JP5087367B2 - Class D amplifier - Google Patents

Description

  The present invention relates to a class D amplifier, and more particularly to a class D amplifier provided with an electrical filter such as a low-pass filter.

  In recent years, class D amplifiers using class D amplification systems suitable for miniaturization and high efficiency have come into widespread use as amplifying means for audio signals mounted on in-car or portable audio playback devices.

  This class D amplifier converts the audio signal before amplification into a PWM signal using pulse width modulation (PWM), and amplifies the PWM signal, so that it is much higher than the conventional analog amplification method. High efficiency can be obtained.

Here, as shown in FIGS. 8A and 8B, the PWM signal is alternately expressed as a high (high voltage) state signal (Hi) and a low (low voltage) state signal (Low). a signal of the square wave, by adjusting the duty is the percentage of time T ₁ of the high state to the basic period T (percentage), the level of the output signal extracted by integrating the PWM signal (analog signal) It is known as a signal capable of controlling (voltage value). FIG. 8 (a) shows a PWM signal in which the duty for each basic period T is 50%, and when this PWM signal is integrated, as shown in FIG. 8 (a). An output signal having a constant value (0 [V]) can be taken out. On the other hand, FIG. 8B shows a PWM signal in which the duty for each basic period T is a constant value of 50% or more. When this PWM signal is integrated, FIG. It is possible to extract an output signal having a constant value (a positive value larger than in the case of FIG. 8A) as shown in FIG. 8 (a) and 8 (b), since the duty for each basic period T is constant, the output signal also takes a constant value, but the duty for each basic period T is different from each other. Naturally, for example, a sine wave or a distorted wave output signal is taken out. Such amplification of the PWM signal can be performed by increasing the amplitude (high-level signal value).

  As shown in FIG. 9, conventionally, in a class D amplifier 1 using a PWM signal, an LC-type low-pass filter 8 is arranged between an audio signal amplifying unit 2 and a speaker 3 which are amplifier bodies. In addition, the low-pass filter 8 demodulates (integrates) the PWM signal amplified by the amplifying unit 2 to extract an amplified analog audio signal. The low-pass filter 8 shown in FIG. 9 includes an inductor 6 including a coil connected in series to the signal output line 5 on the signal output line 5 toward the speaker 3, the signal output line 5, and the ground line 4. And a capacitor 7 composed of a capacitor or the like connected in parallel to the signal output line 5.

  By the way, in this kind of class D amplifier 1, conventionally, electromagnetic non-interference or electromagnetic environment compatibility (EMC: Electro-Magnetic Compatibility) is caused by electromagnetic radiation of a noise component signal included in a PWM signal. ) May be hindered.

  That is, the PWM signal in a general class D amplification method includes noise having a spectrum distribution as shown in FIG. 10 in addition to a signal (audio signal) in an audible band (20 Hz to 20 kHz) that is an original reproduction target. The component signal is included.

Specifically, the PWM signal when the duty is 50% has a fundamental frequency f ₁ (PWM oscillation frequency when the duty is 50%) as a noise component signal, as shown by the solid line in FIG. A signal of a noise component (first order) (hereinafter referred to as a fundamental frequency component) and an odd harmonic component signal such as a third order, fifth order, and seventh order with respect to the signal of the fundamental frequency component are included. In general, the fundamental frequency is a frequency between 400 kHz and 500 kHz.

  Further, in the PWM signal in the case where the duty is other than 50%, in addition to the signal of the fundamental frequency component and the odd harmonic component as the signal of the noise component, 2 for the signal of the fundamental frequency component as shown by the broken line in FIG. Signals of even-order harmonic components such as second, fourth, and sixth orders are included.

  Among such noise component signals, particularly, the fundamental frequency component signal has a high signal level [dB], which has been a major obstacle in securing EMC.

  Here, the above-described LC low-pass filter 8 generally has a frequency characteristic as shown in FIG. 11, and a signal level of a signal having a frequency higher than the cutoff frequency fc is set to a predetermined attenuation rate ( For example, it can be attenuated according to -12 dB / oct). The cut-off frequency fc and the attenuation factor are determined by the constants of the inductor 6 and the capacitor 7.

  Therefore, by using such frequency characteristics, the signal level of the fundamental frequency component signal can be attenuated to some extent.

However, the low-pass filter 8 is required to have a flat characteristic from 20 Hz to 20 kHz so as not to affect (attenuate) the signal in the audible band. It was necessary to set to about 40 kHz. Since the cut-off frequency fc cannot be set sufficiently low as described above, the frequency characteristic of the low-pass filter 8 is such that the signal level attenuation A (at the fundamental frequency f ₁ of the noise component (see FIG. 11)). It was impossible to obtain a sufficiently large amount (see FIG. 11).

  As a result, the conventional class D amplifier 1 is insufficient for securing EMC.

  So far, various measures have been taken for class D amplifiers to improve the attenuation of the signal level at the fundamental frequency of the noise component.

