ABSTRACT
Numerous textbooks have presented excellent treatments of the design and analysis of broadband bipolar amplifiers. This chapter is concerned with techniques for integrated circuit amplifiers, and is written mainly as a tutorial aimed at the practicing engineer. For broadband polar design, it is first important to identify the key difference between lumped and
distributed design techniques. Basically when the signal wavelengths are close to the dimensions of the integrated circuit, then characteristic impedances become significant, lines become lossy, and we essentially need to consider the circuit in terms of transmission lines. At lower frequencies where the signal
wavelength is much larger than the dimensions of the circuit, the design can be considered in terms of lumped components, allowing some of the more classical low-frequency analog circuit techniques to be applied. At intermediate frequencies, we enter the realms of hybrid lumped=distributed design. Many radio-frequency (RF) designs fall into this category, although every day we see new technologies and circuit techniques developed that increase the frequency range for which lumped approaches are possible. In broadband applications, integrated circuits (ICs) are generally designed without the use of special microwave components, so broadband techniques are very similar to those employed at lower frequencies. However, several factors still have to be considered in RF design: all circuit parasitics must be identified and included to ensure accurate simulation; feedback can generally only be applied locally as phase shifts per stage are significant; the cascading of several local feedback stages is difficult since alternating current (ac) coupling is often impractical; the NPN bipolar transistor is the main device used in silicon, since it has potentially a higher ft than PNP bipolar or MOSFET devices; active PNP loads are generally avoided due to their poor frequency and noise performance and so resistive loads are used instead. The frequency performance of an RF or broadband circuit will depend on the frequency capability of the
devices used, and no amount of good design can compensate for transistors with an inadequate range. As a rule, designs are kept as simple as possible, since at high frequencies all components have associated parasitics.
It is important to describe at the outset a very useful approximation that will assist in simplifying the high-frequency analysis of some of the amplifiers to be described. The technique is known as Miller’s theorem and will be briefly discussed here. A capacitor linking input to output in an inverting amplifier results in an input-referred shunt capacitance that is multiplied by the voltage gain of the stage, as shown in Figure 3.1. This increased input capacitance is known as the Miller capacitance. It is straightforward to show that the input admittance looking into the inverting input of the amplifier is
approximately Yin¼ jvCf (1þA). The derivation assumes that the inherent poles within the amplifier are at a sufficiently high frequency so that the frequency response of the circuit is dominated by the input of the amplifier. If this is not the case, then Miller’s approximation should be used with caution as it will be discussed later. From the preceding model, it is apparent that the Thévenin input signal sources see an enlarged capacitance to ground. Miller’s approximation is often a useful way of simplifying circuit analysis by assuming that the input dominant frequency is given by the simple low-pass RC filter in Figure 3.1. However, the effect is probably one of the most detrimental in broadband amplifier design, affecting both frequency performance and=or stability.
