# Acoustical measurements

## Presentation on theme: "Acoustical measurements"— Presentation transcript:

Acoustical measurements
Iiro Jantunen Nokia Research Center S Licentiate course in measurement science and technology © Nokia Acoustical measurements.ppt / / IJ

Contents Principles of acoustics Acoustics measurements Microphone
Sound pressure level measurements Sound intensity measurements Calibration SoundField © Nokia Acoustical measurements.ppt / / IJ

Principles of acoustics
Sound waves in gas or liquid No shear forces → no transverse waves → purely longitudinal waves Audible sound range 20 Hz – 20 kHz Fully described by 3 variables Pressure Particle velocity Density © Nokia Acoustical measurements.ppt / / IJ

Wave equations of sound
Euler’s equation Newton’s 2nd law (F=ma) applied to fluid Continuity equation Bringing extra air to a volume increases density State equation Relates pressure changes to density © Nokia Acoustical measurements.ppt / / IJ

Wave equation of sound Previous wave equations used pressure, density and particle velocity Eliminating density and particle velocity the wave equation of sound is obtained Two basic solutions: Plane wave Spherical wave © Nokia Acoustical measurements.ppt / / IJ

Free field acoustics Sound propagates to all directions without diffraction, reflection or absorption Spherical waves In principle, infinite, empty space without reflections In practice, anechoic chamber, with near 100% absorptive walls © Nokia Acoustical measurements.ppt / / IJ

Free field microphone Intended to measure the sound pressure as it existed before the microphone was introduced Microphone pointed to source Microphone tip causes an increase in sound pressure Taken care of by internal acoustical damping to achieve flat frequency response The free-field microphone should always be pointed toward the sound source (0° incidence). In this situation, the presence of the microphone diaphragm in the sound field will result in a pressure increase in front of the diaphragm, depending on the wavelength of the sound waves and the microphone diameter. For a typical ½’’ microphone, the maximum pressure increase will occur at 26.9 kHz, where the wavelength of the sound (342 m/s / 26.9 kHz = 2.7 mm = 0.5 in.) coincides with the diameter of the microphone. The microphone is then designed so that the sensitivity of the microphone decreases by the same amount as the acoustical pressure increases in front of the diaphragm. This is obtained by increasing the internal acoustical damping in the microphone cartridge, to obtain a frequency response. The result is an output from the microphone, which is proportional to the sound pressure as it existed before the microphone was introduced into the sound field. The curve in (a) is also called the “free-field correction curve” for the microphone, as this is the curve that must be added to the frequency response of the microphone cartridge to obtain the acoustical characteristic of the microphone in the free field. © Nokia Acoustical measurements.ppt / / IJ

Diffuse field – random incidence microphone
Sound reflects from many directions → sound comes to microphone from every direction In practice achieved in a reverberation room with 100% reflective and unparallel walls Microphone diffracts the sound waves from different directions in different ways Combined influence depends on directional distribution of sound waves Standard distribution based on statistical considerations used for random incidence microphone In some cases, (e.g., when measuring in a reverberation room or other highly reflecting surroundings), the sound waves will not have a well-defined propagation direction, but will arrive at the microphone from all directions simultaneously. The sound waves arriving at the microphone from the front will cause a pressure increase, as described for the free-field microphone, while the waves arriving from the back of the microphone will be decrease to a certain extent due to the shadowing effects of the microphone cartridge. The combined influence of the waves coming from different directions therefore depends on the distribution of sound waves from different directions. For measurement microphones, a standard distribution has been defined, based on statistical considerations, resulting in a standardized random incidence microphone. © Nokia Acoustical measurements.ppt / / IJ

Closed coupler Chamber with small dimensions compared to sound wavelength Special case: standing wave tube Diameter smaller than sound wavelength Source at the end Possible to calculate the sound field Used in calibration Used in microphone calibration © Nokia Acoustical measurements.ppt / / IJ

Pressure microphone Measuring the actual pressure on a wall
Typically used in closed coupler for calibration © Nokia Acoustical measurements.ppt / / IJ

