Cross Coupling Between Power and Signal Traces on Printed Circuit Boards

Similar documents
DEPARTMENT FOR CONTINUING EDUCATION

Chapter 12 Digital Circuit Radiation. Electromagnetic Compatibility Engineering. by Henry W. Ott

Chapter 16 PCB Layout and Stackup

10 Safety earthing/grounding does not help EMC at RF

Experimental Investigation of High-Speed Digital Circuit s Return Current on Electromagnetic Emission

Designing Your EMI Filter

X2Y versus CM Chokes and PI Filters. Content X2Y Attenuators, LLC

Learning the Curve BEYOND DESIGN. by Barry Olney

Introduction to Electromagnetic Compatibility

CHAPTER 6 EMI EMC MEASUREMENTS AND STANDARDS FOR TRACKED VEHICLES (MIL APPLICATION)

EMI. Chris Herrick. Applications Engineer

EMC Simulation of Consumer Electronic Devices

Understanding and Optimizing Electromagnetic Compatibility in Switchmode Power Supplies

The Impact Of Signal Jumping Across Multiple Different Reference Planes On Electromagnetic Compatibility

Top Ten EMC Problems & EMC Troubleshooting Techniques by Kenneth Wyatt, DVD, Colorado Springs Rev. 1, Feb 26, 2007

The number of layers The number and types of planes (power and/or ground) The ordering or sequence of the layers The spacing between the layers

AN IMPROVED MODEL FOR ESTIMATING RADIATED EMISSIONS FROM A PCB WITH ATTACHED CABLE

EMC Overview. What is EMC? Why is it Important? Case Studies. Examples of calculations used in EMC. EMC Overview 1

LISN UP Application Note

FLTR100V20 Filter Module 75 Vdc Input Maximum, 20 A Maximum

Frequently Asked EMC Questions (and Answers)

ELEC Course Objectives/Proficiencies

Top Ten EMC Problems

Signal and Noise Measurement Techniques Using Magnetic Field Probes

Improve Performance and Reliability with Flexible, Ultra Robust MEMS Oscillators

SC5407A/SC5408A 100 khz to 6 GHz RF Upconverter. Datasheet. Rev SignalCore, Inc.

FISCHER CUSTOM COMMUNICATIONS, INC.

Harmonizing the ANSI-C12.1(2008) EMC Tests. Harmonizing the ANSI-C12.1(2008) EMC Tests

Heat sink. Insulator. µp Package. Heatsink is shown with parasitic coupling.

Course Introduction. Content: 19 pages 3 questions. Learning Time: 30 minutes

Common myths, fallacies and misconceptions in Electromagnetic Compatibility and their correction.

Comparison of Various Numerical Modeling Tools Against a Standard Problem Concerning Heat Sink Emissions

esa Space Station Electromagnetic Emission and Susceptibility Requirements International Space Station Revision C May 31, 1996 SSP Revision C

SAW Filter PCB Layout

FLTR75V05 Filter Module 75 Vdc Input Maximum, 5 A Maximum

Microcircuit Electrical Issues

Section 11: Power Quality Considerations Bill Brown, P.E., Square D Engineering Services

Modeling of EM1 Emissions from Microstrip Structures with Imperfect Reference Planes

Advanced Transmission Lines. Transmission Line 1

Predicting and Controlling Common Mode Noise from High Speed Differential Signals

FLTR100V10 Filter Module 75 Vdc Input Maximum, 10 A Maximum

Comparison of IC Conducted Emission Measurement Methods

A VIEW OF ELECTROMAGNETIC LIFE ABOVE 100 MHz

Designing external cabling for low EMI radiation A similar article was published in the December, 2004 issue of Planet Analog.

Electromagnetic Compatibility

SIMULATION of EMC PERFORMANCE of GRID CONNECTED PV INVERTERS

NI PXI/PXIe-2532 Specifications

Verifying Simulation Results with Measurements. Scott Piper General Motors

Oversimplification of EMC filter selection

BASIS OF ELECTROMAGNETIC COMPATIBILITY OF INTEGRATED CIRCUIT Chapter VI - MODELLING PCB INTERCONNECTS Corrections of exercises

BIRD 74 - recap. April 7, Minor revisions Jan. 22, 2009

SC5307A/SC5308A 100 khz to 6 GHz RF Downconverter. Datasheet SignalCore, Inc.