  For example, in the class D amplifier 10 shown in FIG. 12, an LC-type low-pass filter 11 similar to the preceding low-pass filter 8 is connected to the subsequent stage of the LC-type low-pass filter 8 similar to that shown in FIG. Yes. Such low-pass filters 8 and 11 having a two-stage configuration are also disclosed in FIG. These low-pass filters 8 and 11 exhibit frequency characteristics as shown by a solid line graph in FIG. 13 by synthesizing their frequency characteristics. This frequency characteristic is obtained by synthesizing the frequency characteristic of the low-pass filter 11 in the subsequent stage in the band portion (broken line part in FIG. 13) on the higher frequency side than the cutoff frequency fc in the frequency characteristic of the low-pass filter 8 in the previous stage. , The attenuation rate is improved twice as much as that shown in FIG. 9 (for example, −24 dB / oct).

Furthermore, as a more preferable measure, for example, in the class D amplifier 12 shown in FIG. 14, another low-pass filter 14 is provided at the rear stage of the LC-type low-pass filter 8 similar to that shown in FIGS. It is connected. 14 is added to the configuration of the LC low-pass filter 11 shown in FIG. 12 by connecting a resistor 15 and a capacitor 16 in parallel to the inductor 6 together. It has become. These low-pass filters 8 and 14 are configured to exhibit frequency characteristics as shown by a solid line graph in FIG. 15 by synthesizing the mutual frequency characteristics. This frequency characteristic is obtained by synthesizing the frequency characteristic of the low-pass filter 14 in the subsequent stage in the band portion (broken line portion in FIG. 15) on the higher frequency side than the cutoff frequency fc in the frequency characteristic of the low-pass filter 8 in the front stage. The signal level attenuation amount A at the fundamental frequency f ₁ is further improved than that shown in FIG.

  In this way, it is possible to improve the attenuation of the signal level at the fundamental frequency by adding a filter.

JP 2002-330035 A JP 2006-229891 A

  However, in each of the class D amplifiers 10 and 12 shown in FIG. 12 and FIG. 14, it is necessary to provide low-pass filters 11 and 14 having substantially the same size as the low-pass filter 8 in the previous stage in the subsequent stage of the low-pass filter 8 in the previous stage. For this reason, there is a problem that the area occupied by the substrate for arranging the class D amplifiers 10 and 12 on the substrate becomes large, making it difficult to reduce the size and increasing the cost. In particular, the inductor 6 connected in series on the signal output line 5 among the low-pass filters 11 and 14 in the subsequent stage needs to flow a large current including an audible band signal and remove a noise component signal due to heat loss. In addition, he was forced to become large.

  Therefore, the present invention has been made in view of such problems, and an object thereof is to provide a class D amplifier that can sufficiently suppress noise while having a small and inexpensive configuration. .

  In order to achieve the above-described object, a class D amplifier according to the present invention is a class D amplifier that amplifies an audio signal using a class D amplification method and outputs the amplified audio signal to a speaker side. The audio signal input from is amplified after being converted into a PWM signal, and the PWM signal after the audio signal is amplified is output to the subsequent stage, and connected to the subsequent stage of the amplification means, A low-pass filter having a predetermined cutoff frequency formed so as to convert the PWM signal output from the amplifying means into an analog audio signal, and to output the converted analog audio signal to a signal output line directed to the speaker. A first filter comprising: a noise component signal having a predetermined frequency higher than the cutoff frequency included in the PWM signal; A first filter formed so as to exhibit a predetermined frequency characteristic capable of attenuating a signal level in accordance with the frequency of the signal, and connected in parallel to the signal output line at a subsequent stage of the first filter. A second filter formed so as to exhibit a predetermined frequency characteristic capable of attenuating a signal level according to the frequency of the signal of the noise component, and the frequency characteristic of the second filter is: In the state combined with the frequency characteristic of the first filter, the attenuation amount of the signal level at the fundamental frequency of the signal of the noise component is the attenuation amount of the signal level at the fundamental frequency due to the frequency characteristic of only the first filter. It is characterized by having a frequency characteristic that is larger than that.

  According to such a configuration, the signal level of the fundamental frequency component signal can be sufficiently attenuated by the mutually synthesized frequency characteristics of the first filter and the second filter, and the second By connecting the filter in parallel to the signal output line, it is possible to reduce the size of the second filter by avoiding a large current including an audible band signal from flowing through the second filter. However, noise can be sufficiently suppressed.

  In the present invention, the frequency characteristic of the second filter is a signal level in a first frequency band having a predetermined bandwidth including the fundamental frequency in a state where it is combined with the frequency characteristic of the first filter. Is the maximum at the fundamental frequency, and the attenuation of the signal level in the first frequency band is the attenuation of the signal level in the first frequency band due to the frequency characteristics of only the first filter. It is preferable that the frequency characteristics be larger than the above.

  According to such a configuration, the signal of the fundamental frequency component can be attenuated more effectively, and noise with a noise component signal is present as the output level of the sound output from the speaker increases. Even if the bandwidth of the band expands in the first frequency band, the noise component signal can be effectively suppressed in the first frequency band.

  Furthermore, in the present invention, the frequency characteristic of the second filter has a predetermined bandwidth that is continuous with the first frequency band on the high frequency side in a state where it is combined with the frequency characteristic of the first filter. The signal level attenuation in the second frequency band is smaller than the signal level attenuation in the second frequency band due to the frequency characteristics of only the first filter, and within the second frequency band, The frequency characteristic may not include the frequency of the second or higher harmonic component signal with respect to the fundamental frequency component signal in the noise component signal.

