We develop delay macromodels for lossless as well as lossy point-to-point MCM transmission lines using a systematic model construction procedure that includes dimensional analysis. The result for lossless lines confirms earlier work and extends it to the case of nonlinear drivers and capacitive termination. For lossy lines, we show that dimensional analysis allows us to reduce the complexity of the delay expression without any loss in accuracy. A second-order polynomial fit to a four-argument delay function is shown to yield predictions that are within 5% of detailed circuit simulation at a fraction of the computational cost. We illustrate the applications of these macromodels in studying the effects of packaging technology on signal delay and in sensitivity analysis. Zndex T'ms-Macromodeling, dimensional analysis, multichip module, transmission line, interconnect delay, packaging, sensitivity. I. INTRODUC-ITON N THIS PAPER WE DEVELOP delay expressions for I lossless and lossy point-to-point multichip module (MCM)interconnections and use them to study the effect of technology and circuit parameters on transmission line delay. Similar expressions have been developed previously and used to derive wiring rules intended to reduce delay and minimize reflections in package design [4]. We argue in this paper that such "macromodels" can be routinely developed by, and indeed can benefit from, a systematic application of experimental model building procedures in conjunction with dimensional analysis. For the case of a lossless line driven by a CMOS inverter, we obtain a delay function that is identical in form to what would be derived from a theoretical analysis of an ideal transmission line driven by an ideal voltage source. Significantly, the delay expression was obtained without making any simplifying modezing assumptions. For lossy lines, we show that dimensional analysis allows us to reduce the complexity of the delay expression without any loss in accuracy. A secondorder polynomial fit to a four-argument delay function is shown to yield predictions that are within 5% of detailed circuit simulation at a fraction of the computational cost.After briefly reviewing our macromodeling methodology in Section 11, we describe in Section 111 its application to the development of delay equations for lossless and lossy pointto-point MCM interconnections. Section IV describes how the derived delay equations can be used to study the effect of technology parameters on line delay as well as to obtain the sensitivity of the delay to various parameters. We conclude in Section V by suggesting an extension to handle lines with multiple drops, and by reiterating the benefits of the proposed macromodeling scheme.
MACROMODELING METHODOLOGYTypically, the design and optimization of electronic systems requires that we establish the dependence of certain "macroscopic" objectives on a set of relevant designable parameters. For instance, we may be interested in keeping simultaneous switching noise on a printed circuit board below a specified thre...