Home > Basic Circuits >Amplifying Circuits > Single differential simplest OCL amplifier-----simple OCL amplifier

Single differential simplest OCL amplifier-----simple OCL amplifier

Source: InternetPublisher:明天见 Keywords: Power amplifier circuit Updated: 2024/06/28

As the performance of modern power amplifiers continues to improve, the circuit structure is becoming more and more complicated. This is the most headache for amateur producers, especially beginners. Here we introduce a simplest power amplifier circuit to see how simple it can be and what kind of performance it can achieve. This is also an interesting question.

Circuit Principle and Performance

 (1) Circuit analysis

Figure 1 is the circuit diagram of this power amplifier. There are only about 20 components in the power amplifier, including transistors. At first glance, it looks like a schematic diagram, but it is indeed a practical power amplifier, and it can provide ≥50W (120W) output power to 8Ω (4Ω) load with low harmonic distortion. It uses a typical OCL circuit, but necessary improvements have been made to the design according to practical conditions during production.

 

 

The input stage BG1-2 is conventionally a differential amplifier stage, but it is slightly different from the common circuit in that it uses a PNP tube. Compared with the NPN tube, the two tubes are easier to pair and have better consistency and lower noise. For a simple circuit structure, this needs to be considered as much as possible.

The second stage BG3 is the main voltage amplifier stage, which provides most of the voltage gain, but does not use the common "bootstrap" circuit. The use of a "bootstrap" circuit in a high-power amplifier does not have much significance in increasing the output power, and it can save an electrolytic capacitor that affects the sound quality, and is conducive to reducing components and simplifying the circuit. C2 is a phase compensation capacitor.

The final stage is a complementary output stage composed of BG4-7 in the simplest way. There are few components and no adjustments, so the use of a smaller power push tube BG4-5 is sufficient to meet the requirement of pushing the final stage output of more than 100W. The setting of the final stage quiescent current is mainly to reduce the crossover distortion at low output power, usually 40-50mA. As for the crossover distortion at high output power, due to the "masking" effect, the impact is not obvious. There is no thermal compensation for the quiescent current. When working, the quiescent current rises accordingly with the increase of temperature, but there is no out-of-control during the trial. This can simplify the installation process and reduce the debugging procedures. In addition, a slightly larger quiescent current can also reduce some crossover distortion at high output. C3 is used for high-frequency decoupling of the power supply.

The gain of this machine after adding overall negative feedback is about 20 times (26dB), but it can also work well after canceling the overall negative feedback. The full power output waveform is still symmetrical. No waveform distortion is observed by oscilloscope. The harmonic distortion is tested by distortion meter. Compared with adding negative feedback, the increase is not much (only about 0.2%). It can be seen that the open loop performance of this machine is good. The purpose of negative feedback is mainly to compensate for the discreteness of BG3 parameters and ensure the stability of the overall gain, rather than mainly to reduce distortion.

Since most preamplifiers have DC blocking capacitors at the output, the DC blocking capacitors at the input of this unit can be omitted. When the preamplifier (or CD player) adopts a DC output design without output DC blocking capacitors, it can also be directly coupled to this amplifier. Just pay attention to the output midpoint voltage of this unit after the preamplifier and the preamplifier are connected. It will not have any effect on the circuit performance and working reliability. Modern CDs can record ultra-low frequency signals as low as several Hz. If you don't want these signals to be lost during playback, it is necessary to consider the problem of using direct coupling.

