Audio Engineering Explained

Douglas Self

• Publisher: Routledge
• Year: 2012
• eBook ISBN: 9781136121258
• Topic: Technology & Engineering
• Subtopic: Acoustical Engineering
• Index: Technology & Engineering

All the design and development inspiration and direction an audio engineer needs in one blockbuster book! Douglas Self has selected the very best sound engineering design material from the Focal and Newnes portfolio and complied it into this volume. The result is a book covering the gamut of sound engineering. The material has been selected for its timelessness as well as for its relevance to contemporary sound engineering issues.

PART I

Basics

PREFACE

This chapter begins our journey through the world of audio engineering by explaining some fundamental principles. Our hearing, like our other senses, works in a logarithmic fashion, which allows us to cope with the enormous variation in sound levels that occur in the real world. Explanation starts with the decibel, a unit whose logarithmic basis matches the response of our ears, and makes the plotting of graphs and the doing of calculations much simpler and more intuitive than the alternative of grappling with scientific notation and numbers wreathed in zeros.

From considering the mathematics of decibels (which is really very simple), Pat Brown moves on to consider how they are used to measure absolute rather than just relative sound levels. Our ears respond to sound pressure levels. Taking the quietest sound that can be detected, and calling it the Threshold of Hearing, we can assign 0 decibels to this sound pressure level. This is invariably abbreviated to 0 dB. The decibel scale can now be used to establish the loudness of common situations, such as a room with quiet air-conditioning (about 30 dB), normal conversation (about 50 dB), and the traditional jack-hammer at 1 meter (about 100 dB). The threshold of pain is usually defined as 130 dB, a level that hopefully few of us will ever encounter; it is a sobering thought that hearing damage due to long-term exposure can begin as low as 78 dB.

Our perception of frequency, i.e., pitch, is also logarithmic, which is why music has as its foundation the octave, which stands for a doubling of frequency. When drawing graphs of quantities against frequency, it is usual to use decades, or ratios of 10, instead of octaves, as this simplifies the maths. Pat then relates frequency to wavelength and the speed of sound.

It is one of the most important features of human hearing that the frequency response of our ears is not a fl at line when plotted against sound pressure level. Worse than that, the deviation from flat increases as the level falls. This disconcerting situation is summed up by the famous Fletcher and Munson equal-loudness curves. This naturally complicates the business of measuring noise levels, and so standard weighting curves have been defined that allow for this, the best-known being the A-weighting curve, which is often used in the measurement of audio electronics. This curve has a pronounced roll-off at the low-frequency end which takes into account the fact that our hearing is markedly less sensitive at, say, 50 Hz, than it is at the reference frequency of 1 kHz. There is then a discussion of how the ear integrates sound, one example being the Precedence effect, in which late-arriving reflections within 40 milliseconds of the original sound reinforce its subjective volume rather than being heard as separate events.

Pat Brown concludes this chapter with a look at some of the simplest cases of sound propagation: the point source and the line source. These are of course mathematical abstractions rather than real sound sources, but they are the basic foundation of more realistic and therefore more complex models, such as the treatment of a loudspeaker cone as a rigid circular piston.

CHAPTER 1

Fundamentals of Audio and Acoustics

Handbook for Sound Engineers by Pat Brown

INTRODUCTION

Many people get involved in the audio trade prior to experiencing technical training. Those serious about practicing audio dig in to the books later to learn the physical principles underlying their craft. This chapter is devoted to establishing a baseline of information that will prove invaluable to anyone working in the audio field.

Numerous tools exist for those who work on sound systems. The most important are the mathematic al tools. Their application is independent of the type of system or its use, plus, they are timeless and not subject to obsolescence like audio products. Of course, one must always balance the mathematical approach with real-world experience to gain an understanding of the shortcomings and limitations of the formulas. Once the basics have been mastered, sound system work becomes largely intuitive.

