Patent Grant
7327188
1. Field of the Invention
The present invention relates to a power amplifier, and more particularly, to a power amplifier and method for error correcting of the output signals thereof.
2. Description of the Prior Art
The class-D power amplifiers, due to their high efficiency, are being widely applied in various audio amplifying apparatus. However, the output signal quality from the class-D power amplifier is easily deteriorated by noise or by the imperfect nature of the switching power stage thereof.
U.S. Pat. No. 6,768,779 discloses a conventional power amplifier. This conventional power amplifier has a disadvantage that the central point of the waveform of the output signal outputted by this conventional power amplifier will be offset t0/2.
In view of the foregoing, it can be appreciated that a substantial need exists for methods and apparatus to improve the quality and fidelity of signals output from the class-D amplifier.
It is therefore an objective of the claimed invention to provide a power amplifier and method for correcting error of output signals thereof.
An exemplary embodiment of a power amplifier is disclosed comprising a subtracting unit for generating an error signal according to an input signal and an output signal; a noise shaping module executing a noise shaping operation on the error signal to produce a noise-shaped signal; a pulse adjustment unit generating a control signal according to the noise-shaped signal and the input signal; and a power stage generating and adjusting the output signal according to the control signal.
An exemplary embodiment of a method for calibrating a power amplifier is disclosed comprising: generating an error signal according to an output signal from the power amplifier and an input signal; producing a noise-shaped signal by executing a noise shaping operation on the error signal; and generating a control signal according to the noise-shaped signal and the input signal to adjust the output signal.
An exemplary embodiment of a power amplifier is disclosed comprising: a first pulse width modulator for generating a pulse width modulated signal according to an input signal; a power stage generating an output signal according to a control signal; a subtracting unit generating an error signal according to the pulse width modulated signal and the output signal; a noise shaping module producing a noise-shaped signal according to the error signal; an adding unit summing the noise-shaped signal and the input signal to produce a calibration signal; and a second pulse width modulator generating the control signal according to the calibration signal; wherein the first and second pulse width modulation modulators are substantially the same.
These and other objectives of the claimed invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.
Please refer to
In this embodiment, the class-D power amplifier 100 receives an input signal Vin, which is a digital audio signal such as a digital pulse width modulated (PWM) signal. The input signal Vin could be regarded as a DC signal for each pulse duration (Ts=1/fs). Accordingly, the signal output error from the class-D power amplifier 100 can be corrected by compensating an error between the input and output signals of the class-D power amplifier 100 of nth pulse duration in (n+1)th pulse duration. In the following descriptions, operations of the class-D power amplifier 100 will be described as shown in
In step 210, the subtracting unit 110 subtracts an output signal Vout generated by the power stage 140 from the input signal Vin to produce an error signal Ve.
In step 220, the noise shaping module 120 performs a noise shaping operation on the error signal Ve to produce a noise-shaped signal. In the embodiment shown in
In step 230, the pulse adjustment unit 130 produces a control signal according to the noise-shaped signal and the input signal Vin to control the operations of the power stage 140. Specifically, the pulse adjustment unit 130 of this embodiment produces the control signal by adjusting the pulse width of the input signal Vin according to the noise-shaped signal so as to calibrate the error of the output signal Vout in next pulse duration.
In another embodiment, the working frequency Fq of the quantizer 124 may be designed to be twice the frequency fs of the input signal Vin, i.e. Fq=2fs, then the pulse adjustment unit 130 calibrates the error of the output signal Vout of former half pulse duration in the later half pulse duration. The error correction range is ±2m−1·Δ, where m is the bit number of the quantizer 124 and Δ is the resolution of the pulse adjustment unit 130. In other words, by adjusting the working frequency of the quantizer 124, the timing of calibrating the error of the output signal from the class-D power amplifier 100 is changed.
Form formulae (1) and (2), it can be derived that if the filter 122 is implemented with a low-pass filter, then the noise immunity of the class-D power amplifier 100 at the audio frequency (i.e., voice frequency) is significantly improved.
In this embodiment, the noise shaping module 440 comprises a filter 442 and a m-bits quantizer 444. The functions and operations of the filter 442 and the quantizer 444 are substantially the same as the filter 122 and the quantizer 124, respectively, and further details are therefore omitted. Additionally, in the class-D power amplifier 400, the cooperation of the adding unit 450 and the second pulse width modulator 460 realizes the functions of the pulse adjustment unit 130 of the first embodiment. In this embodiment, the error correction range is ±2m−1·Δ, Δ=Ts/2n, where m is the bit number of the quantizer 444 and n is the bit number of the second pulse width modulator 460. Similarly, by utilizing the feedback control, the error from the power stage 420 caused by the noise of the power supply or the imperfect nature of the components thereof and the quantization error of the quantizer 444 are all significantly improved.
As disclosed in the foregoing descriptions, the error calibration method for the class-D power amplifier in the present invention is capable of accurately controlling the error correction range while also avoiding the error correction range from being affected by the process variation of the components thereof so that the undesirable output saturation problems are thereby improved.
Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention. Accordingly, the above disclosure should be construed as limited only by the metes and bounds of the appended claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 94101321 A | Jan 2005 | TW | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 6140875 | Van Den Homberg et al. | Oct 2000 | A |
| 6768779 | Nielsen | Jul 2004 | B1 |
| 6943620 | Andersen et al. | Sep 2005 | B2 |
| 6977546 | Stapleton | Dec 2005 | B2 |
| 7058464 | Mallinson | Jun 2006 | B2 |
| 7248193 | Fujimoto | Jul 2007 | B2 |
| 20050012545 | Mallinson | Jan 2005 | A1 |
| Number | Date | Country | |
|---|---|---|---|
| 20060197590 A1 | Sep 2006 | US |
