JP2004343617A - Digital amplifier - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To obtain a digital amplifier which suppresses shock sound emitted at muting and shortens muting time. <P>SOLUTION: When a muting start signal is off, a comparator 42 compares a gain setting sound signal with a peak sound signal held in a peak hold register 43, outputs the larger sound signal to the peak hold register 43, and the peak hold register 43 holds the sound signal outputted from the comparator 42 as the peak sound signal. When the muting signal is turned off, an attenuation value operation part 44 calculates an attenuation coefficient on the basis of the predetermined maximum amplitude A0 of the digital amplifier, the muting time t0 required for muting the maximum amplitude A0 of the digital amplifier and the amplitude A1 of the peak sound signal. A multiplier 45 then performs multiplication between the attenuation coefficient calculated by the attenuation value operation part 44 and the gain setting sound signal held in an output buffer 41. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a digital amplifier for audio, and more particularly, to a digital amplifier of a pulse width modulation (PWM) system.
[0002]
[Prior art]
2. Description of the Related Art In recent years, notebook computers, portable CD (Compact Disk) players, DVD (Digital Versatile Disc) players, car audio, and other devices that operate on batteries and have built-in speakers have become widespread. These devices are required to have not only high sound quality but also small size and low power. Against this background, digital amplifiers have attracted attention. Among them, a PWM digital amplifier used for an audio preamplifier is configured with a digital circuit from input to output, and is capable of digitally processing all audio signals. In the PWM method, it is possible to convert the voltage amplitude of the audio signal into a digital pulse width and directly drive the speaker, and there is no need for analog processing. Therefore, it is possible to realize a small-sized amplifier with low power and small heat generation.
[0003]
However, since everything is digital processing, if the output audio signal changes sharply, an abnormal sound or a shock sound is generated. In particular, this phenomenon occurs in a BTL (Bridge-Tied Load) connection in which a load is driven by two outputs of a positive polarity and a negative polarity, and large noise occurs when the output of the PWM output circuit changes from zero output to absolute zero output.
[0004]
In order to improve such a problem, in the related art, when adjusting the gain value of the output audio signal, the gain difference between the current gain value and the set gain value is detected, and the gain amount is adjusted to a predetermined amount. When the gain is large, the width of change of the gain amount is divided and divided into a plurality of times, and the gain amount is changed little by little (for example, see Patent Document 1).
[0005]
[Patent Document 1]
JP 10-313224 A
[Problems to be solved by the invention]
It is assumed that the above-described conventional technique is applied to a case where an output audio signal is muted by an external mute start signal, that is, the amplitude of the output audio signal is set to zero. For example, when the mute start signal is turned on, the set gain value is set to 0, a gain difference from the current gain value is detected, and when the gain amount is larger than a predetermined amount, the change width of the gain amount is divided and By changing the gain amount little by little, the shock sound can be suppressed and the output audio signal can be muted.
[0007]
However, when the above-described conventional technique is applied to the mute process, there is a problem that the mute time is determined by the number of divisions of the variation width of the gain amount.
[0008]
The present invention has been made in view of the above, and an object of the present invention is to provide a digital amplifier that suppresses a shock sound during mute and shortens a mute time.
[0009]
[Means for Solving the Problems]
In order to achieve the above object, a digital amplifier according to the present invention is a digital amplifier having a mute function of muting a gain-set audio signal based on a mute start signal. An output buffer for accumulating a signal, a peak detector for detecting a maximum value of a gain-set audio signal until the mute function is enabled by the mute start signal, and a mute function enabled by the mute start signal In the case, a predetermined maximum amplitude of the digital amplifier and a mute time required to mute the audio signal of the maximum amplitude of the digital amplifier and a gain until the mute function detected by the peak detection unit is enabled are set. Attenuator that adjusts the mute time based on the maximum amplitude of the audio signal And a multiplier for multiplying the attenuation coefficient calculated in the attenuation value calculation unit by the gain-set audio signal stored in the output buffer. .
[0010]
According to the present invention, a peak audio signal that is the maximum value of the gain setting audio signal before the start of mute is detected, and when the amplitude of the detected peak audio signal is smaller than the maximum amplitude of the digital amplifier at the start of mute, Calculates the attenuation coefficient based on the maximum amplitude of the digital amplifier, the mute time required to mute the maximum amplitude of the digital amplifier, and the amplitude of the peak audio signal, and stores the calculated attenuation coefficient and the output buffer. The output audio signal is muted by multiplying by the gain setting audio signal.
