How Do Guitar amps work? |Transistors

Transistor power amps Though transistors, like valves, are essentially three-terminal current controllers, with their equivalent electrodes being collector (anode), base (grid), and emitter (cathode), they work at the other end of the voltage/current range, and are vulnerable to high running temperatures – so most power stages operate in virtual class B, where one push-pull output device is cut off over almost the entire driven half-cycle of the other.
Bias current in transistor amps is in milliamps, a small fraction of their full-drive current, and designers need to ensure that cut-off doesn’t occur too soon, causing obtrusive zero-crossing distortion at low levels. This is eased by the more linear behaviour of transistors: unlike valves, there’s no decrease in gain at higher drive levels – their current gain (collector current/base current ratio) is practically constant over their operating range.
However, this is where guitarists begin to get fidgety. Valves gradually ease into saturation-based clipping with a rich crop of lower harmonics; transistors can be driven hard into the brick wall of the power supply, and the resulting squared-off waveform (Fig. 5a) shows the abrupt increase in 3rd, 5th and higher odd harmonics (Fig. 5b). There’s no output transformer, so the output remains at full bandwidth over the whole power range, removing the 'focus' effect of an overdriven valve amp.
Attempts have been made to alter this behaviour by applying soft-clip conditioning to the drive signal, or by non-linear feedback loops, dynamic compression, and other methods, but so far none of these have addressed its real causes, or persuaded the majority of us to give up our expensive valve habits. Bassists are happy to use high-powered solid-state amps, but low distortion is more of a blessing here than a curse, and the size and weight savings for a given output is significant at the power levels in use nowadays.