leak detector, the vacuum test can be carried out as follows: a conductor with a high-frequency potential from a spark leak detector touches one of the electrodes of a lamp or a glass container and observe the nature of the glow. To avoid breakdown, do not touch the glass in one place for more than 2-3 seconds. Also, avoid getting sparks on the junctions of metal with glass.
The degree of vacuum is determined by the following criteria:
a) no glow or a weak surface glow (glass fluorescence) of green or blue indicates the presence of a high vacuum;
c) volumetric glow of blue gas indicates that the lamp is "gas". Such a lamp, before being included in the working circuit, must first be subjected to "hardening";
c) volumetric intense glow of pink gas indicates that air is penetrating the lamp;
d) If a spark jumps between the electrodes inside the lamp, this indicates that the lamp has full atmospheric pressure.
The hardening of the lamp can be done either directly in the amplifier in which the lamp will work, or in a special installation, if any.
Keep the lamp at normal filament voltage (without other supply voltages) for 20-30 minutes.
Turn on the negative grid voltage.
Turn on the anode voltage, not exceeding half the nominal value, hold for 5-10 minutes and then increase it in steps through 150 - 200 V to the nominal value, keeping at each step 5-10 minutes. When approaching the nominal voltage value, the exposure time at each stage should be slightly increased (up to 15-20 minutes).
If a discharge occurs in the lamp when the voltage is increased, the voltage should be reduced by one step, hold for 10-15 minutes and then again increase the voltage in steps to normal. The absence of breakdowns indicates that the vacuum in the lamp has increased.
To protect the lamp from damage in the event of a breakdown in the anode circuit during hardening, it is necessary to include a resistance 3-5 times greater than the usual limiting resistance, which is switched on during normal operation of the lamp. At the end of hardening, in the absence of discharges, the resistance value should be reduced to the nominal value.
When increasing the voltage, it is necessary to ensure that the power dissipated by the electrodes does not exceed the maximum allowable values. The anode current can be adjusted by changing the grid bias voltage.
After the anode voltage is brought to the nominal operating value and there are no discharges or any abnormalities in the lamp operation for 20-30 minutes, it is recommended to increase the anode voltage by 5-10% above the nominal value and hold for 10-15 minutes. After that, in the absence of discharges, the lamp can be turned on.
Stiffening can also be done dynamically. In this case, the lamp is turned on at reduced supply voltages and, after holding for 5-10 minutes, the voltage and load slowly increase in steps to normal values.
Turning on the full voltage of the anode must be done with the circuit configured. Otherwise, the lamp may fail due to breakdown. If the lamp at full adjustment after long-term storage does not give off sufficient power, a short-term (no more than 5 minutes) increase in the filament voltage above the nominal voltage by 15% is allowed.
In any case, for long-term and trouble-free operation, new lamps must be subjected to training. When you turn on a new lamp for the first time or after a long break in operation (more than 10 days), the following procedure for preparing the lamp for normal operation is recommended: the glow is turned on; at normal filament voltage (without other electrode voltages), the lamp is kept for 15-20 minutes. After that, you can turn on the voltage of the anode and grids. It is desirable to keep the lamps for 5-6 hours in transmission mode in the absence of an excitation signal.
NOTE:
- The inclusion of any voltage of the electrodes should be made only after the voltage and filament current have reached the nominal values.
- During lamp operation, the filament voltage must be constant and must not exceed the nominal value. Even a slight increase in filament voltage can significantly shorten lamp life.
- The output power and slope of the characteristics of the lamps can decrease by the end of their service life up to 20% of the lower limit of the norm.
- Exceeding the limit operating modes inevitably entails premature failure of the lamp.
Repeated switching on and off of the filament of the lamps is undesirable, as they contribute to the deformation of the cathode and can shorten the life of the lamp. Therefore, when operating lamps with frequent periodic interruptions in operation, it is recommended not to turn off the heat during the break, but it is even better to reduce its voltage to 80% of the nominal value.
2. 1. SCHEME OF HF POWER AMPLIFIER WITH GROUNDED GRIDS (ON GI-7B, GI-7BT, GI-6B, GS-9B, GS-90B, GI-23B, GI-46B, GU-50, G-811, GK-71)
If the output power of the transceiver is about 30 - 50 W, and the transceiver does not have output power level control, the best option in this case is to build an amplifier according to a common grid (OS) scheme.
Common grid amplifiers can operate in any of the modes. The advantages of such amplifiers are good linearity, high energy performance and stability, linearity of operation over a wide range, since in the OS circuit the control grid is an electrostatic screen placed between the anode and cathode, i.e. between the input and output and, while creating a good isolation, allows you to increase the cutoff frequency of the amplified signals. The disadvantages include low input impedance, as a result of which the circuit has a low power gain (Kp » 10-20 times), therefore, a large input excitation power is required to fully drive the amplifier. It is not rational to use lamps designed for linear amplification of signals in the AB mode in a circuit with an OS, since their main advantage, a high gain, is not used in this case. It is also not recommended to use those tetrodes and pentodes, in which the beam-forming plates or the third grid, respectively, are connected to the cathode inside the lamp, since they are prone to self-excitation in this inclusion.
Lamp GI-7B, GI-7BT, GI-6B, GI-23B, GI-46B, GS-9B (option A). The circuit below is designed to work with transceivers having an output power of 20-40 watts. To work with QRP or QRPP devices at the input of such an amplifier, it is necessary turn on the auxiliary preamplifier. The amplifier itself is made on two GI-7B triodes (since all of the above lamps have approximately the same basic electrical parameters and geometric dimensions, only the amplifier circuit on GI-7B lamps is considered) according to a hybrid circuit with grounded grids. GI-7B lamps in grounded grid circuits operate stably at frequencies up to 500 MHz.
GI-6B lamps differ from GI-7B lamps only in the upper limiting frequency; when working on HF, this does not affect in any way. In addition, the choice of these lamps is due to the following: GI-7B lamps are the cheapest lamps of this class in terms of cost and, therefore, are widely used in the construction of amplifiers. For example, in the Ukrainian markets, their cost is only 1 - 2 USD per unit, while, for example, the cost of GU-72 is 15 USD, GMI-11 is 25 USD, GU-74B is 25 USD, 6P45S is 3- 4USD. (data are given for the summer of 2000).
The use of two lamps connected in parallel in the amplifier makes it possible to obtain a much larger anode current with a relatively low excitation power. The amplifier can be made on one lamp, while maintaining the same parameters (meaning the input and output power), while the load on the lamp increases, the lamp operates at high currents, which can lead to overheating of the cathode and grid, therefore, the durability and reliability of the amplifier will be correspondingly lower, and in addition, to obtain the same output power, it is necessary to increase the excitation power. For one lamp, the quiescent current is correspondingly halved, all other requirements remain the same.
A pre-amplifier on a field-effect (biplanar) transistor VT1 is included in the lamp cathode, which is connected, if necessary, depending on the output power of the transceiver using relay K4, while the power gain increases. With a power gain of about 20 (13 dB), the output power of the transceiver used in conjunction with the amplifier should be 20-40 watts. When the pre-amplifier is turned on, the gain increases to 100 (20 dB), so the required excitation power decreases by an order of magnitude and is only 3.0-5.0 W, i.e. in this case, the amplifier can be operated with almost any QRP transceiver (transmitter). When using this amplifier, there are three options:
a) it is assumed that the power amplifier will be used permanently only with a transceiver having a power of 20-40 W, while there is no need for a pre-amplifier and a K4 relay. In this case, the mounting holes in the chassis of the RF unit for relay K4, transistor VT1, variable resistors R22, R23 are not drilled.
b) assuming constant use of the power amplifier only with a QRP transceiver having a power of 3-5 W, there is no need for a K4 relay. In this case, do not drill. mounting holes for relay K4.
c) It is assumed to use a power amplifier with both a QRP transceiver and a transceiver having a power of 20-40 watts. In this case, all holes are drilled in the chassis. Moreover, if you use the amplifier with a QRP transceiver most of the time, it is better to solder the input and output of the preamplifier to the closed contacts of the K4 relay and, accordingly, vice versa, if you work more often with a powerful transceiver, the preamplifier is soldered to the open contacts of the K4 relay, that is, in any In this case, K4 will be in a de-energized state most of the time.
It should be noted right away that when building a universal amplifier, it must be borne in mind that in the above circuit, in the preamplifier, it is best to use "current" transistors, i.e. transistors that deliver maximum power at low collector (drain) voltages. This is due to the fact that with an anode voltage of 1300 V (1500 V) used in the described amplifier circuit and a lamp quiescent current of 50-90 mA, the bias voltage for GI-7B lamps is only 14-15 V (20 - 22 V ), but the same voltage is also used to power the preamplifier. The normal supply voltage for KP904 is 40-50V, therefore, the resulting bias voltage is not enough to get the maximum power from the transistor. This remark applies to many other transistors as well. Therefore, at a given anode voltage, you do not fully use the advantages of a hybrid stage.
thrones VD1-VD4. In this case, the voltage at the cathode will be about + 80 V, while the lamp is securely locked. When the amplifier is switched to transmission mode by pressing the pedal, relay K2, connected in parallel with VD4, shorts it out, reducing the bias voltage on the control grids, and the lamps open. The current flowing through the relay contacts at the peaks of the anode current can reach 1.0 A, therefore, as this relay, it is necessary to use relays with
| powerful contacts, for example RES-47, RES-48, REN-34, etc. The equivalent resistance of the anode load of the cascade is about 1.3 (1.5) kOhm. The input impedance of the stage is about 30 ohms, so already with an input power of 40 W, the voltage at the input of the amplifier will be about 35 V, and this will lead to the appearance of a grid current at the peaks of the input signal, i.e. the amplifier goes into class AB2, which is quite acceptable for SSB mode , therefore, with a slight excess of the bias voltage, this is not scary, since the grid current is insignificant compared to the total input current of the amplifier and the distortions introduced by it |
Fig.5 PSU preamplifier board. |
|
 Fig.6 Relay REN-34 |
zheniya are insignificant. With a further increase in the signal level at the amplifier input, the nonlinear distortion at the amplifier output increases (the variable component of the anode voltage takes on a pulsed character, so harmonics appear at the output), so it is better to adhere to the calculated mode. In the case of using a hybrid cascade, the excess excitation voltage is easily extinguished by reducing the value of R23. In the same way, with a lack of excitation voltage, the value of R23 can be increased. The variable resistor R22 serves to adjust the quiescent current when replacing lamps. | |
For the GI-7B lamp, the power dissipated by the lamp anode is quite large and amounts to 350 watts. And although some authors write, for example, that in "light mode" the lamps can work without forced airflow, I do not recommend using them in this mode. For the same reason, the wires coming from the incandescent choke to the clamps for fastening the heater and cathode leads must be soldered to them only with refractory solder, and even better, securely screwed through washers, and not soldered, since in case of overheating of the lamp, the wires can just chill out. The same requirements apply to the installation of anode circuits (this is especially true for work in competitions, when most of the time the amplifier is in transmit mode and maximum heat generation occurs).