(2) Performance indicators

The circuit in Figure 1 may look inconspicuous, but its performance is impressive. The actual performance of the prototype is as follows:

When BG6 and 7 select different complementary high-power tubes, their rated output power (RMS) is:

 2N3055/MJ2955 (35V, 8Ω) 50W,
 MJ802/MJ4502 (35V, 4Ω) 75W
 MJ802/MJ4502 (37.5V, 8Ω) 62W
 MJ802/MJ4502 (37.5V, 4Ω) 112W
 MJ802/MJ4502 (35V, 4Ω) 131W
 Total harmonic distortion:
 rated output power 1kHz 0.35%
 10W (RMS) 1kHz 0.015%
 Signal-to-noise ratio 115dB
 Power bandwidth (-3dB) 100kHz
 Frequency response (1W, 1dB) 2Hz-110kHz

 Damping coefficient (8Ω) 90     

It is not difficult to see that the circuit of this machine has strong versatility. As long as the corresponding output tube and power capacity are used, the output power of 50-100W can be obtained without changing the circuit. The circuit is so simple that it is almost impossible to reduce a component, but the performance is above average. Compared with some high-end amplifiers on the market, the sound quality of this amplifier is satisfactory.

How can such a simple circuit have lower distortion and better sound quality? I have not studied it in depth. My initial guess is that it is due to the arrangement of the circuit structure, which makes the distortion of the front and rear stages compensate each other, so the open-loop distortion of this machine is lower.

This lays the foundation for obtaining good sound quality. If this theory is correct, then in a simple power amplifier circuit, some attempts can be made to match the front and rear stage components, which may be a way to reduce open-loop distortion. Interested readers may wish to try it further.

2) Selection of main components

(1) Power supply

The power supply of this machine with different output powers uses the same circuit program. (See Figure 1), only the capacity of the power transformer and filter capacitor is different. Table 1 shows the recommended power capacity and filter capacitor size.

The values ​​in the table refer to the power supply conditions that should be met when a single-channel power amplifier is used. For example, for stereo power amplifiers, the secondary current capacity and filter capacitance of the power transformer should be at least doubled. For example, the secondary capacity of the 050WX2 power transformer should be 28VX2/4.5-5A1100WX2, and the secondary capacity of the power transformer of the power amplifier should be 28VX2/7-8A. If the current capacity is not increased accordingly, then when one power transformer is used to supply two power amplifiers to work at the same time, the output power will not reach the original rated value. Of course, it is best to use a power transformer and corresponding rectifier filter circuit for each power amplifier, which is also conducive to reducing crosstalk between channels.

Table 1

rated powerTransformer powerFilter capacitor (minimum value)No-load voltage
50W28V×2/2A5600μF/40V±40V
100W28V×2/3A10000μF/40V±40V

(2)
The main specifications of the two complementary high-power tubes of other devices are shown in Table 2.
Table 2

Output PowerNPN/PNPIcMVceoPch FE (Ic)
50W2N3055/MJ295515A60V115W20~70
(4A)
100WMJ802/MJ450240A100V200W25~100
(7.5A)