  And according to such a structure, even if it is a case where the attenuation amount of the signal level in a 2nd frequency band deteriorates compared with the case of only a 1st filter on the structure of a 2nd filter, this 1st filter. Since the harmonic component signal can be prevented from appearing within the frequency band of 2, the noise can be more effectively suppressed.

  Furthermore, it is preferable that the first filter is an LC type low-pass filter.

  According to such a configuration, an analog audio signal can be efficiently extracted from the PWM signal.

  The second filter preferably includes an inductor and a resistor connected in parallel to each other, and a capacitor connected in series to the inductor and the resistor.

  According to such a configuration, the second filter can be easily formed, and further miniaturization and cost reduction can be achieved.

  Further, a switch disposed in the second filter and configured to connect or disconnect the second filter body with respect to the signal output line, and an output level for detecting an output level of sound output from the speaker The output level detected by the detection means and the output level detection means reaches the second frequency band from a noise band in which a signal of the noise component whose bandwidth is increased as the output level increases is present. It is preferable to include switch control means for controlling the switch to disconnect the second filter from the signal output line when the output level is such that the second output filter is disconnected.

  According to such a configuration, when the noise band reaches the second frequency band as the audio output level increases, the switch causes the second filter to be disconnected from the signal output line by the switch. By doing so, it is possible to use only the first filter that does not have the second frequency band. Therefore, the frequency characteristic of the first filter among the synthesized frequency characteristics of the first filter and the second filter. It is possible to avoid an increase in the signal level of a signal included in a noise band that has spread to this portion due to a portion where the attenuation level of the signal level is smaller (peak characteristics described later).

  Furthermore, the second filter includes an inductor and a resistor connected in parallel to each other, a plurality of capacitors connected in series to the inductor and the resistor and connected in parallel to each other, and a plurality of the plurality of capacitors. Each of the capacitors is formed by the same number of switches as the capacitors that can be connected to or disconnected from the signal output line, and the class D amplifier body can further change the fundamental frequency. A frequency variable means, and a switch control means for controlling the switch for connecting the predetermined capacitor, the switch control means, when the fundamental frequency is changed by the frequency variable means, As the control of the switch, the frequency characteristic of the second filter is changed to the frequency of the first filter. The frequency at which the signal level attenuation at the changed fundamental frequency is larger than the signal level attenuation at the changed fundamental frequency due to the frequency characteristics of only the first filter in the combined state Preferably, the switch is formed so as to control the switch to obtain the characteristics.

  According to such a configuration, even when the fundamental frequency is changed, a capacitor suitable for signal level attenuation at the changed fundamental frequency can be connected, so that the change in the fundamental frequency can be achieved. Regardless of the noise, the noise can be stably suppressed.

  In addition, before and after the change of the fundamental frequency, the frequency characteristic of the second filter is the same as that before or after the change in the state synthesized with the frequency characteristic of the first filter. The attenuation level of the signal level in the first frequency band including the signal level is larger than the attenuation level of the signal level in the first frequency band due to the frequency characteristic of only the first filter, and is high in the first frequency band. The frequency characteristic is such that the attenuation of the signal level in the second frequency band that is continuous on the frequency side is smaller than the attenuation of the signal level in the second frequency band due to the frequency characteristic of only the first filter, The class D amplifier body further comprises output level detection means for detecting the output level of the sound output from the speaker. The switch control means is configured to set the output frequency detected by the output level detection means to a noise band in which a signal of the noise component whose bandwidth increases with an increase in the output level is present in the second frequency. In the case of an output level that reaches a band, it is preferable that the switch is controlled so that the second filter is disconnected from the signal output line.

  According to such a configuration, when the noise band is expanded to the second frequency band with an increase in the output level of the sound both before and after the basic frequency is changed, the switch is used. By cutting the second filter with respect to the signal output line, it is possible to use only the first filter that does not have the second frequency band. In any case, the portion of the combined frequency characteristic of the first filter and the second filter that has a signal level attenuation smaller than the frequency characteristic of the first filter (a peak characteristic described later) spreads to this part. An increase in the signal level of the signal included in the noise band can be avoided in advance.

  According to the present invention, it is possible to sufficiently suppress noise while having a small and inexpensive configuration.

(First embodiment)
A first embodiment of a class D amplifier according to the present invention will be described below with reference to FIGS.

  Note that portions having the same or similar basic configuration as those in the related art will be described using the same reference numerals.

  As shown in FIG. 1, the class D amplifier 18 in the present embodiment has an amplification unit 2 as an amplification unit, which is an amplifier body, as in the prior art, and the amplification unit 2 is a sound source (for example, a memory). A reader that reads audio information stored in the medium and converts it into an audio signal) The audio signal input from the side is converted into a PWM signal, amplified, and the amplified PWM signal is output to the subsequent stage It is supposed to be. Note that the audio signal input to the amplifying unit 2 may be either an analog audio signal or a digital audio signal.

