Design of super bass in active speakers

Design of super bass in active speakers

Generally, sub-bass refers to the audio frequency below 120Hz, and heavy bass refers to the audio frequency between 220Hz and 100Hz. Whether it is popular music or classical music, whether it is natural music or the sounds in nature, heavy bass and sub-bass are rare, but like MSG, they are indispensable. Without it, the music will lack a sense of presence.

The power of the subwoofer is very strong, but the directionality is not obvious, so the subwoofer system.

It is usually monophonic. Except for the latest digital recording system where the subwoofer is recorded as an independent channel, the subwoofers we hear now are separated from the full-band sound signal. Therefore, the circuit of a general 2.1 active speaker generally includes power supply, preamplifier, frequency division circuit, power amplifier, and speaker (speaker), as shown in Figure 1.

1. The power of the subwoofer channel
is either single-order or multi-order. The frequency curve after the crossover point is inclined instead of a steep right angle. The higher the order, the greater the slope.

In order to obtain a flat frequency response, the frequency coverage between each speaker has a certain overlap.

At the same time, the impedance curve of the cone speaker is also nonlinear, with the maximum at the resonant frequency and the minimum at 100-500Hz. As the frequency increases, the impedance increases accordingly. This means that for signals of different frequencies, even if the power input to the speaker is the same, the sound intensity emitted by the speaker is different. In order to make the sound pressure generated by each speaker sound roughly the same as the original music signal to the human ear, the output power of the power amplifier driving each speaker cannot be equal, and there must be a certain proportional relationship.

For example, a 100W three-way system with a crossover point of 400Hz and 3kHz would require 50W for the bass channel and 35W for the midrange.

The treble is 15W, and this ratio must be used for overall design, and the insertion loss of the RC filter and the gain of the amplifier must be adjusted to achieve balance in the entire system. For a subwoofer system, the power of the subwoofer channel is about 10 times greater than that of the main channel, so the amplifier of the subwoofer channel requires a large power margin.

2. Frequency division network
2.1 The frequency division circuit of active speakers can use active RC frequency division, passive RC frequency division, or active power frequency division.

(1) Passive RC crossover
The passive RC electronic crossover has a simple circuit and has the flattest amplitude-frequency and phase characteristics. The phase distortion and transient distortion are very small. The disadvantage is that the out-of-band attenuation characteristics are not good. It is only 6dB/oct. This requires the speaker to have good linearity outside the frequency turning point. That is, the subwoofer is required to have good linearity above 120Hz and the mid-bass speaker is required to have good linearity below 120Hz. 2.1 speakers use small-caliber speakers, which can make up for this shortcoming to the greatest extent.

　　Figure 2 is a first-order RC filter circuit.

The crossover frequency is selected at 120Hz, and 10dB of attenuation is added to the main channel to balance the main channel and the subwoofer channel. In practice, the crossover frequency and attenuation can be adjusted. The input impedance of the RC crossover network is required to be much larger than the output impedance of the signal source, and the output impedance of the RC crossover network is required to be much larger than the input impedance of the power amplifier.

The output impedance of the signal source is generally less than 1kΩ, the input impedance of the power amplifier is generally 47kΩ, and the input and output impedances of the RC frequency division network are equal to the network resistance R. In Figure 2, the sum of R1-1 and R1-2 is equal to the network resistance R, and the resistance of R2-1 and R2-2 in parallel is also equal to R, which is 10kΩ.

The calculation formula for the crossover frequency is:

f0=1/(2πRC)= 160/(RC)

The network capacitance C in the formula is equal to the sum of C1-1 and C1-2. R1-1 and R1-2 are both network resistors in the frequency division network and attenuation resistors in the attenuation network, which are the same as R3-1 and R3-2 in the right channel. When adjusting the attenuation, the sum of the two should be kept unchanged.

If R1-2 is adjusted to a relatively small value and still cannot meet the needs, the gain of the power amplifier can be adjusted at the same time to meet the needs.

If you want to obtain a steeper out-of-band attenuation characteristic, it is best to use an active RC crossover network.

(2) Active RC frequency division
Active RC frequency division circuits mostly use second-order Butterworth filters with an out-of-band attenuation characteristic of 12dB/oct. If you want to obtain a steeper drop characteristic, you can use a two-stage series connection to obtain an out-of-band attenuation characteristic of 24dB/oct. In audio applications, the Butterworth filter requires a quality factor (Q) of 0.707, and the C value in the RC network is preferably not greater than 0.1μF. Figure 3 is a typical circuit with a gain (A0) not equal to 1, and Figure 4 is a typical circuit with a gain (Ao) equal to 1. To illustrate the problem, the components in the figure are based on a transition frequency (f0) of 120Hz.

For a typical circuit with a gain (Ao) not equal to 1, R1=R2 and C1=C2 can be set for convenience of calculation under amateur conditions; for a typical circuit with a gain (A0) equal to 1, R1=R2 and C1=2C2 can be set. In this way, the Q value of the circuit will be independent of the values ​​of R1 and C2, and the calculation and adjustment of the circuit will be more convenient. A bandpass filter can also be used in the main channel to filter out the influence of super-audio signals.

Figure 5 shows such a second-order bandpass filter circuit.

