WO2016110809A2

WO2016110809A2 - Digital audio amplifying system

Info

Publication number: WO2016110809A2
Application number: PCT/IB2016/050050
Authority: WO
Inventors: Andrus Aaslaid
Original assignee: VASSILKOV, Alfred
Priority date: 2015-01-06
Filing date: 2016-01-06
Publication date: 2016-07-14
Also published as: WO2016110809A3

Abstract

The digital audio amplification system described herein uses a pulse density modulation (PDM) encoded 1-bit digital audio signal as the input signal, which is, for example, encoded based on the DSD (Direct Stream Digital) standard or another analogue method, and reproduces it for an acoustic system in an analogue shape, using low-pass filtering, voltage and/or current amplification of digital signals, and, if necessary, conversion of the frequency and/or specification of the time offset of the input signal. The audio amplification system includes a DSD/PDM signal receiving system and a circuit for correcting the time offset of that signal, which can also work as a frequency converter. The achieved PDM signal is directed to a switching circuit, which copies the shape of the input signal with a bigger output current and a voltage amplitude selected by the user. The signal exiting the switch is suitable as an input signal for acoustic systems (speaker, headphones) after low-pass filtering, and the mechanical or electrical parts of the acoustic system itself can serve as low-pass filters. The audio amplification system may be an autonomous component or form an integrated whole with an acoustic system (speaker).

Description

DIGITAL AUDIO AMPLIFICATION SYSTEM

TECHNICAL FIELD

The invention belongs to the field of audio reproduction systems. More specifically, the invention describes a digital amplification system that can form an integrated whole with an acoustic system, i.e. a speaker.

BACKGROUND ART

Amplifiers that are designed to amplify audio signals can broadly be classified into the following categories: In the case of class A architecture, the amplifying element is constantly conducting electricity. The idle operating point is chosen such that the amplifying element would still operate in a region as linear as possible at the maximum value of the input signal. Although great linearity and, owing thereto, less distortion are achieved in such a case, significant energy loss is a disadvantage of class A since a current is constantly passing through the amplifying element.

In the case of class B amplifiers, two amplifying elements that operate in a push-pull topology are used. Each element only controls signals during the respective half period. Compared to a class A amplifier, the efficiency achieved thus is considerably higher, but the imprecision of the operation of the amplifying elements in the case of small input signals is a disadvantage (the so-called zero crossing distortion).

Currently, the architecture of class AB amplifiers is the most widely used. In this case, transistors that have a circuit similar to that of class B are open to some extent according to the selected bias voltage also when idle (in the case of an opposite-polarity signal). This creates a situation where there is no distortion in the case of small signals. Its efficiency is slightly lower compared to a class B amplifier, but is still close to 50%. The slight non- linearity of the amplifying elements is a disadvantage since no ideal operating points that are chosen in the case of class A are used here. Amplifiers of classes G and H use the output stage architecture of class AB amplifiers but alter the supply voltage applied on the amplifying elements according to the level of the output signal. This greatly increases the efficiency of the output stage since the volume of wasted power passing through the amplifying elements is reduced. Class D audio amplifiers are one of the most efficient audio frequency amplifiers in use today. By using only the completely open or completely closed state of the amplifying element and modulating the states according to the amplitude of the reproduced signal, the level of losses achieved is extremely low since the amplifying element essentially works as a switch. A typical class D amplifier is depicted in Fig 1. The amplifier consists of a modulating stage 1 that converts the input signal into a stream of digital pulses with pulse width modulation. The final stage 2 is steered by the stream of pulses, where the active elements are only in either a completely open or closed state, i.e. they work as a switch. The signal of the final stage is directed into a low-pass filter 3, which reconstructs an analogue signal from the stream of pulses that is in turn reproduced by an acoustic element 4.

A disadvantage of class D amplifiers is losses that emerge if a signal in the value domain (amplitude domain), such as an analogue signal or a PCM (pulse-coded modulation), is converted into a signal in the time domain that is necessary to control a class D output circuit, such as a PWM (pulse width modulation) or PDM (pulse density modulation) signal. In the case of an analogue signal, the precision of the conversion circuit is a limiting factor; in the case of a PCM signal, the selection of filtering algorithms, bit depth limitations, and the computing power are limiting factors since the necessary conversion must be done in real time.