Microphone directionality
Directionality indicates the sensitiveness of a microphone to sound coming from different directions No microphone is perfectly omnidirectional Cardioid or hypercardioid commonly used to record vocals Most ribbon microphones are bi-directional Shotgun directionality used outdoors for TV/film production and wildlife recordings © Nokia Acoustical measurements.ppt / / IJ

Parabolic microphone Parabolic reflector used to collect sound waves to microphone Very directional For eavesdropping in e.g. spying © Nokia Acoustical measurements.ppt / / IJ

Microphone transducers
Condenser microphones Electret capacitor microphones Dynamic microphones Ribbon microphones Carbon microphones Piezoelectric microphones Laser microphones © Nokia Acoustical measurements.ppt / / IJ

Condenser microphone Diaphragm and backplate form a plate capacitor
Charge kept constant → voltage varies as pressure actuates the diaphragm External voltage supply or pre-charged diaphragm Acoustical performance determined by physical dimensions © Nokia Acoustical measurements.ppt / / IJ

Condenser microphone – cont
The larger the diaphragm, the more sensitive the microphone Upper limit is defined by diaphragm touching the backplate The smaller the microphone, the greater the frequency range Increasing tension extends range but decreases sensitivity Optimum size of a measurement microphone is (up to 20 kHz) is about 12.6 mm (1/2’’) Damping effect of air reduced by drilling holes in the backplate The acoustical performance of a microphone is determined by the physical dimensions such as diaphragm area, the distance between the diaphragm and the backplate, the stiffness and mass of the suspended diaphragm, and the internal volume of the microphone casing. These factors will determine the frequency range of the microphone, the sensitivity, and the dynamic range. The sensitivity of the microphone is described as the output voltage of the microphone for a given sound pressure excitation, and is in itself of little interest for the operation of the microphone, except for calibration purposes. However, the sensitivity of the microphone (together with the electric impedance of the cartridge) also determines the lowest sound pressure level that can be measured with the microphone. For example, with a microphone with a sensitivity of 2.5 mV/Pa, the lowest level that can be measured is around 40 dB (re. 20 uPa), while a microphone with a sensitivity of 50 mV/Pa can measure levels down to approximately 15 dB (re. 20 uPa). The size of the microphone is the first parameter determining the sensitivity of the microphone. In general, the larger the diaphragm diameter, the more sensitive the microphone will be. There are, however, limits to how sensitive the microphone can be made by simply making it larger. The polarization voltage between the diaphragm and the backplate will attract the diaphragm and deflect this toward the backplate. As the size of the microphone is increased, the deflection will increase and eventually the diaphragm will be deflected so much that it will touch the backplate. To avoid this, the distance between the diaphragm and the backplate can be increased or the polarization voltage can be decreased. Both of these actions will, however, decrease the sensitivity, so that the optimum size of a practical measurement microphone for use up to 20 kHz is very close to ½’’ (12.6 mm). As the size of the microphone is decreased, the useful frequency range of the microphone is increased. The frequency range, which can be obtained, is determined in part by the size of the microphone. At high frequencies, when the wavelength of the sound waves becomes much smaller than the diameter of the diaphragm, the diaphragm will stop behaving like a rigid piston (the diaphragm “breaks up” — this is not a destructive phenomenon). Different parts of the diaphragm will start to move with different magnitude and phase, and the frequency response of the microphone will change. To avoid this, the upper limiting frequency is placed so that the sensitivity of the microphone drops off before the diaphragm starts to break up. This gives, for a typical 0.5 in. microphone, an upper limiting frequency in the range from 20 kHz to 40 kHz, depending on the diaphragm tension. If the diaphragm is tensioned so that it becomes more stiff, the resonance frequency of the diaphragm will be higher; on the other hand, the sensitivity of the microphone will be reduced as the diaphragm deflection by a certain sound pressure level decreases. The frequency response of the microphone is determined by the diaphragm tension, the diaphragm mass, and the acoustical damping in the airgap between the diaphragm and the backplate. This system can be represented by the mechanical analogy of a simple mass–spring–damper system as in Figure 27.3. The mass in the analogy represents the mass of the diaphragm and the spring represents the tension in the diaphragm. Thus, if the diaphragm is tensioned to become stiffer, the corresponding spring will become stiffer. The damping element in the analogy represents the acoustical damping between the diaphragm and the backplate. This can be adjusted by, for example, drilling holes in the backplate. This will make it easier for the air to move away from the airgap when the diaphragm is deflected, and therefore decrease damping. © Nokia Acoustical measurements.ppt / / IJ

