WO2008001321A2

WO2008001321A2 - A device for and a method of processing an audio signal

Info

Publication number: WO2008001321A2
Application number: PCT/IB2007/052510
Authority: WO
Inventors: Renaud De Saint Moulin
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2006-06-30
Filing date: 2007-06-28
Publication date: 2008-01-03
Also published as: WO2008001321A3

Abstract

A device (100, 200) for processing an audio signal (101), wherein the device (100, 200) comprises an adjustment unit (105) adapted for adjusting a dead time setting for a power stage (102) of an amplifier based on an amplitude of the audio signal (103).

Description

A device for and a method of processing an audio signal

FIELD OF THE INVENTION

The invention relates to a device for processing an audio signal. Beyond this, the invention relates to a method of processing an audio signal. Moreover, the invention relates to a program element. Furthermore, the invention relates to a computer-readable medium.

BACKGROUND OF THE INVENTION

Audio playback devices become more and more important. Particularly, audio systems comprising audio amplifiers become more and more important. US 6,294,954 discloses an apparatus for adaptively reducing dead time in a switching circuit including overlap detection circuitry for measuring the dead time/overlap of the switches, and control circuitry for setting the dead time to the optimum level (generally the minimum possible dead time without any overlap occurring). The dead time/overlap may be detected by measuring the current through the switches, the current into the power supply, the voltage waveform at the switch point, or the average voltage waveform at the switch point. The dead time may be controlled by utilizing delay elements prior to the drivers, or by utilizing circuitry to control the driver timing.

However, conventional audio amplifier systems may suffer from high energy consumption. This holds for Class AB amplifiers particularly in a high power music output mode, and also for Class D amplifiers in an idling operation state.

OBJECT AND SUMMARY OF THE INVENTION

It is an object of the invention to provide an audio system having an efficient amplifier. In order to achieve the object defined above, a device for processing an audio signal, a method of processing an audio signal, a program element and a computer-readable medium according to the independent claims are provided.

According to an exemplary embodiment of the invention, a device for processing an audio signal is provided, wherein the device comprises an adjustment unit adapted for adjusting a dead time setting for a power stage of an amplifier based on an amplitude (like a current value) of the audio signal.

According to another exemplary embodiment of the invention, a method of processing an audio signal is provided, wherein the method comprises adjusting a dead time setting for a power stage of an amplifier based on an amplitude (like a current value) of the audio signal.

According to yet another exemplary embodiment of the invention, a computer- readable medium is provided, in which a computer program of processing audio data is stored which, when being executed by a processor, is adapted to control or carry out a method having the above mentioned features.

According to still another exemplary embodiment of the invention, a program element of processing audio data is provided, which program element, when being executed by a processor, is adapted to control or carry out a method having the above mentioned features. Signal processing and circuit management for reducing switch losses which may be performed according to embodiments of the invention can be realized by a computer program, that is by software, or by using one or more special electronic optimization circuits, that is in hardware, or in hybrid form, that is by means of software components and hardware components. According to an exemplary embodiment of the invention, an audio output current may be monitored in an amplifier system to adjust the dead time of drivers of the amplifier system dynamically to the present output load condition. In other words, an amplitude of an electric signal, for instance a current, indicative of the processed audio signal, is detected and the dead time configuration according to which the power stage of the amplifier is operated may be modified in accordance with this amplitude. Embodiments of the invention may allow to significantly reducing switch losses in an amplifier by taking such measures. A long dead-time may result in low idling losses. However, embodiments of the invention may take measures to reduce the idling losses, without compromising on the audio quality. Dead-time adjustment is applicable to particularly to switching converters.