Lamp GU-50. (option F). Pentode GU-50 in the circuit with OK due to the low steepness is not advisable to use (if a preamplifier is not used at the input). The best option is to use it in a circuit with an OS. In the circuit with the OS, the correct choice of the operating point allows you to reduce the quiescent current of the lamp to 10-15 mA compared to the OK circuit, where it is 40-60 ma, while the heating of the lamp during pauses decreases, and the efficiency of the cascade and, accordingly, the output power increase, the lamp mode approaches mode B. in this case, the lamp gives off the most power - up to 110 W (according to the passport!).
The circuit is made on 3 lamps (you can apply four). The lamp is inconvenient because the anode and grid leads are located together, which creates inconvenience during installation. If you bring the anodes together, it is inconvenient to separate the input circuits and, accordingly, vice versa. The search for a way out of this situation led to the decision to install the basement of the RF unit in two floors (see fig. 12E3 and fig. 12E4). The anode and grid circuits are separated by a screen plate, pos.106, to isolate the anode circuits from the amplifier chassis. serves as a plate poz.106A. Due to the fact that the anodes of the lamp are located at the bottom, the arrangement of the elements of the output P-circuit had to be re-arranged (“turn over” the entire installation), the front panel was also slightly re-arranged. Be careful when performing plumbing work. Otherwise, the scheme has no features.
In a circuit with an OS, there are two options for turning on the lamp:
a) with HF grounded grids (option F2), i.e. with the presence of nominal constant stresses on the grids;
c) all grids are directly connected to the body (option F1), while the lamp turns into a high-gain triode. Since all the grids are connected to the case, the amplifier becomes very stable and its linearity is no different from that of an amplifier with rated DC voltages on the grids. In addition, with such a connection of the pentode, additional sources of stabilized voltage for the screen and control grids are not needed, but this circuit requires more excitation power and the currents of the grids connected together increase, with the main part of the current falling on the control grid.
Lamp G-811. (option H). For a long time I did not want to deal with this option, since the lamp does not fit into the case used for other amplifier options in its size. But at the request of friends, this option had to be made. For the normal placement of four lamps and at the same time maintaining the normal temperature regime inside the lamp compartment, it was necessary to increase its width and height by 30 mm (in all drawings, the dimensions of parts for this version of the amplifier are given in brackets). The amplifier can be made on two, three and four lamps, the input impedance of the amplifier depends on the number of lamps connected in parallel. The fact is that these lamps have a small input capacitance, so it is convenient to use them in parallel connection. In addition, they have a low anode load resistance, which gives a gain in high-frequency ranges. The amplifier circuit is shown in fig. 2D.3. When repeating the amplifier, instead of domestic G-811 lamps, use its foreign counterpart - 811-A lamps.
Lamp GK-71 (option I). The difference of this circuit is that a transformer of the ShPTL type is used to match the high input resistance of the lamp. This circuit solution simplifies the design of the amplifier, eliminating the use of switchable input P-circuits for each range. For a full buildup, a power of the order of 70 watts is required, for this purpose the UW3DI is suitable. To obtain output parameters, it is necessary to use a voltage-by-six circuit to power the anode.
As noted above, amplifiers built according to the OS circuit have a low input impedance Rin., which makes it difficult to match the amplifier input with the output of the transceiver (transmitter) when they are used together. Moreover, Rin. depends on both the range and the number of lamps connected in parallel. With an increase in the number of lamps Rin. decreases. The inconsistency leads to the fact that for the normal buildup of the amplifier, the transmitter must have a power margin. The easiest way out of this situation is to match using an RF autotransformer at the amplifier input (see Fig. 2.19). The transformer is wound on a ferrite ring with a permeability of 50 HF and contains 10-15 turns (usually 12). The ring diameter is 20-30 mm (depending on the input power), the wire diameter is 0.6-0.8 mm. The tap position is selected according to the maximum agreement on all ranges. Initially, the tap is taken from 7-8 turns, counting from the grounded end of the transformer winding. The input of an amplifier made with a transformerless anode voltage source is similarly consistent. In this case, the transformer connection of the windings is used and the matching is carried out by changing the number of turns of the input winding.
2. 2 KV POWER AMPLIFIER ACCORDING TO THE SCHEME WITH A COMMON CATHODE (ON LAMPS GU-72, GMI-11, GU-74B, 6P45S, GU-50, G-807)
The circuits of all driven amplifiers are built according to the common cathode (CC) scheme. The common cathode circuit has a large input impedance, so a small amount of power is sufficient to excite it. This inclusion of the lamp allows you to get a large power gain (Kp), so with an output power of your device of 5 - 20 W, it is better to choose this option. The scheme is easily consistent with the previous cascades. However, too high a value of Kp can lead either to unstable operation of the PA or to its self-excitation, so that when installing the RF part of the amplifier, all requirements must be met. In addition, with an increase in the operating frequency, Kp decreases, so it is necessary to provide a certain power margin in the transceiver to obtain the required output power of the amplifier in the HF bands.
Directly at the input of the amplifier, a load resistance is included equal to the output impedance of the transceiver 75 or 50 Ohm, which improves the amplifier's resistance to self-excitation and at the same time is a load for the transceiver. This resistor drops part of the transceiver power (about 20%). The amplifier works stably even without it, but there may be problems with matching some types of imported transceivers with an ALC system. The power dissipated by the resistance is 8 W, when a power exceeding this value is applied to the input of the amplifier, the power of the resistance should also be increased (by typing them from a larger number of resistors).
Lamp GU-72. (option B) The amplifier operates in class AB1 when operating in SSB, AM mode and class C when operating in CW and RTTY mode. The power required to drive the amplifier is 8-12 watts. The lamp mode, depending on the type of work, is automatically set by selecting a bias on the lamp control grid using relay K2, controlled by the switch for the type of work S4 "SSB - CW". In the receive mode, a negative voltage of -100 V is supplied to the control grid of the power amplifier lamp from the VD14 zener diode, the amplifier lamps are securely locked. When contacts 1 and 2 of the XP3 connector (Pedal) are closed, relays K2 and K3 are activated. Relay K3 with its contacts 4.5 disconnects the antenna from the input of the receiver, and contacts 2.3 puts the transceiver into transmission mode.
Rice. 7 Amplifier with transformer supply on 2 lamps GU-72.
The contacts of relay K2 connect the voltage divider R22 or R23 (depending on the selected radiation mode) and the negative voltage on the control grid is reduced to the desired value, corresponding to the quiescent current of the lamp in this mode.
The main advantage of GU-72 tetrodes is that the lamp anode does not require forced airflow, while the permissible power dissipated by the lamp anode is 85 W, therefore, it is possible to remove power up to 350 W.
If the power of the transceiver you are using is about 25-30 W, and the transceiver does not have output power level control, then in order to prevent the amplifier from pumping at the input, it is better to assemble it according to the OS circuit (in this case, with RF-grounded grids), as shown in Fig. 2.B (option C1). In this variant of turning on the lamp, the output power of the amplifier turns out to be thirty percent more compared to an amplifier made according to the OK scheme. The mounting of the amplifier is shown in Fig. 16.6. 
GMI-11 lamp (option C) The GMI-11 pulse generator tetrode with a sufficiently low filament current (only 1.75 A at Un=26 V) has excellent characteristics. The lamp anode current per pulse is > 14 A, the maximum allowable anode voltage is 10 kV. At the same time, it, like the GU-72 lamp, does not need to be blown. This lamp is difficult to “drive” even for lovers of long “pressing” when setting up their “power¢s” right on the air and experiencing great bliss from this, however, it is also necessary to choose the right frequency, for example, a rare DX, because here many will immediately appreciate the power and the quality of work of your wonderful PA, which, by the way, will be reported to you immediately and immediately in a correct and flattering form.
Rice. 8 Amplifier with transformer supply on the GMI-11 lamp.
The circuitry of the amplifier on the GMI-11 lamp is practically no different from the circuit of option B, only one lamp is used. The location of the lamp terminals completely coincides with the GU-72, and therefore, with some change in the design of the amplifier, assembled according to the scheme of option B, it is possible to use two GMI-11 lamps in it, although you should remember about the thermal regime inside the amplifier case and the power of the anode voltage source .
The lamp can also be used in an amplifier according to the OS circuit, assembling it according to the circuit shown in Fig. 2.B (option C1). The mounting of the amplifier is shown in Fig. 16.11.
Lamp GU-74B (option D) Similar to the previous ones, the circuit was also made on the GU-74B lamp, the difference is that the fan blowing the lamp turns on when the amplifier is turned on. The fan has a capacity of about 120 m³/h, while the lamp requires only 35 m³/h to blow through, which allows it to be placed on the side of the lamp, but there is enough space in the housing to install it from above. This tube is specifically designed to amplify single-sideband (SW) signals, so increasing the bias voltage beyond the optimum to reduce the quiescent current in this circuit is undesirable. In this case, the oscillatory characteristic is bent in the region of small input signals. This mode is similar to limiting the telephone signal from below, which leads to a deterioration in the intelligibility of the signal, an increase in non-linear distortion and out-of-band radiation, so the use of these lamps loses all meaning. Based on this, when setting up the quiescent current of the lamp, it should be remembered that it is 300 mA in SSB mode. This lamp can also be used in the version of the amplifier, assembled according to the scheme with the OS.
Lamp 6P45S, 6P42S, 6P36S (options E). Some hams are wary of using horizontal scanning TV tubes in power amplifiers due to their thermal “brittleness”, others claim that such tubes are not suitable for SSB amplification. Of course, there is some truth in these two statements. Thermal “brittleness” (the inability to withstand increased heat for a long time) can simply be eliminated by tuning the amplifier in short cycles (without holding the key pressed until the lamp turns first crimson and then blue) or by using the “cold tuning” of the output stage. The limitation of the continuous operation time of the lamps is due to the fact that TV lamps are designed for pulsed operation with rather large current amplitudes, but with their short duration, and not with constantly operating currents that keep the lamp open for a long time. Nevertheless, TV lamps are quite satisfactory requirements of both professional and amateur communication equipment, when using “intermittent” (not permanent) signals: CW, SSB.
