Hardware Podcasts

High Current PDN Design: Engineer of the Year Reveals Industry Secrets

OnTrack · 21 Apr 2025 · 42m · Episode 296

What this episode covers

Ben Dannan, founder of Signal Edge Solutions and DesignCon's Engineer of the Year, discusses the engineering challenges of designing 2,000-amp power distribution networks (PDN) for high-power AI processors like NVIDIA's Blackwell platform. He details the industry's struggle with current densities exceeding 100 amps per square centimeter and the debate between vertical and horizontal power delivery architectures. Dannan also highlights the critical issue of inaccurate silicon models provided by vendors, which often lead to hardware overheating, and introduces his upcoming measurement-based model library for large-signal simulation.

Most relevant to

companies scaling up

Topics

Mentioned

Companies
Signal Edge Solutions, Keysight, Ansys, Cadence, DuPont, Marada, HPE, AMD, Zylink, Pico Test, Google, University of Cambridge, TI, ADI, Microchip
Products
NVIDIA Blackwell
Tools
Altium Designer
Also
DesignCon

Related episodes from other shows

Other shows on the same ground, closest first.

  • Oxide and Friends · 2h 07m

    More Tales from the Bringup Lab

    Bryan Cantrill and the Oxide Computer Company team, including Arjen Roodselaar, Eric Aasen, and Matt Keeter, detail the bring-up of Sidecar, a switching board based on the Intel Tofino 2 ASIC, and Gimlet, their compute node. They discuss the extreme difficulty of designing the Power Delivery Network (PDN) to handle 300-amp load steps with only 7 millivolts of margin, requiring a custom 'load slammer' for testing. The team also recounts debugging firmware and hardware issues, including TOML parsing errors, slew rate mismatches on SPI traces, and a critical 500-ohm pull-down resistor error on the T6 NIC that nearly halted production.

    PCB design & layoutDebugging & failure analysis
    10 chapters

  • The Amp Hour · 1h 13m

    Switching Converter Engineering with Carmen Parisi

    Carmen Parisi, Staff Applications Engineer at Renesas (formerly Intersil and TI), explains the engineering challenges of designing high-current switching regulators for Intel and AMD processors. She details the use of multi-phase buck topologies to handle extreme load transients of up to 150 amps, the trade-offs between analog and digital control schemes, and the critical role of test engineers in validating silicon and optimizing yield. Parisi also discusses the evolution of PCB design, the importance of proprietary communication buses like SVI and PMBus, and her career advice for aspiring hardware engineers.

    Analog & power electronicsPCB design & layout
    8 chapters

  • EE Times Current · 18m

    Designing the AI Factory from the Rack Up

    Neeraj Paliwal, Senior Vice President of Silicon IP Product Management at Synopsys, explains why the AI rack has replaced the data center as the fundamental unit of design and economics. He details the shift from 8kW to 130kW+ power densities in systems like NVIDIA's GB200 NVL72, driven by the need to optimize cost per token across distributed GPUs. Paliwal highlights that the industry is moving past single-die limits toward 3D stacking, custom HBM memory integration, and high-speed die-to-die interfaces like UCIe to overcome the memory and I/O bandwidth walls.

    Chips, FPGAs & semiconductorsBatteries & low-power design
    5 chapters

  • The Amp Hour · 1h 11m

    Grow A Superbrain

    Chris Gammell of Contextual Electronics and Dave Jones of EEV Blog discuss the Feynman method for learning, the controversy surrounding Walter Lewin's demonstration of Kirchhoff's current law, and the practical challenges of high-speed PCB design for the BeagleBone Black. They also cover the economic realities of grid-scale battery storage, citing the Tesla Megapack in South Australia, and touch on industry trends like PCB manufacturer astroturfing and the KiCad community.

    PCB design & layoutCareers & education
    8 chapters

  • EE Times Current · 40m

    An All Inclusive Journey to HPC/Data Center Silicon Success

    Charlie Matar, Senior Vice President of Systems Solutions at Synopsys, discusses the escalating complexity of High Performance Computing (HPC) and data center silicon design. He highlights critical challenges including engineer shortages, the high cost of advanced nodes, and the shift toward chiplet architectures and 2.5D/3D packaging. Matar also explains how Synopsys is leveraging its DSO.AI platform to optimize power, performance, and area (PPA) while addressing reliability and security concerns in the semiconductor ecosystem.

    Chips, FPGAs & semiconductorsEDA & CAD tools
    8 chapters

The summary, chapters and topics are generated from the episode’s audio by this directory, not written by OnTrack. The episode itself belongs to its publisher.