[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, preferred embodiments of a digital amplifier according to the present invention will be described in detail with reference to the accompanying drawings.
[0012]
Embodiment 1 FIG.
An embodiment of the present invention will be described with reference to FIGS. FIG. 1 is a block diagram showing a configuration of the digital amplifier according to the first embodiment of the present invention. The digital amplifier according to the first embodiment of the present invention includes a serial / parallel conversion unit 10, a new gain value register 20, a gain setting unit 30, a mute processing unit 40, a delta-sigma conversion unit 50, and a PWM output unit. 60. These circuits shown in FIG. 1 are integrated on a single semiconductor chip.
[0013]
The serial / parallel conversion unit 10 serially inputs a single input terminal provided in an IC (Integrated Circuit) in which the circuit of FIG. 1 is integrated such as a linear PCM (Pulse Code Modulation) signal and a DSD (Direct Stream Digital) signal. The input audio signal is converted into parallel data. For example, a 16-bit input audio signal serially input to the single input terminal is converted into 16-bit parallel data. The 16 bits as one unit represent the amplitude of the sound at a certain time, and the next 16-bit unit subsequently inputted is the time determined by the sampling rate from the time at which the immediately preceding 16-bit unit represents the amplitude. Represents the later audio amplitude.
[0014]
The new gain value register 20 holds a new gain value setting value input from an input terminal for inputting a new gain value of the IC. The gain setting section 30 multiplies the audio signal converted into parallel data by the serial / parallel conversion section 10 by the new gain value set value held in the new gain value register 20 to adjust the gain of the audio signal. .
[0015]
The mute processing unit 40 accumulates a gain-set audio signal that is a gain-set audio signal sequentially input from the gain setting unit 30 for a predetermined time, and turns off the mute start signal (normal operation without performing the mute process). The maximum value of the gain setting audio signal during the period is detected. When the mute start signal turns on (performs mute processing), the amplitude of the maximum value of the gain setting audio signal detected during the period in which the mute start signal is off, the predetermined maximum amplitude of the digital amplifier, and the maximum amplitude of the digital amplifier Based on the mute time necessary to mute the amplitude, the shortest attenuation coefficient that does not generate a shock sound is calculated, and the accumulated gain setting audio signal is muted.
[0016]
FIG. 2 is a block diagram showing a configuration of the mute processing unit 40 shown in FIG. The mute processing unit 40 includes an output buffer 41, a comparator 42, a peak hold register 43, an attenuation value calculation unit 44, and a multiplier 45. It should be noted that the comparator 42 and the peak hold register 43 implement the function of a peak detection unit as defined in the claims.
[0017]
The output buffer 41 has n (0 <n, n is a natural number) FIFO (First-In, First-Out) stages in which each stage can hold parallel data for the number of bits of the audio signal set in the gain setting unit 30. And a shift register configured to sequentially shift a gain setting audio signal in synchronization with a clock. The number of stages n of the shift register is determined by the gain setting unit during a mute time t0 necessary for muting the audio signal of the maximum amplitude A0 of the digital amplifier without generating a shock sound when the audio signal is muted (to make the amplitude 0). The gain setting audio signal input from the control unit 30 is set to a value that can be accumulated. Specifically, since the output buffer 41 is a shift register that operates in synchronization with a clock, if the clock cycle is T, the number n of stages of the shift register is a value that satisfies the relationship n> t0 / T. . In the circuit shown in FIG. 2, the maximum amplitude A0 is predetermined as a maximum specified value of a gain-set audio signal input from the gain setting unit 30 to the output buffer 41.
[0018]
When the mute start signal is off, the comparator 42 compares the gain setting audio signal with the peak audio signal held in the peak hold register 43, and outputs the larger audio signal to the peak hold register 43. When the mute start signal is on, the comparator 42 does not perform the comparison operation.
[0019]
The peak hold register 43 holds the audio signal input from the comparator 42. That is, when the mute start signal is off, the peak hold register 43 holds the peak audio signal that is the maximum value of the gain setting audio signal sequentially input from the gain setting unit 30.
[0020]
When the mute start signal is on, the attenuation value calculation unit 44 determines a shock based on the peak audio signal and the predetermined maximum amplitude of the digital amplifier and the mute time required to mute the maximum amplitude of the digital amplifier. Calculate the shortest attenuation coefficient that does not produce sound. When the mute start signal is off, the attenuation value calculation unit 44 fixes the attenuation coefficient to “1” so as not to change the gain of the gain setting audio signal output from the last-stage shift register in the output buffer 41. .