Class D amplifiers have become popular for the last years thanks to their high efficiency at medium and high output power. Because audio systems get smaller and smaller while output power are increasing (to compensate for the low acoustical efficiency of the small loudspeaker boxes that are delivered with those (multichannel) audio sets), Class D becomes more and more important. According to embodiments of the invention, the efficiency of Class D amplifiers may be improved even when idling or at very low modulation, when the output power is zero (or almost zero). When delivering burst of audio signals at very high power, Class D may also be more efficient than Class AB. In contrast to a dynamic dead-time control aiming to keep the dead-time as short as possible (basically as close as possible to zero without getting destructive shoot-through conditions), embodiments of the invention may not only intend to improve or optimize audio performance, but to take special care for idling losses. Embodiments of the invention may take advantage of the fact that total harmonic distortion (THD) may be independent of dead-time for audio output currents smaller than the ripple current through the output filter.

Embodiments of the invention may provide a very high efficiency amplifier that also has low idling losses (efficiency may be zero at no output power), with essentially no compromise on the audio performance. The amplitude of the audio signal may refer to voltage. Voltage amplitude may be not good enough; it may be desirable to use current when working with a load that is not purely resistive.

Therefore, an exemplary embodiment of the invention intends to escape the idling losses problems that may occur particularly with Class D audio amplifiers. A power stage may be operated with a limited dead time as to avoid cross conduction when switching from one stage to the other. During the dead time, the output current may flow through a body diode of a MOSFET of the amplifier.

At low output current (with resistive loads, this may equal output modulation), the dead time has no or very limited influence on the fidelity of the power stage. A pulse width modulated (PWM) waveform may be appropriately buffered by the output stage. In this condition, the output current through an output coil may make the transition occur automatically. Additionally, the energy stored in the parasitic capacitances across the drain/source connections of the MOSFETs may be recycled during the transition.

According to an exemplary embodiment of the invention, an amplifier (for instance a Class D amplifier) may be provided comprising a dead time setting stage adapted to delay a turn-on pulse to a power switch with regard to a turn-off pulse delivered to the opposite power switch of the amplifier. A current sensing stage may be provided and adapted for sensing an output current of the amplifier, wherein the dead time setting stage may be adapted to set a first (for instance large) dead time for output currents that do not exceed a threshold value (for instance a peak value of a ripple current flowing through an output filter). A second (shorter) dead time may be selected for output currents that do exceed the threshold value (for instance a peak value of the ripple current flowing through the amplitude filter).

According to an exemplary embodiment, the current sensing stage may be adapted as an overcurrent detection/protection circuit for providing output current information. Particularly, for the current measurement circuitry, sense resistors may be used. However, it may be more appropriate to provide a secondary winding wound on the output coil core followed by an integrator. The voltage output by the integrator may be a measure of the current flowing through the primary of the output coil.

Exemplary embodiments of the invention may have the advantage that the amplifier may be prevented from running hot when idling and may still have a low total harmonic distortion (THD).

Exemplary fields of application of exemplary embodiments of the invention are all Class D operated sets (like LX 3900sa, DFR 9000, FWM 589). Furthermore, all Class D ICs may be used like TDA 8920, TDA 8932, etc. However, it might be advantageous to adapt such sets or ICs to the delay time adjustment technology according to exemplary embodiments of the invention.

According to an exemplary embodiment of the invention, an (for instance Class D) audio amplifier with low idling losses, and good THD performance may be provided. For this purpose, an output current dependent variable dead time setting may be implemented in a Class D audio amplifier to reduce the loss and distortion. Such a system may be operated with two dead time settings, and a ripple current may be sensed at the output filter. A relatively large dead time setting may be used for no or very low loads and for output currents that do not exceed a set peak value of the ripple current. This current measurement information may be derived from an (existing) overload/overcurrent detection circuitry which may be based on sense resistors or on a secondary winding wound on an output coil core, followed by the integrator (see Fig. 7, for instance).

According to an exemplary embodiment, a switch may be used for the short dead time setting associated to a time constant to keep the power stage in the short dead time setting for a longer period of time once it was triggered, disregarding whether the instantaneous current exceeds the set filter ripple current.