Audio practitioners must have a general understanding of many subjects. The information in this chapter has been carefully selected to give the reader the big picture of what is important in sound systems. In this initial treatment of each subject, the language of mathematics has been kept to a minimum, opting instead for word explanations of the theories and concepts. This provides a solid foundation for further study of any of the subjects. Considering the almost endless number of topics that could be included here, I selected the following based on my own experience as a sound practitioner and instructor. They are:

1. The Decibel and Levels.
2. Frequency and Wavelength.
3. The Principle of Superposition.
4. Ohm's Law and the Power Equation.
5. Impedance, Resistance, and Reactance.
6. Introduction to Human Hearing.
7. Monitoring Audio Program Material.
8. Sound Radiation Principles.
9. Wave Interference.

A basic understanding in these areas will provide the foundation for further study in areas that are of particular interest to the reader. Most of the ideas and principles in this chapter have existed for many years. While I haven't quoted any of the references verbatim, they get full credit for the bulk of the information presented here.

THE DECIBEL

Perhaps the most useful tool ever created for audio practitioners is the decibel (dB). It allows changes in system parameters such as power, voltage, or distance to be related to level changes heard by a listener. In short, the decibel is a way to express “how much” in a way that is relevant to the human perception of loudness. We will not track its long evolution or specific origins here. Like most audio tools, it has been modified many times to stay current with the technological practices of the day. Excellent resources are available for that information. What follows is a short study on how to use the decibel for general audio work.

Most of us tend to consider physical variables in linear terms. For instance, twice as much of a quantity produces twice the end result. Twice as much sand produces twice as much concrete. Twice as much flour produces twice as much bread. This linear relationship does not hold true for the human sense of hearing. Using that logic, twice the amplifier power should sound twice as loud. Unfortunately, this is not true.

Perceived changes in the loudness and frequency of sound are based on the percentage change from some initial condition. This means that audio people are concerned with ratios. A given ratio always produces the same result. Subjective testing has shown that the power applied to a loudspeaker must be increased by about 26% to produce an audible change. Thus a ratio of 1.26:1 produces the minimum audible change, regardless of the initial power quantity. If the initial amount of power is 1 watt, then an increase to 1.26 watts (W) will produce a “just audible” increase. If the initial quantity is 100 W, then 126 W will be required to produce a just audible increase. A number scale can be linear with values like 1, 2, 3, 4, 5, etc. A number scale can be proportional with values like 1, 10, 100, 1000, etc. A scale that is calibrated proportionally is called a logarithmic scale. In fact, logarithm means “proportional numbers.” For simplicity, base 10 logarithms are used for audio work. Using amplifier power as an example, changes in level are determined by finding the ratio of change in the parameter of interest (e.g., wattage) and taking the base 10 logarithm. The resultant number is the level change between the two wattages expressed in Bels. The base 10 logarithm is determined using a look-up table or scientific calculator. The log conversion accomplishes two things:

1. It puts the ratio on a proportional number scale that better correlates with human hearing.
2. It allows very large numbers to be expressed in a more compact form, Figure 1.1.

The final step in the decibel conversion is to scale the Bel quantity by a factor of 10. This step converts Bels to decibels and completes the conversion process, Figure 1.2. The decibel scale is more resolute than the Bel scale.

The decibel is always a power-related ratio. Electrical and acoustical power changes can be converted exactly in the manner described. Quantities that are not powers must be made proportional to power — a relationship established by the power equation:

(1.1)

where,

W is power in watts,

E is voltage in volts,

R is resistance in ohms.

This requires voltage, distance, and pressure to be squared prior to taking the ratio. Some practitioners prefer to omit the squaring of the initial quantities and simply change the log multiplier from 10 to 20. This produces the same end result.

Figure 1.3 provides a list of some dB changes along with the ratio of voltage, pressure, distance, and power required to produce the indicated dB change. It is a worthwhile endeavor to memorize the changes indicated in bold type and be able to recognize them by listening...

Table of contents

• Cover
• Audio Engineering Explained
• Full Title
• Copyright
• Contents
• Preface
• PART I • Basics
• PART II • Microphones
• PART III • Preamplifiers , Mixers, and Interconnects
• PART IV • Power Amplifi ers
• PART V • Loudspeakers
• PART VI • Digital Audio
• PART VII • Acoustics and Sound Reinforcement
• PART VIII • Recording Studios
• Index