Signals that are necessary to steer the amplifying element are formed in the amplifier from an input signal, which predicates that the input signal is transferred from the value domain to the time domain and that an additional conversion according to the output level is performed. A disadvantage of a conversion circuit of this kind is that the conversion precision between the domains is poor and losses result therefrom.

From a practical viewpoint, lossless conversion between the time and amplitude domains is extremely complex and expensive, which is why numerous compromises have to be made in class D amplifiers upon conversion. Today, class D amplifiers predominantly transmute PCM signals into analogue signals first to transform them, then the analogue signal is used to form a necessary pilot signal that works in the time domain. The side effects of such repeated reconversion inevitably alter the audio image being reproduced, and the sound of class D amplifiers is often considered dry and analytical. Class T amplifiers are a solution developed by the company Tripath. Class T amplifiers use a control signal with a higher than usual frequency of up to 50 MHz to control the switches of the output stage of class D.

In the case of the S-Master architecture, Sony upgraded the technology of class D in the first half of the 2000s. An example of a typical amplifier employing the S-Master architecture is provided in Fig. 2. The timing of a digital signal in the PDM format is first improved with the Clean Data Cycle circuit 1. The signal is then converted into a C-PLM (Complementary Pulse Length Modulation) signal by the use of a conversion circuit 2, and the timing of the signal is then further improved by an S-TACT (Synchronous Time Accuracy Controller) circuit 3. The achieved PWM signal is then directed to the final stage 4, where the voltage and current are amplified. In accordance with the chosen audio volume level, the output voltage of the power source of the switches 5 is changed, and the level of the output signal is thereby altered. The stream of pulses achieved is then directed through a low-pass filter 6, where the pulse width modulation signal is transformed into an analogue signal, which is reproduced by an acoustic element 7. DSD (Direct Stream Digital) is the trademark of a method of digitally encoding an audio signal adopted by Sony when the Super Audio CD (SACD) was introduced to the market. It is a PDM (pulse density modulation) encoding method in which case 1-bit encoding in a time domain with a high clock frequency (2 MHz and above) is used. Unlike the PCM (pulse- coded modulation) technology widely used in recording audio signals, where the value of the signal in the amplitude domain is stored at fixed intervals (at the sample frequency of 44.2 kHz, at 16-bit precision in the case of CDs), only the up or down movements of the signal are registered in the case of a PDM modulation, but it is done with a frequency that is 64-512 times higher than the classic 44.2 kHz PCM sample frequency. The implementation of a frequency as high as this makes it possible to reconstruct the original shape of analogue signals in a very precise manner, using a simple low-pass filter for that purpose. SUMMARY OF THE INVENTION

The aim of the invention is to provide a class D type digital audio amplification system provisionally called class DS, which is free from both the side effects created when the input signal is converted to a stream of pulses necessary for controlling the final stage and from drops in the resolution that occur when the level of the output signal is altered.

The invention presents an amplifier architecture, where a PDM signal according to the DSD or another standard is used in its original shape, and which is free from the deficiencies of inter-domain conversion. Owing thereto, sound reproduction that is essentially lossless is achievable by the reproduction path. The DSD signal format, which was first taken into use by the music industry for SACDs (Super Audio CD) and is now also gaining popularity among digitally downloadable audio formats, is essentially suitable for steering the final stage of class D in an unchanged form, which has been specially adjusted thereto, enabling the reproduction of the signal without conversion between the scalar and time domains as well as the analogue and digital domains. The invention describes changing the level of reproduction, which is carried out in the analogue domain by altering the supply voltage of the final stage, for what reason the resolution of the reproducing signal is not dependent on the output flow of reproduction since the whole audio signal information is located in the time domain, which is not changed.

LIST OF THE DRAWINGS Fig. 1 depicts the architecture of a classic class D amplifier.

Fig. 2 depicts a class D amplifier upgraded based on the S-Master technology by Sony.

Fig. 3 explains the technical nature of the invention and depicts a class DS amplifier according to the first example embodiment.