For audio quality reasons (when considered alone), it might be desirable to operate the power stage of the amplifier with the smallest possible dead time. On the other hand, for idling loss reasons (when considered alone), the power stage should be operated with a large dead time as to guarantee that the transitions at the switching output occur automatically, without the necessity to inject energy to charge/discharge the parasitic capacitances in the power stage.

Making the dead time setting dependent on the audio current flowing through the output coil may solve the issue. At an output current lower than a ripple current through the output filter, the dead time has no large influence on the time accuracy of the power stage and the amplifier may be operated with a large dead time, producing limited dissipation. At higher output currents, the amplifier may be operated with a short dead time, solving the dead time induced distortion. An advantage of such a configuration may be that the amplifier may be prevented from running hot when idling and may still have a low THD at any power as for a fixed dead time setting.

Because the overload protection of the amplifier may be also based on a current measurement of the current flowing to the load, the additional circuitry required to implement the described configuration is very small.

Next, further exemplary embodiments of the device for processing an audio signal will be explained. However, these embodiments also apply for the method of processing an audio signal, for the program element and for the computer-readable medium. The adjustment unit may be adapted for adjusting the dead time setting for the power stage of the amplifier based on a current of the audio signal. However, other electrical signals like a voltage value, etc. may be used as well. The current may be determinable easily and can also be determined simultaneously or synergetically for other purposes (for instance for an overcurrent protection purpose or the like), so that detecting the current may be performed without large effort.

The adjustment unit may be adapted for adjusting the dead time setting for one or more driver units (particularly for two driver units connected in parallel to one another) of the power stage of the amplifier based on the amplitude of the audio signal at an output of the amplifier. Particularly, such a current flowing through an output coil of the amplifier may be an appropriate parameter to define the value of the dead time. Amplitude adjustment (that is particularly adjustment of voltage or current) may work when the load is truly resistive, which is a sub-category of loads. When using loads having other properties than a purely ohmic property, other parameters may be considered for adjustment. When the load is purely resistive, voltage may be an option to define the value of the dead time. For any other load, current may be a preferred option.

The adjustment unit may be adapted for adjusting the dead time setting to a predetermined first dead time value when the amplitude exceeds a threshold value, and for adjusting the dead time setting to a predetermined second dead time value when the amplitude is below the threshold value. Particularly, the first dead time value may be smaller than the second dead time value. By taking this measure, high current amplitudes may be correlated with a small dead time value, and small current signals may be assigned to long dead time values. Such a configuration may allow providing a proper trade-off between

(small) idling losses and (a high) audio quality. Therefore, switching losses can be kept small and simultaneously the audio quality may be maintained at a high level.

The first dead time value may be less than essentially 30 ns, particularly may be in the range between essentially 1 ns and essentially 20 ns, more particularly may be in the range between essentially 5 ns and essentially 10 ns. The second dead time value may be more than essentially 50 ns, particularly may be in the range between essentially 80 ns and essentially 200 ns, more particularly may be in the range between essentially 100 ns and essentially 120 ns. For instance, the first dead time value may be 20 ns and the second dead time value may be 100 ns. According to an exemplary embodiment, "two states" having a large and a short dead time may be appropriate. But, of course, a continuous variation between "large" and "small" is possible, stepwise or stepless. It is also possible to have three or more different values of the dead time.

The threshold value may be selected to be equal to a ripple current (amplitude or peak value) through an output filter of the amplifier. Therefore, the ripple current flowing through such an output filter of the amplifier may be taken as a basis to determine whether the device shall be set in the long dead time operation mode or in the short dead time operation mode. The term "ripple current" may particularly denote a peak value of a current flowing through an inductor coil of a Class D amplifier due to a switching process (particularly for switching two MOSFETs connected in parallel in a power stage of a Class D amplifier).

An overcurrent detection circuit may be provided in the device and may be adapted for sensing the amplitude for detecting, based on the sensed amplitude, whether an overcurrent flows through the device, and for supplying the sensed amplitude to the adjustment unit. Therefore, the overcurrent detection circuit may detect this current value for two purposes simultaneously. On the one hand, to decide whether the current exceeds a predetermined overcurrent threshold value for preventing the amplifier from being destroyed by an overcurrent. On the other hand, this measured value can be also used as a basis for the decision as to how the dead time of the amplifier shall be managed. This synergetic configuration may allow manufacturing the device with small efforts and with low cost.

