14 Acoust The values were established on the|basis of a statistical average obtained by testing many otologically normal people between the ages of 18 and 30 years. A similar method was used to obtain the Minimum Audible Field (MAF) which sets out tHresholds obtained binaurally, in response to sound pressures presented through loudspeakers, ‘Thresholds taken from the MAF curve are an average of about 6dB more sensitive than the normal MAP thresholds. 1.22 THE SOUND LEVEL METER A sound level meter i a precision instruinent used for sound measure- ment. A precision microphone converts the sound signal to an electrical signal, which is amplified by a pre-amplifier before being processed. The signal may be displayed in linear dBSPL, or it may pass through a ‘weighting network. This is an electronic Circuit that varies in sensitivity across the frequency range, to simulate the sensitivity of the human ear (Figure 1.8) The weighting networks are termed A] B, C and D. The A weighting network corresponds to an inverted equal loudness contour (section 1.4.2) at low sound pressure levels. The B network corresponds to ‘medium SPLs and the C network to high SPLs. The D network is used for aircraft noise measurements. 1.3 THE AUDIOMETER Hearing is a subjective sensation and # sured’, Audiometry measures the stimul of hearing, ‘An audiometer is an instrument of comparison, which indicates the difference between the sound pressure lve sequied to produce hear fefore cannot truly be ‘mea- 1.3.1 THE AUDIOMETER AND ITS chur which causes the sensation Aa | Fane | t 58 | Microphone Pre-amplifes pears Amplifier Dspiay Fig. 1.8 Constituent parts of a sound level meter using the A weighting network he audiometer 15 ‘Sitch one conduction headbana Pure tone | attenuator b generator AMpifier “™"Gial_— Air conduction headband Fig. 1.9 Constituent parts of en audtometer. ing in the individual under test and that required to produce hearing in an average normal young person. An audiometer generates pure tones at specified points within a restricted range of frequencies considered important for communication, usually the octave intervals from 125Hz to Bkkiz. ‘An octave is a doubling of frequency. Most audiometers also produce tones at the half-octave intervals, 750H2, 1.5 kHz, 3k#Hz and 6kHz. The sound level can also be varied, usually in 5dB steps, from ~10dBHL to T10dBHL or more. The duration of the tones is controlled by an interruptor switch. The audiometer is provided with two transducers, a pair of headphones and a bone vibrator. A block diagram of a basic audiometer is shown in Figure 1.9. ‘An audiometer must be accurate to be of value and, to this end, all audiometers are calibrated to British Standards when new and the talibration should be re-checked at least once a year. BS 5966 (3980) defines the aspects of pure-tone audiometer accuracy with which manufacturers should comply. ‘The main points of this standard, for our purposes, can be sum- marized as follows: ¢ Frequency accuracy must be 3% ‘© Purity must be such that the total harmonic distortion does not exceed 5% (for air conduction). © The attenuator 54B steps must be correct within +1 dB. © Unwanted sound from the audiometer should be inaudible up to and including the dial setting 504BHL © The narrow band noise filters must be centred on the frequency {according to a given table). ‘© The rise and fall times of the signal tone should be within specifi cation {section 1.3.3}16 Acoustics © The hearing level must be accurate to within +34B from 125Hz to 4KHz and to within £54B at 6KHz and 8kHz. Laboratory calibration is carried out for the complete audiometer, The transducers (headphones and bone vibrator) are a part of the audiometer and should not be exchanged unless the audiometer is recalibrated with the new transducers. In addition to calibration, the audiometer should be checked daily as follows: 1.3.2 DAILY AUDIOMETER CHECKS © Straighten any tangled leads. Ensure that all connections are firm and giving good contact. Flex the leads for possible intermittency. © Check that all knobs and switches are secure and function in a silent, smooth and click-free manner. © Check function of response button. * Check tension of AC and BC headbands. ‘© Check output levels for all tones at 10-15dB above own known threshold for AC (each earphone) and BC. Your own audiogram should be used for this approximate calibration check. ‘© Check at 60dBHL for all tones on AC (each earphone) for noticeable distortion, intermittency, etc. Repeat at 40dBHL for BC. Check masking noise over @ range of outputs, also loudness balance and other facilities, if to be used. ‘* Check battery condition, if appropriate. Generally ensure that the audiometer and all its attachments are clean. Wipe earphones and BC receiver with clean, dry tissues. 1.3.3 RISE AND FALL TIMES The audiometer output signal is not an instantaneous event. The signal rises until it reaches its maximum and falls off in a similar mannet when the tone is switched off (Figure 1.10). Rise time is defined as the time taken for the signal to rise from 6048 (or ~20 dB) to within 1B of its steady state. The time taken for the signal to rise from —60dB to ~1 dB of its steady state should not be more than 200 milliseconds. The time taker for the signal to rise from —2048 to ~14B of its steady state should be at least 20 milliseconds. (BS 5966: 1980). Fall time is defined as the time taken for the signal to decay by 608, from its steady state. Fall time should not exceed 200 milliseconds, nor be less than 20 milliseconds. (BS 5966: 1980). This rise and fall (decay) should be correctly timed or it may affect the results of the audiometric test 1@ psychological properties of sound 17 Switch off ee Switch on Fig. 1.10 The rise and falll envelope of test tones. Between the dotted lines sound should rise or fal in a progressive manner. (Alter BS 5966 (1980).) © Too slow a rise time may result in erroneously poor thresholds, because this does not elicit the maximum on-effect of the ear. # Too brief a rise time produce an overshoot which may be heard as a lick. This may result jn erroneously good thresholds if the patient responds to the click rather than the test tone. An overshoot should never be greater than 1B. 1.4 THE PSYCHOLOGICAL PROPERTIES OF SOUND 14.1 PsYCHOACOUSTICS, The study of the psychological properties of sound is known as sychoacoustics, The way in which we hear sound is subjective and cannot be directly measured. Subjective qualities are compared with reference levels oblained by averaging the judgements of a large number of normally- hearing people. 