|
Flip the switch to turn on the circuit.
The class AB lifts and drops the zero volt reference point (usually referred to as the ground) of the low-voltage class A to let it achieve high voltages while only consuming the power of the low-voltage 5V sources. The extra power to lift/drop the zero volt reference point is provided by the more efficient class AB. This way, it can achieve the fidelity of the class-A while letting the class-AB do most of the heavy lifting. This is what makes this a class-H.
The class A uses about 100W or so regardless of load, and the class AB uses from 10W to 426W depending on the load. At maximum load, about 290W is delivered for an efficiency of 290/(100+426) = 290/526 = 55% which is better than the maximum theoretical efficiency of a pure class A amp, which depending on the design is either 25% efficiency, which corresponds to a constant current source plus a fat output transistor that pulls no current to let the current source push current or pulls a ton of current to overcome the current source and still pull more current after that, or 50% efficiency, which corresponds to a more intelligent design with two fat output transistors with the sum of the current going through them always approximately equal to the maximum output current of the amp.
P.S. This has a THD of 0.000004% (2μV / 50V) in the sim. This is in part due to the class A being the main output stage thereby preventing crossover distortion (except for that tiny bit which leaks through from the class AB and is responsible for most of the THD I measured) and in part from the strong negative feedback from the discrete op-amp which greatly decreases THD from all sources.
|