Such an overcurrent detection circuit may be adapted for sensing the amplitude using one or more sense resistors. Alternatively, the overcurrent detection circuit may be adapted for sensing the amplitude using a secondary winding of an output coil of the amplifier and may use an integrator circuit connected to the secondary winding of the output coil. In such a scenario, the current may be tapped off inductively from the secondary winding and may be analyzed after integration (for instance by an integrator circuit comprising an operational amplifier (opamp) and connected resistors/capacitors). The adjustment unit may be adapted to, when switching from a second (long) dead time value to a first (short) dead time value, maintain the device in an operation state defined by the first dead time value at least for a predetermined time interval regardless of the amplitude during the predetermined time interval (that is to say even if the amplitude decreases again after switching, the short dead time configuration is nevertheless maintained at least for a minimum time interval), wherein the first dead time value may be smaller than the second dead time value. Particularly, when switching from a short dead time to a large dead time, it may be appropriate to maintain the system in the previous operation mode for some additional time, to increase the quality.

The amplifier may be a Class D amplifier. In the context of this application, the term "Class D amplifier" may particularly denote an amplifier based on pulse width modulation (PWM) and comprising two switches that are alternatingly conductive. An example for a Class D amplifier is disclosed in EP 1,500,188 A2. Also Fig. 2 shows a Class D amplifier.

The device may comprise an audio reproduction unit for reproducing the processed audio data to the environment. Such an audio reproduction unit may comprise one or more loudspeakers, a subwoofer, etc. Therefore, the amplifier may be implemented in a complete audio data reproduction system.

The device for processing audio data may be realized as at least one of the group consisting of an audio amplifier, a loudspeaker, an audio surround system, a mobile phone, a headset, a loudspeaker, a hearing aid, a handsfree system, a television device, a video recorder, a monitor, a gaming device, a laptop, an audio player, a DVD player, a CD player, a harddisk-based media player, an internet radio device, a public entertainment device, an MP3 player, a hi-fi system, a vehicle entertainment device, a car entertainment device, a medical communication system, a body- worn device, a speech communication device, a home cinema system, a home theater system, an audio server, an audio client, a flat television apparatus, an ambiance creation device, a subwoofer, and a music hall system.

However, although the system according to an embodiment of the invention primarily intends to improve the quality of sound or audio data, it is also possible to apply the system for a combination of audio data and visual data. For instance, an embodiment of the invention may be implemented in audiovisual applications like a video player or a home cinema system in which one or more speakers are used.

The aspects defined above and further aspects of the invention are apparent from the examples of embodiment to be described hereinafter and are explained with reference to these examples of embodiment.

BRIEF DESCRIPTION OF THE DRAWINGS

The invention will be described in more detail hereinafter with reference to examples of embodiment but to which the invention is not limited. Fig. 1 illustrates a device for processing an audio signal according to an exemplary embodiment of the invention.

Fig. 2 illustrates a Class D audio amplifier according to an exemplary embodiment of the invention.

Fig. 3 to Fig. 6 show diagrams illustrating signals of an audio processing device according to an exemplary embodiment in different dead time setting modes.

Fig. 7 illustrates a part of a device for processing audio signals according to an exemplary embodiment of the invention.

Fig. 8 and Fig. 9 show diagrams illustrating use of an audio data processing device according to an exemplary embodiment of the invention. Fig. 10 illustrates a Class D audio amplifier according to an exemplary embodiment of the invention.

DESCRIPTION OF EMBODIMENTS

The illustration in the drawing is schematically. In different drawings, similar or identical elements are provided with the same reference signs.

In the following, referring to Fig. 1, a device 100 for processing an audio signal 101 according to an exemplary embodiment of the invention will be explained schematically. An audio signal 101 to be amplified by an amplifier (a part of which is shown in Fig. 1) is supplied to an input of a power stage 102 of the amplifier. In the power stage

102, the input signal 101 may be processed so as to generate an output signal 103. This output signal 103, if desired after further processing, may be reproduced to generate acoustic waves using a loudspeaker or the like. However, a current measuring unit 104 measures the value of this output signal 103. The current measurement unit 104 may measure the current value of the output signal 103 and may supply the signal to an overcurrent detection unit 107. When the overcurrent detection unit 107 detects that the current value exceeds a predetermined threshold value, operation of the device 100 may be terminated. However, additionally, the current measurement device 104 provides the measured current signal to an adjustment unit 105 adapted for adjusting a dead time setting for the power stage 102 of the amplifier based on the current value of the output audio signal

103. For this purpose, a dead time setting signal 106 is supplied from the adjustment unit 105 to the power stage 102. In the adjustment unit 105, a current threshold value may be stored which may be compared to the present amplitude of the output signal 103. The adjustment unit 105 then sets the dead time to a small value when the amplitude exceeds the threshold value, and sets the dead time to a larger value when the amplitude of the output signal 103 is below the threshold value. Therefore, the adjustment unit 105 adjusts the dead time based on a trade-off between reducing idling losses (favouring a long dead time) and increasing the audio quality (favouring a short dead time).

The short dead time setting may be required for good THD performance, but it is, in fact, only or preferably applicable for modulation that makes the audio output current to exceed the ripple current through the filter. In the following, referring to Fig. 2, an amplifier 200 of a device for processing an audio signal according to an exemplary embodiment of the invention will be explained.

The audio input signal 101 (which can be an analog or a digital signal) can be supplied to a pulse width modulator (PWM) unit 201. The signal output by the PWM unit 201 is then supplied to a power buffer unit 102. The power buffer unit 102 comprises a first driver unit 202 and a second driver unit 203 connected in parallel to one another. The signal output by the PWM unit 201 is applied to an input of each of the drivers 202 and 203, which drivers 202 and 203 are circuited in parallel to one another. An output of the first driver 202 is coupled to a gate terminal of a first field effect transistor (MOSFET) 204 that comprises a body diode 205. In a similar manner, an output of the second driver 203 is coupled to a gate terminal of a second MOSFET 206 that comprises a body diode 207 as well.

A first source/drain terminal of the first transistor 204 is coupled to an upper supply potential Vdd. A first source/drain terminal of the second transistor 206 is coupled to a lower supply potential Vss. A second source/drain terminal of the first transistor 204 is coupled to a second source/drain terminal of the second transistor 206 and to an output coil 208.

As can be taken from Fig. 1, the driver units 202, 203 receive, at a respective control input thereof, a dead time setting signal 106 from a dead time adjustment unit 105. Outputs of the transistors 204, 205 are coupled to the output coil 208. An optional feedback loop 209 connects an output of the coil 208 with an adder unit 210 for adding the output signal to the input signal 101. Furthermore, a capacitor 211 is shown and a load 212 (a destination to which the power is dissipated) which is, in the present case, an ohmic load. The capacitor 211 and the coil 208 form a filter. Fig. 2 shows the load 212 as an ohmic load. However, this is only one possible configuration, and other embodiments of the invention may involve a non-ohmic load. More generally, the ohmic load 212 may be any (complex, that is real and/or imaginary) load "Z", and may have any kind of imaginary part added to the real ohmic impedance or substituting it. One exemplary aspect of the invention is to escape idling losses problems that one may face with Class D audio amplifiers (similar to the one shown in Fig. 2). The power stage 102 is operated with a limited dead time so as to avoid cross-conduction when switching from one state to the other. During the dead time, the output current is flowing through the respective body diode 205, 207 of the MOSFETs 204, 206. At low output current (with resistive loads 212, this equals low output modulation), the dead time has no or very limited influence on the fidelity of the power stage 102. A PWM pulse is appropriately buffered by the output stage (as will be explained below in more detail referring to Fig. 3 and Fig. 4). In this condition, the output current through the output coil 208 makes the transition occur automatically. Additionally, the energy stored in the parasitic capacitances across the drain/source connections of the MOSFETs 204, 206 is recycled during the transition.

