JP4566566B2 - Power amplifier - Google Patents

Description

  The present invention relates to a switching type power amplifying apparatus for supplying power to an acoustic speaker or the like, and particularly to reducing the power consumed by the power amplifying apparatus itself.

  For driving an acoustic speaker, linear power amplifiers such as class A, class B, class AB and the like having a simple configuration are often used. In such a linear power amplifier, the power loss of the power amplifier itself is large on the principle of operation, and the heat generated by the power consumption of the power amplifier itself increases as the output power increases. For this reason, there exists a subject that the big heat radiator for diffusing this is needed. Therefore, a switching type power amplifying apparatus is used for a power amplifying apparatus having a large output power.

  A switching type power amplifying device switches a positive power supply voltage, a zero voltage, or a negative power supply voltage between output terminals at high speed by turning on or off a power switch of an output stage that supplies power. A power low-pass filter (LPF) provided between the output terminal and the load removes audible high frequency band power, and supplies only audible band power to the load. When the switch is on, current flows, but the voltage across the terminals is very small, and when it is off, voltage is applied but the passing current is almost zero. Get smaller. This is the reason why the power consumption of the switching type power amplifying device is low. The output as described above is, for example, a configuration in which one end of the load is grounded and the positive power supply voltage and the negative power supply voltage are switched to the other end, or a bridge configuration in which zero or a positive voltage is applied to both ends of the load. In the latter case, positive and negative voltages can be applied to the load with only one power source.

  As a conventional technique of such a switching type power amplifying apparatus, one having a configuration as shown in a circuit block diagram of FIG. 14 is known. This is a conventional first power amplifier. This configuration is called a bridge-tide load (hereinafter abbreviated as BTL) or an H-bridge, and includes a pair of parallel circuits of two switches whose connection points are output terminals. A bridge configuration in which a load portion is connected between the output terminals. A series circuit of two switches drawn on the left side of the figure is a first switch circuit, and the right side is a second switch circuit. The operation will be described below.

  In FIG. 14, the power amplification device includes a modulation circuit 810 and a main circuit 180. The main circuit 180 includes a first switch circuit 182, a second switch circuit 183, and a load circuit composed of a series circuit of an inductor 185 and a load 186 for power LPF (a capacitor parallel to the load 186 constituting the power LPF is omitted). And a DC power supply 103 which is a power supply unit. The DC power supply 103 is grounded at a negative electrode serving as a reference potential, and supplies a power supply voltage from the positive electrode to the main circuit 180. Further, the first switch circuit 182 includes a series circuit of a high side switch 1821 and a low side switch 1822, and the second switch circuit 183 includes a series circuit of a high side switch 1831 and a low side switch 1832. A connection point between the high-side switch and the low-side switch is an output terminal of each switch circuit.

  Further, the power amplifying device controls the first distribution circuit 120, the second distribution circuit 130, the switch control circuit 140 that controls the high-side switch 1821, the switch control circuit 150 that controls the low-side switch 1822, and the high-side switch 1831. A switch control circuit 160 that controls the low-side switch 1832. The distribution circuit 120 distributes the modulation signal M1 output from the modulation circuit 810 to the switch control circuit 140 and the switch control circuit 150, and the distribution circuit 130 transfers the modulation signal M2 output from the modulation circuit 810 to the switch control circuit 160. This is distributed to the switch control circuit 170. The modulation circuit 810 includes an input signal source 110 that generates an input signal Vs, a triangular wave generator 111 that generates a triangular wave signal Vt, and a comparator 112 that compares and outputs the input signal Vs and the triangular wave signal Vt. The power supply 102 supplies power to a circuit such as the modulation circuit 810.

  Furthermore, each switch control circuit 140, 150, 160, 170 includes switch drivers 141, 151, 161, 171 for turning on / off each switch and a state detection circuit for detecting an overcurrent state of each switch. 142, 162, 162, 172. Each switch of the main circuit 180 uses an N-channel FET that can be easily integrated in a semiconductor and can obtain a low on-resistance with a relatively small area. In order to turn on the N-channel FET, the gate potential may be 3 to 5 V higher than the source potential, and in order to turn it off, it may be set to zero. Specifically, when the N-channel FET of the low-side switch is turned on, a necessary voltage may be applied to the gate with reference to the ground potential, and it can be directly driven from a circuit supplied with power from the DC power supply 102. On the other hand, the N-channel FET of the high-side switch fluctuates between the ground potential and the power supply voltage of the DC power supply 103 because the source is the output terminal of the switch circuit. For this reason, a specially devised switch control circuit is required to drive the N-channel FET of the high-side switch.

  In FIG. 14, the switch control circuit 140 for driving the high-side switch 1821 includes signals from the floating power supply circuit 144 based on the source potential of the high-side switch 1821 and a circuit based on the ground potential such as the modulation circuit 810. A level shift circuit 143 responsible for replacement is provided. Each potential in the switch control circuit 140 moves following the potential fluctuation of the output terminal of the first switch circuit 182. Therefore, even if the potential of the output terminal of the first switch circuit 182 fluctuates, the switch control circuit 140 always operates based on the source potential of the switch 1821. The floating power supply circuit 144 includes a capacitor 1441 and a diode 1442. When the output terminal of the first switch circuit 182 is at the ground potential (hereinafter referred to as a low potential), the current from the DC power supply 102 charges the capacitor 1441 via the diode 1442, and the switch control circuit 140 Power is supplied to each circuit. When the output terminal of the first switch circuit 182 becomes the power supply voltage of the DC power supply 103 (hereinafter referred to as a high potential), the power stored in the capacitor 1441 is used to cover the power of the circuit in the switch control circuit 140. . The diode 1442 has a role in which the charge of the capacitor 1441 does not flow backward to the DC power supply 102 when the output terminal of the first switch circuit 182 becomes a high potential.

  The level shift circuit 143 has a function of transmitting a switch on / off control signal from the distribution circuit 120 operating with the ground potential as a reference to a circuit based on the potential of the output terminal of the first switch circuit 182 that changes. have. Although not shown, the switch state signal from the state detection circuit 142 includes a circuit that transmits the ground potential to the system. In such signal transmission between fluctuating reference levels, a circuit configuration is generally used in which a signal is replaced with a current on the transmission side and output from a current source, and this current is received as it is or replaced with a voltage on the reception side. is there.

  FIG. 15 is an operation waveform diagram of each part of the conventional first power amplifying device shown in FIG. The operation will be described below with reference to the operation waveform diagram of FIG. The output signal L0 of the comparator 112, which is a comparison result between the input signal Vs from the input signal source 101 and the triangular wave signal Vt from the triangular wave generator 111, is when the input signal Vs is larger than the triangular wave signal Vt (Vs> Vt). Becomes high level (hereinafter referred to as “H”). The time width during which the signal L0 is “H” is ½ of the period T of the triangular wave signal Vt when the input signal Vs is no signal (zero voltage), and the ratio to the period T of the triangular wave signal Vt. When represented by the time ratio δ, δ = 0.5. Further, when the instantaneous value of the input signal Vs is positive, the duty ratio δ is larger than 0.5 (δ> 0.5), and when the instantaneous value of the input signal Vs is negative, the duty ratio δ is 0. Smaller than 5 (δ <0.5).

  The signal L0 of the comparator 112 becomes the modulation signal M1 output from the modulation circuit 810 as it is, and the signal obtained by inverting it with the inverter 1140 becomes the modulation signal M2. Further, the signal M1 passes through the buffer 121 and the inverter 122 of the first distribution circuit 120, passes through the switch control circuit 140 and the switch control circuit 150, and controls on / off of the switches 1821 and 1822. That is, when the signal M1 is “H”, the switch control circuit 140 receives the “H” signal from the buffer 121 by the level shift circuit 143, turns on the switch 1821 via the switch driver 141, and performs switch control. The circuit 150 receives the “L” signal of the inverter 122, turns off the switch 1822 via the switch driver 151, and finally, the output of the first switch circuit 182 becomes a high potential. On the contrary, when the signal M1 is “L”, the output of the first switch circuit 182 has a low potential. Since the signal M2 is logically inverted from the signal M1, when the output of the first switch circuit 182 is at a high potential, the output of the second switch circuit 183 is at a low potential. When the output of 182 is low potential, the output of the second switch circuit 183 becomes high potential.

  The state transition diagram of FIG. 16 shows the state of each switch circuit as described above. The potentials of the output terminal of the first switch circuit 182 and the output terminal of the second switch circuit 183 are opposite to each other, and “the output terminal of the first switch circuit 182 has a high potential and the second switch circuit There are two states in which the output terminal of 183 is at a low potential (HL) or “the output terminal of the first switch circuit 182 is at a low potential and the output terminal of the second switch circuit 183 is at a high potential (LH)”. In addition, there are two possible states of the voltage Vpn applied to the series body of the inductor 185 and the load 186: a positive power supply voltage or a negative power supply voltage.

Here, the relationship between the pulse width (time width of “H”) of the signals M1 and M2 generated by the modulation circuit 810 and the input signal Vs will be described. The instantaneous value of the input signal Vs is compared with the instantaneous value of the triangular wave signal Vt by the comparator 112, and the signal L0 is obtained with the pulse width corresponding to the instantaneous value of the signal Vs. The signal L0 becomes the signal M1 as it is. The logical inversion result of L0 is the signal M2. The triangular wave signal Vt is set so that the voltage change with respect to time is linear, and the period of the triangular wave signal Vt is set sufficiently shorter than the change of the signal Vs, so the pulse width of the signal M1 is the signal Vs. It is proportional to the instantaneous value. Now, assuming that the amplitude of the triangular wave signal Vt is Vpt (absolute value of positive and negative peak voltages) and the ratio of the pulse width in the period T of the signal M1 is the time ratio δ1, the time ratio δ1 is expressed by the following equation (1). expressed.
δ1 = (1 + Vs / Vpt) / 2 (1)

Similarly, the ratio of the pulse width in the period T of the signal M2 is set to the duty ratio δ2. Since the signal M2 is a logical inversion of the signal M1, the duty ratio δ2 is expressed by the following equation (2) obtained by subtracting δ1 from 1.
δ2 = (1−Vs / Vpt) / 2 (2)
However, | Vs / Vpt | <1, and this condition is a necessary condition of a method for realizing the generation of the modulation signal of the switching type power amplifying apparatus with a triangular wave and a comparator, and if | Vs / Vpt | Since the signal L0 of the comparator 112 is fixed to the “H” or “L” state throughout one cycle of the triangular wave signal Vt, the ratio of the time ratio proportional to the magnitude of the signal Vs is exceeded. Pulses cannot be generated and power amplification cannot be performed.

The power supply voltage of the DC power supply 103 is set to Vh. When the signal M1 is “H” and the signal M2 is “L”, the voltage Vpn applied to the series body of the inductor 185 and the load 186 becomes a positive power supply voltage (Vh), and the current passing through the load 186 is caused by the inductor 185. Integrated in the positive direction. On the other hand, when the signal M1 is “L” and the signal M2 is “H”, the voltage Vpn becomes a negative power supply voltage (−Vh), and the passing current of the load 186 is integrated in the negative direction by the inductor 185. Ultimately, the voltage Vo generated in the load 186 is proportional to a value obtained by subtracting the application time of the negative power supply voltage from the application time of the positive power supply voltage, and further becomes a value obtained by multiplying the power supply voltage Vh. The ratio of the application time of the positive power supply voltage in one cycle T is equal to the time ratio δ1 of the signal M1, and the ratio of the application time of the negative power supply voltage in one cycle T is equal to the time ratio δ2 of the signal M2. That is, the voltage Vo generated in the load 186 can be expressed by the following equation (3), and an output proportional to the input signal Vs can be obtained from the equations (1) and (2).
Vo = (δ1-δ2) · Vh = (Vh / Vpt) · Vs (3)

  Thus, in the conventional first power amplifying device, the power of the DC power supply 103 is converted and supplied to the load 186 as power proportional to the input signal Vs.

  FIG. 17 shows a block diagram of a conventional second power amplifier, and FIG. 18 is an operation waveform diagram. In FIG. 17, elements having the same configuration and function as those of the conventional first power amplifying device illustrated in FIG. 14 are denoted by the same reference numerals and description thereof is omitted. The configuration of the modulation circuit 910 is different from the conventional first power amplification device shown in FIG. 14, and a comparator 1122 and a linear inverter 1131 are added to the modulation circuit 810 of the conventional first power amplification device. It is a point that has been. The comparator 1121 corresponds to the comparator 112 of the modulation circuit 810 of the conventional first power amplification device.

  Next, the operation will be described step by step using the operation waveform diagram of FIG. The input signal Vs of the input signal source 101 is applied to the positive input of the comparator 1121, and the polarity is inverted by the linear inverter 1131 to become the signal Vsr, which is applied to the positive input of the comparator 1122. A triangular wave signal Vt having a period T is applied from the triangular wave generator 111 to each negative input of the comparator 1121 and the comparator 1122. The output signal L1 of the comparator 1121 becomes “H” when the input signal Vs is larger than the triangular wave signal Vt (Vs> Vt), and the output signal L2 of the comparator 1122 has a signal Vsr larger than the triangular wave signal Vt. It becomes “H” at the time (Vsr> Vt). As described above, the configuration of the comparator 1121 is the same as that of the comparator 112 of the conventional first power amplification device, and the output signal L1 obtained is the same as the output signal L0 of the comparator 112. The pulse width of the output signal L2 obtained from one of the comparators 1122 is reversed in the polarity of the signal applied to the positive input to that of the comparator 1121, so the tendency of increase / decrease in the pulse width of the output signal L1 is reversed. ing. The time ratio δ1, which is the ratio of the period during which the signal L1 is “H” to one cycle T, is 50% when the input signal Vs is no signal, and the time ratio when the instantaneous value of the input signal Vs is positive. Becomes larger than 50%, and becomes smaller than 50% when the instantaneous value of the input signal Vs is negative. On the other hand, the duty ratio δ2 of the signal L2 is 50% when the input signal Vs is no signal, and the duty ratio is less than 50% when the instantaneous value of the input signal Vs is positive. When the instantaneous value is negative, it becomes larger than 50%.

  The signal L1 becomes the output signal M1 of the modulation circuit 910 as it is, and the signal L2 also becomes the output signal M2. Then, the signal M1 passes through the buffer 121 and the inverter 122 of the first switch circuit side distribution circuit 120, the switch control circuit 140 and the switch control circuit 150, and the switch 1821 and the switch 1822 are turned on / off. That is, when the signal M1 is “H”, the switch control circuit 140 receives the “H” of the buffer 121 by the level shift circuit 143, turns on the switch 1821 via the switch driver 141, and the switch control circuit 150 In response to “L” of the inverter 122, the switch 1822 is turned off via the switch driver 151, and the output of the first switch circuit 182 becomes a high potential. On the contrary, when the signal M1 is “L”, the output of the first switch circuit 182 has a low potential. The signal M2 is also the same as the first switch circuit side. When the signal M2 is “H”, the output of the second switch circuit 183 is at a high potential, and when the signal M2 is “L”. The output of the second switch circuit 183 becomes a low potential.

  FIG. 19 is a state transition diagram showing the on / off state of each switch of the main circuit 180. In the above-described conventional first power amplifying device, the output potentials of the first switch circuit 182 and the second switch circuit 183 operate in opposite polarities at the same timing. In the power amplifying apparatus, the output states of the respective switch circuits change independently of each other, and both of them may be at a high potential or a low potential. That is, as shown in FIG. 19, the output states of the first switch circuit 182 and the second switch circuit 183 are “the output terminal of the first switch circuit 182 is at a high potential and the second switch circuit 183 has "Output terminal is low potential (HL)", "Output terminal of first switch circuit 182 is high potential and output terminal of second switch circuit 183 is also high potential (HH)", "Output of first switch circuit 182 “The terminal is low potential and the output terminal of the second switch circuit 183 is high potential (LH)”, “the output terminal of the first switch circuit 182 is low potential and the output terminal of the second switch circuit 183 is also low potential (LL). ) ”. The state of (HH) or (LL) is a state where both ends of the series body of the inductor 185 and the load 186 are short-circuited, and the possible state of the voltage Vpn applied to the series body of the inductor 185 and the load 186 is positive The three power supply voltages, zero V or negative power supply voltage, are obtained.