  Further, a first filter 19 as a first filter is connected to the subsequent stage of the amplifying unit 2, and the first filter 19 includes an inductor 6 connected in series to the signal output line 5, and a signal output line. 5 is an LC type low-pass filter including a capacitor 7 connected in parallel with the capacitor 5. The first filter 19 converts the PWM signal output from the amplifying unit 2 into an analog audio signal, and outputs the converted analog audio signal to the signal output line 5.

  The first filter 19 has a frequency characteristic similar to that shown in FIG. 11. Due to this frequency characteristic, the first filter 19 has a frequency component higher than the cutoff frequency fc included in the PWM signal. The signal level can be attenuated according to the frequency of the signal. However, the frequency characteristic of the first filter 19 alone does not provide sufficient signal level attenuation at the fundamental frequency of the noise component signal.

  Therefore, in the present embodiment, a second filter 20 as a second filter is connected in parallel to the signal output line 5 at the subsequent stage of the first filter 19.

  The second filter 20 is connected in parallel with the signal output line 5 and in parallel with each other, and is connected in parallel with the signal output line 5 and in series with the inductor 22 and the resistor 23. And a capacitor 24. The inductor 22 may be a coil. The capacitor 24 may be a capacitor.

  The second filter 20 configured as described above has a predetermined frequency characteristic capable of attenuating the signal level according to the frequency of the signal of the noise component included in the PWM signal.

That is, as shown in FIG. 2, the frequency characteristic of the second filter 20 is the basic characteristic of the noise component signal in the combined characteristic (the solid line graph in FIG. 2) combined with the frequency characteristic of the first filter 19. attenuation of the signal level at a frequency f ₁ a _{1 + 2,} are largely composed such frequency characteristics than the attenuation a ₁ of the signal level at the fundamental frequency f ₁ by the frequency characteristic of only the first filter 19.

  According to such a configuration, the signal level of the signal of the fundamental frequency component can be sufficiently attenuated by the synthesis characteristic of the first filter 19 and the second filter 20.

  Further, since the second filter 20 is connected in parallel to the signal output line 5, a large current (for example, 5.0 [A]) including an audible band signal is passed through the second filter 20. In addition, only a small current (for example, 0.2 [A]) including a noise component signal can flow. Thereby, the inductor 22 of the second filter 20 can be made smaller than the inductor 6 of the first filter 19, and the resistor 23 and the capacitor 24 can also be made smaller.

  Therefore, in the present embodiment, noise can be sufficiently suppressed while having a small and inexpensive configuration.

In addition to the above configuration, in the present embodiment, the frequency characteristic of the second filter 20 includes a fundamental frequency f ₁ as shown in FIG. 2 in a combined characteristic with the frequency characteristic of the first filter 19. The signal level attenuation in the first frequency band (hereinafter referred to as the first frequency band BW ₁ ) having a predetermined bandwidth is maximized at the fundamental frequency f ₁ and the signal in the first frequency band BW ₁ . The frequency characteristic is such that the level attenuation is greater than the signal level attenuation in the first frequency band BW ₁ due to the frequency characteristic of the first filter 19 alone.

Here, like the composite characteristic in the first frequency band BW ₁ as shown in FIG. 2, the signal level attenuation amount suddenly increases once and then decreases rapidly (hereinafter referred to as a dip (DIP) characteristic). ) Can be realized by providing the second filter 20 including the inductor 22, the resistor 23, and the capacitor 24 described above in the subsequent stage of the first filter 19.

In the present embodiment, as described above, the dip characteristics indicate the attenuation of the maximum signal level at the fundamental frequency f ₁ as described above, the inductance of the inductor 22, the resistance value of the resistor 23, and the capacitor 24. The capacity is set.

  Thereby, the signal level of the signal of the fundamental frequency component can be further effectively attenuated.

By the way, in the separately excited class D oscillation (amplification) system, as shown in FIGS. 3A and 3B, when the output level of the sound output from the speaker 3 (in other words, the volume) increases, It is known that the bandwidth of a noise band, which is a frequency band in which a noise component signal exists, remains unchanged at a high frequency side and a low frequency side with the fundamental frequency f ₁ remaining unchanged. Specifically, in FIG. 3A, since the output level of the sound is small, the noise band NBW ₀ is limited to only one point of the fundamental frequency f ₁ , but in FIG. by levels increased, the noise bandwidth NBW ₁ is formed to spread to the high frequency side and the low frequency side with respect to the fundamental frequency f _1. When the noise band extends from the fundamental frequency f ₁ , the signal level of the noise signal within the expanded noise band is smaller than the signal level when the noise band remains only at the fundamental frequency f _1. Become.

In the present embodiment, thus, with an increase in the sound output level, as the noise bandwidth NBW ₁ measuring over a predetermined range in the first frequency band BW ₁ is formed, the first frequency band The signal level of the noise signal included in the noise band NBW ₁ can be effectively attenuated by the dip characteristic in the BW ₁ as compared with the frequency characteristic of the first filter 19 alone.

On the other hand, in class D oscillation method of self-driven, although not shown, because the fundamental frequency with an increase of the audio output level moves to the low frequency side, the noise band (base in a first frequency band BW ₁ The occupancy ratio of the one spread from the frequency tends to be lower than that of the separate excitation method. However, even in the case of this self-excited method, if the composite characteristic of FIG. 2 is applied, the noise signal included in the noise band formed in the first frequency band BW ₁ is effective as in the case of the separately excited method. There is no substitute for being able to suppress it.

