POLARITY IN ABSOLUTE TERMS

The Wood Effect, by R. C. Johnsen. Modern Audio Association, 100 pp., softbound, $7.95. (Available from The Modern Audio Association, 23 Stillings St., Boston, Mass. 02210.)

The main theme of this book deals with absolute polarity, a very important aspect of sound reproduction which has been much neglected. Because most natural sounds are nonsymmetrical, there is a correct, "true to nature," absolute acoustical polarity of sound.

Natural "live" sound is always heard with the correct acoustical polarity, but when it is picked up by a microphone, amplified, and reproduced by a loudspeaker, it can be reversed. This can be demonstrated with a single-channel system by merely connecting the loudspeaker leads one way or the other; a stereo system would require that both speakers have their leads reversed.

Because there has never been an official standard to control this aspect of sound reproduction, it is easy to see how this reversal of acoustical polarity is completely random. Although the author of this book is mainly concerned about the fact that every electrical recording ever made is either one polarity or the other, the same problem is present in every sound system, even those used for concert sound.

Chapter 1 begins with a brief history of the recording industry, a long quote from an R.I.A.A. brochure, and a shorter quote from the book From Tin Foil to Stereo, by Read and Welch. It shows the many influences which have driven the recording industry. Near the end, Johnsen states his major premise, "Only one concept must be grasped: Electricity can reverse its phase, while music cannot." The problem with this statement, from my viewpoint, is that the author uses the word "phase" instead of "polarity" to describe what can happen to the electrical signal. He does this elsewhere in the book also, which can tend to confuse the very issue that he is trying to clarify. The problem is one of terminology, not of substance. Years ago, during a discussion of this problem, someone asked, "How can we get people to pay attention to absolute phase?" A friend of mine, Ron Wickersham, replied, "When we stop calling it absolute phase and start calling it absolute polarity." When I saw the quizzical look on the face of the man who had asked the question, I said, "Phase is frequency dependent; polarity is not." Polarity is a universal concept--something is either plus or minus, positive or negative, up or down, black or white. There is no in-between or gray area. Phase can be anything from zero to 360° at a given frequency. (Editor's Note: Or even more if the delay is more than a wavelength; again the difference between a repetitive sinewave and music. -E.P.) In Chapter 2, the author lists 10 effects which can be heard when a recording is auditioned with the correct absolute polarity. He also cites two different reviews of the same record, which contain statements about the quality of the sound. Johnsen traced the two reviewers' comments to the different absolute polarities of the records: "'Muffled' was the word both reviewers instinctively and revealingly applied to their reversed-phase conditions." Johnsen also makes the case for correct absolute polarity by comparing photographic negatives and positives: "Negatives are hard to evaluate visually, although all information is present." Chapter 3 consists of excerpts from a large number of journals and magazines which discuss the merit of the concept of maintaining the absolute polarity of the signal; the term absolute phase, however, is used interchangeably here. The title of the book is explained by the description of an experiment conducted by Charles Wood in 1957. Wood used a sinusoidal signal which was clipped on one side only, making it nonsymmetrical. He noticed that the timbre changed when the headphone leads were reversed. This finding led to further investigation by the respected researchers James Craig and Lloyd Jeffress, which appeared in the November 1962 issue of The Journal of the Acoustical Society of America. The trail of comments in print about the audibility of absolute polarity is followed by Johnsen all the way to 1984.

The next three chapters are very short. The lack of a standard for absolute polarity in professional recording equipment is discussed in the four pages of Chapter 4. Chapter 5 is basically an attack on the Audio Engineering Society for not setting appropriate standards for absolute polarity. Chapter 6 presents the author's own experiences in tracking down examples of polarity confusion on various recordings, as well as some anecdotal evidence for the importance of listening to music with the correct polarity. Information about the differences between the way AM and FM radio broadcasters handle the polarity problem sheds light on why correct polarity is easy to hear on AM and almost impossible to discern on FM.

Chapter 7 begins with revelations about the effects of polarity reversal on radio and TV broadcasts, concert sound reinforcement, and even the sale of equipment. The next 13 pages provide an interesting investigation of recordings--from 78s, through LPs, to CDs--with comments about the sound from various published sources. A section is devoted to the audibility of the effects upon music reproduction of the polarity of the a.c. power line, with comments by a number of writers.