Here, the relationship between the pulse width (time width which is “H”) of the signals M1 and M2 generated by the modulation circuit 910 and the input signal Vs will be described. The triangular wave signal Vt is set so that the voltage change with respect to time is linear, and the period of the triangular wave signal Vt is set sufficiently shorter than the change of the input signal Vs, so the pulse width of the signal M1 is the input signal. It is proportional to the instantaneous value of Vs, and the pulse width of the signal M2 is proportional to the instantaneous value of the input signal Vsr. Now, assuming that the amplitude of the triangular wave signal Vt is Vpt, the time ratio δ1 of the signal M1 is expressed by the following equation (4).
δ1 = (1 + Vs / Vpt) / 2 (4)

Further, the duty ratio δ2 of the signal M2 can be expressed by the following equation (5) in which the sign of the input signal Vs of the signal M1 is changed because the sign of the input signal Vs is reversed.
δ2 = (1−Vs / Vpt) / 2 (5)
However, | Vs / Vpt | <1 is the same as in the case of the conventional first power amplifying apparatus.

  When the signal M1 is “H” and the signal M2 is “L”, the voltage Vpn applied to the series body of the inductor 185 and the load 186 becomes a positive power supply voltage (Vh). Integrated in the positive direction. When the signal M1 is “L” and the signal M2 is “H”, the voltage Vpn becomes a negative power supply voltage (−Vh), and the inductor 185 integrates the current passing through the load 186 in the negative direction. Further, when the signal M1 is “H” and the signal M2 is “H”, or when the signal M1 is “L” and the signal M2 is “L”, the voltage between the terminals of the series circuit of the inductor 185 and the load 186 is zero. Since it becomes V, there is no external power supply.

If it is assumed that the change of the input signal Vs is sufficiently slow with respect to the period of the triangular wave signal Vt, the “H” sections of the signals M1 and M2 each have an object shape centered on the dip time of the triangular wave signal Vt. , It is lined up as if the one with the larger duty ratio blocks the smaller one. The time ratio in the state (HL) is a value (δ1-δ2) obtained by subtracting the time ratio δ2 of the signal M2 from the time ratio δ1 of the signal M1. On the other hand, the time ratio of the state (LH) is a value (δ2−δ1) obtained by subtracting the time ratio δ1 of the signal M1 from the time ratio δ2 of the signal M2. If the polarity of the output voltage Vo is added to the duty ratio (δ2-δ1) of this state (LH), it becomes the same as the positive case (δ1-δ2), and considering equations (4) and (5), The voltage generated in the load 186 can be expressed by the following equation (6), and an output proportional to the input signal Vs can be obtained.
Vo = (δ1-δ2) · Vh = (Vh / Vpt) · Vs (6)

  As can be seen from the operation waveform diagram of FIG. 18, when the input signal Vs is no signal or a small signal level, the duty ratio is 50% in the first switch circuit 182 and the second switch circuit 183. Since the output potential of each switch circuit becomes the same value in synchronization, the voltage applied to the series body of the inductor 185 and the load 186 becomes zero. On the other hand, when the level of the small signal is applied, the duty ratio of the first switch circuit 182 and the second switch circuit 183 increases slightly around 50%, and slightly decreases the other, so that the inductor 185 and the load 186 A narrow pulse is applied to the series body in the time difference interval between the rise and fall of the output potential of each switch circuit. In the case of such no signal and small signal, the pulse width of the positive or negative power source voltage applied to the series body of the inductor 185 and the load 186 is narrow, so that the current output from the power source is reduced and the power consumption is reduced. it can. At the same time, the radiation amount of electromagnetic waves can be reduced. In particular, for a signal having a low average level such as a general audio signal, the configuration of the conventional second power amplification device is more effective than the conventional first power amplification device.

  As described above, in the conventional second power amplifying device, the voltage of the DC power supply 103 is applied with a positive direction, a negative direction, or zero to the series body of the inductor 185 and the load 186 under the control of the modulation circuit 910. The power of the DC power supply 103 is converted and supplied to the load 186 as power proportional to the input signal Vs. As a result, when the signal level of the input signal Vs is small, the zero-V application state becomes long, and power consumption and radiation of electromagnetic waves are suppressed to a low level (see, for example, Patent Document 1).

  The difference in the output time ratio when there is no signal in the above two conventional switching type power amplifiers can be distinguished and similar due to the difference in the bias current when there is no signal of the output transistor in the linear amplifier. That is, the bias current at the time of no signal in the class A operation of the linear amplifier is 50% at the maximum output, which is small to several percent in the class AB operation, and becomes almost zero in the class B operation. When these are made to correspond to the duty ratio of the switching type power amplifier, the switching type power amplifying apparatus of the conventional example of FIG. 14 corresponds to the class A operation of the linear amplifier, and the switching type power amplifying apparatus of the conventional example of FIG. It corresponds to the class B operation of the linear amplifier. The purpose of class B operation and class AB operation of this linear amplifier is also to reduce power consumption.

  Some switching-type power amplifying devices do not use a triangular wave generator. For example, it is a power amplifying apparatus called a delta-sigma modulation system, such as Patent Document 2 and Patent Document 3. Basically, the difference signal between the input signal Vs and the feedback pulse signal from the output is obtained, processed by a weighting filter (difference integrator), and then the voltage is output by a comparator (1-bit quantizer). The switch circuit of the main circuit is operated by the pulse signal after the quantization of the time axis and the time axis quantization by the clock generator and the delay unit. Finally, a negative feedback operation is performed so as to reduce the difference signal between the input signal Vs and the feedback pulse signal from the output, whereby a voltage proportional to the input signal Vs is generated in the load connected to the main circuit. It is like that.

Utility model registration No. 852597 JP-A-5-063457 JP-A-5-152867

  The conventional switching type power amplifying device having such a configuration consumes a large amount of power in the drive and signal transmission circuits of the switches in the main circuit, and can increase the power conversion efficiency in a system with a large output power. However, there is a drawback that the efficiency cannot be increased in a portable device that has a small output power and operates on a battery.

  The power consumption of the level shift circuit in the high-side switch control circuit of the main circuit is an example of a large power consumption in the conventional power amplifier. In particular, power consumption in the level shift circuit increases when the output of the switch circuit is at a high potential. In this level shift circuit, even if the reference potential of the high-side switch changes due to the on / off operation, the signal is replaced with a current source in order to perform correct signal transmission. Therefore, main power consumed by the level shift circuit is obtained by multiplying the voltage obtained by adding the voltage of the floating power supply circuit to the output potential of the switch circuit by the current value of the current source. This power consumption increases when the output terminal of the switch circuit is at a high potential. In addition to this, when considering the signal transmission circuit from the state detection circuit in the switch control circuit to the system control unit configured with reference to the ground potential, the power consumption when the output terminal of the switch circuit is at a high potential is even greater. Become.

  An object of the present invention is to provide a power amplifying apparatus suitable for a portable device that consumes less power than a linear type and a conventional switching type and can be further reduced in size even with a small output power.

  In order to solve the above problems, a power amplifying device according to the present invention includes first and second series circuits that are connected in parallel to each other, which includes two switches that are supplied with a power supply voltage from a power supply unit and that have a connection point as an output terminal. And a power amplifying apparatus having a configuration in which a load unit is connected between the output terminals to generate a modulated signal having a pulse width or pulse density corresponding to the input signal to drive each switch, and The voltage of the output terminal is controlled so that the voltage of the output terminal of the second series circuit is any one of the three types of power supply voltage and reference potential, reference potential and power supply voltage, and reference potential and reference potential. .

The power amplifying device according to the present invention is a series circuit of a first high-side switch and a first low-side switch, and has a connection point between the first high-side switch and the first low-side switch as an output terminal. 1 is a series circuit of a switch means, a second high-side switch, and a second low-side switch, and a second switch having a connection point between the second high-side switch and the second low-side switch as an output terminal Means, a power supply for outputting a power supply voltage with respect to a reference potential, and a power supply connected in parallel to the first switch means and the second switch means; an output terminal of the first switch means; and the second switch A load connected between the output end of the means, a first modulation signal for converting the input signal into a pulse width or pulse density and driving the first switch means, and the second switch And a modulating means for generating a second modulation signal for driving the stage. The modulation means is a pulse width modulator. The modulation means includes a triangular wave generator that outputs a triangular wave signal, a first comparator that outputs a first signal that is active when the result of subtracting the triangular wave signal from the input signal is positive, and the input signal and the triangular wave signal. And a second comparator that outputs a second signal that becomes active when the addition result is negative. The first modulation signal is active when the first signal is active and the second signal is inactive, and the second modulation signal is active when the first signal is inactive and the second signal is active It becomes active when The modulation means includes a voltage at the output terminal of the first switch means and a voltage at the output terminal of the second switch means, each of which is a power supply voltage and a reference potential, a reference potential and a power supply voltage, or a reference potential and a reference potential. The modulation signal is controlled to be either of the following.

  The modulating means may further include a linear inverter that inverts the input signal. At this time, the first comparator compares the input signal with the triangular wave signal to generate a first signal, and the second comparator outputs the inverted signal of the input signal output from the linear inverter and the triangular wave signal. The second signal is generated by comparison, the logical product of the logical negation signal of the second signal and the first signal is output as the first modulation signal, and the logical negation signal of the first signal and the second signal The logical product of the above signal is output as the second modulation signal. Alternatively, the second comparator may generate the second signal by comparing the input signal and the inverted signal of the triangular wave signal output from the linear inverter.

The modulation means may further include time offset means for giving a time offset to at least one of the input signal and the triangular wave signal input to the first and second comparators .

  Here, the time offset means includes a rectangular wave generator that outputs a rectangular wave signal synchronized with the triangular wave signal, and an adding means that adds the rectangular wave signal to the triangular wave signal, and outputs the output signal of the adding means to the first signal. You may make it input into any one of a comparator or a said 2nd comparator. Alternatively, the time offset means may further include second addition means for adding the rectangular wave signal to the triangular wave signal with a polarity different from that of the output of the addition means, and the output signal of the first addition means is used as the first comparator. And the output signal of the second adding means may be input to the second comparator.

  The modulation means may have time offset means for giving a time offset to at least one of the first and second signals. Here, the time offset unit may be a delay unit that delays the first or second signal.

The power amplifying device according to the present invention is a series circuit of a first high-side switch and a first low-side switch, and a connection point between the first high-side switch and the first low-side switch is an output terminal. A series circuit of a first switch means, a second high-side switch and a second low-side switch, wherein a second node having an output terminal as a connection point between the second high-side switch and the second low-side switch Switch means, a power supply that outputs a power supply voltage with respect to a reference potential, and a power supply connected in parallel to the first switch means and the second switch means, an output terminal of the first switch means, and a second switch means A load section connected to the output terminal of the first, a first modulation signal for driving the first switch means, and a second switch means for converting the input signal into a pulse width or pulse density And a modulating means for generating a second modulated signal. The modulation means is a pulse width modulator. Modulating means, subtracts a triangular wave generator for outputting a triangular wave signal, a triangular wave signal offset means for offsetting the negative triangular wave signal to the amplitude axis, the output signal of the positive offset triangular wave signal offset means from the input signal A first comparator that outputs a first signal that is active when the result is positive, and a first comparator that is active when the result of subtracting the output signal of the triangular wave signal offset means that is negatively offset from the input signal is negative. And a second comparator that outputs two signals. At this time, the first signal is output as the first modulation signal, and the second signal is output as the second modulation signal. The modulation means includes a voltage at the output terminal of the first switch means and a voltage at the output terminal of the second switch means, each of which is a power supply voltage and a reference potential, a reference potential and a power supply voltage, or a reference potential and a reference potential. The modulation signal is controlled to be one of the following.

  The modulation means includes first minimum pulse width setting means for setting a minimum value for the pulse width of the first modulation signal, and second minimum pulse width for setting the minimum value for the pulse width of the second modulation signal. And setting means.

  Further, the first minimum pulse setting means sets a first minimum pulse setting threshold value that intersects the triangular wave signal and is near a lower limit value of the triangular wave signal, and the triangular wave signal is set to the first minimum pulse setting threshold value. The following period is set to the minimum pulse width of the first signal, and the second minimum pulse setting means crosses the output of the triangular wave signal offset means and is near the upper limit value of the output of the triangular wave signal offset means. A pulse setting threshold value may be set, and a period in which the output of the triangular wave signal offset means is equal to or greater than the second minimum pulse setting threshold value may be set as the minimum pulse width of the second signal. The offset amount of the triangular wave signal offset means is a value smaller than the change width of the triangular wave signal by a predetermined value, so that the first minimum pulse setting threshold and the second minimum pulse setting threshold are equal. Also good.

The modulation means includes first maximum pulse width setting means for setting a maximum value for the pulse width of the first modulation signal, and second maximum pulse width for setting the maximum value for the pulse width of the second modulation signal. And setting means. The first maximum pulse setting means sets a first maximum pulse setting threshold value that intersects the triangular wave signal and is near the upper limit value of the triangular wave signal, and the triangular wave signal is equal to or less than the first maximum pulse setting threshold value. The period may be set to the maximum pulse width of the first signal. The second maximum pulse setting means sets a second maximum pulse setting threshold value that intersects the output of the triangular wave signal offset means and is near the lower limit value of the output of the triangular wave signal offset means, and outputs the triangular wave signal offset means. May be set to the maximum pulse width of the second signal during a period equal to or greater than the second maximum pulse setting threshold. The minimum no-pulse interval between the first modulation signal and the second modulation signal may be larger than the minimum pulse width of the first modulation signal and the second modulation signal.

The modulation means includes first maximum pulse width setting means for setting a maximum value for the pulse width of the first modulation signal, and second maximum pulse width for setting the maximum value for the pulse width of the second modulation signal. And setting means. The first maximum pulse setting means sets a first maximum pulse setting threshold value that intersects the triangular wave signal and is near the upper limit value of the triangular wave signal, and a period during which the triangular wave signal is equal to or less than the first maximum pulse setting threshold value. Is set to the maximum pulse of the first signal. The second maximum pulse setting means sets a second maximum pulse setting threshold value that intersects the output of the triangular wave signal offset means and near the lower limit value of the output of the triangular wave signal offset means, and the output of the triangular wave signal offset means is A period equal to or longer than the second maximum pulse setting threshold is set as the maximum pulse of the second signal.

The power amplifying device according to the present invention is a series circuit of a first high-side switch and a first low-side switch, and a connection point between the first high-side switch and the first low-side switch is an output terminal. A series circuit of a first switch means, a second high-side switch and a second low-side switch, wherein a second node having an output terminal as a connection point between the second high-side switch and the second low-side switch Switch means, a power supply that outputs a power supply voltage with respect to a reference potential, and a power supply connected in parallel to the first switch means and the second switch means, an output terminal of the first switch means, and a second switch means A load section connected to the output terminal of the first, a first modulation signal for driving the first switch means, and a second switch means for converting the input signal into a pulse width or pulse density And a modulating means for generating a second modulated signal. The modulating means has a triangular wave generating means for outputting a triangular wave signal, a feedback signal converting means for outputting the first and second feedback signals, and a difference signal between the first feedback signal and the second feedback signal as an input signal. to have a a feedback input means for adding so the polarity of the negative feedback for. The first feedback signal is transmitted at the time when the output terminal of the first switch means first changes from the power supply voltage to the reference potential or at the first time of the second switch means in one of the falling period and the rising period of the triangular wave signal. output terminal ing and active at a time that varies from the reference potential to the power supply voltage. Further, in the other of the falling period and the rising period of the triangular wave signal, at the time when the output terminal of the first switch means first changes from the reference potential to the power supply voltage, or the output terminal of the second switch means first supplies the power supply voltage. It becomes inactive at the time of changing from the reference potential to the reference potential. In addition, the second feedback signal is transmitted at the time when the output terminal of the second switch means first changes from the power supply voltage to the reference potential or at the first time in one of the falling period and rising period of the triangular wave signal. It becomes active at the time when the output terminal of the means changes from the reference potential to the power supply voltage. Further, in the other of the falling period and the rising period of the triangular wave signal, the output terminal of the second switch means first changes to the power supply voltage from the reference potential or the output terminal of the first switch means first supplies the power supply voltage. It becomes inactive at the time of changing from the reference potential to the reference potential. The modulation means includes a voltage at the output terminal of the first switch means and a voltage at the output terminal of the second switch means, each of which is a power supply voltage and a reference potential, a reference potential and a power supply voltage, or a reference potential and a reference potential. The modulation signal is controlled to be one of the following.