[0021]
The multiplier 45 multiplies the gain setting audio signal output from the last-stage shift register in the output buffer 41 by the attenuation coefficient calculated by the attenuation value calculator 44.
[0022]
The delta-sigma conversion unit 50 performs delta-sigma modulation on the audio signal input from the mute processing unit 40, and moves quantization noise out of the audible band.
[0023]
The PWM output unit performs PWM modulation on the audio signal in which the quantization noise has been moved out of the audible band in the delta-sigma conversion unit 50.
[0024]
Next, the operation of the digital amplifier according to the first embodiment of the present invention will be described. Initially, it is assumed that the mute start signal is set to off. The serial / parallel converter 10 converts an input audio signal input as serial data to a single input terminal such as a linear PCM signal or a DSD signal into parallel data, and converts the converted audio signal into parallel data into a gain setting unit. Output to 30.
[0025]
The gain setting unit 30 multiplies the audio signal converted into the parallel data by the new gain value set value input from the new gain value setting input terminal held in the new gain value register 20. , The gain of the audio signal converted into parallel data is set. Then, the gain setting audio signal that is the audio signal for which the gain has been set is output to the output buffer 41 of the mute processing unit 40 and the comparator 42.
[0026]
The output buffer 41 sequentially shifts the gain setting audio signal. Since the mute start signal is off, the attenuation value calculation unit 44 fixes the attenuation coefficient to “1” and outputs it to the multiplier 45 irrespective of the output from the peak detection unit including the comparator 42 and the peak hold register 43. The multiplier 45 multiplies the gain setting audio signal output from the last-stage shift register in the output buffer 41 by the attenuation coefficient, and outputs the multiplied audio signal to the delta-sigma converter 50. That is, when the mute start signal is off, the attenuation coefficient is “1”, so that the mute processing unit 40 outputs the gain setting audio signal output from the final stage of the output buffer 41 to the delta-sigma conversion unit 50. .
[0027]
On the other hand, the comparator 42 compares the gain setting audio signal input from the gain setting unit 30 with the peak audio signal held in the peak hold register 43, and outputs the larger audio signal to the peak hold register 43. I do. The peak hold register 43 holds the audio signal output from the comparator 42. That is, the comparator 42 detects the maximum value of the gain-set audio signal during the normal operation period, and the peak hold register 43 holds the maximum value of the gain-set audio signal detected by the comparator 42.
[0028]
The delta-sigma conversion unit 50 performs delta-sigma modulation on the audio signal input from the mute processing unit 40, and moves quantization noise out of the audible band. Then, an audio signal obtained by moving the quantization noise out of the audible band is output to the PWM output unit 60.
[0029]
The PWM output unit 60 performs PWM modulation on an audio signal whose quantization noise has been moved out of the audible band, and outputs an output audio signal.
[0030]
Here, it is assumed that the mute start signal is set to ON to mute the output audio signal. When the mute start signal is set to ON, the comparator 42 of the mute processing unit 40 compares the peak audio signal held in the peak hold register 43 with the gain setting audio signal input from the gain setting unit 30. Stop operation. Therefore, the peak hold register 43 holds the peak audio signal which is the maximum value of the gain setting audio signal before the mute start signal is set to ON. That is, the values of the n gain-set audio signals stored in the output buffer 41 are all lower than the peak audio signal held in the peak hold register 43.
[0031]
The attenuation value calculation unit 44, based on a predetermined maximum amplitude of the digital amplifier and a mute time required to mute the maximum amplitude of the digital amplifier, and a peak audio signal held in the peak hold register 43, Calculate the shortest attenuation coefficient that does not generate a shock noise.
[0032]
Assuming that the maximum amplitude of the digital amplifier is A0 and the mute time required to mute the maximum amplitude A0 so as not to generate a shock sound is t0, a mute curve at the maximum amplitude A0 is as shown by a solid line in FIG. Suppose that it has a linear mute curve. Assuming that the amplitude of the peak audio signal is A1, the relationship A1 ≦ A0 holds between the amplitude A1 of the peak audio signal and the maximum amplitude A0 of the digital amplifier.
[0033]
When the amplitude A1 of the peak audio signal is equal to the maximum amplitude A0 of the digital amplifier, the shortest mute time required for mute is t0, and the attenuation coefficient becomes equal to the attenuation coefficient of the mute curve shown by the solid line. Therefore, the attenuation value calculator 44 outputs the attenuation coefficient of the mute curve indicated by the solid line to the multiplier 45.
