Universal truths cannot be disputed - an object at rest stays at rest, energy cannot be created nor destroyed, and a Class A amplifier requires a good heatsink. Okay, so that last one isn’t so much a universal truth, but it certainly will have felt true if you’ve worked with a simple Class-A amplifier delivering decent power. A question was recently posed that went something along the lines of “Why does the Class A get so hot when it has no signal?” Ultimately, this is answered in one key word: bias.
Bias is the idle rate that the transistor sits at while not driving an output. This means that there will always be current going through the transistor even when there is no signal being applied. The power, having nowhere to go, gets dissipated as heat. The higher the potential signal you need to drive ( aka the voltage on the rails), the more power is dissipated at bias. The question then becomes “why does there need to be so much power dissipated at idle? In fact, why do the transistors need to be on at all if there is no signal applied?”
Let’s relate bias to the engine in a vehicle. The term “idle” rate becomes quite relevant here, as the transistor’s bias can be viewed akin to the engine’s idle level. While the vehicle is on but not moving, the engine is still running at a designated RPM. No Miles Per Hour are being delivered, but power is still being dissipated in the engine. The same thing happens in a Class-A amplifier; the bias is set so it will be ready to swing either direction as soon as a signal is applied. Imagine every time you approach a stop light when the vehicle stops moving the engine shuts off. Sure, this would be much more efficient, however when you needed to go again there would be a lag between pressing the accelerator pedal and actually moving (the time for the engine to start up). These do exist in the amplifier world, though, too; they’re called Class-B amplifiers and come with their own challenges.
Here’s where the engine in a vehicle analogy starts to deviate. The engine that you’re most likely used to would be most akin to a Class-AB amplifier, where it still idles, but at a much more efficient rate (so an engine would run at 500RPM when not moving vs 2500RPM when not moving). So why do Class-A amplifiers still exist if the efficiency is so poor? Ultimately, it comes down to fidelity of the signal. If the bias is directly in the middle of the transistor’s current saturation and voltage cut-off points, then it is immediately ready for most frequencies. Is that fidelity worth up over 95% of the power being dissipated as heat? Well, that depends on who you ask!