Before proceeding with the assembly and debugging of this option, a two-tube amplifier was assembled, published in , and two options were tested, both with buildup into the cathode and into the grid. At an anode voltage of 750 V and an amplifier input power of 7–10 W (when driven into the grid), an anode current of 600 mA was obtained in almost all ranges.
As a result of the experiments, it was found that the voltage on the screen grids of the lamps should be 180 - 200 V, as required by the passport for the lamp. With a further increase in the voltage on the second grid, when the amplifier is switched to the transmission mode even without applying the excitation voltage, the lamps begin to open spontaneously, the anode current increases to 1.0 A or more, and the anodes of the lamps instantly become crimson.
Of course, the anode voltage of 1330 V is perhaps a bit too much for 6P45S lamps, but at this voltage, the load resistance (Re) is greater than in the amplifier described by the author, which makes it possible to obtain much lower capacitances of the P-loop. And yet, in the amplifier on 6P45S lamps, the load resistance is quite low, which requires a correspondingly large size of the variable anode capacitor. If it is not possible to purchase such a capacitor, you can make it up from two, “spurring” to the main one on each range (naturally, where the capacity of the main capacitor is not enough) a constant capacitance capacitor, or even replace it with a set of constant capacitance capacitors switched using a range switch . In this case, to fine-tune the anode circuit to resonance, a ball variometer can be used as the inductance of the P-loop. In terms of dimensions and inductance value, the variometer from the R-140 radio station (YAR4.773.022) is very suitable.
In most horizontal-scan lamps, the interelectrode distances are large enough to allow their use at elevated anode voltages. Accumulated practical experience has confirmed that such lamps can be forced with a decrease in their service life. at an anode voltage of 1000 V and above and at currents much higher than the passport allowable values. It's just that the lamps will have to be changed more often than when working within passport modes, but they can be found in almost any market and they are cheaper than generator lamps. In addition, at your discretion, you can always reduce the anode voltage by soldering the taps on the secondary windings of the anode transformers.
When manufacturing an amplifier, it should be borne in mind that the power consumed by each lamp (6P45S) by incandescence is 18 W, therefore, to power four lamps at Un = 6.3 V, it is necessary to obtain 10A from a transformer, which is somewhat problematic while maintaining the small dimensions of the incandescent transformer , therefore, in order to be able to use a standard transformer of the TH series of a suitable size, the filaments of the lamps are connected in pairs in series. No matter what type of tubes you choose for your amplifier, when you connect tubes in parallel, you may (or rather, definitely!) Have specific problems.
The anode current obtained from each lamp in the bundle should be considered as a subject of special attention. Dynamic balance is essential, as it is important that none of the lamps in the combination "set" the others. If, for example, we turn on four 6P45S lamps in parallel, having different slope characteristics, then when transmitting, some of these lamps will be a load for others, others will swing more up to saturation current, which will lead to their overheating, and in general, respectively to reduce the efficiency of the cascade, i.e. to reduce the output power. The result of such work may be overheating of the lamps, their anodes, together with glass cylinders, may melt, and the latter may simply crack.
In the manufacture of this version of the amplifier, the lamps must first be selected before being installed in the circuit, or when setting up the amplifier by adjusting the bias voltage at full buildup of the amplifier, the same anode currents of the lamps are set (each individually). The quiescent currents of the tubes tend to be uneven, but they are too small to affect the linearity of the amplifier as a whole or the durability of the tubes.
This, by the way, also applies to all other lamps if you use more than two in parallel. The ideal case is the selection of lamps also according to the steepness of the characteristic. For an amateur, this solution cannot be called successful, since a large amount of “material” is required from which you can “choose” lamps for use in the RA (as a rule, radio amateurs do not have such stocks), not everyone can afford it.
Another difficulty encountered when using lamps in parallel is a noticeable increase in input and output capacitance. Needless to say, increasing any of these values will increase the RF shunting effect mentioned earlier. Certain and severe restrictions on the value of the upper frequency limit also appear when connecting lamps in RA in parallel.
For example, the passport value of the input capacitance of a 6P45S lamp is 40 pF, the output capacitance is 16 pF. Four lamps connected in parallel will give an input capacitance of 240 pF, an output capacitance of 96 pF. The output capacitance can be absorbed by the anode circuit circuit (included in its circuit, neutralized), but, here, the input capacitance will have to be managed with the help of a matching device, i.e., not in the best way than it is done now in RF power amplifiers on transistors.
Galaxy introduced a 2 kW (PEP) power amplifier (Model 2000+) using 10 horizontal scan tubes connected in parallel. The amplifier worked in the AB1 class, “swinged” through a powerful non-inductive resistor and was made according to the “common cathode” lamp switching circuit.
Since the ray-forming plates of 6P45S lamps (the only one from this series) do not have a connection with the cathode inside the lamp housing, they can also be used in a circuit with an OS, and in both versions: as with grids grounded by HF, i.e. with rated constant voltages on grids; and grids directly connected to the body, as is done, for example, in . The switching circuit is shown in Fig. 2B (options E1 - E2), and the installation of the RF part in Fig. 16.17, Fig. 16.18, respectively..
NOTE: Since the conductor connecting the anode of the lamp with the anode cap inside the 6P45S lamp is made of thin copper wire, which can be soldered or simply melted when using the lamps in the maximum power mode of your RA, especially when working on the HF bands, the amplifier must be equipped with a forced hood, using for this purpose fans from the power supply of the PC.
Lamp G-807 (options G). As long-term practice of using G-807 lamps has shown, they work perfectly both in class C mode when used in telegraph mode, and in class AB1 mode when amplifying a single-sideband signal. So that the lamps do not overheat at the same time, the most favorable operating mode for lamps (meaning for four) = 80-100 per lamp. Re in this case is about 3.3 kOhm. That is, our power supply just satisfies all these requirements. In such modes, the lamps will retain their guaranteed performance for more than 1500 hours.
The scheme for constructing the amplifier is shown in Fig. 2B (options G1 - G2), and the installation of the RF part is shown in Fig. 16.24, Fig. 16.30, respectively.
2. 3 TWO-STROKE HF POWER AMPLIFIER (ON LAMPS GU-72, 6P45S, 6P42S, 6P36S, GU-50, G-807, G-811)
The advantages of building an amplifier circuit according to a push-pull circuit include the following:
a) higher linearity and efficiency compared to single-ended amplifiers;
b) much lower even harmonic emission compared to single-ended amplifiers;
c) serial connection of the input and output capacitances of the lamp to their respective circuits, which reduces the initial capacitance of these circuits;
d) reducing the voltage of the anode source by half compared to the usual switching circuit to obtain equal powers;
e) halving the amplitude of the output signal, which reduces the requirements for the gap of the capacitor of the "hot" end of the output P-loop
Disadvantages of the push-pull scheme:
a) the need to select lamps with similar parameters;
b) doubling the equivalent resistance of the output circuit, which can have a strong effect on the upper ranges.
In an amplifier built according to a push-pull circuit, the excitation voltage to the lamp grids is supplied in antiphase (i.e., with a shift of 180 °) from opposite ends of the input transformer. Lamp anodes are connected in the same way. The output circuit of the amplifier is included in the secondary winding of the output transformer. When the circuit is symmetrical, the odd harmonic currents are added to the load, while the even harmonic currents are compensated. The middle points of the windings of the transformers have zero potential (at high frequency), so the bias voltage and the anode voltage are connected to them, respectively. However, due to the presence of some RF voltage on them, associated with the incomplete symmetry of the circuit, they (middle points) cannot be grounded.
The amplifier, made according to the push-pull circuit, can work both in circuits with OS and in circuits with OK.
A diagram of a push-pull amplifier with OK, made on 4 lamps 6P45S (6P42S, 6P36S) (option E3), is shown in Fig. 2D1. A drawing for mounting the RF amplifier unit is shown in Fig. 16.19 and Fig. 16.20. Lamps 6P45S (only!) Can also be used in a circuit with an OS.
A diagram of a push-pull amplifier with an OS on 4 GU-50 lamps (option F3) is shown in Fig. 2D24. Drawings for mounting the RF block of amplifier options are shown in Fig. 16.21, Fig. 16.22 and Fig. 16.23. GU-50 lamps can also be used in a circuit with OK
At the input of the amplifier, a SPT is switched on, which doubles the amplitude of the input signal and creates anti-phase signals to excite the arms of the amplifier. A similar transformer at the output of the amplifier, on the contrary, reduces the amplitude of the output signal by half. Everything else is similar to the previous schemes.
Similarly, circuits are built on two GU-72 lamps and four G-807 lamps.
2. 4KV POWER AMPLIFIER WITH TRANSFORMERLESS (COMBINED) POWER SUPPLY
If for the manufacture of an amplifier you do not have the opportunity to purchase the necessary anode transformers or you simply need a light but powerful enough PA to work in the field or DXpeditions, where, as you know, every extra kilogram during transportation not only "eats" your money, but also great lengthens the arms, any of the amplifiers described above can be made with a power supply assembled according to a transformerless or combined circuit. Often, in order to obtain the anode voltage, circuits are used to double, triple or quadruple or even increase (I even saw a multiplication circuit by eight) the mains voltage, depending on the required power of the amplifier. The presence of modern small-sized high-capacity electrolytic capacitors with a high operating voltage and the same leakage resistance makes it possible to manufacture transformerless high-voltage anode power supplies for lamp output stages of relatively small power amplifiers, while using an unlimited power resource of such a power source as an industrial power grid. To obtain the heating voltage and the necessary service voltages, you can use a transformer that is small in weight and dimensions. In our case, when using voltage quadrupling, the amplifier turns out to be lighter than the base one by an average of ten to thirteen kilograms. It makes no sense to use a voltage multiplication scheme more than four times, since in this case the weight of electrolytic capacitors used for this purpose, given their total required capacity, and, accordingly, the number, becomes commensurate with the weight, volume and price of anode transformers.
Of course, there are no pros without cons. Some inconveniences also appear, for example, the chassis of the amplifier in this case will no longer be a common negative power bus and must be galvanically isolated from the mains.
It should be immediately warned: for the safety of the life of the operator, as well as to prevent the failure of the equipment connected to the amplifier, the operation of this amplifier is possible only if the radio station has a reliable electrical and technical grounding. Otherwise, the amplifier does not pose any greater danger than any other device that incorporates high-voltage power supplies, the voltages of which are dangerous to human life.