Fig. 4 explains the technical nature of the invention and depicts a class DS amplifier according to the second example embodiment.

Fig. 5 explains the technical nature of the invention and depicts a class DS amplifier according to the third example embodiment. INVENTION' S EMBODIMENT EXAMPLE 1

The embodiment example depicted in Fig. 3 is a single-ended full digital class DS amplifier with a USB connection.

A receiver 2 connected to a computer by means of a USB connection changes a digital audio signal sent in a PDM encoding (e.g. in the DSD encoding) into a bit sequence at the base frequency, the left and right channels separately.

The bit sequence is directed to a single-ended digital switching stage 3, where the current and voltage of the signal are amplified. The switching stage is followed by a low-pass filter 4, in which the bit sequence with a fast frequency is modified into an analogue signal as a result of filtration. Since the scheme is single-ended, in which the level of a digital signal is alternately zero and the maximum supply voltage, the analogue signal created by filtration is directed through a capacitor 5 that insulates direct current. The created signal is directed to an acoustic element 6.

Upgrades to the given example embodiment may include but are not limited to the following solutions and nodes:

- A digital buffer bridge circuit.

- Alteration of the supply voltage of the digital buffer in accordance with the desired audio volume level.

- Improvement of the timing of signals in the DSD/PDM encoding or enhancement of the bit rate before direction into the digital buffer.

- A snubber circuit based on an analysis of the bit sequence and/or output signal by a microprocessor or analogue circuit to identify deviations from the normal signal and to avoid any noise resulting from signal errors and the emergence of a direct current in the output of the switching circuit. INVENTION' S EMBODIMENT EXAMPLE 2

The example embodiment depicted in Fig. 4 is based on the reproduction system of a double- ended audio signal, where the amplifying modules are insulated from an USB connection unit by an optical connection. A receiver 2 connected to a computer by means of a USB connection changes a digital audio signal sent in the DSD encoding into a bit sequence at the base frequency of the DSD format, the left and right channels separately. The description only provides a single receiving channel, but the number of channels is not limited in the actual system.

A bit sequence is used to modulate an optical transmitter 3, which transforms the signal in a DSD-encoded audio file into light impulses and sends them to an optical receiver 5 through an optical fibre cable, where they are re-encoded into the initial bit sequence.

The bit sequence received by the optical receiver 5 is directed to a phase shifter 6, which creates pulse series that are in the phase opposite to the bit sequence. Digital switching stages 7, 8 are steered by the pulse series in the opposite phase, which form a zero potential (G D) and a bridge connection switching between the supply voltage created by a power supply 9, the voltage level of which is adjusted according to the desired audio volume. The supply voltage is directed to the switching stages by a snubber circuit 11 controlled by an analytical circuit 10. The pulse series is directed to a low-pass filter 13 through a smoothing filter 12, where a low frequency audio signal is formed from a high frequency bit sequence as a result of the filtering process, which is reproduced by an acoustic element 14 (speaker). In accordance with the supply voltage, the amplitude domain of the pulses is altered, by which the level of the output signal of the audio signal is changed.

- Improvement of the timing of signals in the DSD/PDM encoding or enhancement of the bit rate.

- Transportation of the signal to the switches (7, 8) by means of any other method, such as: a computer module; a computer module connected to a computer network by any method.

- Inter-synchronisation of the bit sequence in the case of several reproduction modules with the same or different location by employing any method, such as: changing the time signals between the modules using any method, like: • light impulses;

• radio frequency impulses;

- Synchronisation via a computer network;

- Synchronisation via any external signal source, such as:

• a GPS signal

- The addition of an external module which processes signals that are in any other format (PCM, analogue, etc.) into a DSD/PDM signal.

- Receiving a signal in any other manner than by means of a computer, for example:

• from a communications network;

• by Wi-Fi from other devices.

INVENTION' S EMBODIMENT EXAMPLE 3

The embodiment example depicted in Fig. 5 is a double-ended reproduction system of audio signals, where digital audio material in the DSD format located on a computer network or the Internet is used as the input signal of the amplifying modules and which is transported via a wireless computer network. The amplifying modules are integrated with a speaker system and the phase synchronisation between the modules is performed via an optical coupling network that is duplicated with a radio frequency synchronisation circuit. The amplifying system is controlled via a tablet, mobile telephone or other computing device. The example embodiment only provides a single receiving channel, but the number of channels is not limited in the actual system.