The f T of these two pairs of high-power tubes is about 2.5MHz, but because of the large current and the fact that they are both switching power tubes, they have good performance when used as audio power amplifiers. In particular, 2N3055/MJ2955 is the first choice for making high-fidelity power amplifiers below 50W, and is a pair of "quasi-fever tubes" with a very high cost performance. The Vceo specification value of this tube is relatively low, and is often greater than 100V in reality. There should be no problem using it under a voltage of 40V. However, dealers often select and supply it in grades according to Vceo of 60V or 100V. At this time, do not purchase it for use with 60V.
The Pcm of BG1-5 (BC557/546) is 0.5W, and the Icm is 100mA. Among them, BG3 requires Vceo≥50V, and BG4-5 requires Vceo≥100V. The f T is ≥100MHz, and the h FE is 110~220. BC557 and BC546 are difficult to purchase, and corresponding high-frequency low-power tubes can be selected according to the above requirements.
 To improve the open-loop linearity, differential tubes and complementary tubes should be used in pairs, which can also reduce debugging troubles and make it clear. The matching requirements are as follows: Differential tube BG1-2 requires that the h FE difference at 0.5mA is less than 5%, and the V BE difference between the two tubes is less than 20mA . The higher the h FE , the better, try to ≥100. Complementary push tube BG4-5 requires that the h FE
 of the two tubes is as close as possible within 1-80mA . If not, it can also be selected according to the basic consistency of h FE at 60mA . The final stage BG6-7 requires that the h FE  of the two tubes is as close as possible within 100mA-3A (for 50W power amplifier) ​​and 100mA-4A (for 100W power amplifier) . If not, it can also be selected according to the basic consistency of h FE at 3A or 4A . The main voltage amplifier tube BG3 requires that the h FE is not less than 60  at 10-15mA . Of course, it is better if the BG3 of the two-channel power amplifier can be basically consistent. D1-2  is a switching diode 1N914, with a maximum forward current of 75mA and a forward voltage drop of about 1V at 10mA. It can be replaced with 1N4148.  R7-8 uses a 2-5W carburized resistor dedicated to the power amplifier, or it can be replaced with two 0.68Ω (1-2W) metal film resistors in parallel. The rest are all 0.5W metal film resistors. C1-3 uses CBB polyester non-inductive capacitors, C1 can also use general electrolytic capacitors, and C2-3 requires a withstand voltage of ≥100V.  The heat sink uses a profile, and its thermal resistance (to the corpse-path power amplifier) ​​is not inferior to 8-10~C/W. It is recommended to use a larger heat sink as much as possible in production, because the life and reliability of the power amplifier tube are closely related to the tube junction temperature.




3) Installation and debugging
(1) Installation process
 The components of the power amplifier are all installed on the printed board shown in Figure 2. It should be noted that Figure 2 shows the front component installation surface. When making a printed board, do not forget to copy it upside down. In addition, since the size of components selected by each person may not be the same, it is necessary to try to arrange the actual objects on the diagram before making the printed board, and adjust the wiring spacing appropriately to ensure that the components are arranged evenly and beautifully. The production of the power amplifier printed board in the following text is the same.
 The final stage BG6-7 is installed on the printed board through an L-shaped aluminum bracket (60×45×85mm). The installation requirements are shown in Figure 3a. It should be noted that a thermal insulation sheet should be placed between the tube shell and the L-shaped bracket. The tube shell is only connected to the corresponding copper foil on the printed board through the fixing screw A (preferably by welding), and the fastening screw B should not have any electrical contact with the copper foil of the printed board. The heat sink can be directly fastened to the L-shaped bracket, and the thermal contact surface is preferably coated with silicone oil.
 There are not many components in the power rectifier and filter part, which can be directly installed on the chassis with a bracket. Figure 3b is a wiring diagram of a channel, which only points out the following points:
① The "ground" line terminal on the printed board should be directly connected to the common connection point of the secondary center tap of the power transformer and the two filter electrolytic capacitors. This point can be directly welded to the metal chassis nearby or connected to the chassis with another thick wire to achieve "one-point" grounding. The ground terminal of the speaker is directly connected to the above-mentioned common grounding point without passing through the "ground" on the printed board.
② The signal input ground terminal and the power amplifier output ground terminal should be insulated from the metal chassis at the installation location.
⑧ The wiring of the power supply part should be short and thick. It is recommended to use Φ0.2mm×32 multi-strand plastic wire to connect. The speaker terminal is connected to the printed board and the common tapping point with a dedicated feeder for the sound box.

