KR19980073677A - Self-oscillating delta modulation class D acoustic amplifier - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an acoustic amplifier using a class D amplifier. Specifically, the present invention relates to a delta modulation method having a high efficiency, which is an advantage of the class D amplifier, while securing low distortion by adding a voltage negative feedback means to the class D amplifier. It is a self-oscillation class D acoustic amplifier.

The class D acoustic amplifier of the present invention configures a delta modulated class D amplifier to be self-oscillating to secure two degrees of freedom of variable frequency-variable duty, thereby providing a separate modulated carrier signal (sawtooth wave or triangle wave). It is possible to adjust the change of device's element value or characteristics by self switching by variable frequency-variable duty without generating circuit, so that the maximum output voltage is almost externally applied voltage (±) Vdd) up to full swing in symmetry. In addition, energy savings are possible due to low distortion (near 0.1%) and high efficiency (90%), and there is no problem in that the signal is cut off even at full swing of maximum output, and audible frequency from 20Hz to 20kHz While the gain characteristics are reputed in the band, the 3dB band width is also present in the vicinity of 30 kHz, so the response characteristics are good. As the secondary system, the entire system is always stable and there is no stability problem. In addition, the use of a very small number of parts can be manufactured at a low price, thereby improving competitiveness.

Description

Self-oscillating delta modulation class D acoustic amplifier

The present invention relates to an acoustic amplifier using a class D amplifier. Specifically, a self-oscillating delta modulation (self-) having high efficiency, which is an advantage of the class D amplifier, is secured by adding a voltage negative feedback means to the class D amplifier. oscillation delta modulation) Class D acoustic amplifier.

In general, amplifiers used as acoustic amplifiers can be roughly divided into class A, B and AB analog amplifiers and class D digital amplifiers.

In this acoustic amplifier, linearity has been emphasized more than high efficiency for a long time, and acoustic amplifiers have excellent linearity in class A, B, and AB until recently because of the atmosphere and technology lack of consideration of power loss. Class analog amplifier is adopted.

However, implementing a large output amplifier with an analog amplifier is inefficient, resulting in huge power losses. In addition, since the energy of the amplifier is changed to heat and all of the energy other than the energy is converted into heat, the temperature of the amplifier is increased, so that a massive heat sink is inevitably required to force cooling of the amplifier.

The most obvious of these characteristics among analog amplifiers is the Class A amplifier. Class A amplifiers have a loss that is greater than the amplifier's maximum output and have a structural disadvantage that the efficiency does not exceed 25%.

On the other hand, the push-pull class B amplifier, which is adopted to improve this, uses two transistors in the form of emitter follower to reduce energy loss. It can get up to%, but the disadvantage is that crossover distortion occurs at relatively high levels but at a small signal level. Applying the appropriate negative feedback to such a class B amplifier can improve the crossover distortion that occurs when the signal is small to some extent, but the harmonic distortion (THD) becomes worse when a large current must flow at a large voltage. It cannot be solved completely.

The reason depends on the structure of the class B amplifier. In other words, the two transistors in the class B amplifier are alternately turned on and off, so it is easy to turn on and off when a small current flows, but quickly turn on and off when a large current flows. It's hard to be. In other words, since the class B amplifier does not normally flow any bias current at all, especially in a large current region, fast on / off is difficult, resulting in poor harmonic distortion.

Class A amplifiers, which are intermediate between class A and class B, have some current flowing in the absence of a signal. The magnitude of this current is much smaller than that of class A amplifier, and is much larger than class B. Therefore, the more the bias current flows, the closer the characteristic is to Class A. On the other hand, the less the bias current flows, the closer the characteristic is to the amplifier.

