|
This time i won't be presenting specifications and stuff, i can't even begin to measure the parameters of the amp, it's so temperamental i barely got it to stabilise, and still it gives no solution after a while. This is by far the hardest circuit to stabilise i ever encountered. EC is clearly having a very hard time to determine so many variables of the circuit it's going crazy. Just don't try to change anything, it took me half an hour after i finished the original circuit, to stabilise it even to this point, which is also unstable. I'll just point out the major parts of the circuit. The most left part of the circuit is an op amp working as a comparator. It compares the input signal (most left red) with the sampling wave, which is the second major part of the circuit. That is the circuitry after the 3.3nF capacitor, a sawtooth generator. Usually its frequency is much higher than the signal frequency. The higher the better, for less harmonics on the output. Some designs use sampling frequency of up to 50MHz. The third part is the MOSFET switches, which deliver the PWM with high current capacity to the final part, the low pass LC filter. It extracts the audio signal from the digital signal (the PWM), thus restoring the original wave, and feeding it to an 8ohm speaker. Usually more complex bridges are used, like H-bridges, but for the lack of space and stability I only managed to put a single ended stage.
|