Fig. 3 shows a diagram 300.

Along an abscissa 301 of the diagram 300, the time is plotted in ms. Along an ordinate 302 of the diagram 300, an amplitude of an output audio signal (LFout) is plotted. Fig. 3 indicates a first operation state 303 which relates to a condition that the output current I_out is smaller than the ripple current I_rippie. Fig. 3 further shows a second operation mode 304 in which the output current I_out is larger than the ripple current I_rippie. The operation mode 303 is illustrated in more detail in Fig. 4, and the operation mode 304 is illustrated in further detail in Fig. 5.

Fig. 4 shows a diagram 400.

The diagram 400 has an abscissa 401 along which the time is plotted. Along an ordinate 402 of the diagram 400, various signals are plotted. A first curve 403 illustrates a signal provided by the PWM unit 201. A second curve 404 illustrates a switching output. Still referring to Fig. 4, when I_out is smaller than I_rippi_e, the dead time set up has essentially no influence on the fidelity of the output stage that follows accurately the PWM drive.

The dead time Tdead time is indicated in Fig. 4 as well.

As can be taken from Fig. 4, the output stage follows very well the PWM drive, with the dead-time and the associated pn junction drop during dead-time being visible. At higher output currents (basically the condition when the LF output current exceeds the ripple current flowing through the filter at no modulation, no load), the dead time may add distortion. The PWM pulse at the output of the power stage 102 has not the same duty cycle as the PWM pulse that was input to the power stage 102. If the amplifier includes feedback built around the power stage 102, the distortion induced by the dead time may be reduced by an amount that may equal the loop gain of the system.

Now referring to Fig. 5, a diagram 500 will be explained which relates to an operation state 304 Iout>In_PPie.

Along an abscissa 501 of the diagram 500, the time is plotted. Along an ordinate 502 of the diagram 500, various signals are plotted. A first curve 503 relates to the PWM drive 201, and a second curve 504 relates to a switching output.

In the scenario of Fig. 5, the dead time setting may largely influence the fidelity of the output stage. Larger dead time may create larger output stage distortions. As can be taken from Fig. 5, the levels of the "high state" at the output stage output before and after the short "low" time have essentially the same amplitude. Also, the two pn junction drops during dead time (the very short blocks at the beginning and end of the "low" time) may have essentially same duration and same amplitude.

Fig. 6 shows a diagram 600 illustrating operation states I_Out=Iπ_PPie. The diagram 600 relates to a large dead time. Referring to the schematic illustration in Fig. 6, assuming to amplify a sine wave, the distortion induced by dead-time shown looks like a DC voltage that is subtracted/added (positive/negative part of the sine wave) to the undistorted sine wave, for any output current higher than the ripple current through the filter. The "cross-over" region may therefore be a distortion free sine wave. The regions beyond the distortion free zones may also be parts of a sine wave, but which amplitude in absolute value may be modified by a "small" DC value.

A conclusion of the above-mentioned considerations is that, for audio quality reasons, one would like to operate the power stage 102 with the smallest possible dead time, whereas, for idling loss reasons, it would be preferable to operate the power stage 102 with a large dead time, so to guarantee that the transitions at the switching output occur automatically, without having to inject energy to (dis)charge the parasitic capacitances in the power stage 102.

An aspect on which embodiments of the invention are based is to make the dead time setting dependent on the audio current flowing though the output coil 208. At low output current, the amplifier is operated with a large dead time, producing limited dissipation and without suffering from the dead time induced distortion. At higher output currents, the amplifier is operated with a short dead time, solving the dead time induced distortion. An obtainable advantage is that the amplifier does not run hot when idling and still has a low THD.