In the case of the analog amplifiers of the class A, B and AB class, there is a difference in loss amount, but since 21.5% to 75% of the energy applied is eventually lost to heat, a heat sink or a cooling fan is required for heat dissipation. In addition, the addition of the heat sink and the cooling fan may cause a problem of noise pollution of the fan in addition to the problem of increase in volume. Acoustic amplifiers containing these problems are also harmful when used for home use. However, assuming that especially in the case of a closed space, for example a car acoustic device, excessive heat loss can cause the temperature of the sound equipment to rise in the space inside the closed car. This can lead to a shortening of the life of the sound equipment.

Therefore, an important point for all acoustic devices such as home, meeting room, and in-vehicle is not only excellent linearity (or harmonic distortion), but also to be efficient in order to reduce heat loss, and to be as small as possible.

As an amplifier that can solve the problems of the analog amplifier, that is, an amplifier capable of providing good efficiency and a small volume, a class D amplifier, which is a digital amplifier, can be considered.

Class D amplifiers adopt a pulse width modulation (PWM) method that amplifies by switching rather than linear operation. The gate signal of the control pulse, that is, the power field effect transistor (FET), is usually A sawtooth carrier signal having a fixed frequency (several hundreds of kHz) is generated through comparison of a control reference signal (error signal of a speech signal). Distortion exists in acoustic amplifiers using Class D amplifiers because they perform amplification using nonlinear operation by switching, which can be technically corrected to be of sufficient value as an acoustic device by using appropriate negative feedback. Can be.

The principle of an acoustic amplifier using a class D amplifier is the same as that of a switching regulator or a PWM converter, except that the class D amplifier for an acoustic device is about 20 Hz to about 20 Hz to that of the switching regulator or PWM converter. It has a band width of 20 kHz wide audible frequency band.

Class D amplifiers typically employ a high-power FET as a high-power switch, ideally theoretically 100% efficient, but in reality heat loss proportional to the switching frequency and power in various control circuits. As consumption occurs, the efficiency can be expected to be around 90%.

Since the harmonic distortion of the class D amplifier is bad, it is necessary to handle the negative feedback circuit well so that sound quality as an acoustic device can be sufficiently secured.

However, in the acoustic apparatus using the conventional class D amplifier shown in Fig. 1, since the digital circuit portion for performing amplification using pulse width modulation switching and the analog circuit portion accompanying it are mixed, switching noise in the circuit is mixed. There are many, and it is quite difficult to design and add a negative feedback circuit with sufficient stability. Incorrect processing of the negative feedback circuit may cause the circuit to become unstable and cause oscillation.

As an invention for solving the above-mentioned problems included in an acoustic apparatus using a class D amplifier, the present inventor has already filed the invention of a triple negative feedback class D acoustic amplifier (Korean Patent Application No. 96-37905). In the above-described invention, in the acoustic amplifier using the pulse width modulation switching class-D amplifier, it is possible to provide a triple negative feedback acoustic amplifier having improved characteristics of low harmonic distortion (0.1% or less) while improving efficiency and high power. It was for the purpose.

However, a general pulse width modulation class D amplifier having a switching operation by comparing a sawtooth carrier signal having a fixed frequency (several hundreds of kHz) and a control reference signal (error signal of an audio signal) has another problem.

Assuming that the input signal is a sine wave, in order to obtain the maximum output voltage, when the modulation coefficient (MI: Modulation Index = reference signal (audio signal error signal) / sawtooth carrier signal) is set to almost 1, that is, the maximum input voltage In the case of obtaining a maximum output by applying a signal, it is difficult to obtain a symmetrical full swing waveform in which the output waveform reaches the external supply voltage (± Vdd). In other words, when the output voltage is full swing, a clip problem (input voltage problem) occurs in which the output waveform is cut at either the maximum or minimum point of the sine wave output. The reason for this is that the two MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) used in the above-described Class D amplifier have different characteristics on the rising and falling times of the pulse waveform during switching. Another reason is that the switching operation is performed in a fixed frequency-variable duty scheme in the conventional class D amplifier. That is, since the amplification is performed by the switching operation of only duty adjustment (1 degree of freedom), the problem that occurs when the characteristics of the MOSFETs are different as described above affects the output waveform.

An easy way to solve this problem is to increase the external supply voltage (± Vdd) and use the modulation coefficient (MI) in the low range. However, such a simple method not only shortens the lifespan by increasing the voltage stress on the device, but also causes another problem of unnecessary power loss.

Another solution is the individual tuning method. This method is commonly used in expensive audio equipment. For the purpose of full swing output, trial and error can be done by continuously changing the parts that exist in the audio equipment that are considered valid for full swing tuning. It's a cumbersome and unproductive way to adjust.

The present invention has been proposed to solve the above problems, the self-excited delta-modulated D-class sound that can be full swing symmetrically up to the external output voltage (± Vdd) while using the minimum components To provide an amplifier.

To this end, in the present invention, the delta modulated Class D amplifier is configured to be self-oscillating to secure two degrees of freedom of variable frequency-variable duty, thereby adjusting itself even if the device value or various characteristics of the component are somewhat changed. and the ability to self-tuning.

1 is a block diagram of a conventional digital amplifier

2 is a block diagram of an acoustic amplifier of the present invention.

3 is a detailed circuit diagram of FIG.

4A and 4B are input and output waveform diagrams of the acoustic amplifier of the present invention.

5A and 5B are measurement diagrams of harmonic distortion of the acoustic amplifier of the present invention.

Figure 6 is a measure of the efficiency of the acoustic amplifier of the present invention

7a and 7b are measurements of the frequency response of the acoustic amplifier of the present invention.

Explanation of symbols on the main parts of the drawings

10 modulator 20 power switch unit

30: demodulation filter 40: speaker

50: voltage negative feedback means CP: comparator

Q1 to Q4: Transistors M1 and M2: High power switch

D1 to D4: Diodes Z1 to Z4: Zener Diodes

L: Inductor R1-R10: Resistance

RH: Resistor C1 to C5: Capacitor

+ Vdd, -Vdd: External power supply Vin: Input voltage

Vout: Output voltage (speaker input voltage) Vf: Negative feedback voltage

Vcp: comparator output voltage Vq: transistor output voltage

Vm: High Power Switch Output Voltage

As shown in FIG. 2, the self-oscillating delta modulated class D acoustic amplifier of the present invention compares the input signal Vin with the negative feedback voltage Vf inputted from the voltage negative feedback means 50. A modulator 10 for modulating and outputting a pulse waveform of variable frequency-variable duty; A power switch unit 20 amplifying the output of the modulator 10 and performing a switching operation; A demodulation filter (30) which integrally demodulates an output signal of the power switch unit (20); The output voltage Vout of the demodulation filter 30 is attenuated to an appropriate magnitude, and includes a voltage negative feedback means 50 for supplying a negative feedback voltage to the modulator 10 and a speaker 40 which is an external load.

The acoustic amplifier of the present invention having the above configuration will be described with reference to the detailed circuit diagram shown in FIG.

The modulator 10 includes a high pass filter comprising a capacitor C1 and a resistor R1 and blocking direct current and low frequency components of the input voltage Vin; A comparator (CP) for comparing and outputting a signal input from the high pass filter with a signal fed back through the voltage sub-feedback means (50) and a resistor (RH); It consists of a pull-up resistor R2 of the comparator CP and a resistor RH that determines the hysteresis voltage.

The power switch unit 20 includes transistors Q1 to Q4 that form a darlington type emitter follower structure and amplify the output current of the modulator 10; Diodes D1 and D2 for fast reverse recovery of the transistors Q1 to Q4; Distribution resistors R3 and R4 for adjusting the response characteristics; A resistor R5 for slowly turning on the high power switch M1, which is a P-channel MOSFET, and a diode D3 for quickly turning off; A resistor R6 for slowly turning on the high power switch M2, which is an N-channel MOSFET, and a diode D4 for rapidly turning on; Blocking capacitors C2 and C3, pull-up resistors R7, and pull-down resistors R8 for blocking direct current flowing into the high power switches M1 and M2; Zener diodes Z1 to Z4 determine the turn-on and turn-off voltages of the gates of the MOSFETs M1 and M2.