Scheme of a transformerless power supply unit that uses the principle of multiplying the mains voltage to obtain the anode voltage, i.e. not containing scarce high-voltage transformers is shown in Fig. 1D. The circuit is designed to operate from a single-phase AC network with a voltage of 220 V, one of the wires of which is zero.
Considering that with such a construction of the anode voltage source circuit, it does not have galvanic isolation from the primary network, and it is this source that consumes the most power from the network. Therefore, in order to protect against penetration into the network of interference created during the operation of the amplifier (ripples of the anode voltage), there was a need to turn on the input of the radio interference filter, consisting of capacitors C22, C23 and inductor L7.
With such a construction of the circuit, there is no galvanic connection between the lamp electrodes and the amplifier case and, consequently, the case with the mains supply.
If desired, to increase security, you can add an automatic starting device (PU) circuit to the power supply circuit, which is shown in and ensures the correct phasing of the mains when the amplifier is turned on. Such a device is made on the relay K7, K8, the scheme of its inclusion is shown in Fig. 1E.3. The device only works when the radio is connected to electrical ground. When you turn on the PU, the following situations may occur:
a) When properly turned on, the winding of relay K2 is supplied with mains voltage through normally closed contacts K1 and when the relay is activated, the power supply is turned on (relay K1 in this case remains in a de-energized state all the time).
c) If the phasing is broken, then when switched on, relay K1 is activated and its contacts “rephase” the power circuits, then the rectifier works as usual.
c) In the absence of grounding, the power supply circuits of both relays are broken, and the relays will not work, while the PSU simply does not turn on.
Thus, this starting circuit allows you to turn on the power supply at any position of the power plug. True, the RPT-100 relays used in the circuit are quite scarce, therefore, in their absence, the circuit can be performed as shown in Fig. 1.1. Naturally, you can do without it, but then every time you connect the amplifier to the network, you will have to control the correct phasing of the supply network.
Actually, the voltage quadrupler itself is made according to a symmetrical scheme, which has the best dynamic characteristics and a doubled frequency of the rectified voltage pulsations. The circuit includes capacitors C24 - C27, C1 - C8 and diodes VD1 - VD4. To ensure the level of ripples (Up = 0.05% Ua) required for the amplifier to operate in linear mode, the numerical value of the capacitance of the capacitors of each arm of the multiplier in microfarads must correspond to the numerical value of the maximum power of the amplifier in watts. Resistors R1 and R2 are ballast resistors, designed to protect diodes and fuses from overloads that occur when the PSU is turned on. With an anode current of about 600mA (at signal peaks), with the values \u200b\u200bof these resistors indicated on the diagram, only about 4V falls on them and, accordingly, about 2.5 W of power is dissipated, so there is no need to turn them off after charging the capacitors. The PSU, made according to the voltage doubling scheme, works in a similar way. The rest of the power supply circuit and the amplifiers themselves correspond to those described above and do not need explanation.
When repeating the circuit, one should not forget that the cathodes of the lamps will be at a high potential relative to the amplifier case. To increase the reliability of the operation of the transformerless amplifier circuit, in its manufacture it is best to use lamps with insulated cathodes (i.e., lamps with indirect heating), and in the case of using direct incandescent lamps, it is best to use factory-made transformers from the series TN and CCI, having reliable insulation both between the windings and between the windings and the housing. In the manufacture of homemade transformers, special attention should be paid to this issue, since the reliability of your amplifier depends on it.
Almost any of the amplifier circuits shown in this brochure can be used to work with a transformerless power supply. The implementation diagram of the amplifier on two GI-7B lamps is shown in Fig. 2C (option AB).
With an excitation power of 25 W, the amplifier outputs 400-450 W to the antenna at a load of 75 Ohms and about 500 W at a load of 50 Ohms on all amateur bands. The amplifier features excellent gain linearity over the entire frequency range.
Rice. 8 Amplifier with transformerless power supply on 2 lamps GI-7B
In order to isolate and protect the output stage of the transceiver, the excitation voltage is applied to winding III of the inductor L6. Capacitor C14 is needed in case, for any reason, an interturn short circuit L6 occurs. Due to its presence, the transceiver will not suffer.
The permeability of the rings used to wind the inductor may need to be adjusted. The fact is that rings produced by different factories behave differently at different frequencies. Therefore, if possible, it is necessary to make two or three chokes, for example, 2000НН, 1000НН and 600 - 400 НН and drive them in turn through the circuit by ranges, and naturally leave the one in which the output power is more uniform over the ranges, unless of course you want to have somewhere there is a gain increase in one of the ranges (for example, to compensate for the uneven output power of the transceiver).
For decoupling from the antenna and at the output of the amplifier, you can also use the SPT shown in Fig. 2.12, but connected according to the 1: 1 transformer circuit, or to increase the band of satisfactory matching - 2: 1 (the transformer in this case is wound in three wires). The changes that need to be made in this case to the circuit are shown in Fig. XX
When using this amplifier to work with a QRP transceiver, it should be supplemented with a preamplifier made according to any circuit shown in the album, but it is better to use the circuit in Fig. 2.15, this will allow you to simultaneously decouple the transceiver output from a transformerless power supply amplifier. The winding I of the transformer T1 in this case is the winding III of the inductor L6.
When using an amplifier made according to a transformerless power supply scheme, it is better to supplement it with an antenna switch for operation in the field. The switch diagram is shown in Fig.2.17 - Fig.2.18.
Rice. nine Amplifier with transformerless power supply on 4 lamps 6P45S, with built-in antenna switch.
To get by with three toggle switches to switch four antennas (four do not fit into the appearance), I had to use dual MT-3 toggle switches and place them on the rear panel of the amplifier. When you select any of antennas 2 - 4, antenna 1 is automatically disabled. The installation of the switch is shown in Fig. 15AB (the switch relays are installed at the location of the anode transformer on the bracket poz.117, which is attached to the rear panel of the amplifier).
If you still think that with such an amplifier power you are still poorly answered, you can still slightly increase the power of the amplifier assembled according to the transformerless power supply circuit by assembling the anode power supply according to the voltage multiplication circuit by six, as shown in Fig. 1E4.
With this construction of the power supply circuit, the value of the anode voltage will increase to 1800 volts (at idle). In this case, the magnitude of the anode voltage drawdown under load depends only on the capacitance of the capacitors used in the multiplier.
Rice. 10 Amplifier with transformerless power supply on 3 lamps GU-50.
The six voltage multiplication circuit consists of two doubling circuits. Upper - C1, C2, C4-C7, VD1, VD2 and lower - C8, C9, C11-C14, VD5, VD5. Each of these doubling circuits gives 600 V. But since the voltages at the connection points VD1, VD2 and VD5, VD5 are 300 V higher than in the circuit of Figure 5, the input isolating capacitors had to be supplied with the same capacity, but doubled (600 V ) voltage. Both of these doubling schemes are “supported from below” by voltages of “+300 V” and “-300 V”, which are obtained from conventional half-wave rectifiers assembled on VB3, C3 and VD4, C8, respectively. In total, 1800 V is obtained (600 + 600 + 300 + 300).
When applying this scheme, first of all, one should pay increased attention to the insulation of the cathode circuits - in this embodiment, they may have a peak voltage relative to the grounded case up to 1200 V. It is not less than this voltage (and even better - with a two-threefold margin ) the insulation of the filament transformer, as well as (if applicable) of the input RF transformer, must be calculated. The operating voltage for capacitors C19 and Cp, in order to ensure the reliability of the design, should be 2.5 - 3.5 kV. In this case, the use of a starting circuit assembled on R26, C28, K1A is mandatory. The design and installation of the modified power supply are shown in Fig. 12G.
To work together with a transformerless power supply, it is very convenient to use the circuit design of the amplifier itself according to a push-pull circuit. In this case, the galvanic isolation of the amplifier circuit from the AC mains is obtained automatically by using broadband isolation transformers at the input and output of the amplifier (see Fig. 15.3, 15.4 and Fig. 16.3, 16.4, 16.6.
The design modifications of all the above amplifier circuits are shown in Fig. 16 add. 1 - Fig. 16 add. 24
AMPLIFIER PARTS
When designing amplifiers, emphasis was placed on the use of standard factory parts in them, widely used in household equipment and available to many radio amateurs. The exceptions are the anode and filament chokes, the coils of the P-loop of the HF and LF ranges.
The anode choke is one of the most important elements of the circuit, so its manufacture must be taken seriously. So with a small inductance, i.e. commensurate with the inductance of the anode circuit, power distribution occurs, and in the event of a series resonance in one of the operating ranges of the amplifier, power is “sucked off”, while the inductor becomes very hot and may even be charred. The same thing can happen if you make contact leads in the form of a closed coil of magnetic material. Inductor L1 must be rated for current up to 600 mA, its design is shown in Fig. 12C.
Rice. 11. Anode choke
The inductor is made on a frame made of PTFE with a diameter of 20 mm, the length of the frame is selected depending on the lamps used. This is done for ease of installation, it is necessary that the output of its "hot" end is on the same level with the output of the anode of the lamp. Winding is carried out with a PELSHO wire with a diameter of 0.4 - 0.5 mm. For winding, 16 meters of wire is taken.
The choice of length is based on the fact that with such a length of wire, the inductor will not be a half-wave follower in any of the amateur bands. The first 15 turns are wound with a pitch of 2.0 mm, to obtain the required pitch, a spiral groove is cut on the frame, then 40 turns are wound turn to turn, and the remaining wire is wound with a “universal” (option A) so that the turns do not “float”, they are additionally fixed with glue "Moment" or impregnated with varnish. At both ends of the frame of silver-plated wire dia. 1.0 - 1.2 mm, contact leads are made that pass through the frame, and the throttle leads are soldered to them. The resulting choke has an inductance of the order of 500 - 600 uH and works fine on all HF bands. The throttle frame is attached to the chassis with an M4 brass screw, for which a hole for the M4 thread 15 mm deep is drilled from the end of the frame.
When fastened with a steel screw, it should not reach the location of the coil, otherwise the screw will turn into a core. The frame can be taken and the factory ceramic. In the event that you have a problem with the winding of the “universal” type, the inductor is wound turn to turn, and to increase the inductance in the low-frequency part of the inductor, a segment of a round ferrite rod with a diameter of 1.5 mm is inserted. 8 mm from the magnetic antenna of radio receivers 50 mm long (option B). The inductor can also be completely wound on a round ferrite rod from a magnetic antenna of pocket r / receivers or on a ferrite ring with a diameter of 30-40 mm, for example, as it was done in the R-130 radio station. The ring is pre-wrapped with fluoroplastic tape (varnished cloth). In the latter case, it is better to use MGTF wire for winding.