Electret microphone Invented at Bell Labs in 1962 by Gerhard Sessler and Jim West Diaphragm permanently polarized the same way as permanent magnets magnetized (electrostatic magnet) Once considered low price and low quality Now most common microphone type An electret microphone is a relatively new type of condenser microphone invented at Bell laboratories in 1962 by Gerhard Sessler and Jim West [2], and often simply called an electret microphone. An electret is a dielectric material that has been permanently electrically charged or polarised. The name comes from electrostatic and magnet; a static charge is embdedded in an electret by alignment of the static charges in the material, much the way a magnet is made by aligning the magnetic domains in a piece of iron. Electret microphones have existed since the 1920s but were considered impractical, but they have now become the most common type of all, used in many applications from high-quality recording and lavalier use to built-in microphones in small sound recording devices and telephones. Though electret mics were once considered low-cost and low quality, the best ones can now rival capacitor mics in every respect apart from low noise and can even have the long-term stability and ultra-flat response needed for a measuring microphone. Unlike other condenser microphones they require no polarising voltage, but normally contain an integrated preamplifier which does require power (often incorrectly called polarizing power or bias). This preamp is frequently phantom powered in sound reinforcement and studio applications. While few electret microphones rival the best DC-polarized units in terms of noise level, this is not due to any inherent limitation of the electret. Rather, mass production techniques needed to produce electrets cheaply don't lend themselves to the precision needed to produce the highest quality microphones. [2] “Electret Microphone Turns 40” © Nokia Acoustical measurements.ppt / / IJ

Dynamic microphone A movable coil is attached to the diaphragm
An unmovable magnet produces a magnetic field Moving diaphragm moves the coil in the magnetic field, inducing a measurable current Exactly same principle as in loudspeakers, only reversed Poor low-frequency response → reduces handling noise Robust, relatively inexpensive and resistant to moisture → widely used on-stage In a dynamic microphone a small movable induction coil, positioned in the magnetic field of a permanent magnet, is attached to the diaphragm. When sound enters through the windscreen of the microphone, the sound wave vibrations move the diaphragm, When the diaphragm vibrates, the coil moves in the magnetic field, producing a varying current in the coil (See electromagnetic induction). The principle is exactly the same as in a loudspeaker, only reversed. Dynamic microphones are robust, relatively inexpensive, and resistant to moisture, and for this reason they are widely used on-stage by singers. They tend to have a poor low-frequency response, which is advantageous for reducing handling noise as a vocal mic, but tends to exclude them from other uses. © Nokia Acoustical measurements.ppt / / IJ

Ribbon microphones Revolutionized recording and broadcast industry in the 30’s Special type of dynamic microphones Thin metal ribbon between poles of magnet Voltage output typically low compared to normal dynamic microphones Bidirectional Very sensitive and accurate Generally delicate and expensive A Ribbon microphone is a type of dynamic microphone that uses a thin metal ribbon placed between the poles of a magnet and generate voltages by electromagnetic induction. Ribbon mics are typically bidirectional, meaning they pick up sounds equally well from either side of the microphone. In the dynamic microphone, the diaphragm is attached to a light movable coil that generates a voltage as it moves back and forth between the poles of a permanent magnet. Ribbon microphones generate voltages by electromagnetic induction; a current is induced at right angles to both the ribbon velocity and magnetic field direction. As the sound wave causes the ribbon to move, the induced current in the ribbon is proportional to the particle velocity in the sound wave. The voltage output of the ribbon is typically quite low compared to a dynamic moving coil microphone and a step-up transformer is used to increase the voltage output and increase the output impedance. Ribbon microphones are generally the most delicate and expensive microphone. They are often used in pairs to produce the Blumlein Pair recording array. One of the first ribbon microphones was the RCA PB-31. Produced in 1931, it was a breakthrough technology in sound, and revolutionized the recording and broadcast industry, setting a new standard in frequency response. The clarity and realism were unmatched by any of the condenser microphones of its day. © Nokia Acoustical measurements.ppt / / IJ

