a 1ms s to 1gs s ringamp based pipelined adc
play

A 1MS/s to 1GS/s Ringamp-Based Pipelined ADC with Fully Dynamic - PowerPoint PPT Presentation

May 01, 2023 •185 likes •890 views

A 1MS/s to 1GS/s Ringamp-Based Pipelined ADC with Fully Dynamic Reference Regulation and Stochastic Scope-on-Chip Background Monitoring in 16nm Benjamin Hershberg, Nereo Markulic, Jorge Lagos, Ewout Martens, Davide Dermit, Jan Craninckx imec,

  1. A 1MS/s to 1GS/s Ringamp-Based Pipelined ADC with Fully Dynamic Reference Regulation and Stochastic Scope-on-Chip Background Monitoring in 16nm Benjamin Hershberg, Nereo Markulic, Jorge Lagos, Ewout Martens, Davide Dermit, Jan Craninckx imec, Leuven, Belgium 2020 Symposia on VLSI Technology and Circuits CD1.1

  2. Architecture Overview INp c BACKEND STG1 STG2 STG3 STG4 STG5 STG6 STG7 1.5b 1.5b 1.5b 1.5b 1.5b 1.5b 1.5b 1.5b + 3b c INm C TOT = C TOT = C TOT = C TOT = C TOT = C TOT = C TOT = C TOT = master clock c 256fF 128fF 128fF 128fF 128fF 128fF 128fF 256fF VREFP dirty VREFP ext. c VREFP clean VCM ext. c reference VCM VREFM ext. c regulator VREFM dirty VREFM clean Ringamp “deep” pipeline based on: • • New in this work: [Hershberg, ISSCC 2019, Paper 3.6] – Reference Regulation [Hershberg, ISSCC 2019, Paper 3.1] – Stochastic Scope-on-Chip – Fully Dynamic Power Consumption – Asynchronous event-driven timing control – Ringamp Topology – MDAC with passive-hold mode in STG1 – Early quantization in STG2-9 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 1

  3. Outline • Ringamp Topology • Fully Dynamic Reference Regulation • Scope-on-Chip Amplifier Settling Monitor • Performance Summary & Conclusion 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 2

  4. Ringamp Topology • What is a ringamp? – Multi-stage amplifier INp OUTm – Dominant output pole – Dynamic stabilization ☺ High efficiency ☺ High linearity ☺ High speed ☺ Wide output swing ☺ Fully dynamic (switchable) [Hershberg, JSSC 2012] ☺ Scales with Digital 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 3

  5. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN EN C SMALL EN EN EN 7 C BIG 7 C SENSE C FB Dz_nmos[7:0] Dz_pmos[7:0] OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 4

  6. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN EN C SMALL EN EN EN 7 C BIG 7 C SENSE [Lim, JSSC Dec. 2015] C FB Dz_nmos[7:0] Dz_pmos[7:0] OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 5

  7. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN [Lim, JSSC Dec. 2015] EN C SMALL EN EN EN [Hershberg, ISSCC 2019] 7 C BIG 7 C SENSE C FB Dz_nmos[7:0] Dz_pmos[7:0] OUTm (Paper 3.1) 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 6

  8. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN EN C SMALL [Hershberg, ISSCC 2019] EN EN EN 7 C BIG 7 C SENSE C FB Dz_nmos[7:0] Dz_pmos[7:0] (Paper 3.1) OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 7

  9. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN EN C SMALL EN EN EN 7 C BIG 7 C SENSE [Lagos, JSSC Feb. 2019] C FB Dz_nmos[7:0] Dz_pmos[7:0] OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 8

  10. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN [Chen, TCASII 2018] EN C SMALL EN EN EN 7 C BIG 7 C SENSE C FB [Lagos, JSSC Mar. 2019] Dz_nmos[7:0] Dz_pmos[7:0] OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 9