　　Figure 6 is a 2.1 speaker frequency division circuit with input isolation, which uses six dual operational amplifiers NE5532 to complete input buffer isolation, separation of heavy bass signals, separation of main channel signals and subwoofer signals, output buffering, etc.

The performance is perfect, but this circuit is relatively complicated, and if it is not made rigorously, it will be counterproductive.

In Figure 6, IC1 B and IC2B separate and buffer the signals of the left and right channels, synthesize them, and send them to the second-order low-pass filter composed of IC3B to complete the separation of the subwoofer signal.

IC1A and IC2A separate and buffer the left and right channel signals and send them to the bandpass filters composed of IC4A and IC5A respectively to filter out the super-audio signals.

At the same time, it plays a delay role to ensure that the phase is accurately synchronized with the subwoofer channel, which is a major feature of this circuit. Then the signal is sent to the inverter composed of IC4B and IC5B, and finally sent to the subtractor composed of IC6A and IC6B to be added with the separated subwoofer signal. The inverted full-frequency signal and the subwoofer signal are added to obtain the left and right channel signals without the subwoofer signal, which are clean and pure.

This is another major feature of this circuit. This frequency division method basically does not have phase interference phenomenon, and the frequency division curve will not produce the phenomenon of intersecting after the frequency division points fall separately, and will not cause the synthesized curve to have peaks and valleys or peak drums.

The circuit in Figure 6 has two other features. First, a phase conversion circuit composed of IC3A and SW1 is added. IC3A is an inverter. When SW1 is upward, an inverted signal is obtained; when SW1 is downward, an in-phase signal is obtained. This allows for better coordination with the main speaker. Second, a low-frequency signal upper limit frequency cutoff point selection function controlled by SW2 is added. That is, the turning point of the cutoff frequency of the second-order Butterworth low-pass filter composed of IC3B is controlled: 100Hz, 125Hz, and 150Hz. This method improves the circuit's adaptability to the speaker. It is also a way to adjust the musicality. For the independent power receiver that has appeared recently, it should be a good choice.

(3) Active power division
　　  
This is an efficient frequency division method that combines power amplification and active frequency division. This method does not require an op amp as an active device for step-by-step division. Theoretically, as long as the power amplifier IC has two input terminals with opposite phases and adjustable gain, it is feasible. For example, our common TDA2030A, LM1875, etc. It is not suitable for fixed-gain power amplifier ICs such as TDA1517, TD7370, and TDA8946. Figure 7 is its typical circuit. This method actually uses the power amplifier IC as a high-power op amp. According to the component values ​​in the figure.

The frequency turning point is 900Hz, and the out-of-band attenuation characteristic is 18dB/oct. It should be noted that when doing active power division, the resistance value of R6 should not be too large.

Figure 8 is a 2.1 circuit using active power crossover. According to the parameters in the figure, the subwoofer channel can provide about 56W of power. The main channel can provide about 20W of power per channel, and the total harmonic distortion (THD+N) is less than 0.05%. The crossover point is selected at 220Hz, which can be adjusted according to the actual situation of the speaker. The components of the crossover resistor and capacitor network are selected with an error of less than 1%.

3. Power amplifier circuit The
　　
power amplifier circuit used in common 2.1 speakers is mostly an integrated power amplifier. The main performance indicators are shown in Table 1. The working voltage in the table refers to the typical working voltage. The output power column is expressed as "power per channel × number of channels (connected to load impedance)". The loss (THD+N) is calculated based on the nominal value in the power column, not the value under 1W. The data in the table are all from the data files published by the manufacturer. Generally, integrated circuits that work with dual power supplies can work in a single power supply state. Integrated circuits with dual channels can generally be connected in the form of BT1 to obtain greater power.

The author believes that although the sound quality of digital amplifiers cannot completely match that of analog amplifiers, they are at least more suitable for subwoofer channels.

Table 2 contains the most commonly reported digital amplifiers.

Regarding the output LC filter, some products use special technology and do not need coil inductors and ceramic capacitors, but only rely on a magnetic bead.

4. Speakers
　　
In the 2.1 system active speakers, the subwoofer generally uses a 5-inch long-stroke speaker, the box is relatively large, the power supply and circuit parts are installed inside the box, and the stereo mostly uses a small-caliber full-band speaker. There are also a few varieties that use two-way frequency division.

Speakers can be divided into open, closed, inverted, horn, and other types of deformation based on various structures. The author believes that the combination of inverted and labyrinth, and the push-pull form of the speaker can achieve the effect of taking both volume and performance into consideration. The inverted form improves the efficiency of the bass, and the labyrinth extends the path of the sound, which is equivalent to increasing the volume of the speaker in disguise. The push-pull mode of the two speakers undoubtedly increases the energy of the bass. Figure 9 is a conceptual diagram of this speaker. This method is also called the dual drive mode. Mitsubishi Corporation has successfully used it in the subwoofer system of TV sets. Two identical speakers are used in the speaker. Inverted connection. The impedance is half of that of a single speaker, and the resonance frequency is the same as that of a single speaker. In theory, this method can reproduce the same frequency of low frequency. Only half the volume of a single speaker is needed. It can also offset homogeneous nonlinear distortion. The disadvantage is that the design and adjustment are very complicated, so it is easier to achieve mass production by using plastic materials with very good performance consistency.