The L8 inductor used in the cathode of the lamp has much lower requirements. It is wound on a fluoroplastic frame dia. 18 mm, winding is carried out turn to turn also with PELSHO wire with a diameter of 0.4 - 0.5 mm until the space between the leads is filled (see Fig. 12D).
As a choke L2, a factory-made DM-0.1 choke with an inductance of 250 - 500 μH is used, similar chokes are used as L1, L2 SWR meters.
Rice. 12. Choke design L3
The L4 coil is frameless, the coil diameter is 60 mm, the number of turns is 6.5, the winding pitch is 7 mm, it is wound with a copper tube with a diameter of 5 mm. The tube can not be silvered, since the quality factor of the coil turns out to be large and silvering practically does not add anything. The taps on the coil are made from 2¼ vit. - 10 m., 2 ½ vit. - 12 m., 3½ vit. - 15 m. and 4½ vit. - 17m. These data are approximate, since you can slightly reduce the diameter of the coil depending on the size of the anode capacitor, the number of turns will increase, or you can make a mistake in the distance between the turns when making the coils, so it is made with some "margin". sawed off. At the “hot” end of the coil, an M5 thread is cut, with which the coil is screwed into the capacitor C17, a silver-plated wire with a diameter of 10 mm is soldered into the second end of the coil. 1.2 -1.5 mm, with which it is attached to the range switch (it is passed through the switch contacts).
Fig.13. Coil construction L 4
|
Table 4
|
The taps from the coil are also made with a silver-plated wire dia. 1.2-1.5 mm. Considering that HF currents flow only along the surface of the conductors, the L4 coil can also be made from a bimetal. It, like L5, is made according to the recommendations given in the Hand-book magazine for 1986 based on output power (see table 4)
With this wire diameter, they are not overheated by currents flowing through the P-circuit. If you use a thin-walled tube to make an L 4 coil, then when threading it, you need to insert something into it (fill it with solder) so that the die does not tear the edges of the tube. Copper is a “capricious” metal, so only a new die must be used for threading. If the tube diameter is slightly less than 5 mm, the end of the tube should be slightly flattened. In other cases, the end of the coil, which is screwed into the capacitor, can be made in accordance with Fig. 12B.
Fig.14. Coil construction L 5
The L5 coil is wound on a PTFE (ceramic) frame with a diameter of 50 mm, the winding pitch is 2.5 mm (spiral grooves are also cut on the frame to secure the turns and make winding easier. The depth of the grooves must be at least half the diameter of the wire used for winding). For winding, PEL wire is used - 1.2 - 1.5, the number of turns of the coil is 25. The taps are made respectively from the 4th vit. – range 30 m.; 8th vit. - 40m.; 15th vit. -80 m..
As L5, a coil from the matching device of the R-104 radio station, made on a ceramic frame, is suitable. In the absence of a frame of the required diameter, the coil is very easy to count for the existing frame.
For single layer cylindrical coils with
the length of the winding is equal to or greater than half the diameter of the coil, the calculation of the inductance is carried out according to the formula:
L = D²´ n² / (45D + 100l),
where L is the coil inductance, μH; D is the coil diameter, cm;
n is the number of coil turns; l is the length of the coil winding, cm.
For our coil data L5 - D = 5 cm, n = 25, l = 6.25 cm, substituting these values into the formula we get L = 5²´25² / (225 + 625) ≈ 18.38 μH, moreover, if at the same time the diameter of the wire also decreases while maintaining the length of the winding, the inductance will turn out to be 1 - 2% less.
Now let's recalculate the number of coil turns, for example, for a frame diameter of 3.5 cm. In this case, the frame diameter is reduced by 30%, therefore, to maintain the same inductance value, it is necessary to increase the number of turns by 30%, or approximately 8 turns. Thus, the resulting coil will have 33 turns.
Fig.15. Choke design L6
The L6 inductor is wound with two wires folded together on a ferrite ring with a permeability of 400 - 2000, the ring diameter is 40-50 mm. This diameter size is taken for the convenience of winding, it can be smaller. One square of the ferrite section holds 300-500 W of power (different sources give different power values). Much less requirements are imposed on its design than on the anode choke. The inductor windings must be designed for a flowing current of up to 4 A (5 A in the case of using 6P45S lamps). For a transformerless version of the amplifier on GI-7B, GK-71 lamps
The winding is carried out in three wires, and the excitation winding wire has a smaller diameter, because. only the excitation power passes through it, and the ring itself in this case should have a permeability of 400 - 1000, this also applies to the case of using a preamplifier assembled according to Fig. 2.15. The ring before winding must first be wrapped with fluoroplastic tape or varnished cloth. The number of turns is 8-12. (For the GK-71 lamp, the winding III is 20 turns). Their number is not critical and does not have a noticeable effect on the operation of the amplifier, so winding is carried out simply until the core perimeter is filled. But there is no point in winding anymore, because. it simply increases the resistance of the inductor, which leads to losses in it. Wire MGShV-0.75, it is very convenient to wind the choke with a double network wire, which also has double insulation. The choke works well on all HF bands. It is not necessary to block the choke for RF capacitors. During installation, the choke is located near the cathodes of the lamps and is fastened with two washers made of insulating material det.75 and an M4 screw. If you drill holes for the lamp panels using a centrifuge, do not throw away the resulting washers, with their help you will fix the L6 inductor, this also applies to L7. In one of the washers, the hole should be 3.2 mm in diameter, in the other for the M3 thread, this is necessary in order to be able to press the inductor winding to the ring, thus securing it.
The noise filter inductor L7 is wound with two MGSHV-0.35 wires folded together or a double network wire on a ferrite ring with a permeability of 2000NN and contains 20 turns, the ring diameter is 50mm. The design and mounting of the throttle are similar to L6.
The input transformer T1, used in the push-pull amplifier circuit, is wound on a ring core made of HF 50 ferrite with an outer diameter of about 20 mm. In its absence, ferrite with a permeability of 100-600 can also be used without a noticeable deterioration in the parameters of the transformer. The winding of the transformer windings is made with weakly twisted conductors and contains 6 turns. The output transformer T2 is wound on two MH1000 ferrite rings folded together with a diameter of 55 mm, pre-wound with fluoroplastic tape (rings from the transformer of the matching device of the R-130 radio station were used). For winding, twisted wire MGTF-1.5 was used, folded three times (about five twists per centimeter). The windings contain 8 turns. When installing the transformer, special attention should be paid to the correct installation of its terminals.
A PVC tube of the appropriate diameter or a heat shrink tube is put on capacitors C1 -.C8, instead they can simply be wrapped with “adhesive tape”, this will protect the high-voltage rectifier from failure if something (someone HI!) Conductive gets between the capacitor cases. If it is not possible to find washers (pos. 68) for capacitors, they can be made independently by using bare copper wire with a diameter of 1.2-1.5 mm for this purpose, as shown in Fig. 12B (detail 68A). To improve the contact, it is better to pre-tin the wire, this will prevent it from oxidizing in the future. Mounting bracket C24 -.C27 pos.113 can be taken ready-made from power supplies of the EC series or made according to the drawing.
The anode capacitor is a two-section one from old broadcast tube radio receivers with a capacity of 2 x 12-500 pF, the rotor and stator of which are pre-cut through the plate, while the gap between the capacitor plates will be about 2 mm, and the maximum capacitance of the section is 120 pF, the sections are connected in parallel. The breakdown voltage after thinning (constant) is 2500-3500 V (depending on the thoroughness of the assembly after the alteration of the capacitor). A variable capacitor from the R-104 radio station matching device is very well suited for this purpose (the capacity of each section is 12 - 500 pF). You can also take a capacitor from the RSB-5 r / st for this purpose, having a capacitance of 45 - 230 pF. An eccentric is fixed on the axis of this capacitor, which, when the axis of the capacitor rotor is rotated by 180 °, closes the contacts of the switch located on the capacitor housing. To be able to work in the range of 160 m, it is necessary to solder an additional capacitor with a capacity of 150-220 pF of the K15U-1 or KSO-6 type to these contacts, which is connected in parallel with the main capacitor (do not forget to reduce the output power of the amplifier to the permitted 10 W! HI !). In order for the capacitor to work well in the range of 10 m, it is necessary to mill or drill out the bottom and side walls of its case, since in this case its initial capacitance will decrease to 30 pF and the capacitor can already be used. To further reduce its initial capacity, it is necessary to cut off the upper part of the stator plates by 2-3 mm. You can not do this, but turn on a small capacitor in series with it, however, this option complicates the design, because. the case of the main capacitor in this case must be isolated from the chassis of the amplifier. A capacitor from the CHAYKA r / st is also suitable, its capacitance is 6 - 600 pF, as a result of which the tuning in the upper ranges is very sharp, but in parallel with it you can hang a capacitor of the K15U-1 (KSO-6) type with a capacity of 15-20 pF, which will solve this problem. But still, in any case, the anode capacitor should have as little initial capacitance as possible.
If, nevertheless, you did not manage to get any of the above, in this case you can make a capacitor from two capacitors, as shown in Fig. 2.19 - a capacitor of constant capacitance Cp with a nominal value of 5600 - 6200 pF and a two-section variable from a conventional receiver with a capacitance of 2 ´ 12 ¸ 495 pF (The capacitor is pre-thinned through the plate. The maximum capacitance of the capacitor in this case will be 220 pF), connected in series. The capacitance of capacitors connected in series is
C = C1´C2 / (C1 + C2).
In our case, Cmin = 5600´24/ (5600 + 24) = 23.9 pF, in the case of Cmax = 5600´220/ (5600 + 220) = 212 pF, thus getting a capacitor 24 ¸ 212 pF.
It is possible to use for this purpose one two-section capacitor from a conventional receiver 2 ´ 12 ¸ 495 pF, including its sections in series, as shown in Fig. 2.21. With this connection, the capacitor case must be isolated from the chassis. Capacitor mounting is shown in fig.12F1. Capacitor screws M4 flush fastened to the fiberglass plate poz.126, and the plate through three bushings poz.120 screws M4 poz.103 - to the front panel of the amplifier. The axis poz.127 is mounted on the axis of the capacitor and fixed with an M3 screw.