The demodulation filter 30 is composed of an inductor L and a capacitor C4 forming a second instantaneous integral type low pass filter. This demodulation filter can be thought of as a lowpass filter in terms of demodulating a signal in the audible frequency band (20 Hz to 20 kHz) from a modified pulse waveform (hundreds of kHz), and operates instantaneously for pulse signals in the hundreds of kHz. In terms of doing so, it can be thought of as an integrator.

The voltage negative feedback means 50 is composed of resistors R9 and R10 for reducing the output voltage Vout to an appropriate magnitude to obtain the negative feedback voltage Vf and a capacitor C5 for adjusting the response characteristics.

The operation of the present invention having the configuration as described above will be described with reference to FIGS. 2 and 3. For ease of explanation, it is assumed that the input signal Vin is a sine wave.

First, referring to the waveform obtained at each stage in FIG. 2, the sine wave input waveform Vin is input to the modulator 10, and the output of the modulator 10 is a pulse waveform of a variable frequency-variable duty. This pulse waveform is input to the power switch unit 20, and then converted into a pulse waveform of a similar shape and appears as an output of the power switch unit 20.

The output pulse waveforms of the high power switches M1 and M2 capable of providing a large current are demodulated by a sine wave output Vout amplified by several tens of times via the demodulation filter 30. The amplified sinusoidal output Vout is introduced into the speaker 40 and converted to sound, and at the same time is converted into a negative feedback signal Vf by the voltage negative feedback means 50 and input to the negative terminal of the modulator 10. do.

The negative feedback signal Vf input to the negative terminal of the modulator 10 is a sinusoidal signal having substantially the same magnitude as the input signal Vin input to the positive terminal of the modulator 10. The pulse signal appearing at the output of the modulator 10 is used to compare the input signal Vin input to the (+) terminal of the modulator 10 and the negative feedback signal Vf input to the (-) terminal of the modulator 10. It is a pulse waveform obtained by.

It will be described in detail how the signal changes in time sequential through FIG.

The input signal Vin is a sinusoidal wave, and the signal of the sinusoidal wave is about to increase positively, and for the sake of easy explanation, since the resistor RH is large enough, the positive terminal of the comparator CP is connected via the resistor RH. The influence of the hysteresis voltage superimposed on can be ignored, and assume that the initial voltage of the negative feedback signal Vf is zero and the output voltage Vout is also zero.

Then, since the input signal Vin is greater than the negative feedback signal Vf, the voltage of the negative terminal of the comparator CP is greater than the voltage of the positive terminal of the comparator CP. Therefore, the output Vcp of the comparator CP becomes a low voltage (-Vdd). Accordingly, since the transistors Q3 and Q4 are turned on and the transistors Q1 and Q2 are turned off, the output voltage Vq of the transistor becomes the voltage (-Vdd) in the low state. The output voltage (Vcp) waveform of the comparator (CP) and the output voltage (Vq) waveform of the transistor are similar, except that the difference between the emitter follower type to drive the high power switch (M1, M2) of the next stage. The current amplification is performed by the transistors Q1 to Q4.

When the output voltage Vq of the transistor reaches a low voltage (-Vdd), the high power switch M1 is turned on, and as a result, the output voltage Vm of the high power switches M1 and M2 is high. It becomes the phosphorus voltage (+ Vdd), and this voltage (Vm) is instantaneously integrated with a positive voltage by the inductor (L) and the capacitor (C4) which are the second instantaneous integrated low pass filter of the demodulation filter (30). Demodulated to output voltage Vout.

