Sensitivity minimization of networks with operational amplifiers and parasitics

A computational method is described which calculates network sensitivity to the influence of parasitics by simulating only the ideal network. For active networks, a similar sensitivity is defined with respect to the deviations from ideal behavior of operational amplifiers. It is shown that this sensitivity is equivalent to the gain-sensitivity product and depends only on the passive elements and the topology. The procedure was applied to known active networks and, in every case considered, minimization resulted in networks with smaller sensitivities. Compensation for nonideal behavior of the operational amplifiers is outlined. A Monte Carlo simulation is performed on the well-known STAR network and it is shown that optimization can lead to considerable improvement, even in the case of well-designed networks.