Physics-Based Modeling : RLCK Models

Overview

PeakView™ Physics-Based Modeling (PBM™) creates the bridge between frequency-domain electromagnetic models and SPICE friendly time-domain solutions. Unlike popular purely numerical approaches, PBM™, based on positive R, L, C values, generates RLCK models that satisfy passivity constraints and are physically realizable. PBM™ models are compact, DC-accurate, and ready for use in Cadence® Spectre, SpectreRF, Synopsys HSPICE® and other SPICE-based simulators. It is a key technology underlying all of Lorentz’ products (including PeakView EMD™ device synthesis, PeakView LEM™ extraction and PeakView HFD™ interconnect analysis), and rapidly extending itself into the PCB and package co-simulation space.

Benefits

  • Transient Simulations

Driven by PeakView™’s next-generation EM solver, the PBM solution converts any Touchstone formatted model to a Spectre or HSPICE compatible equivalent circuit for use in transient circuit simulations.

  • Passivity

PBM is unique in the sense that these models are passive over a user selectable frequency range and guaranteed to be convergent. Traditional EM tools generate touchstone formatted files (s-parameters, y-parameters, z-parameters) and rational or pole-zero models. With these formulations, often there are issues with passivity, particularly if the data lies on the edge of the Smith Chart (i.e. has very low impedance or very high impedance). PBM circuit equivalent models, on the other hand, are guaranteed to be passive and preserve DC inductance and resistance.

  • Compactness

PBM achieves good quality matching with the EM model generating a very compact model for regular passive devices.

  • Speed

Because of their compact size, PBM models do not markedly expand the circuit simulation matrix leading to faster circuit simulation and higher capacity.

  • Extensibility

The PeakView PBM™  provides a built-in standard library of ready-to-use models for inductors, transmission lines, baluns, transformers and other devices.  PBM can also be customized using the Python programming language and templates provided by Lorentz. Users can create PBM models from scratch or tailor existing models according to their unique design needs.

Applications

PBM is especially useful for simulation of non-linear phenomena such as harmonic balance analysis, phase noise analysis, eye diagrams etc. Any application involving large signal space (i.e. transient simulations for VCOs, linearity of LNAs, PAs, clocking circuits etc.) is a prime candidate for the PBM™ solution.

How it Works

A PBM is derived from a Touchstone formatted small-signal model using a known model topology and constraining the device elements to physical values, therefore guaranteeing passivity. The source of this small-signal model can be from the PeakView™ EM simulation or an imported Touchstone formatted file.

In the PeakView™ GUI, users select an appropriate PBM method resembling the topology of their design. After the EM simulation is finished, PeakView™ generates a lumped model corresponding to the small signal model. The lumped PBM model is optimized at a particular frequency and fitted to cover a desired frequency range.

PeakView™ Physics-Based Model Generation

The Physics-Based model, along with nport (small signal) and other design views (i.e. symbol) are finally synced to the user’s IC design library and ready for use in circuit simulation.

Physics-Based Model in Design Library

Model Options

PeakView™ offers diverse Physics-Based Modeling options generalized into two broad categories. Basic PBM is suitable for classic devices with typical number of ports, i.e. symmetric and asymmetric inductors, transformers, transmission lines, finger capacitors etc. This option is also configurable and customizable for designers’ unique modeling requirements.

PBM2 is an extended feature for arbitrary interconnect modeling. It is suitable for modeling device routing with atypical geometries. It is a key technology that complements PeakView HFD™ in generating RLCK models for interconnect structures of arbitrary  configurations.

PBM and PBM2

Comments are closed