[0034]
When the amplitude A1 of the peak audio signal is smaller than the maximum amplitude A0 of the digital amplifier (A1 <A0), since the mute time required to mute the maximum amplitude A0 is t0, the amplitude A1 of the peak audio signal is muted. Mute time t1 required for
t1 = t0 × A1 / A0 (1)
It becomes. Here, assuming that the time from the start of the mute is t, the attenuation coefficient for muting the amplitude A1 of the peak audio signal at the mute time t1 is:
Damping coefficient = 1−t / t1 (2)
It becomes. Substituting equation (1) into the mute time t1 required to mute the amplitude A1 of the peak audio signal in equation (2), and expressing the attenuation coefficient,
Attenuation coefficient = 1-t / (t0 × A1 / A0) (3)
It becomes. Here, assuming that the number of clocks from the start of the mute is x, the time t from the start of the mute is represented by the number of clocks x from the start of the mute and the cycle T of the clock.
t = xT (4)
Is represented by Substituting equation (4) into the time t from the start of mute in equation (3) to represent the attenuation coefficient,
Attenuation coefficient = 1−xT / (t0 × A1 / A0) (5)
Is represented by For comparison, the mute curve when muting from the maximum amplitude A0 by this attenuation coefficient is as shown by the one-dot chain line in FIG. That is, the smaller the amplitude A1 of the peak audio signal, the faster the attenuation coefficient is set to zero. When the amplitude A1 of the peak audio signal held in the peak hold register 43 is smaller than the maximum amplitude A0 of the digital amplifier, the attenuation value calculation unit 44 counts the number of clocks from the start of mute, and calculates the equation (5). ) Is used to calculate the attenuation coefficient. Then, the calculated attenuation coefficient is output to the multiplier 45.
[0035]
The multiplier 45 multiplies the gain-set audio signal input from the final-stage shift register in the output buffer 41 by the gain coefficient input from the attenuation value calculator 44, and outputs the result to the delta-sigma converter 50. Output.
[0036]
Such an operation is repeated in synchronization with the clock until the mute time t1 required to mute the amplitude A1 of the peak audio signal held in the peak hold register 43, that is, until the attenuation coefficient becomes zero. That is, in synchronization with the clock, the attenuation value calculator 44 calculates 1-T / (t0 × A1 / A0), 1-2T / (t0 × A1 / A0),..., 1-t1 / (t0 × A1 / A0). A0) = 0 are sequentially output. Thus, even when the amplitude of the gain-set audio signal stored in the output buffer 41 is all the amplitude A1 of the peak audio signal held in the peak hold register 43, as shown by the broken line in FIG. The amplitude of the audio output signal becomes 0 at the mute time t1 required for muting without generating a shock sound. That is, the mute time can be reduced by t0−t1 = t0 × (1−A1 / A0) as compared with the case where the mute time is fixed to t0.
[0037]
As described above, in the first embodiment, the maximum value of the gain setting audio signal before the start of the mute is detected, and when the maximum value of the detected audio signal is smaller than the maximum amplitude of the digital amplifier at the start of the mute. Because the attenuation coefficient is calculated based on the maximum amplitude of the digital amplifier, the mute time required to mute the maximum amplitude of the digital amplifier, and the amplitude of the maximum value of the detected gain setting audio signal. In addition, the mute process can be performed in a shorter time than when the gain setting audio signal is muted according to a predetermined mute curve while suppressing the shock sound.
[0038]
In addition, since only the maximum amplitude A0 of the digital amplifier and the mute time t0 required to mute the maximum amplitude A0 of the digital amplifier are stored as data, the time required for mute can be reduced. By changing only these two data, it is possible to support various devices.
[0039]
Furthermore, since the output buffer 41 accumulates the gain setting audio signal from the time when the mute function is enabled by the mute start signal to the time t0 for muting the maximum amplitude A0, the mute is started after the mute operation is started. By inputting the signal accumulated in the output buffer 41 to the multiplier 45 until completion, a signal having an amplitude larger than the peak audio signal is not input to the multiplier 45. Therefore, it is possible to prevent a situation in which a signal having a large amplitude is multiplied by an attenuation coefficient that quickly becomes 0 to a small amplitude and a shock noise is generated.
[0040]
Embodiment 2 FIG.