  11. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB DZN DZN V DD V DD • Trapped-Charge OUTm OUTp EN EN EN EN EN EN EN Biasing V SS V SS DZP DZP • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VDD DZN VSS DZP C FB C SENSE EN EN EN C SMALL EN EN EN 7 C BIG 7 C SENSE [Lagos, JSSC Mar. 2019] C FB Dz_nmos[7:0] Dz_pmos[7:0] OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 10

  12. Ringamp Topology VDD • Fully-Differential INp INm • Multi-Path CMFB • Trapped-Charge OUTm OUTp Biasing • CMOS-Resistor • Bias-Enhanced • Self-Resetting VSS B1 • Slew-Done Detector Trapped Charge CMFB Trapped-Charge Bias Control OUTp VSS DZP VDD DZN C FB C SENSE EN EN EN C SMALL EN EN EN 7 C BIG 7 C SENSE [Lagos, JSSC Mar. 2019] C FB Dz_nmos[7:0] Dz_pmos[7:0] OUTm 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 11

  13. Ringamp Topology INm INp EN EN • Fully-Differential A C + OUTp BN BP BP BN OUTm • Multi-Path CMFB B D slew done detected • Trapped-Charge EN EN Biasing • CMOS-Resistor slew done detector • Bias-Enhanced slew done t d EN • Self-Resetting slew done detected • Slew-Done Detector [Hershberg, ISSCC 2019] (Paper 3.6) 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 12

  14. Ringamp Topology INm INp EN EN • Fully-Differential A C + OUTp BN BP BP BN OUTm • Multi-Path CMFB B D slew done detected • Trapped-Charge EN EN Biasing • CMOS-Resistor slew done detector • Bias-Enhanced slew done t d EN • Self-Resetting slew done Not Mandatory detected • Slew-Done Detector Using a fixed time- delay to generate [Hershberg, ISSCC 2019] slew done also ok. (Paper 3.6) 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 13

  15. Outline • Ringamp Topology • Fully Dynamic Reference Regulation • Scope-on-Chip Amplifier Settling Monitor • Performance Summary & Conclusion 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 14

  16. Fully Dynamic Reference Regulation • Goal: fully dynamic power for complete system (core + reference) – Constant FoM for any clock rate – Enables reconfigurable, multi-standard ADCs • ADC core fully dynamic by design ✓ [Hershberg, ISSCC 2019, Paper 3.6] – Enabler: Asynchronous timing control – Enabler: Ring amplifier with fast on/off switching • Reference Regulation – How can we make this fully dynamic too? 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 15

  17. Fully Dynamic Reference Regulation • Discrete Time Regulator Loop VREF ext. + – Comparator monitors replica level prepare Charge source replica out – Feeds back “charge packet” (C P ) V DD C P update into charge reservoir (C R ) C R R load connect prepare prepare connect [Kull, JSSC 2013] 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 16

  18. Fully Dynamic Reference Regulation • Discrete Time Regulator Loop VREF ext. + – Comparator monitors replica level prepare Charge source replica out – Feeds back “charge packet” (C P ) V DD C P update into charge reservoir (C R ) C R R load connect prepare • Limitations prepare – Must source all charge for an ADC conversion in a single charge packet connect – Reference ripple error a problem – Acceptable for ADCs < 8 bit – C R too large for ADCs > 8 bit (> 1nF) [Kull, JSSC 2013] 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 17

  19. Proposed: Split-Reference Regulation • Key observation: V REF accuracy & Example Settling Waveform current requirements are decoupled w.r.t. time – Initial: large current, low accuracy – Final: small current, high accuracy amplify track IN X OUT Ringamp track track early Example 1.5b flip-around MDAC VREF amplify sub-DAC 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 18

  20. Proposed: Split-Reference Regulation • Key observation: V REF accuracy & Example Settling Waveform current requirements are decoupled w.r.t. time – Initial: large current (98%), low accuracy – Final: small current (2%), high accuracy amplify track IN X OUT Ringamp track track early Example 1.5b flip-around MDAC VREF amplify sub-DAC 2020 Symposia on VLSI Technology and Circuits CD1.1 Slide 19