(2) Debugging
Before debugging, check whether the components, installation and welding are correct and reliable. Pay special attention to whether the polarity of diodes, transistors, and electrolytic capacitors is reversed, and whether the insulation between high-power tubes and heat dissipation brackets is good. After heating, check the power supply part separately. If there is no problem, then connect the amplifier for debugging. According to the load of the amplifier, it is divided into the following three steps:
① No-load debugging In order to reduce the possibility of instantaneous damage to the amplifier, do not connect the load first. After the power is turned on, touch the end tube with your hand. It should be slightly warm, or one tube is hotter and the other tube is cooler. As long as it is not hot and there is no other abnormality, you can safely measure the voltage at various places, the debugging point voltage and the static current. Use the DC voltage (200mV) range of the digital multimeter to measure the output midpoint voltage.
Generally, if it is below ±50mV, it can be considered normal. If the positive bias is too high, increase R2, otherwise reduce R2. As long as the differential tube is selected, it is usually easy to control within ±50m.
 Then test the voltage drop across R7 or R8. Since there is no load, the voltage drop on these two resistors is the same. The static current is 40~50mA, the corresponding voltage drop on R7 or R8 should reach 13~17mV, and the voltage between the bases of BG4-5 is about 2V. If there is no voltage drop on R7 or R8 or it is less than 13-17mV, you can test the voltage drop of D1 and D2 respectively, try to solder off the diode with smaller voltage drop (after disconnecting the power supply) and replace it with a diode with larger voltage drop and retest. If there is no corresponding diode, you can replace it with a 220Ω trimming resistor and fine-tune it until the voltage drop on R7 or R8 reaches 13-17mV. On the contrary, if the voltage drop on R7 or R8 is too large, you can use a 220Ω trimming resistor in parallel with D1 or D2 and fine-tune it until the voltage drop on R7 or R8 reaches 13-17mV, retest the midpoint voltage and adjust R2 so that the midpoint voltage is within 50mV.
② Pure resistance load debugging Connect 8~10Ω 1/4W resistor to the output of the power amplifier, and then measure the voltage drop across R7 and R8. At this time, since BG6 and 7 form their own independent DC circuits through this resistor, the voltage drops on R7 and R8 may be inconsistent. At this time, R2 can be adjusted to make the voltage drops on these two resistors consistent, and the midpoint voltage is close to 0V. If the voltage drops on R7 and R8 are the same but do not reach 13~17mV, D1 and D2 can be replaced or the above 2200 fine-tuning resistors can be fine-tuned 1~2 times. In short, the voltage drops on R7 and R8 should be the same and reach 13~17mV, and the midpoint voltage is generally adjusted.
⑧ Actual load debugging After the above debugging, the sound box can be connected for debugging. The feeder and its length connecting the sound box should be matched according to the actual use in the future as much as possible. After powering on, do not send a signal first, listen to whether there is obvious abnormal sound in the speaker, and touch the final tube shell with your hand. If there is abnormal sound in the speaker and the tube shell is hot, it means that the circuit is self-excited after connecting to the actual load. At this time, as shown in Figure 4, the R1, C1 series compensation network and the L1, R2 parallel compensation network can be connected to the output end of the power amplifier, and a bracket should be used to install it as close to the printed circuit board as possible. Generally, R1 and Cl are used first. If the self-excitation cannot be completely eliminated, L1 and R2 are added. L1 can be made of Φ1.0mm enameled wire wound 10 times on a 12mm frame and then cast. Generally, the self-excitation can be eliminated after the above measures are taken.
Finally, add a signal and listen to it for about 1 hour at a level slightly higher than the normal listening level (pay attention to the temperature of the final tube shell at any time), then re-measure the output midpoint voltage and quiescent current. Generally, the midpoint voltage is <±100mV, the quiescent current is <100mA, and there is no continuous increase. It can be considered that the debugging is completed. This step of debugging is best performed in the summer when the temperature is higher. If the machine is installed in winter, it is hoped that it will be re-tested in the summer.

EEWorld
subscription
account

EEWorld
service
account

Automotive
development
circle

About Us Customer Service Contact Information Datasheet Sitemap LatestNews


Room 1530, 15th Floor, Building B, No.18 Zhongguancun Street, Haidian District, Beijing, Postal Code: 100190 China Telephone: 008610 8235 0740

Copyright © 2005-2024 EEWORLD.com.cn, Inc. All rights reserved 京ICP证060456号 京ICP备10001474号-1 电信业务审批[2006]字第258号函 京公网安备 11010802033920号