The output voltage Vout is converted to the negative feedback voltage Vf by the resistors R9 and R10 and the capacitor C5, which are the voltage negative feedback means 50, and transferred to the positive terminal of the comparator CP. While the P-channel MOSFET high power switch M1 is on, the output voltage Vout continues to increase, and the negative feedback voltage Vf continues to increase. The output voltage of the comparator CP at the instant when the negative feedback voltage Vf transmitted to the positive terminal of the comparator CP exceeds the input voltage Vin delivered to the negative terminal of the comparator CP Vcp) is changed from the low state (-Vdd) to the high state (+ Vdd).

As a result, the output Vq of the transistor also becomes high. This time, the high power switch M2, which is an N-channel MOSFET, is turned on, and the high power switch M1, which is a P-channel MOSFET, is turned off. This time, the high power switch M2 is turned on and the high power switch M1 is turned off. When the high power switch (M2) is turned on, the output voltage (Vm) of the high power switch becomes low (-Vdd). Therefore, the negative voltage is reduced by the inductor (L) and the capacitor (C4), which are the second instantaneous integrated low pass filter. Integrating with the voltage reduces the output voltage Vout and decreases the negative feedback voltage Vf.

(One)

When the negative feedback voltage Vf becomes smaller than the input voltage Vin again, the output voltage Vcp of the comparator CP is changed from the high state (-Vdd) to the low state (-Vdd). By this repetitive operation, the P-channel MOSFET high power switch M1 and the N-channel MOSFET high power switch M2 are alternately turned on / off, thereby amplifying.

Here, the resistors R5 and R6 for slowly turning on the high power switches M1 and M2 are preferably in the range of 47 Ω to 470 Ω, and the blocking capacitors C2 and C2 which block DC components flowing into the high power switches M1 and M2. C3) is suitably in the range of 10 nF to 470 nF.

The reason for this is that in order for the high power switches M1 and M2 to be turned on / off, charging and charging of parasitic capacitances formed between the gates of the switches M1 and M2 and the source terminals must occur. The charging path of the parasitic capacitance in the power switch M2 is routed through the resistor R6 and the blocking capacitor C3.

In this case, the resistor R6 serves to delay the charging speed of the parasitic capacitance. If the resistance R6 is assumed to be extremely small, the resistance R6 cannot be delayed, but on the other hand, it is assumed to be extremely large. As a result, the charging speed of the parasitic capacitance becomes too slow, and the time required for turn-on operation of the N-channel MOSFET high power switch M2 becomes long.

Therefore, the condition that the resistor R6 should not be too large or too small is established. When the N-channel MOSFET high power switch M2 is turned off, the parasitic capacitance is discharged through the diode D4. Since the turn-on resistance of D4) is small, rapid discharge occurs.

Therefore, when the resistance values of the resistors R5 and R6 are in the range of 47? To 470 ?, the blocking capacitors C2 and C3 are suitably in the range of 10nF to 470nF.

As described above, since the acoustic amplifier of the present invention is designed to operate by comparing the input voltage Vin and the negative feedback voltage Vf by the comparator CP, a switching operation by self-oscillation is also performed. The output voltage Vcp of the comparator CP has a variable frequency and a variable duty.

Accordingly, although the rising or falling characteristics of the high power switches M1 and M2 are slightly different in the pulse switching, or the device values of the elements in the power switch unit 20 are slightly different, the variable frequency possessed by the circuit itself is different. Self-adjustment is possible due to the variable duty (two degrees of freedom) characteristics, allowing full swing symmetry on both sides even when the maximum output voltage (Vout) is performed.

In the case of the conventional fixed frequency PWM class D amplifier and the self-excited delta modulated class D amplifier according to the present invention, the difference between the case of trying to swing the output voltage Vout to the maximum is explained. Find out how the output voltage (Vout) full swing problem can be solved.