In addition to the above configuration, in the present embodiment, the frequency characteristic of the second filter 20 is a first characteristic band BW ₁ as shown in FIG. 2 in a combined characteristic with the frequency characteristic of the first filter 19. Signal level attenuation in a _second frequency band (hereinafter referred to as second frequency band BW ₂ ) having a predetermined bandwidth continuous on the high frequency side is a second frequency band based on the frequency characteristics of only the first filter 19. there is a frequency characteristic is smaller than the attenuation of the signal level at the BW _2. Furthermore, the frequency characteristic of the second filter 20 includes the frequency of the second or higher harmonic component signal with respect to the fundamental frequency component signal in the noise component signal in the second frequency band BW ₂ in the composite characteristic state. There is no frequency characteristic. In FIG. 2, the frequency f ₂ of the second harmonic component of the signal is located on the higher frequency side with respect to the second frequency band BW _2.

Here, as the combined characteristic in the second frequency band BW within ₂ as shown in FIG. 2, rapidly increasing property after the attenuation of the signal level is reduced once abruptly (hereinafter referred to as the peak characteristic), inductor 22 described above, the resistance 23 and the configuration of the second filter 20 composed of a capacitor 24, it is forced to be shown as part of the frequency characteristic.

However, in the present embodiment, even if such a peak characteristic occurs, a high-frequency component signal with respect to a fundamental frequency component signal does not appear in the second frequency band BW ₂ where the peak characteristic is indicated. Therefore, the signal of the high frequency component in the noise signal can be effectively suppressed.

  According to the class D amplifier 18 of the present embodiment having the above-described configuration, it is possible to output (reproduce) audio having excellent listening quality in which noise is sufficiently suppressed while being a small and inexpensive configuration.

(Second Embodiment)
Next, a second embodiment of the class D amplifier according to the present invention will be described with reference to FIG. 4 and FIG.

  Note that portions having the same or similar basic configuration as the first embodiment will be described using the same reference numerals.

  As shown in FIGS. 4A and 4B, the class D amplifier according to the present embodiment can exhibit combined characteristics including a dip characteristic and a peak characteristic, as in the first embodiment.

  Further, as shown in FIG. 5, the class D amplifier 26 in the present embodiment is connected to the signal output line 5 in the subsequent stage of the amplification unit 2, the first filter 19 including an LC type low-pass filter, and the first filter 19. On the other hand, the second filter 28 as the second filter connected in parallel is the same as in the first embodiment.

However, as shown in FIG. 4C, the class D amplifier in the present embodiment has a second frequency band BW ₂ (shown in FIG. 4) in which the peak characteristic is shown as the output level of the sound output from the speaker 3 increases. 4 (a) and (b)), a noise band (hereinafter referred to as a strong sound side noise band NBW ₂ ) is formed, so that a signal level of a noise signal included in the strong sound side noise band NBW ₂ is reduced. This is different from the first embodiment in that it is configured to avoid an increase due to peak characteristics.

  More specifically, as shown in FIG. 5, the class D amplifier 26 in the present embodiment has a signal level detection unit 27 as an output level detection means connected to the preceding stage of the amplification unit 2, and this The signal level detection unit 27 detects the signal level of the audio signal input from the sound source side. The signal level detection unit 27 then outputs an audio output level (hereinafter referred to as audio output level) from the speaker 3 based on the detected signal level of the audio signal and the gain of the amplification unit 2 acquired in advance. The sound output level is detected.

  Further, in the class D amplifier 26 in the present embodiment, the second filter 28 has a switch 29 connected in series to the capacitor 24 in addition to the inductor 22, the resistor 23, and the capacitor 24 shown in the first embodiment. The switch 29 is configured to connect or disconnect the second filter 28 with respect to the signal output line 5.

  Furthermore, the class D amplifier 26 in the present embodiment has a switch controller 31 as switch control means connected between the signal level detection unit 27 and the switch 29, and the switch controller 31 has a signal level. An audio output level detection (prediction) result is acquired from the detection unit 27.

Then, when the acquired audio output level is such an audio output level that the noise band whose bandwidth increases as the audio output level increases reaches the second frequency band BW ₂ , The switch 29 is controlled so that the second filter 28 is disconnected from the signal output line 5. On the other hand, the switch controller 31 is acquired voice output level, the noise band, in the case of the audio output level that can not spread until the frequency of the second frequency band BW within ₂ controls the switch 29 second The filter 28 is connected to the signal output line 5.

As a more specific example, the switch controller 31 has a noise band in which the sound output level stops at only one point of the fundamental frequency f ₁ as shown in FIG. 4A (hereinafter referred to as a weak sound side noise band NBW ₀ ). Or a noise band having a width over a predetermined range in the first frequency band BW ₁ as shown in FIG. 4B (hereinafter referred to as a middle-sound noise band NBW ₁ ). In the case of the medium sound level that forms the second filter 28, the second filter 28 is brought into a connected state. Incidentally, in FIG. 4 (b), the maximum frequency of the tone Noise bandwidth NBW ₁ is coincident with the first maximum frequency of the frequency band BW _1. In addition, the switch controller 31 also switches the switch 29 when the sound output level is a sound output level that forms a noise band wider than the weak sound side noise band NBW ₀ and narrower than the middle sound side noise band NBW _1. To close the second filter 28.