Todd H. Hubing Michelin Professor of Vehicular Electronics Clemson University

EMC problems from Common Mode Noise on High Speed Differential Signals

An Introduction to FFT EMI Receivers

Cable Solutions for Servo and Variable Frequency Drives (VFD)

An Introduction to Spectrum Analyzer. An Introduction to Spectrum Analyzer

LM1044 Analog Video Switch

Reducing Motor Drive Radiated Emissions

Using Measured Fields as Field Sources in Computational EMC

NI PXI/PXIe-2541 Specifications 300 MHz Ω Matrix

COMBILOG ANTENNA MODEL AC MHz. rev: 0202

Understanding the Unintended Antenna Behavior of a Product

Techniques to reduce electromagnetic noise produced by wired electronic devices

EMC Immunity studies for front-end electronics in high-energy physics experiments

An Investigation of the Effect of Chassis Connections on Radiated EMI from PCBs

Features. Future Electronics (

Automotive EMC. IEEE EMC Society Melbourne Chapter October 13, 2010 By Mark Steffka IEEE EMCS Distinguished Lecturer

Sources of transient electromagnetic disturbance in medium voltage switchgear

Filters And Waveform Shaping

Technical Criteria for the Accreditation Of Electromagnetic Compatibility (EMC) And Radio Testing Laboratories

MEASUREMENTS OF COUPLING THROUGH BRAIDED SHIELD VIA NEW CONDUCTED IMMUNITY TECH- NIQUE

Radiation from a PCB with Coupling between a Low Frequency and a Digital Signal Traces

A Comparison Between MIL-STD and Commercial EMC Requirements Part 2. By Vincent W. Greb President, EMC Integrity, Inc.

Product Description. Theory of operation

~W~~~ Laboratory Accreditation Program

NI PXI-2532 Specifications

The water-bed and the leaky bucket

Freescale Semiconductor, I

EET 223 RF COMMUNICATIONS LABORATORY EXPERIMENTS

Common myths, fallacies and misconceptions in Electromagnetic Compatibility and their correction.

WT11I DESIGN GUIDE. Monday, 28 November Version 1.1

2620 Modular Measurement and Control System

About the High-Frequency Interferences produced in Systems including PWM and AC Motors

EMC cases study. Antonio Ciccomancini Scogna, CST of America CST COMPUTER SIMULATION TECHNOLOGY

Decoupling capacitor uses and selection

Technical Report Printed Circuit Board Decoupling Capacitor Performance For Optimum EMC Design

Features. Haltronics Ltd (

Parallel Resonance Effect on Conducted Cm Current in Ac/Dc Power Supply

Standardisation and Immunity Tests regarding IEMI

GATE: Electronics MCQs (Practice Test 1 of 13)

TECHNICAL REPORT: CVEL Parasitic Inductance Cancellation for Filtering to Chassis Ground Using Surface Mount Capacitors

Overview of EMC Regulations and Testing. Prof. Tzong-Lin Wu Department of Electrical Engineering National Taiwan University

VSD cables in. Working with. industrial & automation applications

A NEW COMMON-MODE VOLTAGE PROBE FOR PREDICTING EMI FROM UNSHIELDED DIFFERENTIAL-PAIR CABLES

VLSI is scaling faster than number of interface pins

COOLTUBE Radiated Emissions Absorber

Regarding RF Isolation for small Enclosures

PCB Design Guidelines for Reduced EMI

Transfer Functions in EMC Shielding Design

Transcription:

Cross Coupling Between Power and Signal Traces on Printed Circuit Boards Dr. Zorica Pantic-Tanner Edwin Salgado Franz Gisin San Francisco State University Silicon Graphics Inc. Silicon Graphics Inc. 1600 Holloway Ave. 2011 N. Shoreline Blvd., M./S 946 2011 N. Shoreline Blvd., M/S 946 San Francisco, CA 94132 Mountain View, CA 94094 Mountain View, CA 94094 Abstract Switching power supplies and other power switching circuits produce high frequency components in currents on power traces and power grounds in printed circuit boards (PCBs). These time-varying currents can couple energy (noise) to low level signal traces such as those normally encountered with the new low voltage high speed logic families and microprocessors. This can cause the system to malfunction or result in EMI emissions that exceed the levels allowed by regulatory agencies. [l] The effect of proximity and position of power switching circuits with respect to the signal traces will be examined. Sources of DC-DC Converter Noise Many state of the art microprocessors and ASIC integrated circuits are now powered by voltages that are less than the standard 5 VDC. Current technologies operate in the 3-4 volt region, and future devices will operate at voltages in the 2-3 volt range. With this reduced operating voltage, the susceptibility of the signal lines to noise increases. At the same time, these new operating voltages require on-board DC to DC converters that generate a substantial amount of switching noise. This switching noise can couple onto the signal lines, causing the system to malfunction. The noise generated can also have substantial harmonic content that can excite resonant structures within a system causing it to exceed EMC conduced and radiated emission limits. Normally, power is distributed via power and ground planes in printed circuit boards. But because many of the circuits connected to these low voltage microprocessors and ASCII still operate at a nominal 5 volts, a single power plane will not work in the newer dual-voltage systems. Adding an extra low voltage plane adds cost, so this is not always a viable solution. The lowest cost solution is to route the low voltage power for the micro-processors and ASICs through wide traces on the same layers that contain signals, which increases the coupling efficiency between the power traces and adjacent signal traces. In such cases, one can use traditional trace to trace crosstalk coupling models to analyze the problem. [2] As an example, consider the typical printed circuit board DC- DC converter shown in Figure 1. The radiated emissions from this converter can couple onto I/O traces that in turn are connected to long interface cables routed external to the shielded enclosure. If a sufficient amount of common mode noise is coupled onto these traces, the resultant emissions can exceed allowable limits. Referring to Figure 1, one can see that there is a noticeable resonance effect at about 170 MHz where the emissions are 10-20 db higher. The noise generated by this DC-DC converter is shown in Figure 2. The switching frequency, F,, is about 1.2 MHz ( period, T = 833 nsec). Figure 3 shows an expansion in time of a single impulse, showing it to be a ringing type wave form that can be approximated by a damped sinusoid. Modeling the coupling Mechanism Figure 4 shows the model that was used to analyze different parameters of a representative coupling configuration. A power trace and two signal traces are located over a ground plane. The default parameters of the model are as follows: Substrate Thickness: 0.060 Power Trace Width: 0.050 Sig Trace Widths: 0.010 Separation (power to T 1): 0.010 Separation (Tl to T2): 0.010 Dielectric Constant: 4.5 Unless otherwise stated, the termination impedances on all traces were matched to the applicable trace characteristic impedance. FDTD and MOM modeling methods were used to analyze the structure to determine what effects different physical and electrical parameters had on the amount of noise coupled from the power trace to the signal traces.[3] [4] Figure 5 shows the effect that termination impedance has on the crosstalk level. When source and load terminations are not matched to the characteristic impedance of the signal trace, noise coupled onto the signal traces can ring back and forth along the length of the trace. Using a simple trapezoidal pulse to illustrate this phenomena, one can see from Figure 5 that the induced noise is almost twice as high when unmatched termination impedances are used. Figure 6 shows the effect that trace separation has on the crosstalk level. As in previous example, a simple trapezoidal pulse is used to illustrate this phenomena. Referring to Figure 0-7803-5015-4/98/$10.00 1998 IEEE 624