For example, when the rise or fall time of the upper P-channel MOSFET high power switch M1 is longer than the rise or fall time of the lower N-channel MOSFET high power switch M2, that is, the high power switch M1 It is assumed that the effective on time is shorter than the effective on time of the high power switch M2, and since the input voltage Vin is a sine wave, the output voltage Vout is also a sine wave.

Then, in the case of the conventional fixed frequency PWM class D amplifier, the effective on time of the high power switch M1 on the upper side is shorter than the effective on time of the lower high power switch M2. Although the voltage can swing at maximum, the swing of the sinusoid in the upper portion cannot swing enough, so the waveform is truncated during the swing of the sinusoid in the upper portion.

However, in the case of the self-oscillating delta modulation type D amplifier according to the present invention, the frequency is variable and the duty is variable, so that the waveform is not cut even during the full swing. This is because even if the effective on time of the high power switch M1 is shorter than the effective on time of the high power switch M2, the frequency is variable, so that the range of duty that is actually available becomes wider.

In detail, the duty refers to the ratio of the on time to the period of the pulse. When the period of the pulse is shortly fixed (high frequency is fixed), the switching of the high power switches M1 and M2 is performed. If the rise or fall time is too long to be negligible compared to the period of the pulse, the available duty range becomes narrower. However, if the period of the pulse can be extended (lower frequency) or the on time of the pulse can be extended as necessary, even if the rise or fall time of the switching of the high power switches M1 and M2 is slightly different, the effect is almost negligible. It can be reduced to a negligible extent.

That is, if the circuit can adjust the period (frequency) or on time of the pulse itself based on the comparison of the switching frequency (or period) with the input voltage Vin and the negative feedback voltage Vf, It is possible to obtain a symmetrical output voltage (Vout) that is fully swinged to the external power supply (± Vdd).

Since the circuit of the acoustic amplifier according to the present invention belongs to voltage series feedback in terms of the negative feedback technique, the closed loop voltage gain (A) can be expressed simply as Equation (1).

[Equation 1]

(2)

In the acoustic amplifier of the present invention, the switching frequency adjusts the ratio of the resistors RH and R1 that determine the hysteresis voltage, adjusts the capacitor C5, or adjusts the inductor L and the capacitor C4, which are secondary bandpass filters. Set by adjusting.

Here, the hysteresis voltage is obtained as shown in Equation 2 below.

[Equation 2]

Assuming that the resistor R10 is 1 kW, the resistance RH for determining the hysteresis voltage is preferably in the range of 100 kΩ to 1 MΩ. Accordingly, the inductor L has a suitable range of 20 μH to 200 μH, and the capacitor C4 is 0.47. It is preferably in the range of μF to 4.7 μF.

(3)

In addition, the resistor R10 for attenuating the output voltage Vout to an appropriate magnitude is in the range of 1 k? ± 20%, and the capacitor C5 for adjusting the response characteristic is in the range of 10 pF to 470 pF.

In particular, the inductor L and the capacitor C4 of the demodulation filter 30, which demodulate the output voltage Vm of the high power switches M1 and M2 to the output voltage of the audible frequency band, have a frequency characteristic (gain characteristic) of the acoustic amplifier. Since the device has a decisive influence on the determination of phase characteristics, careful selection is required.

The present invention has the following advantages and features as compared to the conventional method. First, as described above, the self-adjusting function is not sensitive to fluctuations in device values. Second, since the sawtooth wave generation circuit is unnecessary, low cost can be realized. Third, the negative feedback method is extremely simple. Fourth, because it is a secondary system, it is always stable. Fifth, there is no use of an operational amplifier, and an acoustic amplifier system is implemented using only a comparator.

Next, look at the performance of the sound device actually produced by the present invention.

For actual experiments with a rated power of 40W and a maximum power of 80W, the external supply voltage ± Vdd is ± 20V, and the speaker load is 4Ω load. Since the input voltage (Vin) is assumed to be 1Vp maximum, the rated output is 40W (= (18 / ) To obtain a ^2/4), so to secure the output voltage (Vout) of 18Vp closed loop gain was adjusted to about 20.