In this way, when the second filter 28 is in the connected state, the second filter 28 functions and exhibits the same synthesis characteristics as in the first embodiment, and the signal of the fundamental frequency component is effective. In addition to suppressing the frequency component signal, it is possible to effectively suppress the signal of the frequency component that spreads in the middle sound side noise band NBW ₁ .

On the other hand, the switch controller 31 opens the switch 29 and opens the second filter 28 when the sound output level is a strong sound level that forms the strong sound side noise band NBW ₂ as shown in FIG. Set to disconnected state.

Thus, when the 2nd filter 28 will be in a cutting | disconnection state, the 2nd filter 28 does not function, but only the 1st filter 19 functions and the frequency characteristic of only the 1st filter 19 is shown. become. In this case, although the attenuation amount of the signal level at the fundamental frequency is reduced as compared with the case of the synthesis characteristic, the peak characteristic shown in the synthesis characteristic in the second frequency band BW ₂ is lost, so the strong sound side It can be prevented in advance that the noise signal included in the noise band NBW ₂ increases due to the peak characteristics.

  Also in the class D amplifier 26 of the present embodiment having the above configuration, as in the first embodiment, it is possible to output a sound with excellent listening quality in which noise is sufficiently suppressed while being a small and inexpensive configuration. it can.

  In the present embodiment, the detection of the audio output level is not necessarily limited to being performed in the preceding stage of the amplification unit 2 as in the signal level detection unit 27 described above, and various changes can be made depending on the concept. Is possible. For example, the audio output level may be detected based on the detection result of the PWM signal output from the amplifying unit 2 toward the first filter 19, or from the second filter 28 toward the speaker 3. This may be done by directly monitoring the output level of the output audio signal.

(Third embodiment)
Next, a third embodiment of the class D amplifier according to the present invention will be described with reference to FIGS.

  Note that parts having the same or similar basic configuration as those of the first and second embodiments will be described using the same reference numerals.

  As shown in FIGS. 6A and 6B, the class D amplifier according to the present embodiment can exhibit composite characteristics including a dip characteristic and a peak characteristic, as in the first and second embodiments. ing.

  Further, as shown in FIG. 7, the class D amplifier 33 in the present embodiment is connected to the signal output line 5 in the subsequent stage of the amplification unit 2, the first filter 19 including the LC type low-pass filter, and the first filter 19. On the other hand, the second filter 34 as the second filter connected in parallel is the same as in the first and second embodiments.

  Further, as shown in FIG. 7, the class D amplifier 33 in the present embodiment includes a switch 29 connected to the capacitor 24 in the second filter 34 and a switch controller 35 that controls the switch 29. This is the same as in the second embodiment.

  However, the class D amplifier in the present embodiment is configured to effectively attenuate the signal level of the signal of the fundamental frequency component after the change even when the fundamental frequency changes as shown in FIG. 6B. This is different from the first and second embodiments.

  More specifically, as shown in FIG. 7, the class D amplifier 33 in the present embodiment has an oscillation frequency controller 36 as frequency variable means connected to the amplifying unit 2, and this oscillation frequency controller 36. The basic frequency can be changed by changing the oscillation frequency of the PWM signal. The oscillation frequency controller 36 may be provided as a countermeasure against radio beats when listening to the radio.

  Further, in the class D amplifier 33 in the present embodiment, the second filter 34 is connected in parallel with the signal output line 5 and in series with the inductor 22 and the resistor 23 in parallel with each other, instead of only one capacitor 24. A plurality of capacitors 24 are provided. The capacitors 24 may have the same capacity, or may have different capacities.

  Further, in the class D amplifier 33 in the present embodiment, the second filter 34 includes not only one switch 29 but a plurality of switches 29 of the same number as the capacitors 24 respectively connected in series to the capacitors 24. These switches 29 are connected to the switch controller 35.

  These switches 29 can be connected or disconnected with respect to the signal output line 5 for each of the corresponding capacitors 24 under the control of the switch controller 35.

  Furthermore, in this embodiment, the switch controller 35 acquires the control result of the oscillation frequency by the oscillation frequency controller 36.

  When the control result of the oscillation frequency indicates that the oscillation frequency, that is, the fundamental frequency is changed, the switch controller 35 controls the switch 29 to change the frequency characteristic of the second filter 34 to that of the first filter 19. In the state of the synthesized characteristic combined with the frequency characteristic, the attenuation amount of the signal level at the changed fundamental frequency is larger than the attenuation amount of the signal level at the changed fundamental frequency due to the frequency characteristic of only the first filter 19. Desired frequency characteristics are obtained.

  The control of the switch 29 is control of the switch 29 for connecting the capacitor 24 (which may be plural) that can set a desired frequency characteristic to the second filter 34.