Johnsen includes a story in which he spent three hours with the late Richard Heyser, discussing whether there could be a preferred a.c. power-plug polarity that would make an audible improvement in the sound. It appears that Richard Heyser didn't think so, but the passage might leave the reader with the idea that Heyser wasn't aware of the audible effects of absolute polarity on the perceived sound. I can attest to the fact that he was very aware of the importance of maintaining absolute polarity--at least as far back as 1979, when we were both at an A.E.S. meeting. When I got up to speak, I said that I thought the absolute polarity of the system was reversed and saw Heyser nod in agreement. I then spoke into the microphone while someone reversed the leads to the loudspeaker system and, sure enough, the reaction of the audience showed me that Heyser and I had been correct; the polarity had been reversed. (Editor's Note: I can attest to the fact that Heyser was aware of absolute polarity and its importance in recording and in testing prior to 1973, when he began testing speaker systems for this magazine. Indeed, the seminal 1967 JAES paper on time-delay spectrometry makes the distinction between polarity and phase plainly. -E.P.) Although I agree with the major premise of this book-that absolute polarity is extremely important--I must say that pages 67 to 74 are full of erroneous and misleading information about loudspeakers. Unfortunately, it is a case of trying to explain, in technical terms, why some loudspeakers behave the way they do, without having the expertise required to do so. Since the author is not a loudspeaker system designer and quotes the writings of others who are not designers either, perhaps this is excusable. Most speaker designers also have difficulty with such concepts as minimum phase, linear phase, phase delay, phase deviation, phase alignment, group delay, etc. Indeed, the design of a coherent loudspeaker is not a trivial task--even for those who are aware of what they are doing.

The last two chapters are an odd mixture and, therefore, difficult to describe. There is an interesting list of recordings, each marked with the author's own polarity convention, which is relative to the first record for which he determined the correct polarity. It would have been better if he had determined the absolute polarity of his own system before he began marking his collection. As it is, his "normal" and "reverse" designations might be reversed! Oh well, at least they are consistent, which is more than the whole audio industry can say for itself. The Epilogue contains additional press comment on the importance of absolute polarity, while the Appendix consists of tape-recorded comments about absolute polarity, from exhibitors at the 1987 summer C.E.S. The tone of Johnsen's book is rather quixotic, and I don't think the author will mind my saying so. Rather, I suspect he will take this as the compliment it is meant to be. This is a potentially controversial book, and it is quite clear the author intends it to be so. I found it fun to read.

-Edward M. Long

==============

Loudspeaker and Headphone Handbook, edited by John Borwick. Butterworth & Co., 573 pp., hardbound, $97.50.

The years 1987 and 1988 have seen the publication of some very important books on sound. In 1987, there was a new edition of Sound System Engineering by Don and Carolyn Davis, then the Handbook for Sound Engineers edited by Glen Ballou, which is subtitled "The New Audio Cyclopedia." Then in 1988 came the Audio Engineering Handbook edited by K. Blair Benson, and now we have the Loudspeaker and Headphone Handbook edited by John Borwick. Although the book reviewed here was published in England and most of the contributors are English, there are also contributors from Austria, Canada, and the U.S. The book is divided into 14 chapters and includes an Index. As the reviewer's favorite cliché says, "It is profusely, illustrated," with many charts, graphs, tables, and schematics. Because each chapter covers a rather broad topic, there is bound to be some overlap, but I consider this an advantage. For example, when I looked in the Index for "Positioning of loudspeaker," I found four different pages listed. These pages are in Chapters 3, 7, 10, and 11, which were written by four different authors. (Each author touches upon different aspects of the positioning of a loudspeaker.) Another advantage of having different authors cover the 14 different sections of the book is that each is able to concentrate on a subject and present it in great detail. Every chapter includes references; Chapters 4 through 9 also include a bibliography.

Chapter 1, written by R. D. Ford of the University of Salford, England, is titled "Principles of Sound Radiation." Ford begins with a brief explanation of such topics as sound waves, loudness and hearing, sound pressure, and decibels. The following sections-on sound radiation from a simple source and the relationship of sound intensity and power-are explained with mathematical equations. Radiation from a flat, rigid, circular piston in an infinite baffle is covered graphically and mathematically. Radiation from a rectangular source is similarly treated, and then the author covers the acoustic impedance of an enclosure. A section on the radiation from multiple drivers has graphic examples as well as the appropriate mathematical formulas. The section on horns left me wishing that more information had been included.

Formulas are given only for the exponential horn, although a graph of the acoustical resistance and reactance versus frequency for parabolic, conical, exponential, and hyperbolic shapes are shown. The author argues in favor of the exponential flare as being probably the best compromise as well as easy to analyze mathematically. He also includes a brief but effective discussion of constant-directivity horns, but there is only one reference cited. Since constant-directivity horns have become so popular, more references on the subject would have greatly enhanced this section. The last section presents electrical circuit analogs of a loudspeaker driver.

I consider Chapter 2, written by the legendary Stanley Kelly, worth the price of the book all by itself. It covers transducer drive mechanisms and their effects in tremendous detail, and includes information not found in other books. It begins with an historical treatment of the subject and includes such original designs as the Western Electric WE555W horn driver and the legendary lonophone, complete with construction details and even the schematic for the lonophone high-frequency oscillator and power supply. The next section has a chart and diagram which show the relationship between electrical, mechanical, and acoustical terms and their schematic representations, followed by an explanation of the Helmholtz resonator, which is the basis of all bass-reflex or ported enclosures.

The next two sections cover direct-radiator diaphragms and motors in great detail, showing the effects of various design parameters on performance.

Horn drivers are well covered in the next section, while the last section contains excellent information about ribbon speakers. This is not unusual because the author is the world's foremost expert on this type of design. The ribbon driver--which has become popular in the last few years, especially in audiophile systems--is not even included in most other books.

The next chapter, by designer and consultant Peter Baxandall, is a must for anyone interested in electrostatic loudspeaker design. Electrostatic drive theory is covered in extreme detail, with appropriate schematic representations, graphic relationships, and mathematical formulas. Radiation characteristics are also extensively covered.

Practical designs are examined next, with the Quad Mark I and ESL-63 used as examples. Baxandall acknowledges the assistance of Peter Walker, the designer of these two legendary ESLs, so the information can be considered not only reliable but very practical as well. In fact, the practical considerations are covered so thoroughly that I am left with admiration for anyone who has produced a successful ESL system.

Multiple loudspeaker system design is covered in Chapter 4 by Laurie Fincham of KEF Electronics. The first section considers the design of multiple-driver systems from a theoretical viewpoint and deals mainly with the target-function approach, which requires that driver and filter characteristics be considered together to arrive at a desired acoustical-output filter shape. Both amplitude and phase characteristics of filter functions are considered. The main crossover types used as examples are the Butterworth, Linkwitz-Riley, and time-delay derived. These are described in terms of their transfer functions, and they are used again later, in the section on practical design procedures. This section would have been more useful if it had shown the method to be used to generate actual parts' values for crossovers. It does give a schematic representation, used to model a speaker driver, which should be used as a load for the crossover, rather than a simple resistive dummy load, as shown in other books on crossover design. The author cautions that in order to obtain the desired acoustical target function for a driver/ crossover combination, the crossover network must also provide for driver-response equalization.

Chapter 5, written by the famous equipment reviewer and designer Martin Colloms, covers from a number of angles the amplifier/loudspeaker interface and its effects upon performance.

Included are typical dynamic and electrostatic speakers with data on impedance and phase versus frequency. Zobel compensation for the rise in dynamic-driver impedance is demonstrated. The graph of impedance versus frequency is shown for a commercial loudspeaker system, which also uses conjugate impedance compensation. The crossover schematic, complete with parts' values for this system, makes clear how complicated such compensation can be. Colloms also presents a case for using the amplifier output resistance, rather than the damping factor, as a criterion. He also shows that, from a loudspeaker designer's or user's standpoint, it would make more sense to rate amplifiers on a system of available level in dB than power in watts. He then covers active loudspeakers-that is, those with built-in amplifiers. Next, he describes motional feedback and digital loudspeaker concepts. The last section is devoted to cables, fuses, protection devices, and connectors. The fact that the resistance of cable used to connect the loudspeaker to the amplifier can effect the amplitude-versus-frequency response is shown graphically.