Carbon microphones Invented by David Hughes in 1878
Very important in the history of telephone Sound pressure (AP) presses the diaphragm (2) to a bed of carbon granules (1). Contact resistance depends on the pressure → resitance R changes Also an amplifier Extremely low-quality sound reproduction Very limited frequency range Very robust A carbon microphone, formerly used in telephone handsets, is a capsule containing carbon granules pressed between two metal plates. A voltage is applied across the metal plates, causing a current to flow through the carbon. One of the plates, the diaphragm, vibrates in sympathy with incident sound waves, applying a varying pressure to the carbon. The changing pressure deforms the granules, causing the contact area between each pair of adjacent granules to change, and this causes the electrical resistance of the mass of granules to change (lose contact). Since the voltage across a conductor is proportional to its resistance, the voltage across the capsule varies according to the sound pressure. Carbon microphones were once commonly used in telephones; they have extremely low-quality sound reproduction and a very limited frequency response range, but are very robust devices. Unlike other microphone types, the carbon microphone is also a class of amplifier, using a small amount of sound energy to produce a larger amount of electrical energy. Carbon microphones found use as early telephone repeaters, making long distance phone calls possible in the era before vacuum tubes. These repeaters worked by mechanically coupling a magnetic telephone receiver to a carbon microphone: the faint signal from the receiver was transferred to the microphone, with a resulting stronger electrical signal to send down the line. © Nokia Acoustical measurements.ppt / / IJ

Piezo microphones Piezoelectric material
Diaphragm moves the armature to bend piezoelectric crystal over a fulcrum Small size, cheap, low quality Have replaced carbon microphones Often used as contact microphones to sound instruments underwater or other unusual environments A piezo microphone uses the phenomenon of piezoelectricity—the ability of some materials to produce a voltage when subjected to pressure—to convert vibrations into an electrical signal. Piezo transducers are often used as contact microphones to amplify acoustic instruments for live performance, or to record sounds in unusual environments (underwater, for instance.) An example of this is Rochelle salt (potassium sodium tartrate), which is a piezoelectric crystal that works as a transducer both ways; it is also commonly used as a slimline loudspeaker component. © Nokia Acoustical measurements.ppt / / IJ

Laser microphones Window of a room acting as diaphragm
Reading with laser beam reflected from the window Two laser beams for common mode rejection of large window movements and path disturbances For eavesdropping Works best with one-glass windows Laser microphones are microphones with a laser beam. They detect vibrations with a laser and convert it to a digital signal. Lasers are usually bounced off a window, or off any object near to the conversation monitored. Any object which can resonate/vibrate (for example, a picture on a wall) will do so in response to the pressure waves created by noises present in a room. The minute differences in the distance travelled by the light to pick up this resonance is detected interfermoetrically. Light that is subject to the varying distance is mixed with light that travels a constant distance. The interferometer converts the variations in distance to intensity variations and electronics are used to convert these variations to digital signals that can be interpreted as sound. This technology can be used to secretly eavesdrop on people, with minimal chance of exposure unless specialized light sensors are used to detect the light from the beam. Supposedly this type of microphone does not work on older "wavy" glass and modern reproductions of it, since it is an irregular surface. © Nokia Acoustical measurements.ppt / / IJ

