Active feedback#

We have seen that the sign of the transfer of the single-loop passive feedback amplifiers from Fig. 227 is indissolubly connected to the type of their transfer. The voltage amplifier and the current amplifier have a noninverting transfer, while the transadmittance amplifier and the transimpedance amplifier have an inverting transfer. The sign of the transfer is the result of the feedback topology and the noninverting transfer of the passive feedback network. The inverting voltage amplifier, an inverting current amplifier, as well a noninverting transadmittance amplifier and a noninverting transimpedance amplifier, all require an inverting feedback network.

At a first glance, this may be a strange conclusion. Many experienced designers know the configuration depicted in Fig. 245 as the inverting voltage amplifier. Looking at this circuit from a conceptual point of view, one might more reasonable call it an inverting transimpedance amplifier (the one from Fig. 227C) with an input series impedance for (brute-force) voltage-to-current conversion. In cases in which the source impedance is very small and the noise performance is not critical, this solution may be good, but in other situations, brute force is not the way to go.

Inverting feedback networks can be constructed with the aid of the transadmittance amplifier or the transimpedance amplifier. An inventory of inverting voltage networks and inverting current networks is shown in Fig. 246. Since these feedback networks comprise an amplifier, they are active circuits. Amplifiers that use active feedback networks are called active feedback amplifiers.

We will discuss the design of single loop active feedback amplifiers in section Single-loop active feedback and that of multiple loop active feedback configurations in section Multiple-loop active feedback.

../_images/InvAttn.svg

Fig. 246 Inverting voltage attenuators (A,B) and inverting current attenuators (C, D). These attenuators are realized with the inverting passive-feedback transimpedance configuration (A, C) or the inverting passive-feedback transadmittance configuration (B, D).#

Single-loop active feedback#

Fig. 247 shows an inverting voltage amplifier, using transimpedance feedback. An alternative solution using transadmittance feedback is left as an exercise to the reader. The\ inverting current amplifier, the noninverting transadmittance amplifier and the noninverting transimpedance amplifier can be similarly realized. These configurations are not drawn; they are also left as an exercise for the reader. The reader is also invited to evaluate the influence of the feedback elements on the noise behavior and the power efficiency for these single loop active feedback configurations.

Let us now consider the inverting voltage amplifier from Fig. 247. We may generate a new circuit with similar properties by pairing the nullators and norators differently. This is because a network with nullors has a unique solution if the input conditions set by \(n\) nullators, can be satisfied with \(n\) norators. Theoretically, there exists no transfer from a nullator to a norator: a nullator sets a network condition by adding an equation and a norator adds the required extra dependent variable such that the condition can be satisfied.

After different pairings of the nullators and norators, a new circuit is found as the cascade connection of two single loop feedback amplifiers. This configuration is shown in Fig. 248.\medskip

../_images/Two-loop-alternative.svg

Fig. 248 Alternative solutions for the active-feedback inverting voltage amplifiers from Fig. 247.#

Multiple-loop active feedback#

Let us assume we need to design a negative feedback amplifier that has an accurately fixed finite nonzero input impedance and zero output impedance. We tried this already in section Dual-loop passive feedback configurations, but with a single nullor and passive feedback, this resulted in a negative input impedance because \(A\) and \(C\) had different signs. Equal signs for both transmission parameters requires a signal inversion in one of the feedback loops. Hence, there are two different realizations for each dual-loop feedback configuration.

../_images/t-loop-in-active-config.svg

Fig. 249 Examples of dual-loop active feedback configurations. (A) \(A\) and \(C\) fixed, \(B=D=0\), inverting transfer. (B) \(A\) and \(C\) fixed, \(B=D=0\), noninverting transfer.#

Fig. 249 shows two of those dual-loop configurations using active feedback. Both configurations have \(A\) and \(C\) fixed to a nonzero value while \(B\) and \(D\) are zero. The sign of their source-to-load transfer, however, is opposite.