This is my idea but I don't have the money to develop it so I declare it open source. May not work any different to a normal amp my ears are too old to tell..I may buy a broken one in 10 years for next to nothing..So here goes..
You have to use 3 class b outputs phase shifted connected with parallel with resistors to a speaker.
The theory is you have backup current when the inductance of the speaker wants it..
I have done some 3 phase tests with a mp3 running 3 x 4 bit's D/A (not convincing quality too low) and 3 phase simulations with circuit maker 2000 student edition. The download link if you want to play..
http://my.ece.ucsb.edu/York/Bobsclass/2C/simulation/cmstudnt.zip
The circuit
View attachment 11492
I have attached one circuit, you can play with values to your harts content.
The values in the circuit are as close to real world as I can find if I am wrong let me know..
Plus
You can eliminate class B crossover distortion at low levels.
It seems you can push power into the inductive load of a speaker as well.
Is not consuming 50% of power supply at Idle Like class A
Negatives:
The Amps efficiency ranges from 16% to 66 % depending on how you torture with the phase shift, the power supply and output devices do suffer.
Audio DSP failure or short circuit output = blow 6 output devices instead of 2... I don't care I will just buy another faulty receiver off the internet as long as 3 channels work.
Have no idea: need help!
How low the parallel resistors can go before Amps negative feedback blows outputs with respect to phase shift.
I have some 2 phase recordings but there is a copyright problem if I share but still have to do sine wave tests yet..
Thanks in advance for any comments