Sound level measurements
Measurement of sound pressure filtered by frequency (A-weighting) time-domain (RMS) Mimics response of human ear to noise The human ear basically hears the sound pressure, but the sensitivity varies with the frequency. The human ear is most sensitive to sound in the frequency range from 1 kHz to 5 kHz, while the sensitivity drops at higher and lower frequencies. This has led to the development of several frequency weighting functions, which attempt to replicate the sensitivity of the human ear. Also, the response of the human ear to time-varying signals and impulses has led to the development of instruments with well-defined time weighting functions. The output of the sound level meter is, in principle, assumed to be an approximate measure of the impression perceived by the human ear. The sound level meter can be functionally divided into four parts: microphone and preamplifier, A-weighting filter, rms detector and display (Figure 27.9). The microphone should ensure the correct measurement of the sound pressure within the frequency range for the given class. Also, the standard gives requirements for the directionality of the microphone. The frequency response of the instrument, including the weighting filter, is given for sound waves arriving at the microphone along the reference direction. For sound waves arriving from other directions, the standard allows wider tolerances at higher frequencies, taking into account the inevitable reflections and diffraction occurring at higher frequencies. The preamplifier converts the high-impedance output signal from the microphone to a low-impedance signal, but has in itself no or even negative voltage amplification. The signal from the preamplifier is then passed through an A-weighting filter. This is a standardized filter which, in principle, resembles the sensitivity of the human ear, so that a measurement utilizing this filter will give a result which correlates with the subjective response of an average listener. The filter, with the filter characteristic as in Figure 27.10, attenuates low and high frequencies and slightly amplifies frequencies in the mid-frequency range from 1 kHz to 5 kHz. There are a number of other weighting curves, denoted B-weighting, C-weighting and D-weighting, which may give better correlation with subjective responses in special cases, such as for very high or very low levels, or for aircraft noise. The signal from the A-weighting filter is subsequently passed through an exponential rms detector, with a time constant of either 125 ms (“fast”) or 1 s (“slow”). These time constants simulate the behavior of the human ear when subjected to time-varying signals. Especially when the duration of the sound stimuli to the human ear becomes shorts (e.g., around 200 ms), the sound is subjectively judged as being lower compared to the same sound heard continuously. The same effect is obtained using the “fast” averaging time. This will however give a higher statistical uncertainty on the level estimate than when using the time constant “slow,” so this should be chosen if the sound signal is continuously. Other standards describe special sound level meters such as integrating sound level meters or impulse sound level meters intended for special purposes. © Nokia Acoustical measurements.ppt / / IJ

Human hearing frequency response
The signal from the preamplifier is then passed through an A-weighting filter. This is a standardized filter which, in principle, resembles the sensitivity of the human ear, so that a measurement utilizing this filter will give a result which correlates with the subjective response of an average listener. The filter, with the filter characteristic as in Fig, attenuates low and high frequencies and slightly amplifies frequencies in the mid-frequency range from 1 kHz to 5 kHz. There are a number of other weighting curves, denoted B-weighting, C-weighting and D-weighting, which may give better correlation with subjective responses in special cases, such as for very high or very low levels, or for aircraft noise. A-weighting curve For subjective responses in special cases there are B-, C- and D-weighting curves very high or low level special noise, e.g., of aircraft © Nokia Acoustical measurements.ppt / / IJ

Sound level measurements
IEC International Standard 651 ”Sound Level Meters” Tolerances per frequency band defined for 4 classes of accuracy Type 0: precision laboratory use Type 1: general purpose Type 2: low price Type 3: not used in practice (too wide tolerances) The resulting measurement instrumentation is the sound level meter, as defined in for example by the IEC International Standard 651, “Sound Level Meters” [2]. The standard defines four classes of sound level meters for different accuracy’s (Table 27.1). Type 0 is the most accurate, intended for precision laboratory measurements, while Type 1 is most widely used for general-purpose measurements, see Figure Type 2 is used where low price is of importance, while Type 3 is not used in practice because of the wide tolerances, making the results too unreliable. © Nokia Acoustical measurements.ppt / / IJ