The feedback input means may be configured to receive the input signal in balance and receive the feedback signal in balance. Alternatively, the feedback input unit may include a balanced LPF, and may receive the input signal in a balanced manner and receive the feedback signal output via the balanced LPF.

The power amplifying device of the present invention is configured so that the output terminal of each switch circuit stays at the power supply voltage of the power supply unit, that is, the high potential, so that the amount of power supplied from the floating power supply circuit is reduced, The power consumed by the switch control circuit can be reduced. In addition, there is an advantageous effect that the capacity of the charging capacitor of the floating power supply circuit can be reduced.
Further, according to the power amplification device of the present invention, it is possible to detect the current passing through the load only by the low-side switch portion of both switch circuits. In the second conventional example, there is a state in which the output terminals of the switch circuits of the main circuit are both “H”. However, when the current monitor function is considered as one of the state detection circuits of the switch circuit, this (HH ) Requires a current monitor circuit for the high-side switch. For this reason, the high-side switch driving unit requires a detection circuit and a level shift circuit for transmitting the detection result to the ground level, and the power of the high-side switch must be supplied from the floating power supply circuit, resulting in a large power loss. There are drawbacks. On the other hand, the power amplifying device of the present invention has an effect that the passing current can be detected only by the low-side switch part because there is no state where the output terminals of the switch circuits of the main circuit are both “H”.
Furthermore, if the output of both switch circuits is fixed at a low potential when the input signal is not present, the power supply from the power source of the standby main circuit can be cut off, and the power consumption can be suppressed. In this standby state, the floating power supply of the high-side switch control circuit is in a charged state, so that it can operate promptly upon startup. Further, since the standby output potential is at the reference level, no DC potential is generated in the load even when the operation of the main circuit starts, and pop noise does not occur. Also, when the power supply voltage of the main circuit is supplied from a boost converter or the like, if the outputs of both switch circuits are fixed at a low potential in the startup process of this converter, the load current is reduced, so that startup is easy and quick. Can be launched. Furthermore, if such a standby or normal operation state is intermittently operated in accordance with an input signal or the like, the power consumption can be further reduced. Furthermore, by gradually increasing the power supply voltage of the main circuit from the low voltage state to the specified value, the driving power supplied to the load by the main circuit can be smoothly raised.
Furthermore, since the output signal amplitude is determined by the time ratio between the power supply voltage and the modulation signal, if the power supply voltage is adjusted here, the output signal amplitude can also be adjusted, and the volume can be adjusted. In the conventional example, if this is applied, the volume can be adjusted similarly. In this case, since the reference level of the output voltage of each switch circuit is one half of the power supply voltage, the power supply voltage is adjusted. Noise may be mixed in due to fluctuations in the reference level. However, in the power amplification device of the present invention, since the reference level is fixed to the reference potential (ground potential in the embodiment), there is no mixing of noise due to the power supply voltage.
Furthermore, if a feedback circuit is included, if the amplitude information is also included in the feedback signal, the voltage gain of the modulation circuit and the main circuit is determined by the negative feedback circuit. A voltage obtained by multiplying the input signal by the gain is output as the output voltage of the apparatus. Although the gain of the main circuit portion fluctuates due to the fluctuation of the power supply voltage, the gain change as a power amplifying device does not appear on the surface because the gain of the modulation circuit is adjusted by negative feedback. In the power amplifying apparatus having such a configuration, even when the power supply voltage of the main circuit is reduced when the output level is low, the gain does not change. Therefore, by dynamically changing the power supply voltage as described above, Even if a mechanism for reducing power consumption is introduced, it is not particularly necessary to provide a mechanism for compensating for the gain. Similarly to the above, even if the power supply voltage is fluctuated, an effect such as no noise is obtained.
Further, when the switch state of the main circuit of the present invention is set to 3 states, the pulse width becomes narrow in the vicinity of the zero cross, and the change of the pulse width may not be linear due to the switch operation delay or the dead time setting ( This can be solved by setting a minimum pulse width.
In addition, since either the positive or negative modulation waveform of the present invention is output by each switch circuit, the average voltage changes greatly due to the increase or decrease of the signal level. Therefore, in the path from the synthesis to the negative feedback, Therefore, a configuration that operates stably from the DC range is necessary. Further, when configuring negative feedback, it is compared with each input signal separately or after combining positive and negative feedback signals into one signal. Since a positive or negative signal waveform has an acute angle portion near the zero cross, an operational amplifier that bears these needs to have a high slew rate. In addition, when performing positive / negative combining, it is necessary to consider the offset when combining near the zero cross. However, by performing this synthesis in the digital signal range, problems associated with these analogs do not occur. Therefore, the output voltage including the output state depending on the commutation generated in the main circuit added to the modulation waveform is fed back to the modulator, and the distortion can be reduced. Also, a simple LPF configuration can be used to make the input unit a balanced type and perform negative feedback with a cross from the output unit, or to generate an analog waveform from the modulated digital waveform.

  Hereinafter, preferred embodiments of the power amplifying device of the present invention will be described with reference to the accompanying drawings.

(Embodiment 1)
FIG. 1 is a block diagram of a power amplifying apparatus according to Embodiment 1 of the present invention. The power amplifying device includes a modulation circuit 110, first and second distribution circuits 120, 130, first and second high-side switch control circuits 140 and 160, and first and second low-side switch control circuits 150. , 170.

  The main circuit 180 includes a first switch circuit 182, a second switch circuit 183, and a load circuit composed of a series circuit of an inductor 185 and a load 186 for power LPF (a capacitor parallel to the load 186 constituting the power LPF is omitted). And a DC power supply 103 which is a power supply unit. The DC power supply 103 is grounded at a negative electrode serving as a reference potential, and supplies a power supply voltage from the positive electrode to the main circuit 180. The first switch circuit 182 includes a series circuit of a first high side switch 1821 and a first low side switch 1822. The second switch circuit 183 includes a series circuit of a second high-side switch 1831 and a second low-side switch 1832. A connection point between the high-side switch and the low-side switch is an output terminal of each switch circuit.

  The first high side switch control circuit 140 controls the first high side switch 1821. The first low side switch control circuit 150 controls the first low side switch 1822. The second high side switch control circuit 160 controls the second high side switch 1831. The second low side switch control circuit 170 controls the second low side switch 1832. The first distribution circuit 120 distributes the first modulation signal M1 output from the modulation circuit 110 to the first high-side switch control circuit 140 and the first low-side switch control circuit 150, and the second distribution circuit 130. Distributes the second modulation signal M2 output from the modulation circuit 110 to the second high-side switch control circuit 160 and the second low-side switch control circuit 170.

  The modulation circuit 110 includes an input signal source 110 that generates an input signal Vs, a linear inverter 1131 that inverts the polarity of the input signal Vs in an analog manner, a triangular wave generator 111 that generates a triangular wave signal Vt, a comparator 1121, and 1122, inverters 1141 and 1142, AND gates 1143 and 1144, and a DC power supply 102 that supplies power to a circuit such as the modulation circuit 110.

  Each of the switch control circuits 140, 150, 160, and 170 includes switch drivers 141, 161, 161, and 171 for turning on / off each switch, and a state detection circuit 142 for detecting an overcurrent state of each switch. 162, 162, 172. Each switch of the main circuit 180 uses an N-channel FET that can be easily integrated in a semiconductor and can obtain a low on-resistance with a relatively small area. In order to turn on the N-channel FET, the gate potential may be 3 to 5 V higher than the source potential, and in order to turn it off, it may be set to zero. Specifically, when the N-channel FET of the low-side switch is turned on, a necessary voltage may be applied to the gate with reference to the ground potential, and it can be directly driven from a circuit supplied with power from the DC power supply 102. On the other hand, the N-channel FET of the high-side switch fluctuates between the ground potential and the power supply voltage of the DC power supply 103 because the source is the output terminal of the switch circuit. For this reason, a specially devised switch control circuit is required to drive the N-channel FET of the high-side switch.

  In FIG. 1, the switch control circuit 140 that drives the high-side switch 1821 exchanges signals between the floating power supply circuit 144 based on the source potential of the high-side switch 1821 and a circuit based on the ground potential such as the modulation circuit 110. Is provided with a level shift circuit 143. Each potential in the switch control circuit 140 moves following the potential fluctuation of the output terminal of the first switch circuit 182. For this reason, even if the potential of the output terminal of the first switch circuit 182 fluctuates, the switch control circuit 140 can always operate based on the source potential of the switch 1821. The floating power supply circuit 144 includes a capacitor 1441 and a diode 1442. When the output terminal of the first switch circuit 182 is at the ground potential (hereinafter referred to as “low potential”), the current from the DC power supply 102 charges the capacitor 1441 via the diode 1442 and the switch control circuit. Power is supplied to each circuit in 140. When the output terminal of the first switch circuit 182 becomes the power supply voltage of the DC power supply 103 (hereinafter referred to as “high potential”), the energy stored in the capacitor 1441 is used for the circuit in the switch control circuit 140. Provide power. The diode 1442 has a role in which the charge of the capacitor 1441 does not flow backward to the DC power supply 102 when the output terminal of the first switch circuit 182 becomes a high potential.

  The signal exchange between the distribution circuit 120 having the different reference potential and the level shift circuit 143 is performed by replacing the signal with a current on the transmission side and outputting the current from the current source, and using the current as it is or a voltage on the reception side. A circuit configuration that is received by replacement is common.

  FIG. 2 is a block diagram showing the distribution circuit 120 and the level shift circuit 143 in more detail. In FIG. 2, the distribution circuit 120 includes a balanced conversion circuit 1211 having a balanced output, high-voltage FETs 1212 and 1213, and the level shift circuit 143 includes resistors 1431 and 1432 and a differential receiving circuit 1433. ing. The other distribution circuit 130 and level shift circuit 163 have the same configuration. The state detection circuit 152 is a circuit that detects a current of the low-side switch 1822, includes a current shunting switch 1521 and a current detection resistor 1522, and is connected to the low-side switch 1822 in parallel. The amplification device 1523 amplifies the detection value of the current detection resistor 1522 to a level that is easy to handle.

  The first modulation signal M1 output from the modulation circuit 110 and referenced to the ground potential is first converted into a balanced current signal pair by the balanced conversion circuit 1211 of the buffer 121 and the high breakdown voltage FETs 1212 and 1213. This current signal pair flows to the resistors 1431 and 1432 of the level shift circuit 143, respectively, and generates a balanced signal voltage between its terminals. Since one ends of the resistors 1431 and 1432 are connected to the floating power supply circuit 144, the reference of the balanced signal voltage generated by the resistors 1431 and 1432 changes following the source potential of the switch 1821. This balanced signal voltage is converted into an unbalanced switch on / off control signal that follows the source potential of the switch 1821 by the differential receiver circuit 1433 that follows the same operation, and between the gate and source of the switch 1821. Become a voltage. By the way, the high withstand voltage FETs 1212 and 1213 operate at a constant current with little fluctuation in the drain current even when the voltage between the source and the drain varies due to the characteristics of the element. Therefore, the current value is almost the same regardless of whether the source potential of the switch 1821 is high or low, and the state can be determined by the differential receiving circuit 1433 in the subsequent stage. Further, since the signal transmission is balanced, even if noise caused by the on / off of each switch is mixed, this is canceled out and it is difficult to malfunction. In the state detection circuit 152, a current shunting switch 1521 having a similar element structure and a current detection resistor 1522 are arranged adjacent to each other in parallel with the low side switch 1822, and a value proportional to the passing current of the low side switch 1822 is obtained. It is configured to detect.

  FIG. 3 is an operation waveform diagram of each part of the switching type power amplifying apparatus shown in FIG. The operation of the power amplifying apparatus will be described with reference to the operation waveform diagram of FIG.

  The input signal Vs from the input signal source 101 is applied to the positive input of the comparator 1121, and the polarity is inverted by the linear inverter 1131 to become the signal Vsr, which is applied to the positive input of the comparator 1122. A triangular wave signal Vt having a period T is applied from the triangular wave generator 111 to the negative inputs of the comparators 1121 and 1122, and the triangular wave signal Vt is set so that a voltage change with respect to time is linear. The first signal L1 output from the comparator 1121 becomes a high level (“H”) when the input signal Vs is greater than the triangular wave signal Vt (Vs> Vt), and the first signal L1 output from the comparator 1122 is output. The second signal L2 becomes “H” when the signal Vsr is larger than the triangular wave signal Vt (Vsr> Vt). The configuration so far is the same as the configuration of the conventional second power amplifying apparatus of FIG. 16, and the waveforms of the first signal L1 and the second signal L2 output from each comparator with respect to the waveform of the signal Vs are also the same. become.

  Now, assuming that the triangular wave signal Vt has an average value of zero and the absolute values of both positive and negative peaks are the same, and the signal Vs is a sine wave and the average value is zero, the signal L1 in one period T The time ratio, which is the ratio of the “H” period, is 50% when the signal Vs is no signal, is greater than 50% when the instantaneous value of the signal Vs is positive, and the instantaneous value of the signal Vs is negative. In the case of, it becomes smaller than 50%. On the other hand, the time ratio of the signal L2 is 50% when the signal Vs is not present, 50% when the instantaneous value of the signal Vs is positive, and 50 when the instantaneous value of the signal Vs is negative. %, Which is the reverse of L1.

  Next, the first signal L1 and the logically inverted signal of the second signal L2 are ANDed by an AND gate 1143 to generate a first modulation signal M1. Similarly, the second signal L2 and the logically inverted signal of the first signal L1 are ANDed by the AND gate 1144 to generate the second modulation signal M2. The waveforms of the signal M1 and the signal M2 are as shown in the time chart of FIG. That is, the signal M1 has a waveform obtained by cutting the “H” section of the signal L2 from the “H” section of the signal L1, and the signal M2 has the “H” section of the signal L1 from the “H” section of the signal L2. It becomes the waveform which cut off the section. The change in the “H” period of the signals L1 and L2 spreads so that the side where the duty ratio exceeds 50% falls below the side where the rising / falling timing is hidden. The side will be extracted from the center. Accordingly, the value of the time ratio of the signal M1 is zero when the time ratio of the signal L2 is subtracted from the time ratio of the signal L1 and the time ratio of the signal L2 is larger than the time ratio of the signal L1. The value of the time ratio of the signal M2 is zero when the time ratio of the signal L1 is subtracted from the time ratio of the signal L2, and the time ratio of the signal L1 is larger than the time ratio of the signal L2.

  For example, when the signal Vs is no signal, the time ratio of the signal L1 is 50% and the time ratio of the signal L2 is also 50%. Therefore, if there is no determination delay in the signal L1 and the signal L2, each rise and fall Are the same, the signal M1 and the signal M2 are cut off from each other, and the duty ratio is 0%. On the other hand, when the instantaneous value of the signal Vs becomes positive, the time ratio of the signal L1 is larger than 50% and the time ratio of the signal L2 is smaller than 50%. Therefore, the time ratio of the signal M1 is changed from the time ratio of the signal L1 to the signal L2. The value is obtained by subtracting the time ratio, and the time ratio of the other signal M2 becomes zero. Conversely, when the instantaneous value of the signal Vs becomes negative, the time ratio of the signal L1 is smaller than 50% and the time ratio of the signal L2 is larger than 50%. Therefore, the time ratio of the signal M1 becomes zero, and the other signal M2 The time ratio is a value obtained by subtracting the time ratio of the signal L1 from the time ratio of the signal L2. These signals M1 and M2 are configured to control each switch of the main circuit via each distribution circuit and switch control circuit.