An audio file in the DSD format located on a server 1 is transported via a wireless computer network 2 to a computer module 3. The computer modules are inter-synchronised via a synchronisation circuit 4 that mediates phase and time information via a radio frequency module 5 and/or an optical module 6.

The computer module 3 separates the bit sequence of an audio channel from the DSD information, which is timed with other computer modules. From the computer module 3, the bit sequence is directed to a precision circuit 7 for timing the signal. Once the timing of the signal is adjusted, it is directed to a phase shifter 8, which creates pulse sequences in the opposite phase. Digital switches 9, 10 are steered by the pulse sequences in the opposite phase, which form a zero potential (GND) and a bridge connection switching between the supply voltage created by a power supply 1 1, the voltage level of which is adjusted according to the desired audio volume. The supply voltage is directed to the switches by a snubber circuit 13 controlled by an analytical circuit 12. The pulse series with its voltage and current amplified is directed to a low-pass filter 15 through a smoothing filter 14, where a low frequency audio signal is formed from a high frequency bit sequence as a result of the filtering process, which is reproduced by an acoustic element 16 (speaker). In accordance with the supply voltage, the amplitude domain of the pulses is altered, by which the level of the output signal of the audio signal is changed. Audio material is selected and the system controlled by a tablet computer 17.

- Enhancement of the bit rate of signals in the DSD/PDM encoding.

- Addition of an external module, which processes signals in any other format (PCM, analogue, etc.) into a DSD/PDM signal and transports them via a wireless computer network or by other means to the amplifying module.

- Transport of audio material to the amplifying module by other means than a wireless computer network.

- The timing is corrected after the phase shifter, in the close proximity of a digital switching circuit.

Claims

1. A digital audio amplification system for controlling a speaker system, which comprises a PDM signal receiving circuit, a switching stage and an electrical filter, characterised in that a PDM signal received via the receiving circuit is, in its unprocessed form, used for controlling the switching stage.

2. A digital audio amplification system according to claim 1 , characterised in that the PDM signal is a DSD-encoded signal.

3. A digital audio amplification system according to claims 1 to 2, characterised in that the electrical filter is a part of the speaker system.

4. A digital audio amplification system according to claims 1 to 2, characterised in that the audio amplification system is integrated with the speaker system.

5. A digital audio amplification system according to claims 1 to 4, characterised in that an electrical filter is formed through the electroacoustic and mechanical properties of the speaker system.

6. A digital audio amplification system according to claims 1 to 5, characterised in that the timing of single bits of the digital input signal has been corrected.

7. A digital audio amplification system according to claims 1 to 6, characterised in that the bit rate of the digital input signal has been altered.

8. A digital audio amplification system according to claims 1 to 7, characterised in that it includes a receiving circuit for digital audio signals that are not in the PDM format and a conversion circuit for converting digital audio signals that are not in the PDM format into a PDM signal for steering the switching stage, without converting it into an analogue signal beforehand.

9. A digital audio amplification system according to claims 1 to 7, characterised in that it contains a receiving circuit for analogue audio signals, which is connected to a conversion circuit used to convert analogue audio signals into a PDM signal that is necessary for steering the switching stage.

10. A digital audio amplification system according to claims 1 to 7, characterised in that it contains a receiving circuit for receiving audio input signals in a digital format via a data communication channel.

1 1. A digital audio amplification system according to claims 1 to 10, characterised in that the supply voltage circuit of the switching stage comprises a tuned filter.

12. A digital audio amplification system according to claims 1 to 11, characterised in that the supply voltage is controllable to change the level of analogue signals reproduced in the acoustic system.

13. A digital audio amplification system according to claims 1 to 12, characterised in that separately located amplification systems are interconnected by means of a low latency communication channel to improve the timing of the bits of the modulation signal of such amplification systems.

14. A digital audio amplification system according to claim 13, characterised in that the low latency communication channel is an optical fibre cable, radio communication or optical communication is.