At no signal input, the self-oscillating switching frequency is approximately 350kHz. However, the acoustic amplifier of the present invention has a feature that the switching frequency decreases as the magnitude of the input voltage increases, so in practice, the average switching frequency is about 200kHz to 250kHz.

4A to 4B illustrate waveforms of an externally applied voltage (+ Vdd) and an output voltage Vout when the input voltage Vin is a sine wave of 1 kHz having a magnitude of 1 Vp, and an input voltage Vin having a magnitude of 1 Vp. The output voltage Vout in the sine wave of 5 kHz and the output voltage Vm of the high power switch are shown. It can be seen that the waveform is clearly amplified well with a symmetrical full swing state, and the output voltage (Vm) of the high power switch operates as a variable frequency-variable duty.

FIG. 5A shows the relationship between the static output power at 1 kHz and harmonic distortion, and FIG. 5B shows the relationship between the frequency at 1 kHz, 10 W, 40 W and harmonic distortion. In the case of Figure 5a, the shape is like a 'U' shape. The harmonic distortion at lower output levels is worse than the harmonic distortion at intermediate levels because noise affects the signal more seriously at smaller output levels.

On the other hand, the harmonic distortion at a very large output level is worse because the swing of the output voltage Vout becomes larger than the external supply voltage (± Vdd), causing a clip problem at the output voltage (Vout). . It can be seen from FIG. 5A that harmonic distortion of about 0.1% is obtained. Also in Fig. 5B, the harmonic distortion property is generally good from 500 Hz to 5 kHz, which is relatively full of acoustic energy.

6 shows the relationship between the energetic output power versus efficiency. The calculation of the efficiency was defined and applied as in Equation 3.

[Equation 3]

Here, Po (x), Pdd (x) and x represent the rated output power, the total supply power and the arbitrary output level, respectively. It can be seen from FIG. 6 that an efficiency of around 90% is achieved at full power.

It can be seen from Figs. 7A and 7B showing the frequency characteristics that the gain characteristics are generally flat in the audible frequency range, that is, 20 Hz to 20 kHz, and the 3 dB band width is also located near about 30 kHz.

Based on the above experimental results, it can be seen that the acoustic amplifier of the present invention can be sufficiently commercialized as a low-cost, low-cost digital acoustic amplifier.

It also saves energy by matching the green rounds. For example, suppose that 50W of static output power is required. In the case of a class A amplifier (assuming 20% efficiency), 200W of power is wasted, but the amplifier of the present invention (90% assumption) In this case, there is only about 5W loss. Therefore, when comparing the losses, it can be seen that the energy loss of the present invention is reduced to about 1/40 (= 5/200) in comparison with the class A amplifier. Assuming that all or most of the existing analog amplifiers are replaced by the acoustic amplifier technology of the present invention, the energy saving is indeed enormous, which in turn helps to protect the earth from environmental pollution, as well as green round environmental technology. It will be possible to preoccupy the world's environmental technology in the infinite competition.

As described above, the present invention configures a delta modulated Class-D amplifier to be self-oscillating to secure two degrees of freedom of the variable frequency-variable duty so that the variable frequency-variable without a separate modulated carrier signal (sawtooth wave or triangle wave) generation circuit. By amplifying by switching operation by duty, it is possible to adjust the change in device value or various characteristics of components by itself.

In addition, since the harmonic distortion is around 0.1% and the efficiency is about 90%, it can not only save energy but also have no problem of cutting off the signal even when full swing of maximum output, and audible frequency from 20Hz to 20kHz While the gain characteristics are reputed in the band, the 3dB band width is also present in the vicinity of 30 kHz, so the response characteristics are good. As the secondary system, the entire system is always stable and there is no stability problem.