  Now, the first modulation signal M1 turns on / off the switches 1821 and 1822 via the buffer control circuit 140 and the switch control circuit 150 via the buffer 121 and the inverter 122 of the first distribution circuit 120. That is, when the signal M1 is “H”, the switch control circuit 140 receives “H” of the buffer 121 by the level shift circuit 143, turns on the high-side switch 1821 via the switch driver 141, and performs switch control. In response to the “L” level of the inverter 122, the circuit 150 turns off the low-side switch 1822 via the switch driver 151, and the output of the first switch circuit 182 becomes a high potential. On the contrary, when the signal M1 is “L”, the output of the first switch circuit 182 has a low potential. The second modulation signal M2 is the same as that on the first switch circuit side. When the signal M2 is “H”, the output of the second switch circuit 183 is at a high potential, and the signal M2 is “L”. ", The output of the second switch circuit 183 is at a low potential.

FIG. 4 is a state transition diagram showing the on / off states of the switches of the main circuit 180 of the power amplification device according to the first embodiment. The output states of the first switch circuit 182 and the second switch circuit 183 include only the following three states.
(HL): The output terminal of the first switch circuit 182 is at a high potential (H), and the output terminal of the second switch circuit 183 is at a low potential (L). Refer to FIG.
(LH): The output terminal of the first switch circuit 182 is at a low potential (L), and the output terminal of the second switch circuit 183 is at a high potential (H). Refer to FIG.
(LL): The output terminal of the first switch circuit 182 is low potential (L), and the output terminal of the second switch circuit 183 is also low potential (L). Refer to FIG.

  That is, in the power amplifying device of the present embodiment, “the output terminal of the first switch circuit 182 has a high potential (H), as shown in the state transition diagram of the conventional second power amplifying device shown in FIG. There is no state (HH) in which the output terminal of the second switch circuit 183 is also at a high potential (H). The state of (LL) is a state in which both ends of the series body of the inductor 185 and the load 186 are short-circuited, and a possible state of the applied voltage Vpn is 3 of positive power supply voltage or zero V or negative power supply voltage. It becomes a state.

Here, the relationship between the pulse width (time width which is “H”) of the first modulation signal M1 and the second modulation signal M2 generated by the modulation circuit 110 and the input signal Vs will be described. Before that, first, the relationship between the input signal Vs and the pulse widths of the first signal L1 and the second signal L2, which are the basis thereof, is confirmed. Since the signal L1 and the signal L2 have the same circuit configuration as the above-described conventional second power amplifying device, the signal L1 and the signal L2 have the same relationship as the signal M1 and the signal M2. That is, the triangular wave signal Vt is set so that the voltage change with respect to time is linear, and the period of the triangular wave signal Vt is set sufficiently shorter than the change of the signal Vs. Similarly, the pulse width of the signal M2 is proportional to the instantaneous value of the signal Vsr. Now, assuming that the amplitude of the triangular wave signal Vt is Vpt, the time ratio δ1 ′ of the signal L1 is expressed by the following equation (7).
δ1 ′ = (1 + Vs / Vpt) / 2 (7)

Further, since the sign of the input signal Vs is reversed, the duty ratio δ2 ′ of the signal L2 can be expressed by the following equation (8) in which the sign of the input signal Vs of the signal L1 is changed.
δ2 ′ = (1−Vs / Vpt) / 2 (8)
However, | Vs / Vpt | <1 is the same as in the case of the conventional power amplifying apparatus.

From the operation described above, the value of the time ratio δ1 of the first modulation signal M1 is changed from the time ratio δ1 ′ of the first signal L1 to the time ratio δ2 ′ of the second signal L2 as shown in Expression (9). (Δ1 = δ1′−δ2 ′) and becomes zero when the time ratio of the signal L2 is larger than the time ratio of the signal L1.
δ1 = δ1′−δ2 ′ = Vs / Vpt,
However, when δ1′−δ2 ′ <0, δ1 = 0 (9)

Similarly, the value of the time ratio of the second modulation signal M2 is obtained by subtracting the time ratio of the first signal L1 from the time ratio of the second signal L2 (δ2 = δ2 ′) as shown in Expression (10). −δ1 ′) and becomes zero when the time ratio of the signal L1 is larger than the time ratio of the signal L2.
δ2 = δ2′−δ1 ′ = − Vs / Vpt,
However, when δ2′−δ1 ′ <0, δ2 = 0 (10)

  When the signal M1 is “H” and the signal M2 is “L”, the voltage Vpn applied to the series body of the inductor 185 and the load 186 becomes a positive power supply voltage (Vh). Integrated in the positive direction. When the signal M1 is “L” and the signal M2 is “H”, the voltage Vpn becomes a negative power supply voltage (−Vh), and the inductor 185 integrates the current passing through the load 186 in the negative direction. Further, when the signal M1 is “L” and the signal M2 is “L”, the voltage between the terminals of the series circuit of the inductor 185 and the load 186 becomes zero V, so that no power is supplied from the outside.

As shown in the operation waveform diagram of FIG. 3, the state (HL) is when the time ratio δ1 ′ of the signal L1 is 50% or more, and the time ratio becomes the time ratio δ1 of the signal M1. Further, the state (LH) is set when the time ratio of the signal L1 is 50% or less, and the time ratio becomes the time ratio δ2 of the signal M2. From equations (8) and (10), since δ1 and δ2 are zero when one is positive, the other is zero. From another viewpoint, a positive power supply voltage is applied to the series body of the inductor 185 and the load 186. The amount of applying (Vh) is controlled by the time ratio δ1, and the amount of applying the negative power supply voltage (−Vh) is controlled by the time ratio δ2. From the above, the voltage Vo generated in the load 186 can be expressed by the following equation (11), and an output proportional to the input signal Vs can be obtained.
Vo = (δ1-δ2) · Vh = (Vh / Vpt) · Vs (11)

  In FIG. 3, when the input signal Vs is no signal, both the first switch circuit 182 and the second switch circuit 183 have a duty ratio of 0%. This is a significant difference from 50% of the duty ratio. Next, when the input signal Vs increases slightly positively or negatively, one of the time ratios of the first switch circuit 182 and the second switch circuit 183 slightly increases from 0%, and the other remains 0%. In order to maintain a low potential, only a narrow pulse from the switch circuit on one side is applied to the series body of the inductor 185 and the load 186.

  Thus, when there is no signal and a small signal, the pulse width applied to the series body of the inductor 185 and the load 186 is narrow, so that the current flowing when each switch is in a conductive state is reduced, and the on resistance of each switch Loss and loss due to the equivalent series resistance of the inductor 185 can be reduced. At the same time, the amount of electromagnetic waves radiated due to the current can be reduced. These are particularly effective for a signal with a low average level such as an audio signal because the average pulse width becomes narrow. The above effects are the same as those of the conventional second power amplification device. This effect is obtained by reducing the conduction loss of the main circuit 180, and these switching type power amplifying devices have a state in which the voltage Vpn applied to the series body of the inductor 185 and the load 186 can be This is because there are three states of positive power supply voltage, zero V or negative power supply voltage.

  By the way, when the impedance of the load 185 is high as in a portable device and the power to be extracted is small, the load current is also small, so that the conduction loss of the main circuit is also low and the loss of other circuits becomes conspicuous. A problem arises in that the efficiency of the system decreases. Generally, the power loss in the switching type power amplifying device includes one generated in the main circuit 180, one generated in the switch control circuit, and one generated in the modulation circuit. We have been talking about the conduction loss that occurs in The loss of the main circuit 180 includes a switching loss in addition to the conduction loss, which is mainly related to charging / discharging such as parasitic capacitance of each switch due to a high / low voltage change of the output of each switch circuit. Therefore, it is proportional to the number of on / off times.

  In order to solve such a problem, the power amplifying apparatus according to the first embodiment of the present invention reduces the loss generated in the switch control circuit when the on / off control of each switch of the main circuit 180 mentioned above is performed. This is realized by paying attention to. That is, for the loss generated in the switch control circuit, the power for charging / discharging the gate capacitance of the switches 1821, 1822, 1831 and 1832 of the N-channel FET and the level shift circuit of the high side switches 1821 and 1831 are used for signal transmission. It is largely due to current. The power due to charging / discharging of the former gate capacitance is intermittent, and depends on the capacitance viewed from the gate end of the FET, the required voltage value, and the number of switchings. The latter is, for example, power consumption due to the current that the level shift circuit of FIG. 2 constantly flows, and is due to the current that the resistors 1431 and 1432 take in from the floating power source side. Multiplied by the potential (potential difference from the source potential of the high breakdown voltage FETs 1212 and 1213) is a power loss.

  In the level shift circuit shown in FIG. 4, the current flowing through the resistors 1431 and 1432 is generally a constant current because of the circuit configuration, so that the loss generated in this portion increases when the positive potential of the floating power supply is high. In the configuration of FIGS. 1 and 2, the potential on the floating power source side of the resistors 1431 and 1432 is the potential of the voltage supplied from the power source 102 when the low-side switch 1822 is on. When the high-side switch 1821 is on, the voltage potential from the power source 102 temporarily stored in the floating power supply 144 and the output terminal voltage of the first switch circuit (corresponding to the power supply voltage of the DC power supply 103) Will be added. Such loss due to the signal transmission current for level shift in the switch control circuit is further increased when the maximum output voltage of the power amplifier is increased, that is, the voltage Vh of the DC power supply 103 of the main circuit 180 is increased. Then it will increase further.

  However, this loss can be reduced by shortening the period during which the output of each switch circuit is at the high potential as described above. The power amplifying device of the present embodiment is configured such that the period during which the output of each switch circuit is at a high potential is shorter than that of the conventional second power amplifying device, so that power consumption can be kept low. it can.

  The power consumption of a level shift circuit (not shown) that transmits a state signal from the state detection circuit 142 provided in the first high-side switch control circuit 140 to a system control unit configured with reference to the ground potential is also included. Considering this, the configuration of the power amplifying apparatus of this embodiment is effective for reducing loss. Further, since the period during which the output of each switch circuit is at a high potential is short, the average current discharged from the capacitor 1441 of the floating power supply circuit 144 is small, so that the capacitance value can be reduced.

  Even in the method of supplying power to the series body of the inductor 185 and the load 186 using the three states as described above, the output pulse voltage is positive during the positive period by the inductor 185 as in the conventional switching type power amplifier. Since the current is integrated in the positive direction, the negative period is integrated in the negative direction, and in the zero voltage period, the current is integrated in the direction of decreasing the absolute value in the zero direction. A proportional voltage is obtained. In addition, when the input signal Vs is no signal or small signal, the pulse width applied to the series body of the inductor 185 and the load 186 is narrow, so that the current output from the power source is reduced and the consumed power is reduced. And the amount of radiation of electromagnetic waves can be reduced.

  In addition, the state of the BTL 181 is “the output terminal of the first switch circuit 182 is at a high potential and the output terminal of the second switch circuit 183 is at a low potential (HL)”, “the first switch circuit 182 The output terminal of the second switch circuit 183 has a low potential, the output terminal of the second switch circuit 183 has a high potential (LH), “the output terminal of the first switch circuit 182 has a low potential, and the output terminal of the second switch circuit 183 has It includes only the three states of “low potential (LL)” and does not include the state of “the output terminal of the first switch circuit 182 is high potential and the output terminal of the second switch circuit 183 is high potential (HH)”. . For this reason, the period during which the output of each switch circuit is at a high potential is shortened when the signal Vs is no signal or very small. For this reason, the loss due to the current for signal transmission for level shift in the switch control circuit can be suppressed lower than that of the conventional switching type power amplifier. In particular, when driving a load with low impedance and low impedance, or when driving voltage is high like a piezoelectric device, power consumption can be suppressed more effectively.

  Furthermore, when a capacitance type load such as a piezoelectric device is driven with a switching type power amplifying apparatus having a bridge configuration as in the present embodiment, the electric charge temporarily accumulated in the piezoelectric device is converted into power again. Due to the characteristic of returning to the source (regeneration), the power supplied from the power source is further reduced. Therefore, although the piezoelectric device has a drawback that the drive voltage is high, there is almost no power loss due to its own capacitance, and in terms of power consumption, there is an advantage of saving power compared to a device using electromagnetic force. is there. Such an application is suitable for portable devices with limited power sources. Further, the point that the capacitance value of the capacitor of the floating power supply circuit can be reduced is also suitable for portable devices.

  By the way, in the conventional second power amplifying device, there is a state where the output terminals of the switch circuits of the main circuit are both at “H”. In this case, a current monitor circuit is required on the high-side switch side in order to detect the current in this (HH) state. For this reason, the high-side switch drive unit requires a detection circuit and a level shift circuit that transmits the detection result to the ground level, and the power of the high-side switch must be supplied from the floating power supply circuit, resulting in a large power loss. There were drawbacks. On the other hand, the power amplifying device according to the first embodiment of the present invention has the effect that the output terminal of each switch circuit of the main circuit is not in the “H” state, so that the passing current can be detected only by the low-side switch unit. Play.

  Furthermore, according to the present embodiment, if the output terminals of both switch circuits are fixed at a low potential when waiting for no input signal, the power supply from the power source of the main circuit during standby can be cut off. Power consumption can be reduced. In this standby state, the floating power supply of the high-side switch control circuit is in a charged state, so that it can operate promptly upon startup. Further, since the standby output potential is at the reference level, no DC potential is generated in the load even when the operation of the main circuit starts, and no pop noise is generated. In addition, when the power supply voltage of the main circuit is supplied from a boost converter or the like, if the outputs of both switch circuits are fixed at a low potential during the startup process of this converter, the load current is reduced, so that startup is easy and quick. Can be launched. Further, if such a standby or normal operation state is intermittently operated according to an input signal or the like, the power consumption can be further reduced. Furthermore, by gradually increasing the power supply voltage of the main circuit from the low voltage state to the specified value, there is an effect that the drive power supplied to the load by the main circuit can be smoothly raised.

  Further, since the output signal amplitude is determined by the time ratio between the power supply voltage and the modulation signal, if the power supply voltage is adjusted here, the output signal amplitude can also be adjusted, and the volume can be adjusted. In the conventional example, if this is applied, the volume can be adjusted similarly. In this case, since the reference level of the output voltage of each switch circuit is one half of the power supply voltage, the power supply voltage is adjusted. Noise may be mixed in due to fluctuations in the reference level. However, in the power amplification device of the present invention, since the reference level is fixed to the reference potential (ground potential in the embodiment), there is no mixing of noise due to the power supply voltage.

  In the present embodiment, the second signal L2 is generated by comparing the input signal Vs with the signal Vsr obtained via the linear inverter 1131 and the triangular wave signal Vt. The second triangular wave signal obtained by inverting the phase of the triangular wave signal Vt and the input signal Vs can be similarly obtained.

(Embodiment 2)
FIG. 5A is a block diagram of a second embodiment of the power amplifying apparatus of the present invention. In FIG. 5A, components having the same functions and configurations as those of the power amplifying device of the first embodiment shown in FIG. The difference between the second embodiment and the first embodiment is the configuration of the modulation circuit. Hereinafter, the configuration of the modulation circuit 210 of the present embodiment will be described.

  The modulation circuit 210 differs from the modulation circuit 110 of the first embodiment in the waveform of the triangular wave signal input to the comparators 1121 and 112. That is, in the first embodiment, the output signal Vt of the triangular wave generator 111 is input to the comparators 1121 and 112 as it is or with the polarity inverted, but in the present embodiment, the output signal Vt of the triangular wave generator 111 is input. A signal obtained by adding or subtracting the rectangular wave signal Vq to the output signal Vt is input to the comparators 1121 and 112. Therefore, the modulation circuit 210 includes a rectangular wave generator 22111 that generates a rectangular wave signal Vq synchronized with the triangular wave signal Vt, and a linear inversion that generates a polarity inversion signal Vqr by inverting the polarity of the rectangular wave signal Vq in an analog manner. And an adder 2151 and linear adders 2151 and 2152. The linear adder 2151 adds the signal Vq and the triangular wave signal Vt to generate and output a first triangular wave signal Vt1. The first triangular wave signal Vt1 is applied to the negative input of the comparator 1121. The linear adder 2152 adds the polarity inversion signal Vqr and the triangular wave signal Vt to generate and output a second triangular wave signal Vt2. The second triangular wave signal Vt2 is applied to the negative input of the comparator 1122. FIG. 5B shows the waveform of the first triangular wave signal Vt1 generated by adding the signal Vq and the triangular wave signal Vt.