Recommend

Ringamp: The Scalable Amplifier We ve All Been Waiting For? Benjamin Hershberg Imec March 22

Ringamp: The Scalable Amplifier We ve All Been Waiting For? Benjamin Hershberg Imec March 22

Ringamp: The Scalable Amplifier We ve All Been Waiting For? Benjamin Hershberg Imec March 22 nd 2020 NANOSCALE CMOS ADAPTATION AND SURVIVAL Some ADC architectures have thrived and expanded. SAR VCO-based CT 2 NANOSCALE CMOS

1.64k views • 135 slides
Review: FP Pipeline Model 4-stage fully pipelined adder, Non-pipelined multiplier and divider A1

Review: FP Pipeline Model 4-stage fully pipelined adder, Non-pipelined multiplier and divider A1

Review: FP Pipeline Model 4-stage fully pipelined adder, Non-pipelined multiplier and divider A1 A2 A3 A4 EX IF MEM WB ID/REG DIV (6 cycle non pipelined) MUL (4 cycle non pipelined) 1 Review: Summary If instructions A and B are

511 views • 33 slides
DLX Pipeline 2-stage fully pipelined Adder 4-stage fully pipelined Multiplier 5-cycle

DLX Pipeline 2-stage fully pipelined Adder 4-stage fully pipelined Multiplier 5-cycle

DLX Pipeline 2-stage fully pipelined Adder 4-stage fully pipelined Multiplier 5-cycle non-pipelined Divider EX MEM A1 A2 IF ID WB M1 M2 M3 M4 DIV (5 cycle non pipelined) 9 Structural Hazard: WB stage 1 2 3 4 5 6 A IF ID +

753 views • 23 slides
- A Tutorial - Based on Slides from Dr. Bibhudatta Sahoo University of Illinois at

- A Tutorial - Based on Slides from Dr. Bibhudatta Sahoo University of Illinois at

Department of Electrical and Computer Engineering Pipelined ADC Design - A Tutorial - Based on Slides from Dr. Bibhudatta Sahoo University of Illinois at Urbana-Champaign Slides by Bibhudatta Sahoo -1- Outline Review of Pipelined ADCs

1.42k views • 124 slides
Chapter 6: Designing a Pipelined CPU What are our resources? 1 washer, 1 dryer, 1 folder

Chapter 6: Designing a Pipelined CPU What are our resources? 1 washer, 1 dryer, 1 folder

Chapter 6: Designing a Pipelined CPU What are our resources? 1 washer, 1 dryer, 1 folder (you), 1 put awayer (roommate) What % of the time are they idle? 1 2 Chapter 6: Designing a Pipelined CPU Chapter 6: Designing a Pipelined CPU

560 views • 16 slides
LECTURE 9 Pipeline Hazards PIPELINED DATAPATH AND CONTROL In the previous lecture, we

LECTURE 9 Pipeline Hazards PIPELINED DATAPATH AND CONTROL In the previous lecture, we

LECTURE 9 Pipeline Hazards PIPELINED DATAPATH AND CONTROL In the previous lecture, we finalized the pipelined datapath for instruction sequences which do not include hazards of any kind. Remember that we have two kinds of hazards to worry

975 views • 62 slides
A 0.5-to-1 V 9-bit 15-to-90 MS/s Digitally Interpolated Pipelined-SAR ADC Using Dynamic Amplifier

A 0.5-to-1 V 9-bit 15-to-90 MS/s Digitally Interpolated Pipelined-SAR ADC Using Dynamic Amplifier

A 0.5-to-1 V 9-bit 15-to-90 MS/s Digitally Interpolated Pipelined-SAR ADC Using Dynamic Amplifier James Lin, Zule Xu, Masaya Miyahara, and Akira Matsuzawa Tokyo Institute of Technology, Japan b. Matsuzawa &amp; Okada Lab y Outline

328 views • 30 slides
Pipelined Scheduling of Acyclic SDF Graphs using SMT Solvers P. Tendulkar, P. Poplavko, J.

Pipelined Scheduling of Acyclic SDF Graphs using SMT Solvers P. Tendulkar, P. Poplavko, J.