An antenna capacitor with four sections (the capacity of each section is 12 - 510 pF), from the ARC-5 or ARC-7 aviation radio compass, or from the r / st. R-104 or from a matching device of the same r / st., if you use the amplifier in the maximum possible output power mode, it is better to thin it out for greater reliability (the capacitor from the r / st R-104 does not need to be thinned, it has a sufficient gap ). If at the same time it turns out that the maximum capacitance of the antenna capacitor is small (since it is thinned out) or it was not possible to find such a capacitor, you can put a three- or two-section one, and in parallel with it, depending on the range, connect capacitors of constant capacitance, using two free sections to connect them range switch.
In this case, in the process of tuning the amplifier on those ranges where there is not enough capacitance C21, its rotor is set to the middle position, and an auxiliary capacitor of variable capacitance is connected in parallel with C21 and it is tuned, then the value of its capacitance is measured, and it is replaced by a capacitor of constant capacitance, required value. For this purpose, it is best to use capacitors of the KVI or KSO-6 type, which have sufficient allowable reactive power and operating voltage. These capacitors are soldered to the side of the antenna capacitor C21 (see Figure 12C).
On the drawing of the front panel of the amplifier, the holes for mounting the capacitors C20, C21 are not indicated, since their location depends on the specific type of capacitor used.
To switch the taps of the P-loop coil when switching from range to range, a 11P-5N switch was used. Three of its jacks, connected in parallel for greater reliability, actually serve to switch taps, although due to the possibility of a “cold” setting, the overvoltage mode of the output stage is practically absent. The two remaining biscuits, also connected in parallel, are used to connect, if necessary, additional constant capacitors to the antenna capacitor. Before installing the switch, it must be modified. The fact is that the taps of the ranges of 10-18 meters are made with a silver-plated wire dia. 2.2 mm, which is wider than the holes in the switch contacts and does not fit into it. They need to be made wider. For this purpose, an awl or "gypsy" needle is used. By inserting the awl into the contact hole, and rotating it, they gradually achieve such a diameter that the wire enters. This is done carefully so as not to tear the edges of the contact and not damage the lamella itself.
As a transformer Tr.3 for amplifier options B and C, you can use TA-163 220/127-50, or CCI-287. For amplifier options A, E, - use TN-53 220/127-50 TN-55 220/127-50, TN-56 220/127-50, TN-57 220/127-50, (or any of the TN series corresponding in current and power, or CCI-287, according to tables 2-3). For option D - TN-57 220 / 127-50 (in this case, the filament circuits of the lamps VL1, VL2 and VL3, VL4 will need to be connected in pairs in series).
Possible options for replacing Tr.1-Tr.2 without loss of amplifier power are shown in Table 1. For winding transformers, when they are independently manufactured, tape iron of the PL 20´40 - 80 type is used.
Button S1 - PKN41-1-2, buttons S2 - S6 P2K with independent fixation, installed on a common bar, and S6 is installed only when using the "BYPASS" mode in the amplifier.
Rice. 16. The design of the relay TKE52PKT and RP-2
As K3, a relay from the UM-3 power amplifier of the R-105 radio station (its old name RP-2) was used. Instead, you can use a relay from the TKE52 series, best of all TKE52PKT. In this case, a bracket det.21V is used for its fastening, and holes dia. 3.2 mm according to fig.10.
Flashlights for signal lamps are used from power supplies, engineering consoles and other devices that are part of the EC-1022, EC-1045, etc.
Instead of light bulbs for indication, you can also use LEDs, for example, AL307, which are powered from a source to power the relay. In this case, the LEDs are switched on through MLT-0.25 resistors with a nominal value of 2.7 - 3.0 kOhm (at a voltage of 24 V) or 1.2 -1.5 kOhm (at a voltage of 12 V). The LEDs are mounted on a printed circuit board, which is attached to the front panel using bushings pos.74, similar to buttons. For this purpose, additional holes are drilled in the front panel dia. 3.2 mm. To fit the LEDs tightly into the holes, adjust. A drawing of a printed circuit board for installing LEDs is shown in Fig. 13 (Pl. 4), and the switching circuit in Fig. 2.13 is respectively drilled for their mounting in the front panel
Shunts RSH1 and RSH2 are wound with nichrome wire on MLT-2 resistors with a resistance of at least 100 kOhm. If possible, it is best to use ready-made resistors of the C5-16T type of the required resistance, or, if there is a C5-16T of a higher rating, make the required ones from them. As you know, resistance is a linear quantity, so C5-16T is disassembled, the length of the wire with which it is wound is measured, and a piece is cut off, the length of which corresponds to the desired resistance (see Chapter 1).
GI-7B lamps are fastened with homemade clamps to the mesh leads in the holes cut in the chassis; standard panels are used to mount lamps of other types, which naturally does not exclude the use of homemade ones. In their manufacture, it should be borne in mind that the terminal contacts must be reliable (this is especially true for the heater and cathode terminals, where large currents flow and in the presence of transient resistance they will be very hot).
Rice. 18. Fee 1
Any transistor used in the cathode of the lamp, with a cutoff frequency of at least 100 MHz and a collector (drain) current of at least 2 A, the operating voltage of the transistor collector is 30 V.
Power supply elements R7-R13, C9-C10, VD5-VD6 are located on the printed circuit board Pl.1. Elements C13 - C14, VD15 - VD16 - on the board Pl.3 Drawings of boards and placement of elements on them are shown in Fig.13.
Bushings pos.71 - pos.73 are used ready-made - from the switch 11P-5N, or home-made can be used.
In order to reduce the level of noise generated by the fan, it is advisable to install the fan on a bracket through rubber bushings or completely on rubber.
Washers for capacitors C1 - C8 poz.69, which serve to insulate capacitor cases, installed on both sides of the chassis - polystyrene, they, like washers poz.68, are used at the factory. In the absence of these washers, the chassis of the PSU can be made of fiberglass 4 mm thick (when using a thinner material, the chassis will sag), but it will be necessary to adjust the position of the holes on the front wall of the PSU, which serve to attach transformers Tr.1, Tr.2 to it .
The screws and nuts used to fasten the clamps on the anodes and cathodes of the lamps, the C19 capacitor, the L5 coils are brass.
Rice. 19. SWR meter board
SWR meter. As a variable resistor R19 in the SWR meter circuit, it is better to use a paired potentiometer, in which both halves have closer characteristics, for example PP3, since the accuracy of the SWR meter readings depends on this. When used as P1 and P2 devices with a total deviation current of less than 1.0 mA, a resistor R3 'is installed in series with the resistor R3 on the SWR meter board, the value of which is selected depending on the sensitivity of the device used. When using a 1.0mA device, a jumper is placed instead of R3 '.
The diodes used in the circuit can be either germanium or silicon, for example, GD507, KD522A (it is better to use germanium).
Trimmer capacitors - KPK, KPVM, the current transformer is made on an annular core of size K12´6´4.5 from M50VN-14 grade ferrite. The primary winding is a piece of silver-plated wire with a diameter of 0.8 - 1.0 mm, threaded through the ring, the secondary winding is 30 turns of PEV-2 0.25 wire. The SWR meter circuit is mounted on a fiberglass printed circuit board, the board drawing is shown in Fig. 13B. The board is installed in the basement of the PSU chassis and is separated by a shielding partition from the rest of the amplifier mounting.
4. STRUCTURE AND ASSEMBLY OF THE AMPLIFIER
Now that you have carefully read the description of the amplifier options and the corresponding drawings, only after you have read and specifically selected the amplifier option depending on your requirements and capabilities, as well as the transceiver used for operation, feel free to get to work. This will save you from unnecessary work and mistakes when marking and drilling holes (and me from nightmares at night).
All drawings are made in natural size, this is done with the aim that in the absence of a certain size in the drawing, it could easily be removed from the drawing.
It should be recognized that when designing the amplifier, one of the basic laws of design was violated - the separability of all its constituent parts. “It mainly concerns the rear panel. This is explained by the fact that the placement of connectors on the front side of the panel allows you to hide all the flaws made when drilling holes in the panel during its manufacture at home, and also to abandon the false panel. Otherwise, in order not to spoil the appearance of the amplifier, all work would have to be done with great care.
The assembly drawing of the amplifier *, made on a scale of 1: 1.34 (the scale is taken in such a way that the assembly drawing fits completely on a standard A3 sheet), is shown in Fig. 15 - Fig. 16. Resistors R15, R17 in Fig. 16A, B, D, E and F are located under resistors R14 and R16 respectively. A drawing of the layout of the basic amplifier harness (when manufacturing various modifications of amplifiers, it is necessary to correct the table of wires), which matches in scale with the scale of the assembly drawing, is shown in Fig. 14.
The harness drawing is made on paper with a grid applied to it, the grid spacing is 1 cm. For ease of reading the assembly drawing, the harness is not shown on it, but if you have any difficulties during assembly, you can always combine these drawings. For the same reason, the taps from the coils L4 and L 5 are not shown.
The design of transformerless amplifiers differs only in that the chassis of the RF unit (pos. 5) and the PSU are made of fiberglass 4 mm thick, and a subchassis of duralumin with thicknesses is installed on top of the chassis of the RF unit. 2mm (pos.5A), the dimensions and configuration of which depend on the type of lamps used.
The dimensions of the front panel of the amplifier case were chosen at one time in relation to the dimensions of the RADIO-77 transceiver, and for the EFIR-M transceiver, for example, it is better to make it 380 mm wide. Everyone can solve that question in their own way. The body of the amplifier is divided into three compartments. In the first compartment - the power supply compartment, there are transformers Tr.1 - Tr.3, electrolytic capacitors C1 - C8, C12; in the 2nd compartment there is an anode circuit block, where the separating, anode and antenna capacitors, range coils, range switch, relay K3 for switching the antenna from reception to transmission are located. The third compartment is a high-frequency unit (tube), where the amplifier lamps, an anode choke, a fan for blowing lamp anodes, and pointer devices are located. The high frequency block is replaceable, its design depends on the type of tubes used in the amplifier. The anode circuits of the HF block are separated from the grid circuits and filament circuits by a horizontal chassis.
Before assembly, all panels are carefully cleaned with fine-grained sandpaper ("zero"), if possible, they are better chemically treated instead.
The false panel, the front panel of the amplifier, the power supply panel and the rear panel are interconnected with round-section ties pos.8 - pos.13, and the false panel and the front panel for beauty are attached to the screeds with half-hidden M5 chrome screws. At the installation site of the front panel of the power supply, the ties are interconnected with M5 studs, poz.14. The chassis and partition of the RF unit, the chassis of the power supply are attached to the couplers using M3 screws. The cross section of the ties can also be square, it’s just that the design was used ready-made from power supplies of the Minsk-32 computer.
