Stefaan Sercu

Modeling differential via holes

In this paper, we present a method to characterize differential via holes in printed circuit boards in a both fast and accurate way. The via hole is modeled as a cascade of capacitances and inductances. We use FASTCAP to compute the values of the capacitances, and a closed form formula to obtain the inductance values. The numerical predictions are compared with experimental data.

Modelling complex via hole structures

Derives a physics-based circuit model for complex via hole structures in printed circuit boards. The via hole is modeled as a cascade of capacitance and inductance matrices. Capacitance values are computed using a three-dimensional electrostatic solver and inductance values are computed from a two-dimensional quasi-TEM solver. This model is valid at frequencies up to a few gigahertz for typical via hole geometries, where the return current follows a well defined path.

Characterizing N-port packages and interconnections with a 2-port network analyzer

In this paper a technique is described for the measurement of the correct S-parameters of an N-port package or interconnection using a 2-port network analyzer with 50 /spl Omega/ system impedance and N imperfect terminations. The technique is fully general and can be applied using arbitrary terminations. Broadband 50 n loads are not required. The method is illustrated on a coupled microstrip line structure.

High-frequency circuit modeling of large pin count packages

In this paper, a technique is presented for the high-frequency circuit modeling of coupled conductor structures. The method is, in particular, very useful for the modeling of structures with a varying signal/ground configuration. Structures with a large number of conductors (N+1, N>100) are also easy to model, as the method reduces the modeling of the 2N-port to the modeling of two- and four-port structures. Two- and four-port structures are much easier to model since their equivalent circuit model has fewer parameter values. Examples of multiconductor structures are high-density connectors and large pin count electronic packages. The model accurately simulates the electrical properties, such as reflection and transmission of all conductors, and the backward and forward crosstalk to all other conductors.

Modelling differential via holes

With the ever-increasing frequencies of printed circuit board (PCB) interconnections, the role played by via holes is no longer negligible. Thoughtless design of via holes on a board may seriously degrade signal integrity due to reflections, ground bounce, etc. However, the characterisation of via holes has not drawn much attention until now. This paper presents a method for the characterisation of differential via holes as a cascade of capacitances and inductances. Capacitances are computed using FASTCAP and inductances using simple analytical models.

A new approach for the experimental circuit modeling of coupled interconnection structures based on causality

In this paper, a time-frequency-domain technique for the experimental circuit modeling of coupled interconnection structures and discontinuities is presented. The technique models and de-embeds all discontinuities and coupled substructures of the device under test (DUT) one by one, and is based on the principle of causality. Validation of each part of the model is done in the time domain, while all calculations are performed in the frequency domain. To validate the accuracy of the circuit models, measured reflection, transmission, and near- and far-end crosstalk are compared with simulated results.

A new algorithm for experimental circuit modeling of interconnection structures based on causality

A new algorithm is described for the circuit modeling of two-port interconnection structures starting from S-parameter measurements or simulations. The presented procedure is particularly useful in modeling interconnections with multiple discontinuities. The theory behind the method is based on the principle of causality. Step by step, a circuit model is proposed for each discontinuity and the corresponding parameter values are determined using the time-domain reflection and transmission response derived from the S-parameters. In this way, it is possible to obtain a circuit model for which each element is related to a certain physical part of the passive structure under test. The results show good-agreement between measured and modeled reflection and transmission waveforms.

Parameter extraction for circuit models of electronic packages without optimization

In this paper, we derive circuit models for electronic packages starting from Z- or Y-matrix descriptions. If we propose a T- or a /spl Pi/-circuit, a direct relation is found between the admittance and impedance values and the circuit parameters. No optimization is needed. The T- and /spl Pi/-circuit models are valid for a general class of interconnections and packages. We illustrate the method on an example of a TAB interconnection structure.

Circuit modelling of two-port interconnection structures based on causality

In this paper a new mixed time-frequency domain procedure is described for the construction of circuit models of interconnection structures. This method extracts hybrid circuit models from reflection and transmission measurements and is based on the principle of causality. The measurements can be performed in the time or the frequency domain. To validate the accuracy of the circuit models, SPICE simulated results for the derived circuit models are compared with the measured data.

Experimental circuit modeling of coupled interconnection structures based on causality

In this paper a time-frequency domain technique for the experimental circuit modeling of coupled two-port interconnection structures and discontinuities is presented. The technique models and de-embeds each appearing discontinuity and coupled substructure of the DUT one by one, and is based on the principle of causality. To validate the accuracy of the circuit models, measured reflection, transmission, near-end and far-end crosstalk are compared with simulated results.