Because the overload protection of the amplifier is also based on the current measurements of the current flowing to the load 212, the additional circuitry required to implement the proposed architecture is simple.

Embodiments of the invention may solve the trade-off issue between low idling losses (Class D is expected to run cool and basically dissipates more than most linear amplifiers at no load) and low THD. This may be achieved by an output current varying dead time setting.

Dead time may be created by delaying turn-on pulses to a power switch with regard to the turn-off pulse delivered to the opposite power switch. Typically, the dead time is set in practical implementation by a small signal current source.

An idea is to use the output current information delivered by the overcurrent detection/protection circuit as to trigger a value change of the dead time setting current. Of course, the trigger output current that will be used for the feature is in many cases well below the threshold current of the overcurrent protection, requiring a second trigger circuit but making possible the re-use of the current measurement circuit. Practically, for the current measurement circuit, sense resistors may be used.

However, a more preferred system is shown in Fig. 7 and may comprise a secondary winding 701 wound on an output coil 208 core 702 followed by an integrator 705. The voltage output by the integrator 705 may be a measure of the current flowing through the primary coil 704 of the output coil 208.

A signal 703 coming from the power stage 102 is sent to the output coil 208, and from there to the load 212. Apart from this, a signal present at the secondary winding 701 is supplied to the integrator 705 formed by an opamp, resistors, and capacitors. Along an abscissa 801 of a diagram 800 shown in Fig. 8 and illustrating idling, the time is plotted. Along an ordinate 802, a current is plotted. The ripple current I_rippie is shown for a frequency of approximately 300 kHz.

In the diagram 900 shown in Fig. 9 and illustrating audio modulation, operation points 901 are shown at which it is necessary or advantageous to adjust/switch between different dead time settings. Therefore, Fig. 9 shows an audio modulation signal and different points of time 901 at which the dead time setting mode is modified, namely when passing threshold values Ithreshoid-

Referring to the schematic illustrations in Fig. 8 and Fig. 9, the HF current bounded by the "Iripple" amplitude is a triangular current, which frequency is the PWM repetition rate (for instance 300kHz).

It may be advantageous to operate the amplifier with two discrete dead time values:

A large dead time setting will be used at no load and for output currents that do not exceed a peak value of the ripple current flowing through the output filter. A short dead time setting may be used for output currents that do exceed the peak value of the ripple current flowing through the output filter.

Optionally, the switch to the short dead time setting may be associated to a time constant as to keep the power stage in the "short dead time setting" for a longer time once it was triggered. A main purpose of this option would be to operate the amplifier at a short dead time setting at any output modulation, disregarding whether the instantaneous output current exceeds the filter ripple current but still keeping the idle losses low.

Fig. 10 illustrates a portion of a Class D audio amplifier 1000 according to an other exemplary embodiment of the invention. Many of the components of the apparatus 1000 equal to the corresponding components of the apparatus 200.

As can taken from Fig. 10, the bridge of Fig. 2 can be mirrored to drive the load 212 symmetrically. In that case, the load 212 is driven differentially, with two bridges, like sketched in Fig. 10. Two PWMs are provided which may be configured complementary, effectively doubling the voltage across the load 212 as compared to the "half bridge" topology of Fig. 2.

Other modulation schemes are possible, however, which produce the same or a similar LF result (doubling the voltage) but demonstrating different HF content (PWM repetition rate and multiples).

Of course, the same idea as for the half bridge topology applies for the two bridge topology: measuring the current flowing through the filter is a proper way to proceed to control the variable dead-time.

It should be noted that the term "comprising" does not exclude other elements or features and the "a" or "an" does not exclude a plurality. Also elements described in association with different embodiments may be combined.

It should also be noted that reference signs in the claims shall not be construed as limiting the scope of the claims.

Claims

CLAIMS:

1. A device (100, 200) for processing an audio signal (101), wherein the device (100, 200) comprises an adjustment unit (105) adapted for adjusting a dead time setting for a power stage (102) of an amplifier based on an amplitude of the audio signal (103).