  FIG. 6 is an operation waveform diagram of each part of the power amplifying apparatus of the present embodiment shown in FIG. 5A. The operation of the power amplifying apparatus will be described using the operation waveform diagram of FIG.

  The input signal Vs from the input signal source 101 is applied to the positive input of the comparator 1121, and the polarity is inverted by the linear inverter 1131 to become the signal Vsr, which is applied to the positive input of the comparator 1122. The first triangular wave signal Vt1 obtained by adding the rectangular wave signal Vq to the triangular wave signal Vt is applied to the negative input of the comparator 1121, and the polarity of the rectangular wave signal Vq to the triangular wave signal Vt is applied to the negative input of the comparator 1122. A second triangular wave signal Vt2 to which the inverted signal Vqr is added is applied. Therefore, as shown in FIG. 6, the first triangular wave signal Vt1 and the second triangular wave signal Vt2 have waveforms in which the rising period and the falling period of the triangular wave signal Vt are level-shifted in a complementary manner.

  The first signal L1 output from the comparator 1121 becomes a high level (“H”) when the input signal Vs is greater than the first triangular wave signal Vt1 (Vs> Vt1), and is output from the comparator 1122. The second signal L2 becomes “H” when the signal Vsr is greater than the second triangular wave signal Vt2 (Vsr> Vt2). For this reason, the signal L1 and the signal L2 have the same time ratio as the power amplification device of the first embodiment, but the phase of the signal L1 is advanced by a predetermined amount ΔT (+ ΔT), and the signal L2 is slightly delayed. (−ΔT) waveform. From the above operation, the configuration including the triangular wave generator 111, the rectangular wave generator 2111, the linear inverter 2150, and the linear adders 2151 and 2152 is substantially regarded as a time offset circuit that advances and delays the time axis of the triangular wave signal Vt. be able to.

  The configuration in which the first modulation signal M1 and the second modulation signal M2 are generated from the signal L1 and the signal L2 in the modulation circuit is the same as in the case of the power amplification device of the first embodiment. Therefore, the signal M1 has a waveform obtained by cutting the “H” section of the signal L2 from the “H” section of the signal L1, and the signal M2 has the “H” section of the signal L1 from the “H” section of the signal L2. It becomes the waveform which cut off the section.

  However, since there is a phase shift (± ΔT) between the signal L1 and the signal L2, the small time ratio side is clipped by a shift of 2 times ΔT (2ΔT) from the center of the large time ratio side. Therefore, the value of the time ratio of the signal M1 is zero when the time ratio of the signal L2 is subtracted from the time ratio of the signal L1 and the time ratio of the signal L2 is somewhat larger than the time ratio of the signal L1. The value of the time ratio of the signal M2 is zero when the time ratio of the signal L1 is subtracted from the time ratio of the signal L2 and the time ratio of the signal L1 is somewhat larger than the time ratio of the signal L2. When the signal Vs is no signal, the time ratio between the signal L1 and the signal L2 is 50%, but a narrow pulse corresponding to the phase shift (2ΔT) is generated in each of the signal M1 and the signal M2. To do.

  The following operation of main circuit 180 is the same as that of the power amplifying apparatus of the first embodiment. That is, when the first modulation signal M1 is “H”, the output of the first switch circuit 182 is at a high potential, and when it is “L”, the output is at a low potential. When the second modulation signal M2 is “H”, the output of the second switch circuit 183 is at a high potential, and when it is “L”, the output is at a low potential.

The states of the outputs of the first switch circuit 182 and the second switch circuit 183 include only the following three states.
(HL): the first output terminal is the power supply voltage, the second output terminal is the reference potential (LH): the first output terminal is the reference potential, and the second output terminal is the power supply voltage (LL): the first output The terminal is the reference potential and the second output terminal is also the reference potential

  Possible states of the voltage Vpn applied to the series body of the inductor 185 and the load 186 are three states of a positive power supply voltage (Vh), zero, or a negative power supply voltage (−Vh).

Here, the relationship between the pulse width (time width which is “H”) of the first modulation signal M1 and the second modulation signal M2 generated by the modulation circuit 210 and the input signal Vs will be described. Now, assuming that the amplitude of the triangular wave signal Vt is Vpt, the time ratio δ1 ′ of the signal L1 is expressed by the following equation (12).
δ1 ′ = (1 + Vs / Vpt) / 2 (12)

Further, since the sign of the input signal Vs is reversed, the duty ratio δ2 ′ of the signal L2 can be expressed by the following equation (13) in which the sign of the input signal Vs of the signal L1 is changed.
δ2 ′ = (1−Vs / Vpt) / 2 (13)

When the ratio of the average pulse width in the period T of the signal M1 is a time ratio δ1, and the ratio of the phase shift of the signals L1 and L2 to the period T is δx (δx = 2ΔT / T), the following equation (14) is obtained. Thus, when the value obtained by subtracting the time ratio δ2 ′ of the signal L2 from the time ratio δ1 ′ of the signal L1 is equal to or larger than twice (2 · δx) of δx, the time ratio δ1 is the difference (δ1 = δ1′− δ2 ′), and is between −2 · δx and 2 · δx, a value obtained by adding twice δ3 to half the difference (δ1 = (δ1′−δ2 ′) / 2 + δx), and −δx It is zero in the following cases.
When δ1′−δ2 ′ ≧ 2 · δx, δ1 = δ1′−δ2 ′ = Vs / Vpt,
-2 · δx <δ1'-δ2 '<2 · δx
δ1 = (δ1′−δ2 ′) / 2 + δx = Vs / Vpt / 2 + δx
When δ1′−δ2 ′ ≦ −2 · δx, δ1 = 0 (14)

Similarly, when the ratio of the average pulse width in the period T of the signal M2 is the time ratio δ2, the following equation (15) is obtained.
When δ2′−δ1 ′ ≧ 2 · δx, δ2 = δ2′−δ1 ′ = − Vs / Vpt,
-2 · δx <δ2'-δ1 '<2 · δx
δ2 = (δ2′−δ1 ′) / 2 + δx = −Vs / Vpt / 2 + δx
When δ2′−δ1 ′ ≦ −2 · δx, δ2 = 0 (15)

The voltage Vo generated in the load 186 can be expressed by the following equation (16), and an output proportional to the input signal Vs can be obtained.
Vo = (δ1-δ2) · Vh = (Vh / Vpt) · Vs (16)

  As described above, the power amplifying apparatus of the present embodiment compares the first triangular wave signal Vt1 obtained by adding the rectangular wave signal Vq to the triangular wave signal Vt and the input signal Vs in the modulation circuit 210, and The second triangular wave signal Vt2 obtained by subtracting the rectangular wave signal Vq from the triangular wave signal Vt is compared with the inverted input signal Vsr, and the first signal L1 and the second signal L2 are generated. A minute phase difference is provided. Therefore, even when the input signal Vs is no signal, a narrow pulse corresponding to this phase difference is generated in each of the first modulation signal M1 and the second modulation signal M2. As a result, the power amplifying device according to the present embodiment has the characteristics of the power amplifying device according to the first embodiment in which the period during which the output terminal of each switch circuit is at a high potential is shortened and power consumption is kept low. Further, non-linear distortion near the zero crossing of the signal or low signal level can be reduced. That is, since the pulse width becomes narrow near the zero cross, the proportion of the pulse width lost due to the switch operation delay or dead time setting increases, and the linearity is lost. However, in the above-described configuration, since the pulse is generated even near the zero cross, the voltage applied to the load is the difference between the positive voltage and the negative voltage, so that the influence can be reduced.

  In the power amplifying device of this embodiment, the rectangular wave signal Vq is added to the triangular wave signal Vt to obtain the first triangular wave signal Vt1, and the rectangular wave signal Vq is subtracted from the triangular wave signal Vt to obtain the second triangular wave. Although the signal Vt2 is obtained, the first and second triangular wave signals may be reversed, or one of the triangular wave signal Vt1 and the triangular wave signal Vt2 and the original triangular wave signal Vt may be used. . Furthermore, a voltage offset may be given to the input section of the comparator that generates the signal L1 or the signal L2.

  The gist of the present embodiment is to provide a minute phase difference between the first signal L1 and the second signal L2 in order to secure the pulse width of the modulation signals M1 and M2 near the zero cross of the input signal Vs. The time offset is obtained by adding / subtracting the DC voltage corresponding to the offset amount to / from the triangular wave signal Vt using the linear relationship between the time and the instantaneous value of the triangular wave signal Vt.

(Embodiment 3)
FIG. 7 is a block diagram of a power amplifying apparatus according to the third embodiment of this invention. 7, components having the same functions and configurations as those of the power amplifying device of Embodiment 1 are given the same reference numerals, and descriptions thereof are omitted. The power amplification device of the present embodiment is different from the power amplification device of the first embodiment in the configuration of the modulation circuit.

  That is, the modulation circuit 310 of this embodiment is different from the configuration of the modulation circuit 110 in FIG. 1 in that a delay circuit 316 is provided between the output of the comparator 1122 and the input of the AND gate 1144. The output signal L2 of the comparator 1122 becomes a signal L3 through the delay circuit 316 and is input to the inverter 1141 and the AND gate 1144.

  FIG. 8 is an operation waveform diagram of each part of the power amplifying device shown in FIG. The operation of the power amplifying apparatus will be described using the operation waveform diagram of FIG. The input signal Vs from the input signal source 101 is applied to the positive input of the comparator 1121, and the polarity is inverted by the linear inverter 1131 to become the signal Vsr, which is applied to the positive input of the comparator 1122. A triangular wave signal Vt is applied to the negative inputs of the comparator 1121 and the comparator 1122. The first signal L1 output from the comparator 1121 becomes a high level (“H”) when the input signal Vs is greater than the first triangular wave signal Vt1 (Vs> Vt1), and is output from the comparator 1122. The second signal L2 becomes “H” when the signal Vsr is greater than the second triangular wave signal Vt2 (Vsr> Vt2). The signal L2 is further output as the third signal L3 via the delay circuit 316. Let this delay time be ΔT.

  The configuration in which first modulation signal M1 and second modulation signal M2 are generated from first signal L1 and third signal L3 in modulation circuit 310 is the same as in the case of the power amplifying device of the first embodiment. . Accordingly, the signal M1 has a waveform obtained by cutting the “H” section of the signal L3 from the “H” section of the signal L1, and the signal M2 has the “H” section of the signal L1 from the “H” section of the signal L2. It becomes the waveform which cut off the section.

  However, since there is a phase shift (ΔT) due to delay in the third signal L3, the small time ratio side is cut off by ΔT shift from the center of the large time ratio side. Therefore, the value of the time ratio of the signal M1 is zero when the time ratio of the signal L2 is subtracted from the time ratio of the signal L1 and the time ratio of the signal L2 is somewhat larger than the time ratio of the signal L1. The value of the time ratio of the signal M2 is zero when the time ratio of the signal L1 is subtracted from the time ratio of the signal L2 and the time ratio of the signal L1 is somewhat larger than the time ratio of the signal L2. When the signal Vs is no signal, both the signal ratios of the signal L1 and the signal L2 are 50%, but a narrow pulse corresponding to the phase shift (ΔT) is generated in each of the signal M1 and the signal M2. To do.

  The following operation of main circuit 180 is the same as that of the power amplifying apparatus of the first embodiment. That is, when the first modulation signal M1 is “H”, the output of the first switch circuit 182 is at a high potential, and when it is “L”, the output is at a low potential. When the second modulation signal M2 is “H”, the output of the second switch circuit 183 is at a high potential, and when it is “L”, the output is at a low potential. The output states of the first switch circuit 182 and the second switch circuit 183 are “the output terminal of the first switch circuit 182 is at a high potential and the output terminal of the second switch circuit 183 is at a low potential (HL)”. , “The output terminal of the first switch circuit 182 is a reference potential and the output terminal of the second switch circuit 183 is a high potential (LH)”, “the output terminal of the first switch circuit 182 is a low potential and the second switch The output terminal of the circuit 183 is also in three states of “low potential (LL)”, and the possible state of the voltage Vpn applied to the series body of the inductor 185 and the load 186 is positive power supply voltage (Vh), zero V or negative The three power supply voltages (−Vh) are obtained.

Here, the relationship between the pulse width (time width which is “H”) of the signals M1 and M2 generated by the modulation circuit 310 and the input signal Vs will be described. Now, assuming that the amplitude of the triangular wave signal Vt is Vpt, the time ratio δ1 ′ of the signal L1 is expressed by the following equation (17).
δ1 ′ = (1 + Vs / Vpt) / 2 (17)
Further, the duty ratio δ3 ′ of the signal L3 can be expressed by the following equation (18) in which the sign of the input signal Vs of the signal L1 is changed because the sign of the input signal Vs is reversed.
δ3 ′ = (1−Vs / Vpt) / 2 (18)

Assuming that the ratio of the average pulse width in the period T of the signal M1 is the time ratio δ1, and the ratio of the phase shift of the signal L3 to the period T is δy (δy = ΔT / T), as shown in the following equation (19), When the value obtained by subtracting the time ratio δ3 ′ of the signal L2 from the time ratio δ1 ′ of the signal L1 is equal to or greater than twice (2 · δy) of δy, the time ratio δ1 is the difference (δ1 = δ1′−δ3 ′). If it is between -2 · δy and 2 · δy, the value obtained by adding twice δ3 to the half of the difference (δ1 = (δ1'-δ3 ') / 2 + δy), which is -2 · δy or less In case of, it becomes zero.
When δ1′−δ3 ′ ≧ 2 · δy, δ1 = δ1′−δ2 ′ = Vs / Vpt,
-2 · δy <δ1'-δ3 '<2 · δy
δ1 = (δ1′−δ3 ′) / 2 + δy = Vs / Vpt / 2 + δy
When δ1′−δ3 ′ ≦ −2 · δy, δ1 = 0 (19)

Similarly, when the ratio of the average pulse width in the period T of the signal M2 is the time ratio δ2, the following equation (20) is obtained.
When δ3′−δ1 ′ ≧ 2 · δy, δ2 = δ3′−δ1 ′ = − Vs / Vpt,
-2 · δy <δ3'-δ1 '<2 · δy,
δ2 = (δ3′−δ1 ′) / 2 + δy = −Vs / Vpt / 2 + δy
When δ3′−δ1 ′ ≦ −2 · δy, δ2 = 0 (20)
The voltage Vo generated in the load 186 can be expressed by the following equation (21), and an output proportional to the input signal Vs can be obtained.
Vo = (δ1-δ2) · Vh = (Vh / Vpt) · Vs (21)

  As described above, in the power amplifying apparatus of the present embodiment, the delay circuit 316 is provided in the modulation circuit 310, and the second signal L2 output from the comparator 1122 becomes the third signal L3 via the delay circuit 316. By adopting a configuration in which the signal is input to the inverter 1141 and the AND gate 1144, a minute phase difference is provided between the two signals. Therefore, a narrow pulse corresponding to this phase difference is generated in each of the first modulation signal M1 and the second modulation signal M2 even when the input signal Vs is no signal. As a result, the power amplifying device according to the third embodiment has the characteristics of the power amplifying device according to the first embodiment in which the period during which the output terminal of each switch circuit is at a high potential is shortened to reduce power consumption. Furthermore, non-linear distortion near the zero crossing of the signal or low signal level can be reduced. That is, since the pulse width becomes narrow near the zero cross, the proportion of the pulse width lost due to the switch operation delay or dead time setting increases, and the linearity is lost. However, in the above-described configuration, since the pulse is generated even near the zero cross, the voltage applied to the load is the difference between the positive voltage and the negative voltage, so that the influence can be reduced.

  In the power amplification device according to the third embodiment of the present invention, the second signal L2 is delayed to generate the third signal L3. However, the first signal L1 is delayed to generate the third signal L3. You may do it. Further, a time delay may be provided for the analog signal in the previous stage of the comparators 1121 and 1122 that generate the signal L1 or L2.

  The gist of the present embodiment is that the time offset for providing a minute phase difference between the first signal L1 and the second signal L2 in order to secure the pulse width in the vicinity of the zero crossing as described above. This is obtained by using a delay circuit. The time delay circuit may be provided either before or after the comparator that obtains the signal L1 or the signal L2.