All controls, indication and pointer devices are displayed on the front panel. On the rear panel there is only a “2 - 3” switch, which practically does not have to be used during operation. All connectors are on the rear panel. The drawing of the front panel does not show the mounting holes for variable capacitors, since their location depends on the specific type of capacitors used.
The top and bottom housing covers (option A) are attached to the front and rear panels with M3 screws with washers and engravers using bosses pos.15 or corners pos.15A. When manufacturing the case according to option B, the covers are interconnected on both sides with M3 screws with washers and engravers using straps pos.109, pos.110.
To give a marketable appearance, the false panel should be engraved in accordance with the drawing Fig.3. When making an amplifier at home, this work is done as follows: the panel is first thoroughly degreased, then heated on a regular home electric stove (tile), and finally, hot painted in the selected color with ML paint (preferably any color other than black, because no white ink). True, the inscriptions can also be made in red ink, but the contrast is lost). If possible, before painting, it is best to pre-apply a layer of primer. It is undesirable to use PF enamel for painting the front panel, it reacts with zapon varnish, and the coating begins to swell. Through the stencil with ink "Kalmar" inscriptions are applied in accordance with the option of the amplifier you have chosen. The inscriptions are then fixed with a colorless zapon varnish. It turns out a good look. The inscriptions can also be made using a transfer font (by the way, there is a white font). In this case, before applying a layer of zapon-lacquer, they must be fixed with hairspray (for example, "Charm"). There is another way. The drawing of the front panel with all the inscriptions is made in mirror image on a computer and printed on a laser printer, then it is applied to the false panel and carefully ironed through the fabric with a hot iron. Printed circuit boards are produced in a similar way. To get a good drawing on the board, the drawing must be run through the printer 2-3 times.
To ensure that there are no defects in the inscriptions, the panel must be smooth, and, in addition, it is best to practice beforehand. Painting the body can also be done with spray paints, which are sold in the automotive markets for touching up cars, but this is much more expensive.
The case is painted with PF enamel or another at your discretion. Figure 17 shows two options for making the case, you can use any of these options if you wish (by the way, the second option was born due to the lack of metal of great length).
No matter how great your desire is, do not rush to start the installation (I know from my own experience, but apart from extra work later, this did not work). Assembly and installation of the amplifier must begin only after the full completion of the entire scope of plumbing work and painting all the necessary parts and panels.
First, bushings poz.71 and mounting racks Ms3, Ms4 poz.62 are screwed to the PSU chassis from below with M3 screws with heads, then ties poz.13 are installed on top of the chassis, they are fastened from the bottom of the chassis with M3 screws through two bushings det. bushings are then installed printed circuit boards 1 and 2). Further, transformers Tr.1 - Tr.3 are installed on top of the chassis, and Tr1.- Tr2. fastened with M6 screws both to the front panel of the power supply unit and to the chassis, which is the basis for the rigidity of the entire structure, transverse rigidity is created by ties. For fastening to the chassis of the PSU bracket poz.113, with capacitors C24 - C27 installed on it in a transformerless version, the same holes are used as for fastening transformers. The Tr.3 transformer is fixed with M5 screws to the chassis. After that, the ends of the bundle are soldered to the terminals of the transformers and the bundle is refueled in the basement of the chassis, on top of the chassis through the insulating washers det. From the bottom of the chassis on the brackets det.21 (det.21A), the relay K1 (K1A) RES34 is (are) installed, then the back panel of the UM is screwed to the couplers with M4 screws, with connectors and fuse holders pre-installed on it (of course it should already be painted) .
The front panel of the power supply unit is installed, the strips poz.22 - 2 pcs., poz.23 (23A) with diodes and zener diodes, mounting racks Ms5 and Ms6 poz.62, bracket poz.78 and, if necessary, the K3 relay mounted on the bracket are pre-attached to it pos.105. The panel itself is attached to Tr.1 - Tr.2. In the upper part of the structure, the front panel of the PSU and the rear panel of the PA are interconnected using ties det.12. Next, the harness is desoldered in the basement of the chassis and to Pl1 and the SWR meter board, which are then fixed on the bushings pos.71, pos.73. After that, the screen of the SWR meter poz.77 is installed.
Couplers pos.8 - pos.11 are screwed to the ties pos.12, pos.13 with studs pos.14, to which the chassis of the RF unit is horizontally attached from below, with a panel pre-installed on it for a lamp (lamps), a choke L3 together with a capacitor C13 , variable resistors R22 and R23. Capacitor C13 is soldered at one end to the output of the inductor L1, and at the other end to the mounting tab poz.61, which in turn is attached to the chassis with an M3 screw. In versions C and E, the lamp panels are mounted on plates pos.76, pos.107, which are fastened to the chassis with M3 screws through bushings 74 from below, Ms1, Ms2 pos.60 and four mounting petals pos. .61. The partition of the RF unit is vertically mounted with a bracket poz.19 with relay K2 (RES9), relay K3 and bushing poz.20, which is also installed in advance, the bundle is unfolded, the wires of the bundle going to K3 are passed through the hole in the partition and soldered to the relay. The installation of the harness and hinged elements of the basement of the RF unit is in progress. The front panel is fastened with buttons S1 - S5, devices P1, P2 and lamps pre-installed on it with the help of bushings pos.72. On the P1 device, the PL.3 board is fixed, and the installation is carried out.
Capacitor C17 is attached to the partition of the RF block through the bushing with an M5 screw (preferably brass), the “hot” end of the L4 coil is screwed into it, the second end of the coil is fixed on the range switch, then variable capacitors are installed, after which the entire RF part of the amplifier is mounted.
The lamp blower fan is mounted on the bracket pos.76 (except for option D) through rubber gaskets, which reduces the fan noise level. The bracket itself is attached to the couplers pos.8 and pos.9 with M3 screws. For this purpose, holes for the M3 thread are additionally drilled in the screeds, two holes in position 8 and one in position 9 (one is already available). In Fig. 9A, the dimensions of the mounting holes are indicated for the installation of the VVF-71M fan; when using fans of other types, they must be corrected.
The legs are attached to the bottom cover of the housing with flush M4 screws. The top and bottom housing covers are attached to the front and rear panels, as well as to each other with the help of bosses pos.15 (15A). The lugs are screwed through washers with M3 screws.
An example of an assembly drawing of an amplifier having the "BYPASS" mode and the "3 - 2" mode is shown in Fig. 15BM and Fig. 16BM, respectively.
When making various personal changes and adjustments to the design of the amplifier, remember: RF installation is carried out along the shortest path, the installation site for blocking capacitors in any case should be located directly at the lamp grids.
Note. When manufacturing a transformerless amplifier with a built-in antenna switch, the following additional work must be performed:
a) in the board of the SWR meter, make a semicircular cutout with a radius of 3-5 mm for the output of the coaxial cable (see Fig. 16AB);
c) in the chassis of the power supply unit, the outer hole with a diameter of 6.2 mm, intended for fastening Tr.1, is drilled to a diameter of 10 mm (so that cable pos. 70 can pass through this hole);
c) drill a hole dia. 5mm through which jumper 38 will pass (see wiring table).
5. SETTING THE AMPLIFIER
I would like to write: "A device made from known good parts does not need to be adjusted." But, alas.
Before proceeding with the setup of the amplifier, it is NECESSARY TO MAKE SURE THE CORRECT INSTALLATION IS DONE. Acquaintance with high voltage, as a rule, does not bring much joy, and the observation of a long short circuit, too, again, you have to ventilate the room, and, as a rule, the most valuable burns out, and in addition, located in the most inaccessible place.
When setting up transformerless circuits, it should be remembered that they have two common wires. One is for the DC circuit, it is indicated on the diagram by the dot “0V”. All DC measurements should be made with respect to this point. Given that these circuits do not have galvanic isolation from the supply circuit, it is necessary to follow the rules of electrical safety during measurements (this, by the way, also applies to all other work). The common wire for the RF signal is the amplifier housing, and accordingly, all RF voltage measurements, if necessary, are made relative to it.
Amplifier tuning has no special features. Her sequence is this.
Pre-trained lamps are inserted into the sockets. Initially, the amplifier is configured without turning on the power supply. This is done with the help of a GSS and an RF voltmeter, or with the help of a GIR, or simply by ear using a receiver. The GSS is connected to the XP7 (Ant) connector, and the RF voltmeter is connected to the lamp anode. First of all, it is necessary to “lay down” the number of turns of the coils, so the setting starts from a range of 20 meters. If the resonance in this range is "somewhere close", try squeezing or pushing the turns of L4, otherwise it will be necessary to reduce the number of turns. On 20 meters, the L4 coil must be fully engaged. After that, on a range of 160 meters, the number of turns of the L5 coil is adjusted. Then the position of the coil taps on the remaining ranges is clarified and the possibility of tuning the P-loop is checked, and as the frequency decreases from range to range, resonance should be observed with ever-increasing values of capacitances C20 and C21.
At the next stage, the operation of the high-voltage part of the power supply is checked. To do this, contacts 1 and 4 of relay K1 are supplied with a reduced mains voltage (about 60 volts) through LATR, which will eliminate all sorts of surprises, and sequentially measure the voltage on pairs of capacitors C7, C8 and C5, C6; C3, C4 and C1, C2, if the voltages between adjacent pairs have a large spread, this means that either training (forming) of the capacitors is necessary, or their replacement. To train the capacitors, the PSU is kept on at an input mains voltage of 60 V for 5-6 hours, then measurements are taken again, if the spread has decreased, the voltage is increased to 150 volts, etc. If the voltage difference between any of the pairs of capacitors has not changed and is 20-30 volts (at U mains = 60 V) and accordingly increases with increasing mains voltage, then the capacitors, on which the voltage is less, should be replaced (or perhaps one from a pair), otherwise in the future they will still let you down. I had a case in the manufacture of the first amplifier, when a capacitor that fired pierced through six tickets for the Dynamo Kyiv international match, which were lying on a shelf above the amplifier (the amplifier was in the running-in stage and stood without a case). For this reason, in order to ensure the symmetry of the arms, it is advisable to purchase capacitors for use in the multiplication circuit in one batch and with a certain margin in quantity, and it is even better to check them for leakage beforehand.
Turn on the power according to the normal circuit and check the compliance of the voltage at the electrodes of the lamp (lamps). At the anode of the lamp should be about +1330V (+1260V for transformerless version), on the screen grid - +300 V, on the control grid - minus 100 V. If there is nothing to measure the high voltage, it is enough to measure it on capacitors C7, C8 and multiply the reading by four. By switching the amplifier to transmit mode, resistors R22 and R23 set the required quiescent current of the lamps in SSB and CW mode, respectively.