2. The device (100, 200) according to claim 1, wherein the adjustment unit (105) is adapted for adjusting the dead time setting for the power stage (102) of the amplifier based on a voltage or based on a current of the audio signal (103).

3. The device (100, 200) according to claim 1, wherein the adjustment unit (105) is adapted for adjusting the dead time setting for the power stage (102) of the amplifier based on the amplitude of the audio signal (103) at an output of the amplifier.

4. The device (200) according to claim 1, wherein the adjustment unit (105) is adapted for adjusting the dead time setting for one or more driver units (202, 203) of the power stage (102) of the amplifier based on a current of the audio signal (103) flowing through an output coil (208) of the amplifier.

5. The device (100, 200) according to claim 1, wherein the adjustment unit (105) is adapted for adjusting the dead time setting to a predetermined first dead time value when the amplitude is above a threshold value, and for adjusting the dead time setting to a predetermined second dead time value when the amplitude is below the threshold value.

6. The device (100, 200) according to claim 5, wherein the first dead time value is smaller than the second dead time value.

7. The device (200) according to claim 5, wherein the first dead time value is less than essentially 30 ns, particularly is in the range between essentially 1 ns and essentially 20 ns.

8. The device (200) according to claim 5, wherein the second dead time value is more than essentially 50 ns, particularly is in the range between essentially 80 ns and essentially 200 ns.

9. The device (200) according to claim 5, wherein the threshold value equals to a peak value of a ripple current flowing through an output filter (208) of the amplifier.

10. The device (100, 200) according to claim 1, wherein the adjustment unit (105) is adapted for adjusting the dead time setting based on a trade-off between a reduction of idling losses, and an increase of an audio quality.

11. The device (100, 200) according to claim 1, comprising an overcurrent detection circuit adapted for sensing the amplitude for detecting, based on the sensed amplitude, when an overcurrent flows through the device (100, 200), and for supplying the sensed amplitude also to the adjustment unit (105).

12. The device (100, 200) according to claim 11, wherein the overcurrent detection circuit is adapted for sensing the amplitude using one or more sense resistors.

13. The device (100, 200) according to claim 11, wherein the overcurrent detection circuit is adapted for sensing the amplitude using a secondary winding (701) of an output coil (208) of the amplifier and using an integrator circuit (705) connected to the secondary winding (701) of the output coil (208).

14. The device (100, 200) according to claim 1, wherein the adjustment unit (105) is adapted to, when switching from a second dead time value to a first dead time value, maintain the device (100, 200) in an operation state defined by the first dead time value at least for a predetermined time interval regardless of the amplitude during the predetermined time interval.

15. The device (100, 200) according to claim 14, wherein the first dead time value is smaller than the second dead time value.

16. The device (100, 200) according to claim 1, wherein the amplifier is a Class D amplifier.

17. The device (100, 200) according to claim 1, realized as at least one of the group consisting of an audio amplifier, a Class D audio amplifier (200), a loudspeaker, an audio surround system, a mobile phone, a headset, a loudspeaker, a hearing aid, a handsfree system, a television device, a video recorder, a monitor, a gaming device, a laptop, an audio player, a DVD player, a CD player, a harddisk- based media player, an internet radio device, a public entertainment device, an MP3 player, a hi-fi system, a vehicle entertainment device, a car entertainment device, a medical communication system, a body-worn device, a speech communication device, a home cinema system, a home theater system, an audio server, an audio client, a flat television apparatus, an ambiance creation device, a subwoofer, and a music hall system.

18. A method of processing an audio signal (101), wherein the method comprises adjusting a dead time setting for a power stage (102) of an amplifier based on an amplitude of the audio signal (103).

19. A computer-readable medium, in which a computer program of processing an audio signal (101) is stored, which computer program, when being executed by a processor (105), is adapted to carry out or control a method according to claim 18.

20. A program element of processing an audio signal (101), which program element, when being executed by a processor (105), is adapted to carry out or control a method according to claim 18.