More specifically, as shown in FIG. 6 (a), (b) , with the change of the oscillation frequency by the oscillation frequency controller 36, FIG fundamental frequency from the frequency f ₁ shown in FIG. 6 (a) 6 ( When the frequency f ₁ ′ shown in b) is changed, the switch controller 35 changes the frequency characteristic of the second filter 34 such that a new (changed) synthetic characteristic shown in FIG. 6B is obtained. Control of the switch 29 for setting is performed. In the new composite characteristic shown in FIG. 6B, the dip characteristic and the peak characteristic part move from the state of FIG. 6A to the f ₁ ′ side, and the signal level in the dip characteristic changes. The frequency at which the amount of attenuation is maximum matches the changed fundamental frequency f ₁ ′. In addition, the first frequency band BW ₁ and the second frequency band BW ₂ are also moved to the changed fundamental frequency f ₁ ′ side.

Thus, in a new synthetic characteristic shown in FIG. 6 (b), _'the attenuation of the signal level at the fundamental frequency f ₁ after the change due to the frequency characteristic of only the first filter _19' fundamental frequency f ₁ after the change in It can be made larger than the attenuation of the signal level.

  Therefore, according to the present embodiment, even when the fundamental frequency is changed, the capacitor suitable for the attenuation of the signal level at the changed fundamental frequency can be set in the connected state. Regardless, noise can be stably suppressed.

In the present embodiment, similarly to the second embodiment, the audio output level is detected by the signal level detection unit 27, and the detected audio output level reaches the high sound level, and the noise band is increased. When the fundamental frequency spreads into the second frequency band BW ₂ where the peak characteristic is shown, the second filter 34 is disconnected from the signal output line 5 by the control of the switch 29 by the switch controller 35. May be. However, in this case, it is necessary to open all the switches 29. The switch controller 35 controls the switch 29 based on such audio output level may be performed similarly after changing the fundamental frequency.

  In addition, this invention is not limited to embodiment mentioned above, A various change is possible as needed.

1 is a block diagram showing a first embodiment of a class D amplifier according to the present invention. The graph which shows a frequency characteristic in 1st Embodiment of Class D amplifier concerning this invention In the first embodiment of the class D amplifier according to the present invention, (a) illustrates the state of the noise band when the audio output level is low in the separately excited class D oscillation system, and (b) Explanatory drawing for demonstrating the state of a noise band when the audio | voice output level increases in the separately excited class D oscillation system In the second embodiment of the class D amplifier according to the present invention, (a) shows the formation state of the weak-side noise band, (b) shows the formation state of the mid-side noise band, and (c) shows The figure which shows the formation state of the strong sound side noise band Block diagram showing a second embodiment of a class D amplifier according to the present invention In the third embodiment of the class D amplifier according to the present invention, (a) is a graph showing the frequency characteristics before the fundamental frequency is changed, and (b) is a graph showing the frequency characteristics after the fundamental frequency is changed. Block diagram showing a third embodiment of a class D amplifier according to the present invention It is explanatory drawing for demonstrating a PWM signal, Comprising: (a) shows a PWM signal of duty 50%, (b) shows a PWM signal in case a duty is larger than 50%. Block diagram showing an example of a class D amplifier having an LC type low-pass filter that has been conventionally employed The figure which shows the spectrum distribution of the signal of the noise component contained in the PWM signal of a general class D amplification system A graph showing the frequency characteristics of a general LC low-pass filter Block diagram showing a class D amplifier as one example of conventional measures for improving the attenuation of the signal level at the fundamental frequency of the noise component Graph showing frequency characteristics of class D amplifier shown in FIG. FIG. 12 is a block diagram showing a class D amplifier different from FIG. 12 as one example of conventional measures for improving the attenuation of the signal level at the fundamental frequency of the noise component. Graph showing frequency characteristics of class D amplifier shown in FIG.

Explanation of symbols

2 amplifying section 3 speaker 5 signal output line 18 class D amplifier 19 first filter 20 second filter

Claims (7)