(Embodiment 4)
FIG. 9 is a block diagram of a power amplifying apparatus according to Embodiment 4 of the present invention. In FIG. 9, components having the same functions and configurations as those of the power amplification device of the first embodiment are given the same reference numerals, and descriptions thereof are omitted. The power amplifier device according to the fourth embodiment of the present invention differs from the power amplifier device according to the first embodiment in the configuration of the modulation circuit.

  The modulation circuit 410 of this embodiment includes an input signal source 101 that generates an input signal Vs, a triangular wave generator 111 that generates a triangular wave signal Vt, a constant voltage source 4153 that outputs a DC voltage V4153, and a constant voltage that outputs a DC voltage V4127. A source 4127, a constant voltage source 4128 that outputs a DC voltage V4128, a linear inverter 2150, linear adders 2151 and 2152, comparators 1121, 1122, 1123, 1124, 4125 and 4126, OR gates 4145 and 4146, an AND gate 1143 and 1144 is included.

  An input signal Vs from the input signal source 101 is applied to the positive input of the comparator 1121 and the negative input of the comparator 1122. The DC voltage V4153 from the constant voltage source 4153 is applied to the linear adder 2151, the polarity is inverted by the linear inverter 2150, and the output voltage (−V4153) is applied to the linear adder 2152. The linear adder 2151 adds the triangular wave signal Vt and the voltage V4153 and outputs a triangular wave signal Vtp (= Vt + V4153). The linear adder 2152 adds the triangular wave signal Vt and the output voltage (−V4153) of the linear inverter 2150 and outputs a triangular wave signal Vtn (= Vt−V4153). Here, assuming that the triangular wave signal Vt is increased or decreased with the amplitude Vpt around the ground potential zero V, the voltage V4153 is set to a value slightly smaller than the triangular wave amplitude Vpt.

  The positive input of the comparator 1123 and the negative input of the comparator 1124 are grounded. The DC voltage V4127 from the constant voltage source 4127 is applied to the positive input of the comparator 4125. DC voltage V4127 is set to a value slightly smaller than twice the triangular wave amplitude Vpt. The DC voltage V4127 from the constant voltage source 4127 is applied to the positive input of the comparator 4126. DC voltage V4128 is a negative voltage, and its absolute value is equal to DC voltage V4127, and is set to a value slightly smaller than twice the triangular wave amplitude Vpt. A triangular wave signal Vtp is applied to each negative input of the comparator 1121, the comparator 1123, and the comparator 4125. A triangular wave signal Vtn is applied to the positive inputs of the comparator 1122, the comparator 1124, and the comparator 4126.

  The first signal L1 output from the comparator 1121 and the third signal L11 output from the comparator 1123 are input to the OR gate 4145. The output signal of the OR gate 4145 and the fifth signal L12 output from the comparator 4125 are input to the AND gate 1143. The second signal L2 output from the comparator 1122 and the fourth signal L21 output from the comparator 1124 are input to the OR gate 4146. The output signal of the OR gate 4146 and the sixth signal L22 output from the comparator 4126 are input to the AND gate 1144. The first modulation signal M1 output from the AND gate 1143 and the second modulation signal M2 output from the AND gate 1144 are input to the distribution circuits 120 and 130 as output signals of the modulation circuit 410, respectively.

  FIG. 10 is an operation waveform diagram of each part of the power amplifying device shown in FIG. The operation of the power amplifying apparatus will be described using the operation waveform diagram of FIG.

  The triangular wave signal Vtp is shifted to a high potential by a voltage V4153 that is slightly smaller than the amplitude Vp. Therefore, the triangular wave signal Vtp crosses the ground potential just before turning from falling to rising, and crosses the voltage V4127 just before turning from rising to falling. Therefore, the third signal L11 output from the comparator 1123 is a narrow pulse generated when the triangular wave signal Vtp changes from falling to rising, and the fifth signal L12 output from the comparator 4125 is a triangular wave. This is a pulse having a narrow gap when the signal Vtp changes from rising to falling. On the other hand, the triangular wave signal Vtn is also shifted to a low potential by a voltage V4153 that is slightly smaller than its amplitude Vpt. Therefore, the triangular wave signal Vtn intersects with the ground potential immediately before turning from rising to falling and intersects with the voltage V4128 just before turning from falling to rising. Therefore, the fourth signal L21 output from the comparator 1124 is a narrow pulse generated when the triangular wave signal Vtn changes from rising to falling, and the sixth signal L22 output from the comparator 4126 is the triangular wave signal Vtn. Becomes a pulse having a narrow gap as it goes from descending to rising. For simplicity of explanation, the widths of these narrow gaps and narrow pulses are all set to ΔT.

  Since the first signal L1 is a comparison result between the input signal Vs and the triangular wave signal Vtp, the first signal L1 has a pulse width when the input signal Vs is slightly higher than the ground potential, and the pulse width is the input signal Vs. The higher the value, the wider. Similarly, since the second signal L2 is a comparison result between the input signal Vs and the triangular wave signal Vtn, the second signal L2 has a pulse width when the input signal Vs is not higher than the ground potential, and the pulse width is The lower the input signal Vs, the wider.

  The first modulation signal M1 is a logical product of the logical sum of the first signal L1 and the third signal L11 and the fifth signal L12. Therefore, the signal M1 is provided with a narrow pulse generated when the triangular wave signal Vtp turns from rising to rising and a narrow gap when turning from rising to falling in the signal L1. That is, the time ratio of the signal M1 reflects the time ratio of the signal L1 within a range in which the time ratio of the signal L11 is the minimum value and the time ratio of the signal L12 is the maximum value. On the other hand, the second modulation signal M2 is a logical product of the logical sum of the second signal L2 and the fourth signal L21 and the sixth signal L22. Therefore, the signal M2 is provided with a narrow pulse generated when the triangular wave signal Vtn turns from rising to falling and a narrow gap when turning from rising to falling. That is, the time ratio of the signal L2 reflects the time ratio of the signal L2 within a range in which the time ratio of the signal L21 is the minimum value and the time ratio of the signal L22 is the maximum value.

  Since the operation of the main circuit 180 is the same as that of the power amplifying device of the first embodiment, description thereof is omitted here.

Here, the relationship between the pulse widths of the first modulation signal M1 and the second modulation signal M2 generated by the modulation circuit 410 and the input signal Vs will be described. Now, assuming that the time ratio between the signal L11 and the signal L12 is δz (= ΔT / T), the time ratio δ1 of the first modulation signal M1 is expressed by the following equation (22).
When Vs <0, δ1 = δz
When 0 ≦ Vs / Vpt ≦ 2-δz, δ1 = δz + Vs / Vpt / 2
When Vs / Vpt> 1-2δz, δ1 = 1−δz (22)
Further, the time ratio δ2 of the second modulation signal M2 is expressed by the following equation (23).
When Vs> 0, δ2 = δz
When δz−2 ≦ Vs / Vpt ≦ 0, δ2 = δz−Vs / Vpt / 2
In the case of Vs / Vpt <δz−2, δ2 = 1−δz (23)

The voltage Vo generated in the load 186 can be expressed by the following equation (24), and an output proportional to the input signal Vs can be obtained.
Vo = (δ1-δ2) · Vh = (Vh / Vpt) · Vs (24)
However, | Vs / Vpt | ≦ 2-δz.

  As described above, in the fourth embodiment of the present invention, in the modulation circuit 410, two triangular waves shifted from the triangular wave signal Vt of the triangular wave generator 111 positively or negatively by the first DC voltage V4153 that is smaller than the triangular wave amplitude Vpt by a predetermined amount. Signals Vtp and Vtn are generated. A first signal L1 that is a comparison output between the first triangular wave signal Vtp and the input signal Vs, a third signal L11 that is a comparison output between the triangular wave signal Vtp and a zero voltage, and a second voltage that is less than twice the voltage V4153. A fifth signal L12 that is a comparison output with the DC voltage V4127 is generated.

  A logical product of the logical sum of the first signal L1 and the third signal L11 and the fifth signal L12 is output as the first modulation signal M1. Further, the second signal L2 which is a comparison output between the second triangular wave signal Vtn and the input signal Vs, the fourth signal L21 which is a comparison output between the triangular wave signal Vtn and the zero voltage, and a negative potential of the voltage V4153. A sixth signal L22 that is a comparison output with the third DC voltage V4128 that is twice or less the absolute value is generated. By configuring the logical product of the logical sum of the second signal L2 and the fourth signal L21 and the sixth signal L22 as the second modulation signal M2, the ratio between the signal L11 and the signal L12 is increased. Pulse signals M1 and M2 having a minimum time ratio and a maximum time ratio are obtained. Therefore, even when the signal Vs is no signal, a narrow pulse corresponding to the minimum duty ratio is output to each of the signal M1 and the signal M2.

  As a result, the power amplifying device according to the fourth embodiment has the characteristics of the power amplifying device according to the first embodiment in which the period during which the output terminal of each switch circuit is at a high potential is shortened to reduce power consumption. Further, non-linear distortion near the zero crossing of the signal or low signal level can be reduced. That is, near the zero cross, the pulse width is narrowed, so that the ratio of the pulse width lost due to the switch operation delay and the dead time setting increases, and the linearity is lost. However, in the above-described configuration, since the pulse is generated even near the zero cross, the voltage applied to the load is the difference between the positive voltage and the negative voltage, so that the influence can be reduced.

  Further, when the amplitude of the input signal Vs exceeds twice the amplitude Vt of the triangular wave, the first signal L1 is always “H” and the second signal L2 is always “L”. However, since the first modulation signal M1 is ANDed with the narrow gap of the pulse of the fifth signal L12 and output, the period of this narrow gap is “L”. At the same time, the pulse width of the fourth signal L21 is output to the second modulation signal M2. By setting the maximum duty ratio as described above, a situation where the state is fixed to the “H” or “L” state throughout one cycle of the triangular wave signal Vt does not occur.

  In the above maximum duty ratio setting mechanism, the pulse widths of the third signal L11 and the fourth signal L21 and the narrow gaps of the pulses of the fifth signal L12 and the sixth signal L22 are both ΔT. Therefore, the narrow gap of the first modulation signal M1 matches the pulse width of the second modulation signal M2. The actual setting needs to be set so that the narrow gap of the first modulation signal M1 is wider than the pulse width of the second modulation signal M2 in consideration of variations. If the pulse width of the second modulation signal M2 becomes wider than the narrow gap of the first modulation signal M1, both the output of the first switch circuit and the output of the second switch circuit become “H”. A period occurs, and the basic idea of the present application cannot be realized. This is the reason why the magnitudes of the second DC voltage V4127 and the third DC voltage V4128 are less than or equal to twice the first DC voltage V4153. Further, it is preferable that the minimum no-pulse interval between the first modulation signal M1 and the second modulation signal M2 is larger than the minimum pulse width of the first modulation signal M1 and the second modulation signal M2.

  The third signal L11 is the output of the comparator 1123 that compares the triangular wave signal Vtp with the zero voltage, and this zero voltage corresponds to the reference potential of the input signal Vs when the AC amplitude of the input signal Vs is zero. To do. Now, when the AC amplitude of the input signal Vs is zero, since the triangular wave signal Vtp and the input signal Vs having an AC amplitude of zero are also input to the comparator 1121, the first signal L1 that is the output is the first signal L1. 3 is the same as the signal L11. Next, when the instantaneous value of the input signal Vs increases to the positive side of the reference potential, the pulse width of the first signal L1 becomes larger than the pulse width of the third signal L11. On the other hand, when the pulse width decreases to the negative side, the pulse width of the first signal L1 becomes smaller than the pulse width of the third signal L11. When the instantaneous value decreases further to the negative side, no pulse is generated. The first signal L1 and the third signal L11 are ORed by an OR gate 4145. The pulse width output from the OR gate 4145 follows the pulse width of the first signal L1 when the input signal Vs increases to the positive side, and decreases to the negative side when the input signal Vs increases to the positive side. Fixed at the pulse width. The operation when the input signal Vs on the triangular wave signal Vtn side is negative is the same as this.

  Thus, since the pulse width generation mechanism is switched near the zero amplitude of the input signal Vs, linearity is lost if there is discontinuity or delay in the vicinity of this switching. Therefore, in the configuration like the power amplifier according to the fourth embodiment of the present invention, it is necessary to reduce the input offset voltage of each comparator.

  When the above-mentioned maximum time ratio setting mechanism is mounted on a conventional power amplifier, even if the amplitude of the input signal Vs exceeds the amplitude Vt of the triangular wave, the equations (1), (2), (4), (5) | Vs / Vpt | <1 is satisfied, and the output of each comparator is not fixed to the “H” or “L” state throughout one cycle of the triangular wave signal Vt. . Therefore, there is an advantage that a mechanism for preventing an excessive signal from being applied to the input unit becomes unnecessary.

  In the first to fourth embodiments, the triangular wave generator and the input signal are compared, and each signal is applied to a two-input comparator. As an alternative, the triangular wave generator and the input signal may be added in consideration of their signs and then applied to a one-input comparator incorporating a reference value.

(Embodiment 5)
FIG. 11 is a block diagram of a power amplifying apparatus according to Embodiment 5 of the present invention. In FIG. 11, components having the same functions and configurations as those of the power amplifying device of Embodiment 1 are given the same reference numerals, and descriptions thereof are omitted. The power amplifying apparatus of the present embodiment is different from the power amplifying apparatus of the first embodiment in the configuration of the modulation circuit.

  The modulation circuit 510 of this embodiment is an input signal source 101 that generates an input signal Vs, linear adders 5171 and 5193, a weighting filter 5172 such as an integrator, a determiner 518, and first and second latch circuits D. Latches 5191 and 5192 and a clock generator 5194 are provided.

  The determiner 518 includes comparators 5181 and 5182, a linear inverter 5183, and a constant voltage source 5184. The clock generator 5194 quantizes the rising or falling time of the pulse applied to the D input of each D latch 5191, 5192.

  The configuration of the modulation circuit 510 is for a power amplification device of a type that does not use a triangular wave generator as disclosed in Patent Document 2 and Patent Document 3. This is to obtain the difference signal between the input signal Vs and the feedback pulse signal from the output, pass it through a weighting filter (difference integrator), quantize the voltage in the comparator, The switch circuits 182 and 183 of the main circuit 180 are operated by the pulse signal after the time axis is quantized by the delay unit. A negative feedback operation is performed so as to reduce the difference signal between the input signal Vs and the feedback pulse signal from the output, so that a voltage proportional to the input signal Vs is supplied to the load connected to the main circuit. .

  The comparators in the power amplifying devices of Patent Document 2 and Patent Document 3 have one input threshold value and two comparison outputs. In the fifth embodiment of the present invention, the input threshold value is 2. The difference is that the comparison output is ternary. Further, it differs from Patent Document 2 and Patent Document 3 in that the three values output from the determiner 518 are allocated so as to be in three states of the main circuit. That is, “the output terminal of the first switch circuit is a high potential and the output terminal of the second switch circuit is a low potential (HL)”, “the output terminal of the first switch circuit is a low potential and the second switch circuit By assigning the output terminal to the three states of “high potential (LH)” and “the output terminal of the first switch circuit is low potential and the output terminal of the second switch circuit is also low potential (LL)”. The same effects as those of the first to fourth embodiments can be obtained.

Details of the modulation circuit 510 will be described below. First, the input signal Vs from the input signal source 101 is subtracted by the pulse-modulated output signal fed back by the linear adder 5171, and the output of the linear adder 5171 is pulse-modulated with the input signal Vs. An error signal of the output signal is obtained. This error signal is input to the determiner 518 after the high frequency component of the pulse is suppressed by the weighting filter 5172. In the determiner 518, the comparator 5181 and the comparator 5182 are set with thresholds having different polarities.
When the polarity of the error signal output from the linear adder 5171 is positive and larger than the threshold value of the comparator 5181, the first signal L1 output from the comparator 5181 is “H”, and the comparator 5182 The output second signal L2 is “L”. When the absolute value of the error signal is smaller than the threshold value, both the signal L1 and the signal L2 are “L” regardless of the polarity. When the error signal is negative and its absolute value is larger than the threshold value of the comparator 5182, the signal L1 is “L” and the signal L2 is “H”.

