Abstract :
In communications and control practice increasing use is being made of active-RC networks instead of the conventional passive RLC types. The network theory literature contains a number of different procedures for the synthesis of such networks but so far there has been very little correlation of these procedures either with each other or with the well-known methods of passive synthesis, which are generally based on the lattice. The object of this paper is to show that such correlation may be achieved. The treatment is comprehensive and is believed to have some tutorial value in the earlier sections but there are also some original features. As a preliminary step an original development is made of the structure of the lattice in order to bring out the distinction between the restricted (i.e. convertible to bridged-T form) and the general cases of the lattice. The unbalanced equivalents of the latter case generally contain an ideal unity-ratio inverting transformer and, as a transition to active form, it is shown that this may be realized formally by a (+R, ?R) network. Hence, the general lattice may be realized as an active bridged-T consisting of two negative impedance converters (NIC´s) and the usual inverse pair of passive impedances. Then, when the latter are realized in active ?RC form using the standard procedure employed in the literature the same pair of NIC´s may be employed. However, there are two special cases of the active transformer replacement, for which one or other of the inverse impedance arms are nullified and the active bridged-T reduces to active T or ? form. Then, when the three impedances of the T or ? are separated into + and ?RC parts, the three negative ones may be brought together and realized by the corresponding network placed in cascade between a pair of NIC´S. When the networks so obtained are operated between ideal terminations, for example an ideal voltage source and an open-circuit load, certain simplifications may be made. One of- these is identified as the form of realization used (without derivation) by T. Yanagisawa in his wellknown paper. The principal conclusion of the paper is that one can develop from the classical lattice a broad class of active-RC networks that includes useful forms, not all of which appear yet to have been employed in the literature.
