A Systematic Analog Verification Approach Using Analog FastSPICE Harmonic Balance

In this white paper, we focus on the application of Analog FastSPICE (AFS) Harmonic Balance (HB) engine that provides improved convergence, enhanced runtime and memory reduction via three case studies of circuits commonly simulated using HB. We will provide a summary of performance and memory advantages of AFS HB. We conclude the paper with a summary of the latest advancements for post-layout designs with Siemens EDA’s AFS eXTreme (AFS XT) and a summary of the latest suite of AFS HB analyses.

Analog FastSPICE RF Harmonic Balance Offerings

The AFS Harmonic Balance suite comprises of large signal HB analysis and small signal noise (HBNOISE), AC (HBAC), stability (HBSTB), scattering
parameter (HBSP) and transfer function (HBXF) analyses. While single tone is supported for driven circuits and autonomous circuits, multi-tone (2-4) is supported for driven circuits.

For superior accuracy and performance on linear
and moderately nonlinear circuits, the AFS HB solvers are further optimized for performance and convergence via automated intelligence algorithms, with no compromise in accuracy. Nanometer RF
circuit design requires extensive sweeps, corners, and Monte Carlo analysis. AFS HB utilizes multithreaded capability for an additional performance speedup versus single core operations. Integrated with Solido™ Variation Designer software, the AFS platform addresses comprehensive corner variation
results efficiently. It allows a more effective verification that provides accuracy with a lesser number of simulations across all environmental corners,
reducing the RF simulation complexity.

The Harmonic Balance method computes the periodic steady state of the circuit utilizing the
frequency-domain, similar to the Shooting Newton method, which utilizes the time-domain. After a periodic steady-state solution is found, small signal analyses such as HB noise can also be run.