Joint Filtering, Dead-Time, and Loss Bounds on Switching Frequency in Class-D Power Stages

Selecting the switching (carrier) frequency of a Class-D power stage couples reconstruction-filter attenuation, dead-time distortion, pulse-width headroom, and switching loss. This paper derives closed-form joint bounds on the carrier of a full-bridge stage and, from them, sufficient carrier-independent screening conditions on the blanking interval, shortest pulse, and switching energy for a specification and filter order. An extended commutation model links dead-time distortion to switch-node charge and inductor ripple, defining when the square-wave formula holds; a current-dependent switching-energy model supplies the loss bound. Switching-level simulations with silicon and gallium nitride (GaN) datasheet parameters and resistive and loudspeaker loads verify the predictions: at modulation index 0.8, the extended model agrees within 7% for blanking above 1%, whereas the square-wave formula overestimates distortion by 30–51% at 2% blanking. For a 20 kHz band with 1 dB passband and 40 dB carrier attenuation, a second-order filter requires at least 280 kHz (347 kHz with tolerances and guard band), limiting the effective blanking to 71 ns (58 ns) at a 2% allocation; the modeled Zobel-compensated loudspeaker raises the bound to 502 kHz. Under these second-order targets, the 25–40 kHz carriers and 660 ns transition interval (taken as blanking) measured on a 14 V amplifier would support only about 2 kHz of bandwidth.

Authors

Institutions

Publication Details

Journal
Electronics
Published
2026-11-12
DOI
https://doi.org/10.3390/electronics15225255
Primary Topic
Advanced Power Amplifier Design
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Joint Filtering, Dead-Time, and Loss Bounds on Switching Frequency in Class-D Power Stages

Ahmad Alzahrani
Electronics
Advanced Power Amplifier Design
article

Joint Filtering, Dead-Time, and Loss Bounds on Switching Frequency in Class-D Power Stages

Ahmad Alzahrani
article en

Abstract

Selecting the switching (carrier) frequency of a Class-D power stage couples reconstruction-filter attenuation, dead-time distortion, pulse-width headroom, and switching loss. This paper derives closed-form joint bounds on the carrier of a full-bridge stage and, from them, sufficient carrier-independent screening conditions on the blanking interval, shortest pulse, and switching energy for a specification and filter order. An extended commutation model links dead-time distortion to switch-node charge and inductor ripple, defining when the square-wave formula holds; a current-dependent switching-energy model supplies the loss bound. Switching-level simulations with silicon and gallium nitride (GaN) datasheet parameters and resistive and loudspeaker loads verify the predictions: at modulation index 0.8, the extended model agrees within 7% for blanking above 1%, whereas the square-wave formula overestimates distortion by 30–51% at 2% blanking. For a 20 kHz band with 1 dB passband and 40 dB carrier attenuation, a second-order filter requires at least 280 kHz (347 kHz with tolerances and guard band), limiting the effective blanking to 71 ns (58 ns) at a 2% allocation; the modeled Zobel-compensated loudspeaker raises the bound to 502 kHz. Under these second-order targets, the 25–40 kHz carriers and 660 ns transition interval (taken as blanking) measured on a 14 V amplifier would support only about 2 kHz of bandwidth.

ElectronicsVol. 15(22)
Najran University (SA)
Najran University
Advanced Power Amplifier Design
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon