Spectrum and Network Measurements, 297 Pages

Author: Robert A. Witte
Publisher: Noble Publishing, 2001
Originally Published: 1993 by Prentice-Hall

Most EMC engineers are familiar with spectrum analyzers as they use them to verify product compliance to emission requirements. However, the spectrum analyzer is a versatile troubleshooting tool to the EMC engineer if the spectrum analyzer capability and limitations are understood. This also applies to the network analyzer, which is often used to measure various characteristics of two port networks. However, many engineers have learned how to use these instruments by using one or some combination of the following methods: (1) reading the voluminous instruction manual, (2) watching and later imitating a fellow technician or engineer, or (3) reading application notes, provided they exist.

Some of the problems with this approach to spectrum and network measurement are: (1) The initial investment of time to read the instruction manual far out weighs the immediate short term goal of making a measurement, (2) the person whose ÒexperienceÓ you are depending upon may have limited capability, and (3) the versatility and limitations of the instrument may be unknown to the user.

This book helps to bridge the gap between theory and practice to enable a better understanding of how spectrum and network measurements relate to theory. Most instruction manuals assume some minimal level of knowledge of specific instrument types, so they are usually very light on the theory of operation. This book describes spectrum, wave, network and FFT analyzers. It covers the following topics related to the FFT analysis: coherence, correlation, cross correlation and auto correlation. These topics are explained in a practical sense, with just enough mathematics, something that is missing when you are first introduced to this topic in college.

This book consists of 17 chapters, which I have arbitrarily broken down into six parts. The first part is comprised of the first three chapters. Introductory material on spectrum and network measurements, decibels and Fourier Theory are covered. The brief review of Fourier Theory bridges the gap between theoretical calculations and how to relate these to actual measurement values. It compares the Fast Fourier Transform (FFT) with the Discrete Fourier Transform (DFT) and how it is implemented into analyzers. The FFT is computationally more efficient than the DFT as the FFT requires less than 1% of the computation time that the DFT requires for the same record length.

The second part is made up of chapters 4 and 5, which are devoted to FFT analyzers and swept spectrum analyzers. Basic theory of operation, instrument architecture, window functions and FFT functions are covered in these chapters. The FFT window functions covered are hanning, flat top, uniform and exponential. Other FFT functions such as coherence, auto correlation and cross correlation are covered. Equations for computing the minimum sweep rate for specific spectrum analyzer settings are given. If the selected sweep rate does not meet the criteria of the equation (frequency span and resolution bandwidth), the measurement will be in error. Other areas covered are: LO feed thru, detectors, tracking generators, differences between FFT analyzers, swept analyzers and hybrid analyzers.

Chapters 6 to 10, which are the third part, are basic theory and descriptive material for the various measurements that a spectrum analyzer is used for: modulation, distortion, noise and pulsed power. The role of averaging and filtering on measurement sensitivity and accuracy are also covered. Amplitude modulation and frequency modulation measurements are described in the time and frequency domain. Useful equations to calculate and measure the modulation indices for AM and FM are presented. The spectrum analyzer zero span function can be used to examine the modulation characteristics of a carrier in the time domain. This chapter on modulation also explains Bessel functions, Carson's Rule and carrier nulls in a practical manner.

The chapter on distortion clearly explains the effects of distortion in amplifiers and how to measure amplifier distortion. The fundamental concepts of gain compression, second and third order intercept points, as well as second and third order distortion products are clearly illustrated. It describes how to perform single tone and two-tone distortion measurements, which are useful concepts to understand when performing CS103 (Intermodulation), and CS105 (Cross-Modulation).

The chapter on noise and noise measurements is a brief introduction to noise theory. Theory is reviewed before more practical information is presented. Noise units, noise measurements, noise floor and noise floor corrections are explained. Phase noise, phase noise terminology and phase noise measurements are briefly introduced. For those that require more detailed information, a list of references is presented.

The chapter on pulse power measurements presents useful equations to determine the optimum spectrum analyzer settings (resolution BW, sweep time) provided the duty cycle and PRF of the signal are known. The average power of the pulsed signal can also be calculated. The amplitude of pulsed signals, due to the pulse desensitization factor, are actually higher than measured. The pulse desensitization factor and how to calculate it are presented. The role of averaging and filtering to enhance measurement sensitivity and accuracy is discussed. The role resolution bandwidth, video bandwidth, predetection and post-detection filtering play on accuracy and signal stability is described.

The fourth part of the book, chapters 11 and 12, examines the effect transmission lines and measurement connections have on the measurement accuracy of spectrum analyzer measurements. Those with limited knowledge often ignore these effects and this can result in erroneous measurement values. Basic transmission line theory is reviewed. Also presented are the fundamental concepts of characteristic impedance, reflection coefficient, return loss, standing wave ratio and how these items are related. The role that transmission line characteristics play in measurement accuracy is quantified and specific examples are given. The loading effect of measurement connections to the unit under test is described for some common connection devices such as high impedance probes, attenuating probes, high impedance inputs, characteristic impedance inputs and input connectors. Some rules of thumb are presented along with some of the common limitations of each connection type.

Two port network theory, network analyzers, transmission measurements and reflection measurements are covered in the fifth part of the book in chapters 13 to 16. The concept of a transfer function applied to a two-port network is developed prior to the presentation of the more common network parameters. The network parameters described are Z, admittance, hybrid, transmission and scattering. The basic concept of a network analyzer is illustrated from the simplistic model of an oscilloscope with a sweep generator to the more sophisticated spectrum analyzer with a tracking generator configuration. The S-parameter test set is described. The minimum sweep speed required for accurate network analyzer measurements is described. Advanced network analyzer functions such as amplitude sweep, gain compression point, and frequency offset are described.

The chapter on transmission measurements illustrates the following concepts: insertion loss and gain, phase error, group delay, line stretch, measurement plane, normalization, and how they must be used for accurate measurements.

Reflection measurements, as described in chapter 16, are the measure of the return loss or reflection coefficient of a two-port network. The concepts of directivity, coupling factor, directional coupler error model, error correction, two-term error correction, and three-term error correction are explained. The use of directional couplers and how the directivity specification limits the accuracy and limits the overall dynamic range of the measurement are discussed.

The final part of the book, chapter 17, covers analyzer performance and specifications. This chapter discusses how the instrument manufacturer will describe performance. This insight will aid the EMC engineer in selecting an appropriate instrument for a particular measurement type.

Closing Comments

I would recommend this book to anyone that performs spectrum and network measurements. The material presented in this book is easy to understand for those new to spectrum and network measurements. It is also useful to those already familiar with these measurements as it can help one to perform more accurate measurements by understanding how the instrument works, limitations of the measurement technique, limitations of the instrument, relating theory to actual measurements and understanding the potential impairments to accurate measurements and how to avoid them.

Ray Adams, of Hughes Space and Communications, is currently Chairman of the Los Angeles Chapter of the IEEE EMC Society. He may be reached at ray.adams@hsc.com.

EMC