Sigma-delta based class D audio power amplifier with high power efficiency

A method of driving an analog load with a digital signal, the method comprising:

• receiving a digital input signal;
• converting the digital input signal to a plurality of noise shaped output pulses;
• processing at least a portion of the plurality of noise-shaped output pulses to ensure that each of the noise shaped output pulses in the portion contains an amount of energy that is as close as possible to the amount of energy in the other noise-shaped output pulses;
• obtaining a sample of energy delivered by at least a portion of the processed noise-shaped output pulses;
• generating an error signal based on the difference between the sampled processed noise-shaped output pulses;
• and
• using the error signal during the conversion of the digital input signal to a plurality of noise-shaped output pulses.

A system for driving an analog load with a digital signal, the system comprising:

• means for receiving a digital input signal;
• means for converting the digital input signal to a plurality of noise shaped output pulses;
• means for processing at least a portion of the plurality of noise-shaped output pulses to ensure that each of the noise shaped output pulses in the portion contains an amount of energy that is as close as possible to the amount of energy in the other noise-shaped output pulses;
• means for obtaining a sample of at least a portion of the processed noise-shaped output pulses;
• means for generating an error signal based on the difference between the sampled processed noise-shaped output pulses;
• and
• means for using the error signal during the conversion of the digital input signal to a plurality of noise-shaped output pulses.

An article of manufacture, comprising:

• a computer useable medium and computer readable code embodied on the computer useable medium for driving an analog load with a digital signal, the computer readable code comprising:
• computer readable program code devices configured to receive a digital input signal;
• computer readable program code devices configured to convert the digital input signal to a plurality of noise shaped output pulses;
• computer readable program code devices configured to process at least a portion of the plurality of noise-shaped output pulses to ensure that each of the noise shaped output pulses in the portion contains an amount of energy that is as close as possible to the amount of energy in the other noise-shaped output pulses;
• computer readable program code devices configured to obtain a sample of at least a portion of the processed noise-shaped output pulses;
• computer readable program code devices configured to generate an error signal based on the difference between the sampled processed noise-shaped output pulses;
• and
• computer readable program code devices configured to use the error signal during the conversion of the digital input signal to a plurality of noise-shaped output pulses.

A signal path, comprising:

• a sigma delta modulator (SDM) receiving a digital input signal and producing a plurality of noise-shaped output pulses based upon the digital input and an error signal;
• a pulse processing circuit, the pulse processing circuit processing at least a portion of the plurality of noise-shaped output pulses to ensure that each of the noise-shaped output pulses in the portion contains an amount of energy that is as close as possible to the amount of energy in the other noise-shaped output pulses;
• an output stage coupled to the pulse processing circuit, the output stage having a first state wherein the output stage is capable of providing analog noise-shaped output pulses to a load and a second state where the output stage is delivers essentially zero output energy to the load;
• and
• a feedback loop coupled between an output of the output stage and an input of the SDM, the feedback loop sampling at least a portion the analog noise-shaped output pulses provided by the output stage when the output stage is in the second state, and generating an error signal based on the difference between the sampled portion of the analog noise-shaped output pulses and the input digital signal to the amplifier.