1.4.2 LOUDNESS Loudness is the subjective perception of sound in terms of intensity. The relative loudness of pure tones is normally expressed as equal loudness contours. Each ¢ontour represents sounds that appear equally Joud. Each contour or has a loudness value given in phons, which can be defined as the sound pressure level of a TkHz tone judged to be of equal loudness. The 40phon curve, for example (Figure 1.7), represents pure tones across the frequency sage, which are judged equally loud as a 1 Kz122 assessment procedure configuration, which again is not standardized. The following terms are sometimes used: © Flat, to indicate a loss that does octave. © Gradually sloping, to indicate a l octave. © Precipitously or sharply falling, t ‘mote per octave. ‘© Abruptly falling, to indicate a loss that is flat or gradual in the low frequency region, but then falls sharply. This configuration is also often referred to as a ‘ski-slope’, but this term is not recommended, ‘© Rising or reverse audiogram, to indicate a loss that increases by 5dB or more per octave. © Trough, to indicate a loss that falls in the midfrequency range (1kH2-2kFz) by 20dB or moré| in comparison with the loss at 500 Hz and 4kHz. Although not recommended, the American term ‘cookie bite’ is very descriptive for this shape of loss. tise or fall more than 54B per 3s that falls by 5B to 10dB per indicate a loss that falls 15dB or 6.6 ROOM REQUIREMENTS FOR AUDIOMETRY Background noise can have a significant effect on audiometric results, particularly those obtained by bone conduction. It is therefore necessary to reduce background noise to a level that will not elevate the hearing thresholds of the person under test. Ideally the room is specially constructed, but offen audiometric tests are undertaken in less than ideal conditions, for example in the home situation. Background noise should be reduced as far as is practicable and the use of specially designed earphones or of insert receivers may further assist in its re- duction. Where background noise is ufficiently high to have a possible effect on the thresholds, this information should be recorded on the audiogram. A sound-proof room is riot a necessity for audiometry, but the noise level should be below that which would cause masking and threshold shift in someone with normal hearing. Where a room is, specially constructed for the purpose of audiometry, there are design recommendations that should be followed wherever possible (Department of Health and Social Security, 1974). The room should have an ambient noise level below 30dBA. A prefabricated sound booth is sometimes used to guarantee a particular degree of sound attenuation. The booth musi, however, be situated within a quiet room since it ensures only a certain degree of attenuation, not a Particular sound level | ‘A booth that is to be used only for adults should be a minimum size of 2.2m by 1m with « height of 24, but preferably the room size | i i i i ‘ ! i ; | should be at least 2.4m by 2.4m, with a height of 2m. If children are to be tested, a room that is a minimum of 6.3m by 4.8m, with a height of 2m is required. This allows space for free-field testing, and also for parents or other adults to be present. ‘A sound-treated room should be situated away from obvious sources of noise and should be isolated from the xest af the building. This can be achieved by building the room of brick with a concrete skin and cavity, thus using the effect of mass and of dead-air space. Vibrations from the building can be deadened by mounting the whole interior on rubber shock absorbers. Solid double doors should be used to exclude noise. Preferably, the doors should be lead-lined and should close with magnetic seals. Ventilation will be needed because the room will rapidly become hot and airless, but the noise from the ventilation ‘system must be minimized using acoustic filters to act as baffles. Tungsten lighting is preferable, to fluorescent, which produces a hum, and if the latter is used the choke and starter should be mounted away from the room. If there are windows, double glazing should be used to reduce noise from outside. A much larger gap between the panes of glass is necessary than is required for heat retention. If the glass is 3mm thick the gap should be at least 100mm, and if the glass is 6mm the gap should be at least 50mm. In addition, where there is more than one window in the room, these should not be positioned opposite each other, as the creation of standing waves is more likely. This is important where free-field testing is carried out. A room for audiometry must not only be quiet, but also have a low reverberation time, that is, there must be very little ‘echo’. Rooms in which reverberation is exceptionally low are called anechoic chambers; these are very quiet and unnaturally ‘dead’, and are used only for research purposes. For audiometric purposes a reverberation time between 0,2 and 0.25 seconds is ideal. This should be achieved by covering the walls and floor with soft, sound-absorbent materials, such a acoustic tiles fixed on battens to the walls and ceiling and a thick foam-backed carpét on the floor. These will absorb sound and limit reverberation. 6.7 SUMMARY Tuning fork tests provide a quick indication of type of loss. They should not be considered as 2 substitute for pure-tone audiometry. Pure-tone tests are used to provide accurate hearing threshold levels by air and bone conduction, together with determination of ULLs, which are important for hearing aid fitting Audiometric test results are recorded on a graph known as an audiogram. The audiologist must be able to interpret the results and to