The technology behind the beautiful devices that allow us to enjoy music can be quite mysterious for the audiophile/music lover. The numerous technical terms used by manufacturers to describe the circuits in their creations often remain vague concepts for the uninitiated. Uncertainty is the enemy of informed choices, so this month we begin a series aimed at demystifying the concept of the class of operation (A, B, AB, C, D, etc.) of power amplifiers. This topic is gaining particular importance with the current influx of class D amplifiers across all price levels on the market.
– Sketches Credit: Normand Daigle
To start, it would be helpful to clarify the terminology and some concepts we will encounter along the way. The primary goal of active electronic components (tubes or transistors of various families) in an analog audio circuit is to provide signal amplitude gain, either in voltage (symbol V, unit = Volts or V) or current (symbol I, unit = Amperes or A), and in some cases both. This amplification allows a source as weak as a few millivolts in some cases to reach nearly a hundred volts for a high-power amplifier. Added to these notions is that of impedance (Z, in ohms or Ω), which in audio will transition from high at the start of the chain to low at the interface with the loudspeakers.
The circuits we will examine will be composed of different types of transistors, namely bipolar (BJT for Bipolar Junction Transistor), field-effect (FET for Field Effect Transistor), or MOSFET (Metal Oxide FET). These are three-lead components, with one lead controlling the current flowing from the second to the third. For BJTs, these leads are called the Base (B), Collector (C), and Emitter (E). For FETs and MOSFETs, the equivalents are Gate (G), Drain (D), and Source (S). Each type of transistor is available in different and opposite polarities, NPN or PNP for BJTs, and N or P for FETs and MOSFETs, allowing for the design of symmetrical and complementary circuits that offer better linearity. Each type has its advantages and limitations that must be considered in circuit design. They also come in low and high-power versions.
Types of Transistors
In addition to active components, there will be passive elements such as resistors (R, quantified in Ohms or Ω) and capacitors (C, quantified in Farads or F). The symbols for these components are sometimes preceded by a prefix indicating a division or multiplication factor of the value, such as “p” (pico, or 10 to the power of -12, e.g., pF), “n” (nano, or 10 to the power of -9, nF), “u” (micro, or 10 to the power of -6, uF), “m” (milli, or 10 to the power of -3, mΩ), “k” (kilo, or 10 to the power of 3, kΩ), or “M” (mega, or 10 to the power of 6, MΩ). These passive components are intended to determine the operating parameters of the circuits, shape their performance, and ensure their operational quality and stability.
Passive Components and Other Circuit Elements
A circuit using active components can be configured in different ways, each configuration named based on the common lead and having specific characteristics. For BJTs, there is the common base configuration, which offers voltage gain only, the common collector that provides only current gain, and the common emitter that allows both voltage and current gain. The equivalents for FETs and MOSFETs — common gate, common drain, and common source — have roughly the same characteristics.
Amplifier Configurations
BJTs are the most commonly used, partly because they are less expensive. Their relatively low base impedance makes them current-controlled components, while the high gate impedance of FETs and MOSFETs allows for voltage control, similar to tubes whose performance they resemble. In high-power use, BJTs’ advantages include low internal resistance between the collector and emitter in saturation mode and good linearity. However, they have relatively slow switching speeds and some thermal stability issues that require compensation circuits to prevent thermal runaway, which could destroy the output stage. FETs and MOSFETs have poorer linearity and higher internal resistance between the drain and source but benefit from greater speed and good thermal stability, facilitating parallel mounting to increase potential power.
To simplify the approach and understanding, a simplified representation will sometimes be used. For example, a circuit might be illustrated by a triangle with a “+” and “-” sign on its vertical face, representing the inputs, inverting for the “-” and non-inverting for the “+”, with the output being the opposite point. In simplified versions, only one input and one output are sometimes shown. This symbol represents an OPAMP (Operational Amplifier), essentially a black-box representation of a high-gain amplifier with differential input (the “+” and “-” signs). This differential input is one of the most important and versatile points of such amplification circuits as it allows the use of feedback, a crucial element of any amplifier circuit as it largely determines its performance, especially its gain, frequency response curve, and operational stability. Some amplification circuits do not have a differential input, but we will not consider them in our overview since they are practically no longer used for new designs due to their inferior performance compared to OPAMPs, which also offer the possibility of a symmetrical input (also called balanced,), an important asset in the professional field and where immunity against interference and the possibility of using long cables are particularly important.
Operational Amplifier and Configurations
Let’s now transfer this amplification element to the world of high fidelity. In a typical traditional stereo system, the preamplifier will receive sound sources with varying requirements and normalize them to attack the power amplifier. For example, a turntable with a MM (Moving Magnet) cartridge will provide a signal of a few millivolts (mV) or less, with an impedance around 50 kΩ and very specific frequency response requirements conforming to RIAA (Recording Industry Association of America) standards. Most other sources will have a much higher output level, ranging from the most common standard of -10 dBV (316 mV, referenced to 1 V) to the professional standard of +4 dBu (1.228 V, referenced to 775 mV), except for CD players, which will be closer to 2 V. The impedance will usually be around a hundred ohms for semiconductor devices, with tubes being a special case we will not address here. The selected source will then be affected by the volume control and the equalization circuit, if present, before passing to the output stage that provides the final amplification sufficient for the least sensitive amplifiers to reach their full power if needed. Here again, standards vary, and some amplifiers will reach their saturation threshold with only 500 mV at the input, while others will require 2 V or more.
Gain Structure of the Hi-Fi Chain
Aside from a very low output impedance (on the order of fractions of an Ohm), what differentiates power amplifiers from other devices using an amplification factor is precisely this notion of power we use to define their most obvious characteristic, which is to provide work, quantified in Watts (W). This is where things get complicated and manufacturing costs increase. Voltage gain (Av) will then be accompanied by a very significant current gain (Ai), and the product will be the power (P in Watts) delivered to the speaker (V x I = P). This work can be very energy-consuming and requires some design strategies for the output stage if we want to limit thermal dissipation by increasing efficiency, which leads us directly to the concept of the class of operation.
Although we usually talk about power amplifiers in class A, B, or others, it is only the output stage that operates in this mode. To facilitate understanding, it would be helpful to consider the circuit in two sections. The first performs the voltage gain and contains the most components, though they are small. The second is the output stage that provides the current gain and includes the most massive parts of the assembly: power transistors, heat sinks etc. There are configurations where the output stage also provides some voltage gain, but they are not the norm, although some are notable for their performance level.
Simplified Circuit of an OPAMP Attacking a Power Stage
This sets the stage for addressing our topic, but other points will need to be covered before we can address all the factors that will determine the overall performance of a power amplifier, which we will do in our next meeting. We will cover, among other things, the power supply, a particularly important element that takes on a specific level of complexity due to the power involved. We will go from the essential to the most sophisticated: single or double polarity, regulated or not, etc. The last important point before moving on to the analysis of class A will be the coupling of the amplifier to the load (the speaker) depending on the technology used (semiconductors or tubes) and the type of power supply.
Let’s get technical-Part 2 : CLICK HERE
