A review of the book

This is a review originally published in a newsletter of the Scandinavian Vibration Society (SVIB). It is reproduced here by the kind permission of Dr. Plunt.

Review of a new textbook, Brandt A,”Noise and Vibration Analysis: Signal Analysis and Experimental Procedures”, 1st edition 2011, John Wiley and Sons (ISBN 978-0-470-74644-8).

First, it is a true pleasure to review a good textbook on noise and vibration technology written by a Scandinavian author with a strong reputation as an excellent teacher. Anders Brandt is renowned for teaching hard digested subjects, as practical measurements of random data or frequency response functions, digital signal processing and analysis, modal analysis and other advanced experimental methods, in a readily accessible manner. The book is well bound with 400+ pages of relevant content. It is accompanied by a downloadable toolbox and examples, which were however not present at the announced website at the time of this review. I got a sample of the software from the author and it is really ambitious with script variants for Matlab as well as the free Octave. In addition to student needs, it seems also to be quite valuable for practicing engineers who e.g. may want to experiment with the influence of different signal processing parameters.

Do we need another book that covers test-based noise and vibration analysis and procedures, and what is then missing in existing textbooks? As the author points out, the book is aimed as compre�hen�sive literature for university graduate courses on the subject and he hopes that it also will be a useful handbook for engineers and researchers measuring and analyzing noise and vibration signals.

The first five chapters (115 pages) are introductory content necessary for a course book and probably well known to practicing experimentalists: 1) Introduction, 2) Dynamic signals and systems, 3) Time data analysis, 4) Statistics and random processes and 5) Fundamental mechanics.

Although this is basic content, Anders Brandt makes some usually rather difficult theory easily understandable like the discussion on convolution in Chapter 2 or the thorough discussion in Chapter 3 on the fundamental properties of digital time data analysis. Although well known and subjects learnt decades ago, I actually found those chapters very entertaining to read.

A large part of the book treats measurement of frequency response functions and modal analysis. Theory of modal analysis is presented in Chapter 6, which is a well written and explained summary of the most important basics of modal analysis. It is a good introduction to experimental as well as computational modal analysis of dynamical systems with references to some classic textbooks. More advanced aspects of experimental modal analysis are saved to the last chapter of the book.

Transducers are introduced in Chapter 7 including accelerometers, force transducers, the impedance head, the instrumented impact hammer and shakers. Principles as well as calibration and mounting of accelerometers are commented. It is when I read this chapter that I start to suspect that the book will cover more of vibration than acoustic (noise) analysis. Measurement microphones are treated in less than one page and I found the first minor detail where I disagree with the author which is the short note on calibration of microphones. Latest accredited calibration values should be used and the procedure presented in the book shall be used only to check the microphone calibration. The same applies to the use of portable, handheld accelerometer calibrators.

Chapters 8-10 are a “must read” even if you think you know everything you need to know about the frequency analysis of signals. It starts ground up with the theory for periodic, random and transient signals and proceeds with a very thorough treatise on experimental, digital frequency analysis including DFT, FFT, leakage, the picket fence effect, time windows, corrections for applied windows, oversampling, circular convolution, aliasing and estimation of spectra and correlation functions in great detail. Again, the presentation is a good mix of practical consequences and theoretical details.

Chapter 11 is a short, neutral discussion about commercial measurement and analysis systems and general specifications for good test systems.

Chapter 12 treats rotating machinery analysis, and introduces the use of tachometers, order analysis and tracking including synchronous sampling, again easily understood. The Vold-Kalman filter method is shortly commented but the reader has to get more details from the references.

Frequency response measurements are treated in Chapter 13 and 14, single-input and multiple-input respectively. This is one of the very strong and detailed parts of this book and fills 65 pages. The theory and practical measurement issues are comprehensively discussed and well illustrated both using impact hammer and shaker excitation.

Chapter 15 introduces Principal Component Analysis as a technique to orthogonalize signals from different sources as well as their use for data reduction with a very nice and illustrative example of image compression by using only the largest principal components. Virtual signals and virtual coherence functions are introduced and typical use for noise source identification with an automotive example is given.

The final Chapter 16, “Advanced Analysis Methods” includes short introductions to various subjects not treated previously, like shock response spectra, the Hilbert transform, Cepstrum analysis, the envelope spectrum, operating deflection shapes and a very short introduction to experimental modal analysis methods for the identification of modal parameters.

So what are my main impressions? I think it is a very good and up to date course book for advanced signal analysis and vibration measurement classes on graduate level. Maybe it is a bit heavy for undergraduate courses. It is of course also well suited as a textbook for advanced, comprehensive courses for practicing engineers. It is also good reading for everyone that makes this type of measurements professionally.

Do I miss anything essential? Well, I think it is the special analysis techniques and procedures that need be used in acoustics. An acoustic cavity (e.g. a room) is typically a dynamic system with many overlapping modes already at relatively low frequencies. The high generic variability of detailed system properties like e.g. frequency response functions was realized by acousticians already in the 1950’s and has to be understood also today especially when properties of ensembles of “identical” objects are needed. This inherent uncertainty seems not to be well known among practicing engineers, especially those who entered the field with a structural dynamics background. This large uncertainty is the main reason why vibro-acoustic systems are preferably analyzed and tested at higher frequencies with methods different than the ones covered by this book. The book may eventually be disappointing for a reader who is more interested in noise issues than vibration analysis. Hopefully this void can be filled in a future edition of this book.

Gothenburg 2011-07-07
Juha Plunt, PhD
Principal Specialist
Muller-BBM Scandinavia AB