A class D amplifier that amplifies an audio signal using a class D amplification method and outputs the amplified audio signal to a speaker side,
Amplifying means for amplifying the audio signal input from the sound source side after converting it to a PWM signal, and outputting the amplified PWM signal to the subsequent stage;
Connected to the subsequent stage of the amplifying unit, the PWM signal output from the amplifying unit is converted into an analog audio signal, and the converted analog audio signal is output to a signal output line toward the speaker. A first filter composed of a low-pass filter having a predetermined cutoff frequency according to a frequency of the signal with respect to a signal of a noise component having a predetermined frequency higher than the cutoff frequency included in the PWM signal. A first filter formed to exhibit a predetermined frequency characteristic capable of attenuation of the signal level;
In a subsequent stage of the first filter, the filter is connected in parallel to the signal output line so as to exhibit a predetermined frequency characteristic capable of attenuating a signal level according to the frequency of the signal of the noise component. A second filter, and
When the frequency characteristic of the second filter is combined with the frequency characteristic of the first filter, the attenuation of the signal level at the fundamental frequency of the signal of the noise component is the frequency characteristic of only the first filter. wherein the frequency characteristic is larger than the attenuation of the signal level at the fundamental frequency by,
In addition, the frequency characteristic of the second filter is a signal level attenuation amount in a first frequency band having a predetermined bandwidth including the fundamental frequency in a state where it is combined with the frequency characteristic of the first filter. The maximum attenuation at the fundamental frequency and the signal level attenuation in the first frequency band is greater than the signal level attenuation in the first frequency band due to the frequency characteristics of only the first filter. And frequency characteristics like
Further, the frequency characteristic of the second filter is a second frequency band having a predetermined bandwidth continuous to the first frequency band on the high frequency side in a state where it is combined with the frequency characteristic of the first filter. The signal level attenuation amount in the second frequency band due to the frequency characteristics of only the first filter, and the noise component of the second frequency band is within the second frequency band. The frequency characteristic is such that it does not include the frequency of the second or higher harmonic component signal relative to the fundamental frequency component signal in the signal,
A switch disposed in the second filter and configured to connect or disconnect the second filter body with respect to the signal output line;
Output level detection means for detecting the output level of the sound output by the speaker;
The output level detected by the output level detecting means is an output that brings the noise band in which the signal of the noise component whose bandwidth increases with the increase in the output level to the second frequency band. Switch control means for controlling the switch to disconnect the second filter from the signal output line in the case of a level;
Class D amplifier characterized by comprising a. A class D amplifier that amplifies an audio signal using a class D amplification method and outputs the amplified audio signal to a speaker side,
Amplifying means for amplifying the audio signal input from the sound source side after converting it to a PWM signal, and outputting the amplified PWM signal to the subsequent stage;
Connected to the subsequent stage of the amplifying unit, the PWM signal output from the amplifying unit is converted into an analog audio signal, and the converted analog audio signal is output to a signal output line toward the speaker. A first filter composed of a low-pass filter having a predetermined cutoff frequency according to a frequency of the signal with respect to a signal of a noise component having a predetermined frequency higher than the cutoff frequency included in the PWM signal. A first filter formed to exhibit a predetermined frequency characteristic capable of attenuation of the signal level;
In a subsequent stage of the first filter, the filter is connected in parallel to the signal output line so as to exhibit a predetermined frequency characteristic capable of attenuating a signal level according to the frequency of the signal of the noise component. The second filter
With
When the frequency characteristic of the second filter is combined with the frequency characteristic of the first filter, the attenuation of the signal level at the fundamental frequency of the signal of the noise component is the frequency characteristic of only the first filter. The frequency characteristic is larger than the signal level attenuation at the fundamental frequency by
The second filter includes an inductor and a resistor connected in parallel to each other, a plurality of capacitors connected in series to the inductor and the resistor and connected in parallel to each other, and each of the plurality of capacitors. Is formed by the same number of switches as the capacitor that can be connected to or disconnected from the signal output line,
The class D amplifier main body further includes frequency variable means capable of changing the fundamental frequency, and switch control means for controlling the switch for connecting the predetermined capacitor.
When the fundamental frequency is changed by the frequency variable means, the switch control means combines the frequency characteristic of the second filter with the frequency characteristic of the first filter as control of the switch. In order to obtain a frequency characteristic such that the signal level attenuation amount at the fundamental frequency after the change is larger than the signal level attenuation amount at the fundamental frequency after the change due to the frequency characteristic of only the first filter. Be configured to control the switch
Class D amplifier characterized by When the frequency characteristic of the second filter is combined with the frequency characteristic of the first filter, the amount of signal level attenuation in the first frequency band having a predetermined bandwidth including the fundamental frequency is The maximum at the fundamental frequency, and the amount of attenuation of the signal level in the first frequency band is larger than the amount of attenuation of the signal level in the first frequency band due to the frequency characteristics of only the first filter. The class D amplifier according to claim 2 , wherein the class D amplifier has frequency characteristics. The frequency characteristic of the second filter is a signal in a second frequency band having a predetermined bandwidth connected to the first frequency band on the high frequency side in a state where it is combined with the frequency characteristic of the first filter. The level attenuation amount is smaller than the signal level attenuation amount in the second frequency band due to the frequency characteristic of only the first filter, and the noise component signal is within the second frequency band. 4. The class D amplifier according to claim 3 , wherein the class D amplifier has a frequency characteristic not including a frequency of a second or higher harmonic component signal with respect to a fundamental frequency component signal. 2. The class D amplifier according to claim 1, wherein the second filter includes an inductor and a resistor connected in parallel to each other, and a capacitor connected in series to the inductor and the resistor. Before and after the change of the fundamental frequency, the frequency characteristic of the second filter includes the fundamental frequency before or after the change in the state combined with the frequency characteristic of the first filter. The signal level attenuation amount in one frequency band is larger than the signal level attenuation amount in the first frequency band due to the frequency characteristic of only the first filter, and the first frequency band has a higher frequency side. The signal level attenuation amount in the second frequency band that continues in the frequency band characteristic is smaller than the signal level attenuation amount in the second frequency band due to the frequency characteristic of only the first filter,
The class D amplifier body further includes output level detection means for detecting the output level of the sound output by the speaker,
The switch control means is configured such that the output level detected by the output level detection means includes a noise band in which a signal of the noise component whose bandwidth increases with an increase in the output level is present in the second frequency band. 5. The device according to claim 2, wherein the second filter is configured to be disconnected with respect to the signal output line by controlling the switch . The class D amplifier according to item 1 . The first filter is an LC type low-pass filter.
The class D amplifier according to any one of claims 1 to 6, wherein:

Images