IDEA 2015 Investigating Dataflow in Embedded computing Architecture Pipelined Scheduling of Acyclic SDF Graphs using SMT Solvers P. Tendulkar, P. Poplavko, J. Maselbas, and O. Maler Verimag Lab (CNRS, University of Grenoble), France

266 views • 15 slides
Designing a Pipelined Processor Computer System Architecture Go back and examine your

Designing a Pipelined Processor Computer System Architecture Go back and examine your

Designing a Pipelined Processor Computer System Architecture Go back and examine your datapath Pipelining Part II and control diagram associated resources with states ensure that flows do not conflict, or Chalermek Intanagonwiwat

328 views • 8 slides
Pipelined Computations In the pipeline technique, the problem is divided into a series of tasks

Pipelined Computations In the pipeline technique, the problem is divided into a series of tasks

Pipelined Computations In the pipeline technique, the problem is divided into a series of tasks that have to be completed one after the other. In fact, this is the basis of sequential programming. Each task will be executed by a separate process

529 views • 27 slides
CSEE 3827: Fundamentals of Computer Systems, Spring 2011 9. Pipelined MIPS Processor Prof.

CSEE 3827: Fundamentals of Computer Systems, Spring 2011 9. Pipelined MIPS Processor Prof.

CSEE 3827: Fundamentals of Computer Systems, Spring 2011 9. Pipelined MIPS Processor Prof. Martha Kim (martha@cs.columbia.edu) Web: http://www.cs.columbia.edu/~martha/courses/3827/sp11/ Outline (H&amp;H 7.5) Pipelined MIPS processor

617 views • 30 slides
High Performance Pipelined Process Migration with RDMA Xiangyong Ouyang, Raghunath

High Performance Pipelined Process Migration with RDMA Xiangyong Ouyang, Raghunath

High Performance Pipelined Process Migration with RDMA Xiangyong Ouyang, Raghunath Rajachandrasekar, Xavier Besseron, Dhabaleswar K. (DK) Panda Department of Computer Science &amp; Engineering The Ohio State University Outline Introduction

593 views • 36 slides
A 61.5dB SNDR Pipelined ADC Using Simple Highly-Scalable Ring Amplifiers Benjamin Hershberg 1 ,

A 61.5dB SNDR Pipelined ADC Using Simple Highly-Scalable Ring Amplifiers Benjamin Hershberg 1 ,

A 61.5dB SNDR Pipelined ADC Using Simple Highly-Scalable Ring Amplifiers Benjamin Hershberg 1 , Skyler Weaver 1 , Kazuki Sobue 2 , Seiji Takeuchi 2 , Koichi Hamashita 2 , Un-Ku Moon 1 1 Oregon State University, Corvallis, OR, USA 2 Asahi Kasei

680 views • 53 slides
Performance and energy optimization of concurrent pipelined applications Anne Benoit, Paul

Performance and energy optimization of concurrent pipelined applications Anne Benoit, Paul

Framework Complexity Experiments Conclusion Performance and energy optimization of concurrent pipelined applications Anne Benoit, Paul Renaud-Goud and Yves Robert Institut Universitaire de France ROMA team, LIP Ecole Normale Sup

1.18k views • 102 slides
Mapping pipelined applications with replication to increase throughput and reliability Anne Benoit

Mapping pipelined applications with replication to increase throughput and reliability Anne Benoit

Framework Complexity Practical Conclusion Mapping pipelined applications with replication to increase throughput and reliability Anne Benoit 1 , 2 , Loris Marchal 2 , Yves Robert 1 , 2 , Oliver Sinnen 3 1. Institut Universitaire de France 2.