Let the lamps warm up for at least 5 minutes. After warming up, an antenna equivalent and an RF voltmeter (for example, type VK7-9) are connected to the output of the amplifier; in the detuned state of the output circuit, it should be 400 - 500 mA, and when the circuit is adjusted to the maximum output voltage, it should decrease to 300 - 350 mA, and the lamp used as a load should burn almost at full heat. If the anode current does not reach this value in any of the ranges, then the excitation power at the amplifier input is low. If, on one of the ranges, the anode current is normal, and the output power is low, although the anodes of the lamps turn red at the same time, and besides, there is no “excitation”, then the design of your anode choke turned out to be unsuccessful, the number of turns of the choke must be changed up or down by 10 - 15%.
When setting up (in the case of self-excitation) an amplifier made according to a transformerless circuit, you may have to experimentally select the installation location of the capacitor Cp, or dial it from several, placing them around the lamp panels.
At the next stage, turning on the transceiver in the tuning mode and gradually increasing the excitation voltage, the linearity of the amplifier is checked, i.e. correspondence of the increase in the output power of the transceiver to the increase in the anode current and the output power of the RA. The cessation of the growth of the output power of the RA with the continued growth of the anode current indicates "saturation", i.e. occurrence of grid current. In this case, you need to reduce the excitation power. Do not forget about this when working on the air, there will be no complaints about your "tails" from friends - colleagues on the air, and neighbors who love TVI will not climb the roof with wire cutters.
The SWR meter is adjusted with a dummy antenna connected. Switch S5 is set to the “SWR” position, a buildup is applied from the transmitter and by adjusting the capacitance of the capacitor C1, the division coefficient of the capacitive divider C1, C2 is changed so that the voltage amplitudes across the capacitor C2 and resistor R1 are equal. Since these voltages are reversed with respect to the VD1 diode, the current through the diode must be zero. If, by adjusting C1, it is not possible to set the arrow of the device P1 to zero division of the scale, then the terminals of the winding II of the transformer T1 of the SWR meter should be swapped. Then the connection points of the output of the PA and the equivalent are interchanged, and by adjusting C3, the arrow of the device P2 is set to zero division. Next, the connections are restored, the load is connected, the buildup is applied, and with the resistor R19 the arrow of the P1 device is set to the last division of the scale (if the readings of the device go off scale, it is necessary to reduce the number of turns of the secondary winding of the transformer T1 and, conversely, with a slight deviation of the device arrow - increase the number of turns). With a load resistance of 75 (50) Ohm, the pointer of the device P2 should be at zero division of the scale, which corresponds to SWR = 1.0. The load resistance is changed and on the scale of the instrument P2 the SWR value corresponding to this resistance is noted, etc. Everyone sets the upper limit for measuring SWR at their own request. Do not forget in the future, with each measurement by resistor R19, set the arrow of the device P1 to the last division of the scale.
Now you can start calibrating the device P2 for measuring power. To do this, turn on the transceiver in the setup mode, “squeeze” the maximum output power from the amplifier (in our case, it is 350 - 200 W), measure the voltage at the load and, using Table 5, find the maximum power of your amplifier corresponding to this voltage. By rotating the slider of the resistor R3 of the SWR meter, set the arrow of the device P2 to the last division. This division will correspond to the maximum power of the amplifier, while it is possible that R3 will have to be selected. Further, reducing the buildup voltage and controlling the voltage at the output of the amplifier, calibrate the rest of the scale of the device. In the future, when measuring, it should be remembered that the accuracy of the instrument reading when measuring power in a real antenna will be the higher, the better the SWR, i.e. the closer the resistance of the antenna you are using is to 75 (50) ohms.
The amplifier assembled according to option A is first configured without connecting a preliminary stage. Initially, the quiescent current of the lamps is set within 60-100 mA by selecting the number of zener diodes (VD11-VD14). For example, when using three D815A zener diodes, the quiescent current turned out to be 30mA, we short-circuit VD11 with a jumper, the quiescent current increases to 150 mA. We remove VD11 completely, and set the D815B zener diode as VD12, Io becomes 75mA, then again we change VD12 to D815A, and instead of VD11 we use a diode of the KD202 type or similar, connected in the forward direction, the quiescent current becomes 100 mA, if we add one more less diode KD202 (a place for installing this diode is provided on the bar det. 23A).
When using other types of zener diodes, it must be borne in mind that the maximum current through them at signal peaks can reach 1.0 A. The use of D815A zener diodes is due to the fact that in this case they can be used even without radiators. D815A zener diodes can be replaced with D815E-Zh zener diodes connected in parallel through limiting resistors with a resistance of 3 - 4 Ohm (the letter will be determined during configuration). Diodes are required only for forward current (it must be more than 1.0 A), since the voltage applied to them is negligible.
ONLY AFTER THIS OPERATION, you can start setting up the preliminary stage. This order allows you to set up quickly and without frustration. It can be seen from the diagram that the transistor is connected in parallel to the L6, VD11-VD13 chain, therefore, by slightly opening the transistor VT1, you can adjust the quiescent current of the lamps. Initially, the variable resistance engine R22 is placed in the position of maximum resistance. After that, turn on the amplifier and let the lamps warm up for at least 5 minutes. After warming up, R22 can be used to set the rated quiescent current. After tuning, if desired, the resistor can be replaced with a constant one of the corresponding rating. Further, turning on the transceiver in the setup mode and, gradually increasing the excitation voltage, check the linearity of the amplifier operation, with "saturation", i.e. the appearance of the grid current, you need to reduce the value of R23. If the anode current does not reach 0.5 A in any range, then the excitation power is low and the resistance R23 can be increased. When tuning, it should be remembered that tuning is carried out according to the maximum RF voltage on the antenna equivalent, or in the absence of such, using the simplest field strength indicator directly in the antenna itself. I do not recommend setting the amplifier to the maximum anode current, while you simply switch the stage to the DC amplifier mode, which will not correspond to the maximum output power mode.
The matching of the transceiver used for operation with the input of an amplifier built with a transformerless power supply circuit for anode circuits (for 2 GI-7B) is carried out by changing the number of turns of the III L6 winding according to the minimum SWR at the amplifier input and the maximum return.
When adjusting an amplifier assembled according to a push-pull scheme, first of all, it is necessary to achieve symmetry of the arms, i.e. equality of RF voltages on grids (cathodes) of lamps. If necessary, this is done by moving the midpoint of the windings I and II of the input transformer T1. Further, by changing the transformation ratio of the same transformer, an acceptable SWR value is adjusted at the input of the amplifier.
Table 5
During the tuning process of any of the amplifier circuits, it may turn out that the band of satisfactory matching with the load is insufficient to cover the working range of the range without adjusting the elements of the P-loop. The width of the matching band depends on the transformation ratio of the resistance in the matching circuit, and is the ratio:
N \u003d Re / Rn, where Re is the equivalent resistance of the output stage lamp, and Rn is, respectively, the load resistance.
Thus, when using a coaxial cable with a standard wave impedance of 50-75 Ohm to power the antenna, in our case we obtain a transformation ratio of approximately 20. Its value can be reduced (accordingly, the band of satisfactory matching will expand) by increasing the load resistance, i.e. using a transformer with a transformation ratio of 4: 1 at the output of the P-loop to match the load with the SHP (Fig. 2.12). However, it should be borne in mind that with an increase in the value of Rn, the voltage on C21, while maintaining the output power, will also increase, so the gap between its stator and rotor plates must be increased. This transformer is made similar to the L6 choke (it is better to take rings for winding it with a permeability of 600-1000, rings from self-propelled guns for r / st. R-130 are suitable) and is installed on the front panel of the PSU (Fig. 15BM, Fig. 16BM). The mounting of the transformer is also similar to that of the L6 choke.
Now some tips:
1. Amplifier does not "like" work with high SWR (≥2.5), in this case, the power is redistributed and part of this power goes to the throttle, breaking it.
- The amplifier should preferably be grounded. During operation, there were cases when, during strong lightning discharges, the mains fuses “knocked out”. This happened when open type antennas (LW, Inv.Vee, V-Beam, etc.) were used with the amplifier, which were not turned off during a thunderstorm.
- Do not try to get more output power from the amplifier than the overall power of the power transformers you used allows, although the transformers used in this design are naturally made with a margin, but according to the existing (and not canceled by anyone yet) Murphy laws - just at the most crucial moment (for example when you call P5 or, in extreme cases, FO0), they will order you to live long (and he is about to answer, and it even seems to you!). In addition, due to the anode voltage drop, this will primarily affect the signal quality of your radio.
- Do not be lazy, do all the work, and then you will always have something to present and something to be proud of in front of the amateur radio fraternity, because their amplifier only works without a case, and even standing on its side, and before turning it on, you still need to hit it in a certain place fist (and maybe more than once!).
- The amplifier is intended for carrying out intercontinental communications, tk. for local it is necessary to use at least GU-5B, well, in extreme cases, GU-81 is allowed (but at 3 kV at the anode). The essence of this is as follows: all the gain knobs on the receiver are immediately set to zero, while, firstly, even the most primitive detector receiver absolutely stops making noise, and secondly, no matter how much the rest of the “wise men” break at the frequency, they will not interfere with your pleasant conversation about the prospects for the harvest, etc. And all the others, because of the “offal”, will themselves scatter in the direction of about fifty or a hundred kHz, they also need to hear someone.
- If the amplifier you assembled with love and knowledge of the matter did not work, it immediately becomes clear that naturally the author is a fool and his scheme is stupid, and genetics, again, of course, has nothing to do with it. Of course, it's all humor.
And most importantly, what should never be forgotten - the amplifier will only be a good helper if it is an attachment to a good antenna.
The above is not a dogma, but a practical guide to work, suggesting the direction of design, in the process of creativity, you can increase the overall dimensions, change the layout depending on the details you used, your imagination, capabilities, experience, etc., in short - create like we do do better than us. Most importantly, I ask you to write about all the wishes, criticisms and errors you have found, both theoretical, technical, and spelling. The material was prepared and edited by virtually one person, so it was almost impossible to keep track of everything. Thanks a lot in advance.
In conclusion. I would like to express my deep gratitude to A.S. Rozhnov UY0UZ and V.V. Yurchenko UT5UAO for the great technical assistance provided by them in the manufacture of individual prototypes of amplifiers and conducting experiments to develop their individual components.
Viewed: 18 525
Loading...