  The signal L1 and the signal L2 are respectively applied to the data inputs of the D latch 5191 and the D latch 5192, and the respective input states are output and held at the timing of the clock generator 5194 in the respective Q outputs. The first modulation signal M1 and the second modulation signal M2 are output. The signals M1 and M2 control the output state of the main circuit 180 via the distribution circuits 120 and 130. For example, when the signal M1 is “H” or “L”, the first state of the main circuit 180 is controlled. The output of the switch circuit is also “H” or “L”, respectively, and when the other signal M2 is “H” or “L”, the output of the second switch circuit of the main circuit 180 is also “H”. Or it is comprised so that it may become "L".

  Therefore, depending on the circuit configuration, depending on the polarity and magnitude of the output signal of the weighting filter 5172, a positive voltage, a zero voltage, or a negative voltage is applied to the load connected to the main circuit 180. As a result, the same three states as in the first to fourth embodiments are output. On the other hand, the first modulation signal M1 and the second modulation signal M2 are also supplied to the linear adder 5193, respectively, so that the polarity is such that the signal M2 is subtracted from the signal M1, and (M1-M2) is linearly added. Returned to one side of vessel 5171.

  Here, the behavior when the input signal Vs from the input signal source 101 suddenly increases will be described. When the input signal Vs increases suddenly, the output of the linear adder 5171 increases because the feedback signal from the linear adder 5193 has not yet changed. For this reason, since the output of the weighting filter 5172 that is cascade-connected thereto also increases, the values of the outputs (L1, L2) of the determiner 518 attempt to transition in the following directions, respectively. That is, when the output value of the weighting filter 5172 is lower than the threshold value of either the comparator 5181 or the comparator 5182, the value first changes to a value between the threshold values of the comparator 5181 and the comparator 5182. Therefore, the output (L1, L2) changes from (L, H) to (L, L). From this point, when the output value of the weighting filter 5172 further increases, the output (L1, L2) changes from (L, L) → It changes to (H, L). Since these outputs (L1, L2) appear at the outputs (M1, M2) of the modulation circuit at the clock time due to the configuration of the D latch 5191, the D latch 5192, and the clock generator 5194, the timing is Although delayed, the outputs (M1, M2) are (L, H) → (L, L) and (L, L) → (H, L), respectively, similarly to the outputs (L1, L2). To change. Therefore, (M1-M2) appears in the output of the linear adder 5193 due to the circuit configuration, and changes in the direction of increasing voltage, such as “negative → zero” and “zero → positive”, respectively. These changes increase the feedback signal from the linear adder 5193 and induce the absolute value of the output of the linear adder 5171 to decrease. Similarly, when the input signal Vs suddenly decreases, the output (M1-M2) of the linear adder 5193 changes in a decreasing direction such as “positive → zero” and “zero → negative”. Is also guided in the direction of decreasing the absolute value of the output of the linear adder 5171. In this way, the feedback signal (M1-M2) from the linear adder 5193 performs a negative feedback operation that changes following the signal Vs input to the modulation circuit 510.

  The output of the linear adder 5171 obtained as described above includes an analog signal obtained by subtracting a digital feedback signal (M1-M2) having positive, zero, and negative three values from the input signal Vs that is an analog signal. A digital mixed error signal appears. This high frequency component is removed by the weighting filter 5172 and then obtained as an error signal obtained by analogizing the digital component. When the gain of the loop transfer function of the feedback system is increased, the amplitude of the error signal in the band of the weighting filter 5172 is reduced. Therefore, the feedback signal (M1-M2) which is a digital signal having three values is integrated, It is proportional to the input signal Vs. On the other hand, the voltage applied to the series body of the load 186 and the inductor 185 connected to the main circuit 180 is proportional to the feedback signal (M1-M2) output from the linear adder 5193 because of the circuit configuration. Thereafter, the voltage applied to the load 186 is proportional to the input signal Vs. Since the high frequency component of the voltage applied to the load 186 is removed by the inductor 185, the voltage returns to an analog signal proportional to the input signal Vs. In particular, when the magnitude of the input signal Vs is small, the feedback signal output from the linear adder 5193 closely follows this, so that the signal M1 and the signal M2 are frequently operated and the pulse width becomes narrow.

  Even in the method using the modulation circuit 510 that outputs such three states, the current is integrated in the positive direction by the inductor 185 during the period when the pulse voltage of the output is positive, as in the conventional switching type power amplifier. Since the negative period is integrated in the negative direction and the zero V period is integrated in the direction in which the absolute value of the current decreases, a voltage proportional to the input signal Vs is obtained at the load 186, and no signal and At the time of a small signal, the pulse width applied to the series body of the inductor 185 and the load 186 is narrow, the current output from the power source is reduced, the consumed power can be reduced, and the radiation amount of electromagnetic waves is also reduced. can do.

  Further, the state of the BTL 181 is “the output terminal of the first switch circuit is high potential and the output terminal of the second switch circuit is low potential (HL)”, and “the output terminal of the first switch circuit is low potential and the first terminal is low. 2, the output terminal of the second switch circuit is at a high potential (LH), and “the output terminal of the first switch circuit is at a low potential and the output terminal of the second switch circuit is also at a low potential (LL)”. Therefore, since the period during which the output of each switch circuit is at a high potential is shortened when the input signal Vs is no signal or very small, the loss due to the signal transmission current for level shift in the switch control circuit is The switching type power amplifying apparatus can be kept lower. Thereby, especially when driving a load with a small power to be taken out or when the driving voltage is high like a piezoelectric device, the power consumption can be reduced.

  As described above, the power amplifying device according to the fifth embodiment has the advantage that it is not necessary to provide a low-distortion triangular wave with good linearity required for the power amplifying devices according to the first to fourth embodiments using a triangular wave generator. , Has the effects as described above.

  The weighting filter 5172 is set so that the one-round transfer function is stable, including the influence of the delay of time quantization by the D latches 5191 and 5192 and the clock generator 5194. The weighting filter 5172 removes quantization noise such as harmonics included in the ternary feedback signal from the linear adder 5193. In such a power amplifying apparatus, in order to increase the bandwidth and dynamic range of the passing signal, it is necessary to increase the frequency of the clock generator 5194 or to increase the resolution of the feedback signal. The former is not preferable because the switching frequency of the main circuit becomes high, and the loss increases when the control signal of the main circuit is directly used. Further, as a method for increasing the latter resolution, there is a further increase in the number of comparators after the weighting filter 5172, but the consistency with the three states of the main circuit is not good as it is.

  As a method for matching them, for example, in the former case, a time resolution expressed by a pulse with a high repetition frequency generated with a high clock is used, and a ternary value having a resolution with a pulse width finely adjusted with a low repetition frequency. It is conceivable to perform signal processing in the digital domain in which the pulse width is expressed as follows. Similarly, the latter expresses the amplitude resolution expressed by the multi-value amplitude generated by the multi-valued comparator as the ternary pulse width having the resolution with the pulse width finely adjusted at a low repetition frequency. To do. As described above, the modulation circuit 510 configured as shown in FIG. 11 as in the fifth embodiment of the present invention may use a higher clock frequency or may include a comparator for determining more levels. In particular, if the state representations (M1, M2) of the signals M1 and M2 are three states (H, L), (L, L), and (L, H), respectively, the above effects can be obtained. .

  By the way, in order to perform time-quantized processing after the linear adder 5171, an anti-alias filter (not shown) is required between the input signal source 101 and the linear adder 5171. However, a high clock like the former is required. When the frequency is used, the shoulder characteristic of the filter can be relaxed, so that the configuration can be simplified. Then, the transfer characteristic of the weighting filter 5172 affects the spectrum of the error signal input to the determiner 518, and the reciprocal characteristic of the transfer function including time quantization by the D latches 5191 and 5192 and the clock generator 5194 is In the form of noise spectrum, which is an error signal left after negative feedback. Therefore, the noise spectrum can be manipulated by the frequency characteristics of the weighting filter 5172, and the noise level of the reproduction bandwidth can be lowered by moving this noise energy to the upper limit of the reproduction bandwidth.

  Further, in the configuration of FIG. 11, the negative feedback to the linear adder 5171 is (M1−M2), which is the difference between the first modulation signal M1 and the second modulation signal M2, which is the main circuit 180, respectively. This may be the difference between the output terminal of the first switch circuit and the output terminal of the second switch circuit. In this case, distortion due to delay or the like generated in the main circuit can also be reduced.

  In the modulation circuit having the configuration of FIG. 11, the linear adders 5171 and 5193, the weighting filter 5172, the comparators 5181 and 5182, and the like are represented by analog circuits. You may implement | achieve by the logic operation and algebra operation of.

  In this embodiment, the threshold value of the determiner is set to have the same absolute value with different positive and negative polarities because the average value of the output signal of the weighting filter 5172 is zero. This value may be set in any way.

(Embodiment 6)
FIG. 12 is a block diagram of the power amplifying apparatus according to the sixth embodiment. 12, components having the same functions and configurations as those of the power amplifying device of Embodiment 1 are given the same reference numerals, and descriptions thereof are omitted. The power amplifying apparatus of the present embodiment is different from the power amplifying apparatus of the first embodiment in the configuration of the modulation circuit.

  The modulation circuit 610 of the present embodiment includes a feedback input circuit 1010 that uses an input signal as a differential input, a feedback signal conversion circuit 6100 from an output terminal of each switch circuit of the main circuit 180 to the feedback input circuit 1010, and a modulation signal generation circuit. 6200. The rectangular wave generator 2111 generates a rectangular wave signal Vq synchronized with the triangular wave signal Vt of the triangular wave generator 111.

  The feedback input circuit 1010 constitutes a balanced LPF that removes the harmonic components of the feedback signals Q1 and Q2 from the feedback signal conversion circuit 6100 and the operational amplifier 1011 that receives the differential input signal via the resistors 1012 and 1013. Resistors 1018 and 1019, a capacitor 1020, a resistor 1014 that applies this balanced output to the non-inverting input terminal of the operational amplifier 1011, a resistor 1015 that applies the inverting input terminal thereof, and an output terminal and a non-inverting input terminal of the operational amplifier 1011 And a resistor 1017 connected between the inverting input terminal of the operational amplifier 1011 and the ground, and outputs the output of the operational amplifier 1011 as a signal Vs.

  The feedback signal conversion circuit 6100 includes a level converter 6115 that converts the output terminal of the first switch circuit 182 to a logic level, an inverter 6101 to which the output signal is input, a NOR gate 6102, and an AND gate 6103. The output of the inverter 6101 is input to a NOR gate 6102 and an AND gate 6103. The feedback signal conversion circuit 6100 includes a level converter 6116 that converts the output voltage of the second switch circuit 183 into a logic level, an inverter 6104 to which the output signal is input, a NOR gate 6105, and an AND gate 6106. The output of the inverter 6104 is input to a NOR gate 6105 and an AND gate 6106.

  The feedback signal conversion circuit 6100 includes an OR gate 6107 to which outputs of the NOR gate 6102 and the AND gate 6106 are input, an OR gate 6108 to which outputs of the NOR gate 6105 and the AND gate 6103 are input, and an output of the OR gate 6107. And an AND gate 6109 to which the rectangular wave signal Vq is input, an AND gate 6110 to which an output of the OR gate 6108 and an inverted signal of the rectangular wave signal Vq are input, an output of the OR gate 6108 and the rectangular wave signal Vq are input. The AND gate 6111, the AND gate 6112 to which the output of the OR gate 6107 and the inverted signal of the rectangular wave signal Vq are input, the output signal R1 of the AND gate 6109 is input to the reset terminal, and the output signal S1 of the AND gate 6110 is set. RS latch 6113 input to the terminal and AND Output signal R2 of the RS gate 6111 is input to the reset terminal, and the output signal S2 of the AND gate 6112 is input to the set terminal. The RS latch 6114 has the output Q1 of the RS latch 6113 and the output Q2 of the RS latch 6114. The input to the feedback input circuit 1010 is configured.

  Now, to simplify the explanation, consider the case where there is no influence of the feedback signals Q1 and Q2. In such a feedback input circuit 1010, when the resistor 1012 and the resistor 1013 have the same resistance value R1, the resistor 1016 and the resistor 1017 have the same resistance value R2, respectively, and the amplification factor of the operational amplifier 1011 is sufficiently large, A difference voltage (Vsp−Vsn) between the signals Vsp and Vsn is output as the signal Vs. The modulation circuit 610 outputs signals M1 and M2 by the signal Vs. The signal M1 drives the first switch circuit 182 via the distribution circuit 120 and the switch control circuits 140 and 150. The signal M2 also drives the second switch circuit 183 via the distribution circuit 130 and the switch control circuits 160 and 170. The above operation is the same as that of the power amplifying apparatus of the first embodiment shown in FIG.

  FIG. 13 is an operation waveform diagram of each part of the power amplifying device shown in FIG. The operation of the feedback signal conversion circuit 6100 will be described below using the operation waveform diagram of FIG.

  The output Vp of the first switch circuit 182 has a waveform in which the amplitude of the signal M1 is enlarged by the main circuit 180 and a delay time due to dead times of the high-side switch 1821 and the low-side switch 1822 is generated. The signal having the timing of the output Vp that has become the logic level by the level converter 6115 is synchronized with the rising and falling of the output Vp of the first switch circuit 182 by the inverter 6101, the NOR gate 6102, and the AND gate 6103. Output a pulse. Similarly, the signal having the timing of the output Vn that has become the logic level by the level converter 6116 is synchronized with the rise and fall of the output Vn of the second switch circuit 183 by the inverter 6104, the NOR gate 6105, and the AND gate 6106. Output one-shot pulse.

  The one-shot pulse synchronized with the rising edge of the signal Vp and the one-shot pulse synchronized with the falling edge of the signal Vn are ANDed with the rectangular wave signal Vq via the OR gate 6107 and input to the RS latch 6113 as the reset signal R1. . The one-shot pulse synchronized with the falling edge of the signal Vp and the one-shot pulse synchronized with the rising edge of the signal Vn are ANDed with the inverted signal of the rectangular wave signal Vq via the OR gate 6108 and input to the RS latch 6113 as the set signal S1. Is done. Therefore, the output signal Q1 of the RS latch 6113 becomes H level in synchronization with the rising of the output Vp of the first switch circuit 182 when the rectangular wave signal Vq is L level, and when the rectangular wave signal Vq is H level. It becomes L level in synchronization with the fall of the output Vp of the first switch circuit 182. The signal Q1 becomes H level in synchronization with the fall of the output Vp of the second switch circuit 183 when the rectangular wave signal Vq is at L level, and the second signal when the rectangular wave signal Vq is at H level. It becomes L level in synchronization with the rise of the output Vn of the switch circuit 183.

  On the other hand, the one-shot pulse synchronized with the rising edge of the signal Vp and the one-shot pulse synchronized with the falling edge of the signal Vn are ANDed with the inverted signal of the rectangular wave signal Vq via the OR gate 6107, and the RS latch 6114 is set as the set signal S2. Is input. The one-shot pulse synchronized with the falling edge of the signal Vp and the one-shot pulse synchronized with the rising edge of the signal Vn are ANDed with the rectangular wave signal Vq via the OR gate 6108 and input to the RS latch 6114 as the reset signal R2. Therefore, the output signal Q2 of the RS latch 6114 becomes H level in synchronization with the fall of the output Vp of the first switch circuit 182 when the rectangular wave signal Vq is L level, and when the rectangular wave signal Vq is H level. Becomes the L level in synchronization with the rise of the output Vp of the first switch circuit 182. The signal Q2 becomes H level in synchronization with the rise of the output Vp of the second switch circuit 183 when the rectangular wave signal Vq is L level, and the second switch when the rectangular wave signal Vq is H level. It becomes L level in synchronization with the fall of the output Vn of the circuit 183.

  The signals Q1 and Q2 obtained as described above are signals having four states similar to the signals L1 and L2 in FIGS. 17 and 18 of the conventional second power amplifying device. This includes delays in each switch circuit and delays such as a set dead time. Thus, the feedback signal conversion circuit 6100 converts the three states of the power amplifying apparatus as in the first to third embodiments of the present invention into four states. The signal obtained by integrating the differential voltage (Q1−Q2) between Q1 and Q2 obtained here has a waveform substantially similar to the voltage supplied to the load 186, except for the amplitude.