In the first embodiment, a linear mute curve is used. However, in the case of a linear mute curve, depending on the mute time, a shock sound may be generated at the start of mute and at the end of mute. In order to improve such a problem, in the second embodiment, as shown by the solid line in FIG. 4, at the start of mute and at the end of mute, a mute curve with a gently changed cosine curve is used. . Also in this case, when the amplitude A1 of the peak audio signal held in the peak hold register 43 is smaller than the maximum amplitude A0 of the digital amplifier, the attenuation coefficient = (1 + cos (xT × π / (t0 × A1 / A0)) ) / 2 is calculated and multiplied by the gain setting audio signal sequentially output from the last-stage shift register in the output buffer 41, thereby comparing with the case using the attenuation coefficient of the mute curve shown by the solid line in FIG. Thus, the mute time can be reduced by t0−t1 = t0 × (1−A1 / A0) while suppressing the shock noise.
[0041]
In the first and second embodiments, the mute time t0 required to mute the maximum amplitude A0 of the digital amplifier has been described as one. However, the attenuation value calculation unit 44 includes, for example, FIG. A plurality of mute times t0 are stored, such as a mute time t0 due to the linear mute curve shown and a mute time t0 due to a gently changed cosine curve at the start and end of the mute shown in FIG. In addition, as shown in FIG. 5, a desired mute time may be selected by adding a mute time selection signal to the mute processing unit 40 of FIG. With this, when an input audio signal of a large volume (a large maximum amplitude) that is likely to be affected by a change in the gain value is input, a mute time is selected so that the slope of the gain value at the time of mute is moderated. When an input audio signal of a small volume (smallest amplitude is small) that is less affected by a change in the gain value is input, a mute time such as steepening the slope of the gain value at the time of mute is selected. A mute time suitable for a sound source input to the digital amplifier can be set.
[0042]
【The invention's effect】
As described above, according to the digital amplifier of the present invention, the peak audio signal which is the maximum value of the gain setting audio signal before the start of the mute is detected, and when the mute starts, the amplitude of the detected peak audio signal is reduced. If less than the maximum amplitude of the digital amplifier, the attenuation coefficient was calculated based on the maximum amplitude of the digital amplifier, the mute time required to mute the maximum amplitude of the digital amplifier, and the amplitude of the peak audio signal. The gain setting audio signal is muted by multiplying the attenuation coefficient by the gain setting audio signal stored in the output buffer. This makes it possible to perform a mute process in a shorter time than muting the gain setting audio signal according to a predetermined mute curve while suppressing the shock noise.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration of a digital amplifier according to a first embodiment of the present invention.
FIG. 2 is a block diagram illustrating a configuration of a mute processing unit illustrated in FIG. 1;
FIG. 3 is a diagram showing a mute curve according to the first embodiment of the present invention.
FIG. 4 is a diagram showing a mute curve according to a second embodiment of the present invention.
FIG. 5 is a block diagram illustrating a configuration of a mute processing unit illustrated in FIG. 1;
[Explanation of symbols]
Reference Signs List 10 serial / parallel conversion unit, 20 new gain value register, 30 gain setting unit, 40 mute processing unit, 41 output buffer, 42 comparator, 43 peak hold register, 44 attenuation value calculation unit, 45 multiplier, 50 delta-sigma Conversion unit, 60 PWM output unit.

Claims (3)

In a digital amplifier having a mute function for muting a gain-set audio signal based on a mute start signal,
An output buffer for accumulating a gain-set audio signal for a predetermined time;
A peak detection unit that detects the maximum value of a gain-set audio signal until the mute function is enabled by the mute start signal,
When the mute function is enabled by the mute start signal, a predetermined maximum amplitude of the digital amplifier and a mute time required to mute the audio signal of the maximum amplitude of the digital amplifier are detected by the peak detection unit. An attenuation value calculation unit that calculates an attenuation coefficient that adjusts a mute time based on an amplitude of a maximum value of an audio signal whose gain is set until the mute function is enabled,
A multiplier for multiplying the attenuation coefficient calculated by the attenuation value calculation unit and the gain-set audio signal stored in the output buffer;
A digital amplifier comprising: The attenuation value calculation unit,
The digital amplifier has a plurality of mute times required to mute the audio signal having the maximum amplitude, and sets a mute time required to mute the audio signal having the maximum amplitude of the digital amplifier selected by the mute time selection signal. 2. The digital amplifier according to claim 1, wherein an attenuation coefficient for adjusting a mute time is calculated using the attenuation coefficient. The output buffer comprises:
The digital amplifier according to claim 1, wherein a gain-set audio signal is stored during a mute time required to mute the audio signal having the maximum amplitude of the digital amplifier.

Images