Abstract
Two general methods for realizing high-order delay approximation functions by RC active networks are described. One is to cascade simple low-order sections, each section being terminated by a suitable amplifier for direct cascading; the other is to apply additional feedforward paths to a cascaded network. The second method has particular advantages over the first for realizing a high-order transfer function having transmission zeros in the right half S-plane, The amplifiers used here are unity-gain and operational amplifiers. Some basic transposition properties of an active network have been discovered and applied to the synthesis procedures. A new RC active method of realizing an RL admittance and its application to synthesis, give the minimum sensitivity of the denominator coefficients of a transfer function to variations in the parameters of the active element. A new two-operational-amplifier method is derived from Millman's theorem, and a realization of an RL admittance using two operational amplifiers is derived as an extension of methods using one finite-gain amplifier.