In addition, the use of a very small number of parts can be manufactured at a low price, thereby improving price competitiveness.

Claims (10)

In digital (D class) acoustic amplifier using switching, A self-oscillating delta modulated Class D acoustic amplifier characterized in that amplification is performed by switching operation by a variable frequency-variable duty without a separate modulation carrier signal (sawtooth wave or triangle wave) generation circuit. The method of claim 1, wherein the input signal Vin is modulated into a pulse waveform of a variable frequency-variable duty by comparing the input signal Vin with the negative feedback voltage Vf input from the voltage negative feedback means 50. A modulator 10; A power switch unit 20 amplifying the output of the modulator 10 and performing a switching operation; A demodulation filter (30) which integrally demodulates an output signal of the power switch unit (20); Including a voltage negative feedback means 50 for supplying a negative feedback voltage (Vf) to the modulator 10 by attenuating the output voltage (Vout) of the demodulation filter 30 to an appropriate magnitude Self-oscillating delta modulation class D acoustic amplifier, characterized in that the configuration. 3. The apparatus of claim 2, wherein the modulator (10) comprises: a high pass filter comprising a capacitor (C1) and a resistor (R1) and blocking a direct current component and a low frequency component of an input voltage Vin; A comparator (CP) for comparing a signal input from the high pass filter with a signal input feedback through the voltage sub-feedback means (50) and a resistor (RH) and outputting a pulse waveform of a variable frequency-variable duty; A self-oscillating delta modulation class D acoustic amplifier comprising a pull-up resistor (R2) of the comparator (CP) and a resistor (RH) for determining a hysteresis voltage. 4. The self-oscillating delta modulated Class A acoustic amplifier according to claim 3, wherein the resistance (RH) for determining the hysteresis voltage is in a range of 100 k? 3. The power switch unit (20) of claim 2, further comprising: transistors (Q1 to Q4) forming a Darlington type emitter follower structure and amplifying the output current of the modulator (10); Diodes D1 and D2 for fast reverse recovery of the transistors Q1 to Q4; Distribution resistors R3 and R4 for adjusting the response characteristics; A resistor R5 for slowly turning on the high power switch M1, which is a P-channel MOSFET, and a diode D3 for turning off the fast; A resistor R6 for slowly turning on the high power switch M2, which is an N-channel MOSFET, and a diode D4 for rapidly turning on; Blocking capacitors C2 and C3, pull-up resistors R7, and pull-down resistors R8 for blocking direct current flowing into the high power switches M1 and M2; A self-oscillating delta modulation class D acoustic amplifier comprising a zener diode (Z1 to Z4) for determining the turn-on / turn-off voltage of the gates of the MOSFETs (M1 and M2). 6. The self-oscillating delta modulated Class A acoustic amplifier according to claim 5, wherein the resistors (R5, R6) range from 47Ω to 470Ω and the capacitors (C2, C3) range from 10nF to 470nF. The self-oscillating delta modulated Class A acoustic amplifier according to claim 2, wherein the demodulation filter (30) comprises an inductor (L) and a capacitor (C4) forming a second instantaneous integral low pass filter. 8. The self-oscillating delta modulated Class A acoustic amplifier according to claim 7, wherein the inductor (L) is in the range of 20 mu H to 200 mu H, and the capacitor C4 is in the range of 0.47 mu F to 4.7 mu F. 9. The capacitor (C5) of claim 8, wherein the voltage negative feedback means (50) adjusts the resistances (R9, R10) and the response characteristics to obtain a negative feedback voltage (Vf) by attenuating the output voltage (Vout) to an appropriate magnitude. Self-oscillating delta modulation class D acoustic amplifier, characterized in that consisting of. 10. The self-oscillating delta modulated Class A acoustic amplifier according to claim 9, wherein the capacitor C5 is set within a range of 10pF to 470pF when the resistor R10 is set within a range of 1kΩ ± 20%.