1.01k views • 78 slides
Pipelined Multithreading Generation in a Polyhedral Compiler January 22nd 2020, IMPACT20,

Pipelined Multithreading Generation in a Polyhedral Compiler January 22nd 2020, IMPACT20,

Pipelined Multithreading Generation in a Polyhedral Compiler January 22nd 2020, IMPACT20, HiPEAC, Bologna, Italy Harenome Ranaivoarivony-Razanajato, Cdric Bastoul, Vincent Loechner University of Surasbourg and Inria Nancy Grand Est Team

721 views • 37 slides
Corporate Presentation June 2019 1 Agenda The Problem Diluent Value Loss with Pipelined

Corporate Presentation June 2019 1 Agenda The Problem Diluent Value Loss with Pipelined

Corporate Presentation June 2019 1 Agenda The Problem Diluent Value Loss with Pipelined Heavy Oil/Bitumen Egress for Western Canadian Oil Production How Rail Can Help Already Exists (Established and Operational System)

667 views • 20 slides
Pipelined Compressor Tree Optimization using Integer Linear Programming International Conference

Pipelined Compressor Tree Optimization using Integer Linear Programming International Conference

Pipelined Compressor Tree Optimization using Integer Linear Programming International Conference on Field Programmable Logic 03.09.2014 Martin Kumm, Peter Zipf University of Kassel, Germany C ONTENTS 1. Introduction to Compressor Trees 2.

461 views • 44 slides
SEU-Hardened Energy Recovery Pipelined Interconnects for On-Chip Networks A. Ejlali*, B. M.

SEU-Hardened Energy Recovery Pipelined Interconnects for On-Chip Networks A. Ejlali*, B. M.

SEU-Hardened Energy Recovery Pipelined Interconnects for On-Chip Networks A. Ejlali*, B. M. Al-Hashimi Electronics and Comp. Science * Computer Engineering Dept. University of Southampton Sharif University of Technology Southampton, UK

440 views • 23 slides
Processor Design Pipelined Processor Hung-Wei Tseng Drawbacks of a single-cycle processor

Processor Design Pipelined Processor Hung-Wei Tseng Drawbacks of a single-cycle processor

Processor Design Pipelined Processor Hung-Wei Tseng Drawbacks of a single-cycle processor The cycle time is determined by the longest instruction Could be very long, thinking about fetch data from DRAM Hardware is mostly idle 3

866 views • 47 slides
Pentium 4 Deeply pipelined processor supporting multiple issue with speculation and

Pentium 4 Deeply pipelined processor supporting multiple issue with speculation and

Pentium 4 Deeply pipelined processor supporting multiple issue with speculation and multi-threading 2004 version: 31 clock cycles from fetch to retire, 3.2 GHZ clock rate (deep pipeline allows higher clock rate) Front end decoder

692 views • 22 slides
Energy Minimization of Pipeline Processor Using a Low Voltage Pipelined Cache Vincent J. Mooney

Energy Minimization of Pipeline Processor Using a Low Voltage Pipelined Cache Vincent J. Mooney

Energy Minimization of Pipeline Processor Using a Low Voltage Pipelined Cache Vincent J. Mooney III, Krishna Palem, Jun Cheol Park, and Kyu-won Choi Georgia Institute of Technology {mooney, palem, jcpark, kwchoi}@ece.gatech.edu Outline

441 views • 26 slides
Contents Introduction Pipelined FPGA DNN accelerators Roof-line Model and optimizing

Contents Introduction Pipelined FPGA DNN accelerators Roof-line Model and optimizing

FPGA implementation of Quantized (Deep) Neural Network (QNN) Accelerators Biruk Seyoum Phd Student, Scoula Superiore SantAnna, Pisa Contents Introduction Pipelined FPGA DNN accelerators Roof-line Model and optimizing FPGA DNN

587 views • 23 slides
ReCPU: a Parallel and Pipelined Architecture for Regular Expression Matching Marco Paolieri,

ReCPU: a Parallel and Pipelined Architecture for Regular Expression Matching Marco Paolieri,

ReCPU: a Parallel and Pipelined Architecture for Regular Expression Matching Marco Paolieri, Ivano Bonesana Marco D. Santambrogio ALaRI, Faculty of Informatics Dipartimento di Elettronica e Informazione University of Lugano, Lugano,

312 views • 6 slides

More recommend