6 one can see that the trace closes to the power trace has more noise induced than the trace that is farther away. Figure 7 shows the effects of frequency. In this case, a sine wave is used for the driving function since the effects of frequency are easier to show. Referring to Figure 7, one can see that as the frequency is increased from 20 MHz to 50 MHz, the amount of noise that is coupled onto the signal traces increases. In real life, however, the DC-DC switching noise does not always have a simple pulse or sinusoidal shaped envelope. For example, the DC-DC converter shown in Figure 1, exhibited a damped sinusoidal shaped envelope. The spectral response of such a waveform exhibits a resonance where the spectral energy is significantly higher than at other frequencies. Such a waveform can be mathematically represented by a damped sinusoid whose voltage, V,, can be expressed as References [l] For example, FCC in the United States and EN regulations in Europe. [2] Analysis and Modeling of Power Distribution Networks and Plane Structures in Multichip Modules and Pi%,, Frank Y. Yan, Timothy K, Postel, and Lawrence M. Rubin, 1995 IEEE International Symposium on Electromagnetic Compatibility. [3] LC FDTD Modeling Code (see http://www.cray.com/lc/ for further details) [4] INCASES EMC Workbench [5] Fourier Transforms and Their Physical Applications, D. C. Champeney, 1973, Academic Press, Inc. Y =Ae(%in(w n r t) (Em 1) Referring to Figure 8a, the values of the constants were determined to be A = 0.75, c1 = 50*106, and w, = 2 n: 170*106. The sin term produces a natural resonant frequency near 170 MHz. The Fourier transform of Eqn 1 is shown in Figure 8b, and clearly shows the resonant peak. [5] In this case, a simple RLC resonant shunt circuit, tuned to 170 MHz can be used to filter out the noise at this frequency. In Figure 8b, the spectrum of the damped sinusoid (peak) and the impedance of the shunt RLC filter are overlayed, showing how the two can cancel each other out. Figure 8c shows the modeling results when the properly tuned RLC filter is inserted into the model. In this case, the actual damped sinusoid model is used as the noise source on the power trace rather than the simplified pulse or sinusoids used earlier. Referring to Figure 8, one can see that with the notch filter, the amount of noise coupled from the power trace to the signal trace is reduced significantly. Summary and Conclusions Based on the analysis described herein, crosstalk between power and adjacent signal traces can be reduced using one or more of the following techniques: - Matching source and load termination impedances - Increasing source-to-susceptor separation distance - Decreasing coupling length - Band-limiting (filtering) the frequency of the noise source 625

Figure 1: Radiated Emissions from DC-DC Converter Switching Noise Coupling onto I/ 0 Traces Figure 2: Switching Transients on Output Figure 3: Detail of Switching Transients Figure 4: Model used to Evaluate Coupling Mechanisms 626

Power Tl T2 Figure 5: Crosstalk Effects when Unmatched Terminations are used Figure 6: Effects of Trace Separation on Crosstalk Figure 7: Effects of Increasing Source Frequency 627

! AsMat e &t( Figure 8: A shunt RLC Notch Filter between the power trace and ground that is tuned to w can be used to filter out the damped sinusoid noise