Sound intensity measurements
ISO Standard 3745 “Acoustics — Determination of sound power levels of noise sources — Precision method for anechoic and semi-anechoic rooms” no. x/r y/r z/r 1 -0.99 0.15 2 0.5 -0.86 3 0.86 4 -0.45 0.77 0.45 5 -0.77 6 0.89 7 0.33 0.57 0.75 8 -0.66 9 -0.57 10 1.00 The calculation in Equation of the sound power from sound pressure measurements is based on Equation This equation, which gives the intensity based on a pressure measurement, is however only valid in a free field, in the direction of propagation. In general, in the presence of background noise or with reflections from walls, etc., it is not possible to calculate the sound intensity from a single pressure measurement. In these cases, it is however possible to measure directly the sound intensity with a two-microphone intensity probe, Figure Sound intensity I is the product of the pressure and the particle velocity: While the pressure p is a scalar and independent of the direction, the particle velocity is a vector quantity and directionally dependent. When the particle velocity is stated as in Equation 27.15, it is implicit that the velocity is in a certain direction and that the resulting intensity is calculated in the same direction. For example, the particle velocity v in the direction of propagation, Figure 27.14(a), gives the intensity radiation away from the point source, while the particle velocity perpendicular to the propagation direction, Figure 27.14(b), is zero. The intensity calculated from Equation will therefore be zero in the direction perpendicular to the propagation direction even though the sound pressure is the same. This means that the sound energy flows away radially from the point source and no energy is flowing tangentially. © Nokia Acoustical measurements.ppt / / IJ

Two-microphone probe Measures the sound intensity in two directions
Pressure is mean of the two measured pressures Air particle velocity calculated from the two pressures All intensity is in radial direction, no intensity in perpendicular Powerful tool to locate noise sources Sound intensity I is the product of the pressure and the particle velocity. While the pressure p is a scalar and independent of the direction, the particle velocity is a vector quantity and directionally dependent. When the particle velocity is stated as in Equation 27.15, it is implicit that the velocity is in a certain direction and that the resulting intensity is calculated in the same direction. For example, the particle velocity v in the direction of propagation, Figure 27.14(a), gives the intensity radiation away from the point source, while the particle velocity perpendicular to the propagation direction, Figure 27.14(b), is zero. The intensity calculated from Equation will therefore be zero in the direction perpendicular to the propagation direction even though the sound pressure is the same. This means that the sound energy flows away radially from the point source and no energy is flowing tangentially. The measurement of the sound intensity according to Equation requires the measurement of the sound pressure and the particle velocity. With the two-microphone intensity probe, the pressure in a position in between the two microphones is calculated as the mean pressure measured by the two microphones: © Nokia Acoustical measurements.ppt / / IJ

Calibration techniques
Reciprocity calibration method Comparison or substitution methods Pistonphone (closed coupler) Sound pressure calibrator Electrostatic actuation © Nokia Acoustical measurements.ppt / / IJ

Reciprocity calibration method
Microphone can be used as a loudspeaker Three test microphones measured against each other alternating the function As a result a set of 3 equations with microphone sensitivities as unknowns Very accurate Rather tedious Requires well-controlled environment Seldom used in practical situations © Nokia Acoustical measurements.ppt / / IJ

Comparison/substitution methods
Microphone measured related to a reference microphone Comparison method: microphone and reference at the same time Substitution method: microphone put in the lace of the reference Sound source stability © Nokia Acoustical measurements.ppt / / IJ

Pistonphone Closed coupler Well-defined sound pressure level
Relatively simple mechanically, very stable Used often as the sound source in comparison/subsitution calibration Accuracy around 0.1 dB Depends on Volume of the coupler Volume displacement Barometric pressure Humidity Heat dissipation © Nokia Acoustical measurements.ppt / / IJ

Sound pressure calibrator
Small, self-contained Comparison calibrator Closed coupler Small loudspeaker produces single-frequency signal Reference microphone gives feedback signal Well-defined, provided that reference microphone and feedback gain are stable For field-calibration of microphones Normally not for laboratory calibrations © Nokia Acoustical measurements.ppt / / IJ

Electrostatic calibration
Direct use of electrostatic actuator to drive the diaphragm 800 V DC V AC signal Generally used to measure frequency response of microphones Widely used as a convenient and accurate test method For production and final calibration of measurement microphones © Nokia Acoustical measurements.ppt / / IJ

SoundField microphone
3D view of the sound with a single device 4-channel measurement of sound: B-format The spatial pattern can be decided later Mono, stereo, 5.1, … Fairly expensive, but replaces effectively a system of many microphones © Nokia Acoustical measurements.ppt / / IJ