  The outputs Q1 and Q2 of the feedback signal conversion circuit 6100 are returned with polarities that become negative feedback through the resistors 1018, 1019, 1015, 1014, and the capacitor 1020 constituting the balanced LPF of the feedback input circuit 1010, respectively. Finally, in the power amplifying apparatus of the sixth embodiment, negative feedback can be applied from the output in a balanced manner without delay, and output distortion caused by dead time and power supply voltage fluctuation can be corrected. .

  Also in the power amplifying devices of the first to fifth embodiments described above, by negatively feeding back the outputs of the output terminal Vp of the first switch circuit 182 and the output terminal Vn of the second switch circuit 183, Although the output distortion can be corrected, in this case, the output terminal Vp of the first switch circuit 182 and the output terminal Vn voltage of the second switch circuit 183 each express only one polarity. Since the average voltage changes greatly due to the increase / decrease, the path from the synthesis to the negative feedback requires a configuration that operates stably from the DC range. Further, when configuring negative feedback, it is compared with each input signal separately or after combining positive and negative feedback signals into one signal. Since a positive or negative signal waveform has an acute angle portion near the zero cross, an operational amplifier that bears these needs to have a high slew rate. In addition, when positive / negative synthesis is performed, there is a problem that it is necessary to consider the offset when synthesizing near the zero cross.

  In the present embodiment, since this synthesis is performed digitally, problems associated with these analogs do not occur. In other words, the feedback signal conversion circuit 6100 generates a balanced voltage proportional to Vpn, which is the differential voltage output from the main circuit, so that the signals Q1 and Q2 have no waveform joints, and the magnitude of the signal is small. Regardless, the average value is the average value of logic H and L levels (H + L) / 2. Therefore, Vpn actually applied to the load is fed back to the input unit again, and the distortion can be reduced with a simple circuit configuration.

  It should be noted that most of the distortion generated in the path from the modulation circuit 610 to Vpn applied to the load is related to the main circuit. The dead time for preventing the through current, the switch operation delay of each switch circuit, and the inductor 185 Such a simple negative feedback is particularly beneficial in a configuration operating in three states as in the present invention.

  In the feedback input circuit 1010 of FIG. 12, since the input unit is a balanced type, the output signal of the parallel type main circuit is only a balanced LPF composed of only a passive element resistor and a capacitor, and is negative without delay. Since feedback can be performed, problems associated with delays and slew rates that are often associated with active devices are not bothered. Further, the balanced type has an advantage that the influence of noise caused by the switching operation of the main circuit 180 can be reduced. Even when the input section is unbalanced, the polarity of either the circuit that calculates the difference between the output Vp and the output Vn of the main circuit or the outputs of the switch circuit 182 and the switch circuit 183 is inverted, and these are overlapped. If a circuit that synthesizes the combined ternary output signals of positive, negative, and ground is provided, a short delay time and noise resistance performance cannot be obtained, but a negative feedback operation can be performed.

  Since the output terminal signal of each switch circuit in FIG. 12 includes a lot of noise at the time of switching, the level converters 6115 and 6116 have a filter function for removing this noise, so that the feedback signal conversion circuit 6100 can be provided. Can be made more stable.

  In addition, the power amplifying devices according to the first to third embodiments are also provided with the feedback signal conversion circuit 6100 and the feedback input circuit 1010 having the functions as in the present embodiment, respectively, so that they are the same as in the present embodiment. The effect of can be obtained.

  The power amplifying device of the present invention is useful as a power amplifying device that amplifies an audio signal or the like and supplies it to an acoustic speaker or the like.

1 is a circuit configuration diagram of a power amplification device according to a first embodiment of the present invention. Detailed circuit configuration diagram of a main part of the power amplification device according to the first embodiment Operation Waveform Diagram in Power Amplifying Device of Embodiment 1 State transition diagram in power amplifying apparatus according to Embodiment 1 The circuit block diagram of the power amplifier of Embodiment 2 of this invention The figure which shows the waveform of the triangular wave with which the rectangular wave in the modulation circuit of Embodiment 2 was added Operation Waveform Diagram in Power Amplifying Device of Embodiment 2 Circuit configuration diagram of a power amplifying device according to Embodiment 3 of the present invention Operation Waveform Diagram in Power Amplifying Device of Embodiment 3 The circuit block diagram of the power amplifier of Embodiment 4 of this invention Operation Waveform Diagram of Power Amplifier Device of Embodiment 4 Circuit configuration diagram of a power amplifying device according to a fifth embodiment of the present invention Circuit configuration diagram of a power amplifying device according to Embodiment 6 of the present invention Operation Waveform Diagram in Power Amplifying Device of Embodiment 6 Circuit configuration diagram of first conventional power amplification device Operation waveform diagram of first conventional power amplification device State transition diagram of first conventional power amplifier Circuit diagram of second conventional power amplification device Operation waveform diagram of second conventional power amplification device State transition diagram of second conventional power amplifier

Explanation of symbols

101 Input Signal Source 102 DC Power Supply 103 DC Power Supply 110 Modulation Circuit 111 Triangular Wave Generator 1121 Comparator 1122 Comparator 1131 Linear Inverter 1141 Inverter 1142 Inverter 1143 AND Gate 1144 AND Gate 120 First Distribution Circuit 130 Second Distribution Circuit 140 First high side switch control circuit 150 First low side switch control circuit 160 Second high side switch control circuit 170 Second low side switch control circuit 180 Main circuit 181 BTL
182 First switch circuit 1821 First high side switch 1822 First low side switch 183 Second switch circuit 1831 Second high side switch 1832 Second low side switch 185 Inductor 186 Load

Claims (21)

A first circuit that is a series circuit of a first high-side switch and a first low-side switch, and has a connection point between the first high-side switch and the first low-side switch as an output terminal;
A series circuit of a second high-side switch and a second low-side switch, wherein the second switch means has a connection point between the second high-side switch and the second low-side switch as an output terminal;
A power supply voltage is output with respect to a reference potential, and a power supply connected in parallel to the first switch means and the second switch means;
A load connected between the output end of the first switch means and the output end of the second switch means;
Modulation means for converting an input signal into a pulse width or pulse density and generating a first modulation signal for driving the first switch means and a second modulation signal for driving the second switch means; Prepared,
The modulating means is a pulse width modulator;
The modulating means includes a triangular wave generator that outputs a triangular wave signal, a first comparator that outputs a first signal that is active when a result of subtracting the triangular wave signal from the input signal is positive, and the input signal And a second comparator that outputs a second signal that becomes active when the result of adding the triangular wave signal is negative,
The first modulated signal is active when the first signal is active and the second signal is inactive, and the second modulated signal is active when the first signal is inactive and the second signal is inactive. Active when the signal is active,
In the modulation means, the voltage of the output terminal of the first switch means and the voltage of the output terminal of the second switch means are respectively 3 of power supply voltage and reference potential, reference potential and power supply voltage, reference potential and reference potential. A power amplifying apparatus for controlling the modulation signal so as to be one of the streets. The modulation means further includes a linear inverter that inverts the input signal.
The first comparator compares the input signal with the triangular wave signal to generate a first signal, and the second comparator outputs an inverted signal of the input signal output from the linear inverter. The triangular wave signal is compared to generate a second signal, a logical product of the logical negation signal of the second signal and the first signal is output as the first modulation signal, and the first signal The power amplifying apparatus according to claim 1, wherein a logical product of the logical negation signal and the second signal is output as the second modulation signal. The modulation means includes a linear inverter that inverts the triangular wave signal.
The first comparator compares the input signal with the triangular wave signal to generate a first signal, and the second comparator outputs the input signal and the triangular wave signal output from the linear inverter. To generate a second signal, and outputs a logical product of the logical negation signal of the second signal and the first signal as the first modulation signal, and the first signal The power amplifying apparatus according to claim 1, wherein a logical product of the logical negation signal and the second signal is output as the second modulation signal.   2. The modulation means according to claim 1, further comprising time offset means for giving a time offset to at least one of the input signal and the triangular wave signal input to the first and second comparators. Power amplification device.   The time offset means includes a rectangular wave generator that outputs a rectangular wave signal synchronized with the triangular wave signal, and an adding means that adds the rectangular wave signal to the triangular wave signal, and the output signal of the adding means is The power amplifying apparatus according to claim 4, wherein the power amplifying apparatus is configured to input to either the first comparator or the second comparator.   The time offset unit is different from the rectangular wave generator that outputs a rectangular wave signal synchronized with the triangular wave signal, a first adding unit that adds the rectangular wave signal to the triangular wave signal, and the first adding unit. A second adding means for adding the rectangular wave signal to the triangular wave signal in polarity; an output signal of the first adding means is input to the first comparator; an output of the second adding means; 5. The power amplifying apparatus according to claim 4, wherein a signal is input to the second comparator. The modulation means further includes time offset means for giving a time offset to at least one of the first and second signals.
The power amplification device according to claim 1.   The power amplifying apparatus according to claim 7, wherein the time offset unit includes a delay unit that delays the second signal.   8. The power amplifying apparatus according to claim 7, wherein the time offset means comprises delay means for delaying the first signal. A first circuit that is a series circuit of a first high-side switch and a first low-side switch, and has a connection point between the first high-side switch and the first low-side switch as an output terminal;
A series circuit of a second high-side switch and a second low-side switch, wherein the second switch means has a connection point between the second high-side switch and the second low-side switch as an output terminal;
A power supply voltage is output with respect to a reference potential, and a power supply connected in parallel to the first switch means and the second switch means;
A load connected between the output end of the first switch means and the output end of the second switch means;
Modulation means for converting an input signal into a pulse width or pulse density and generating a first modulation signal for driving the first switch means and a second modulation signal for driving the second switch means; Prepared,
The modulating means is a pulse width modulator;
The modulation means includes a triangular wave generator for outputting a triangular wave signal, a triangular wave signal offset means for offsetting the triangular wave signal positively or negatively with respect to an amplitude axis, and an output of the triangular wave signal offset means positively offset from the input signal A first comparator that outputs a first signal that is active when the result of subtracting the signal is positive, and a result of subtracting the output signal of the triangular wave signal offset means that is negatively offset from the input signal is negative A second comparator that outputs a second signal that is sometimes active,
Outputting the first signal as the first modulated signal, outputting the second signal as the second modulated signal;
In the modulation means, the voltage of the output terminal of the first switch means and the voltage of the output terminal of the second switch means are respectively 3 of power supply voltage and reference potential, reference potential and power supply voltage, reference potential and reference potential. A power amplifying apparatus for controlling the modulation signal so as to be one of the streets.   The modulation means includes first minimum pulse width setting means for setting a minimum value for the pulse width of the first modulation signal, and second minimum pulse for setting a minimum value for the pulse width of the second modulation signal. The power amplifying apparatus according to claim 10, further comprising a width setting unit. The first minimum pulse setting means sets a first minimum pulse setting threshold value that intersects the triangular wave signal and is near a lower limit value of the triangular wave signal, and the triangular wave signal sets the first minimum pulse. A period less than or equal to a threshold is set to the minimum pulse width of the first signal;
The second minimum pulse setting means sets a second minimum pulse setting threshold value that intersects the output of the triangular wave signal offset means and is near the upper limit value of the output of the triangular wave signal offset means, and the triangular wave signal offset 12. The power amplifying apparatus according to claim 11, wherein a period in which the output of the means is equal to or greater than the second minimum pulse setting threshold is set to a minimum pulse width of the second signal.   The amount of offset of the triangular wave signal offset means is a value smaller than the change width of the triangular wave signal by a predetermined value, and the first minimum pulse setting threshold and the second minimum pulse setting threshold are equal. The power amplification device according to claim 12.   The modulation means includes first maximum pulse width setting means for setting a maximum value for the pulse width of the first modulation signal, and second maximum pulse for setting a maximum value for the pulse width of the second modulation signal. The power amplifying apparatus according to claim 11, further comprising a width setting unit. The first maximum pulse setting means sets a first maximum pulse setting threshold value that intersects the triangular wave signal and is near the upper limit value of the triangular wave signal, and the triangular wave signal sets the first maximum pulse. A period less than or equal to a threshold is set to the maximum pulse width of the first signal;
The second maximum pulse setting means sets a second maximum pulse setting threshold value that intersects the output of the triangular wave signal offset means and is near the lower limit value of the output of the triangular wave signal offset means, and the triangular wave signal offset The power amplifying apparatus according to claim 14, wherein a period in which the output of the means is equal to or greater than the second maximum pulse setting threshold is set as the maximum pulse width of the second signal.   The power amplifying apparatus according to claim 14, wherein a minimum no-pulse interval between the first modulation signal and the second modulation signal is larger than a minimum pulse width between the first modulation signal and the second modulation signal.   The modulation means includes first maximum pulse width setting means for setting a maximum value for the pulse width of the first modulation signal, and second maximum pulse for setting a maximum value for the pulse width of the second modulation signal. The power amplifying apparatus according to claim 10, further comprising a width setting unit. The first maximum pulse setting means sets a first maximum pulse setting threshold value that intersects the triangular wave signal and is near the upper limit value of the triangular wave signal, and the triangular wave signal sets the first maximum pulse. Set the period below the threshold to the maximum pulse of the first signal;
The second maximum pulse setting means sets a second maximum pulse setting threshold value that intersects the output of the triangular wave signal offset means and is near the lower limit value of the output of the triangular wave signal offset means, and the triangular wave signal offset 18. The power amplifying apparatus according to claim 17, wherein a period in which the output of the means is equal to or greater than the second maximum pulse setting threshold is set as the maximum pulse of the second signal. A first circuit that is a series circuit of a first high-side switch and a first low-side switch, and has a connection point between the first high-side switch and the first low-side switch as an output terminal;
A series circuit of a second high-side switch and a second low-side switch, wherein the second switch means has a connection point between the second high-side switch and the second low-side switch as an output terminal;
A power supply voltage is output with respect to a reference potential, and a power supply connected in parallel to the first switch means and the second switch means;
A load connected between the output end of the first switch means and the output end of the second switch means;
Modulation means for converting an input signal into a pulse width or pulse density and generating a first modulation signal for driving the first switch means and a second modulation signal for driving the second switch means; Prepared,
The modulating means includes
A triangular wave generating means for outputting a triangular wave signal;
Feedback signal conversion means for outputting first and second feedback signals;
Feedback input means for adding a difference signal between the first feedback signal and the second feedback signal so as to have a negative feedback polarity with respect to the input signal;
The first feedback signal is
In one of the falling period and rising period of the triangular wave signal, the first time when the output terminal of the first switch means changes from the power supply voltage to the reference potential, or the first time when the output terminal of the second switch means changes from the reference potential. It becomes active at the time when the power supply voltage changes,
In the other of the falling period and the rising period of the triangular wave signal, the output terminal of the first switch means is first changed from the reference potential to the power supply voltage, or the output terminal of the second switch means is first changed from the power supply voltage. It becomes inactive at the time when it changes to the reference potential,
The second feedback signal is
In one of the falling period and rising period of the triangular wave signal, the output terminal of the second switch means is first changed from the power supply voltage to the reference potential, or the output terminal of the first switch means is first changed from the reference potential. It becomes active at the time when the power supply voltage changes,
In the other of the falling period and the rising period of the triangular wave signal, the time when the output terminal of the second switch means first changes from the reference potential to the power supply voltage, or the time when the output terminal of the first switch means first changes from the power supply voltage. It becomes inactive at the time when it changes to the reference potential,
In the modulation means, the voltage of the output terminal of the first switch means and the voltage of the output terminal of the second switch means are respectively 3 of power supply voltage and reference potential, reference potential and power supply voltage, reference potential and reference potential. A power amplifying apparatus for controlling the modulation signal so as to be one of the streets. 20. The power amplifying apparatus according to claim 19 , wherein the feedback input unit is configured to receive the input signal in a balanced manner and receive the feedback signal in a balanced manner. 20. The power amplifying apparatus according to claim 19 , wherein the feedback input means includes a balanced LPF, and is configured to receive the input signal in a balanced manner and to receive the feedback signal output via the balanced LPF. .

Images