Chapter 6, "Loudspeaker Enclosures" by Desmond Thackeray of the University of Surrey, is rather brief and very general. Certain aspects of closed-box and vented enclosures are covered, but readers seeking a guide for designing enclosures will have to look elsewhere. Formulas for the exponential and the hyperbolic exponential, however, are included.

Chapter 7, "The Room Environment" by Glyn Adams of the University of Sydney, has an excellent discussion of standing waves, room modes, and reverberation. The section on speaker placement shows the interaction with room boundaries, the effect of room modes, and even includes a discussion of stereo imaging. The discussion of measuring room acoustics is rather basic and shows only reverberation and third-octave techniques, although the use of time-delay spectrometry, developed by Richard Heyser, is mentioned. A few pages are devoted to listening room design, including the use of sound-absorber panels. The last section touches briefly upon the use of equalization.

Chapter 8, "Sound Reinforcement and Public Address" by Peter Mapp, begins by making a distinction between sound reinforcement and public address, with distributed and central loudspeaker systems covered. A section on auditorium systems is followed by a discussion of the Haas effect, as well as the findings of Wallach and others regarding the precedence effect. Mapp also covers response shaping, speech intelligibility, and climatic effects. Sound masking systems, reverberation systems, electronic architecture, and cinema systems are all dealt with in a clear fashion.

Chapter 9, "Loudspeakers for Studio Monitors and Musical Instruments," was written by Mark Gander of JBL. He begins by laying out a number of performance requirements for studio monitors. A section is devoted to "significant monitor designs" and includes this reviewer's contribution to the genre, the UREI 813 Time Align monitor, as well as monitors by JBL, B & W, and Tannoy. An excellent section follows and is devoted to the construction of high-powered, musical-instrument loudspeaker drivers. Many details are made clear by diagrams and photographs. The chapter ends with a discussion of speaker enclosures that includes photos and drawings.

Chapter 10, "Loudspeaker Measurements," was written by the book's editor, John Borwick. He first lays out the important parameters which should be measured and then mentions the published standards. Borwick then discusses the measurement environment (i.e., free-field, diffuse-field, etc.) before covering other test conditions which should be taken into account.

Various test methods are then discussed, including: Continuous sine wave, using either small, discrete steps or slow sweep; time-delay spectrometry; gated tone burst; impulse, and random noise. Various speaker parameter measurements are covered, including directional response, sensitivity, efficiency, impedance, and large-signal distortion.

Chapter 11, "Subjective Evaluation," covers its topic in great detail. It was written by Floyd Toole, who has become an acknowledged expert in this area. Toole discusses various aspects, including the room, program material, selection of listeners, procedures, and rating schemes.

Chapter 12, "Headphones" by C. A. Poldy of AKG Acoustics, is a small book in itself. Since in-depth information on headphones is so difficult to find, this chapter alone is extremely valuable. Just about every aspect of 'phones is covered in detail, including the different types of transducer elements, the effects of ear cushions, and sound insulation. The hearing mechanism and the many aspects of out-of-head localization are discussed, including left-right, front-back, cone of confusion, elevation effect, and more.

The problems of defining a standard for headphone measurement are made very clear, as are the problems of testing headphones, covered at the end of this chapter and followed by a section containing 178 references! Phase effects, binaural recording and reproduction, artificial heads, compatibility with loudspeaker listening, and blending circuits are also well covered.

Chapters 13 and 14, "International Standards" and "Terminology," both by J. M. Woodgate, discuss the various measurement standards published by the I.E.C. and by various countries and give definitions for the terms used in this book.

To conclude, I must say that editor John Borwick has done an excellent job and can be justly proud. This book should remain a valuable asset for years to come. I recommend it to anyone interested in speakers, especially those who want to know more about headphones.

--Edward M. Long

(adapted from Audio magazine, Aug. 1989)

= = = =