HEALTH RISKS DUE TO EXPOSURE OF LOW-FREQUENCY NOISE : HÄLSORISKER VID EXPONERING AV LÅGFREKVENT BULLER

(1)

Örebro universitet Örebro University

Akademin för naturvetenskap och teknik School of Science and Technology 701 82 Örebro SE-701 82 Örebro, Sweden

Examensarbete 15 högskolepoäng C-nivå

HEALTH RISKS DUE TO EXPOSURE OF

LOW-FREQUENCY NOISE

Richard Storm

Ljudingenjörsprogrammet 180 högskolepoäng Örebro vårterminen 2009

Examinator: Dag Stranneby

HÄLSORISKER VID EXPONERING AV LÅGFREKVENT BULLER

(2)

SUMMARY R.STORM

i

Summary

This paper was made with an aim to prepare the risk assessment of operators’ noise exposure in Atlas Copco’s existing machines. This work is divided into three major parts:

low-frequency noise, hearing protection and sound propagation in mines. Several studies,

regulations and books have been examined in order to collect data of relevance to this work. The existence of man-made low-frequency noise (20-200 Hz) has been reported in many environments as a critical pollution problem. Symptoms like hearing loss, tinnitus, annoyance, disturbance of rest and sleep, fatigue, lower performance and social orientation, feeling of pressure on the eardrum and head, headache, disorientation, nausea, and balance disturbance, have among other ailments been reported by exposure to low-frequency noise.

The most effective and economic way to control noise is at its source but sometimes the noise may still consists such high levels that the hearing may be at risk. By the proper use of

hearing protectors, the hearing can be protected even in noisy environments. It is important though that the hearing protectors make an airtight seal with the ear to get maximum

protection. Noise would still reach the inner ear by bone conduction even if hearing protectors were complete effective in blocking airborne sound. When noise levels exceed 100 dBA for an 8-hour period, the use of wearing double hearing protection is necessary and when the noise is being attenuated by both plugs and muffs, hearing protection normally can’t be used to limit the effects of disturbance due to the poor attenuation of lower frequencies.

Mine workers, due to the reflection of machine-generated noise that would otherwise dissipate in an aboveground situation, are exposed to additional noise levels underground. Mining machines operates in an acoustic environment that is a critical factor which is affecting the sound pressure level exposure for mining machine operators.

(3)

ACKNOWLEDGEMENTS R.STORM

ii

Acknowledgements

This paper was presented as an examination report during the last period of the sound engineering program at Örebro University, organized and supported by Atlas Copco Rock Drills AB.

The author wishes to thank his mentors:

Mattias Göthberg, Noise and Vibration Specialist, Atlas Copco Rock Drills AB Jonas Karlsson, Assistant Master, Örebro University

And special thanks to:

Åsa Skagerstrand, Reg Audiologist, Örebro University Bengt Lindström, Librarian, Örebro University

Dag Stranneby, Examiner, Örebro University

Correspondence mail address:

richard.storm@live.se

Örebro, Sweden, 2009-10-12

____________________________ Richard Storm

(4)

CONTENTS R.STORM iii

Introduction

Aim

The aim of this report is to prepare the risk assessment of operators’ noise exposure in Atlas Copco’s existing machines following three cases where the machines are in operation:

1. Low-frequency sound (< 200 Hz) in cabins

2. Low-frequency sound in mine without cabin (with hearing protection) 3. Low- and high-frequency sound by radio-controlled machines

This work is divided into three major parts: low-frequency noise, hearing protection and sound propagation in mines.

Noise

Noise is defined as unwanted sound that causes annoyance and can lead to hearing loss [4], Noise has gained recognition as one of our critical environmental pollution problems where the pollution increases with population density, just like air and water pollution. People are affected by noise in many different ways and the effects on individuals may vary, which applies to both the disruptive effects and the risk of hearing damage. Noise-induced hearing loss is a major health problem for many people working in noisy environments and beside hearing loss, people are affected in many other ways by high levels of noise. Several physiological effects are resulting from exposure to noise, such as increased heart rate and blood pressure, vascular contractions, increasing size of pupils, effect on respiration and secretion of stress hormones [4, 35]. In addition, noise interferes with speech communication, interrupts sleep and reduces human efficiency [2].

Even during sleep, the environment is controlled by the hearing and sleep disturbances may follow as a result from noise exposure [35]. The body’s defense mechanism starts when sudden, unexpected or unfamiliar sound at high sound pressure levels appear, which leads to increased blood pressure, increased heart rate and muscle tension, and contractions of the skin’s blood vessels [5, 35].

Low-frequency noise

There’s no international determined definition for low-frequency noise (LFN) but in this context, with approval of several sources, LFN is defined as noise with dominated sound-energy in the frequency range between 20-200 Hz [4, 9-11, 33]. As will be mentioned later in this paper, exposures to this type of noise can lead to fatigue, sleepiness, headache, and disturbances of different kinds. Risk of disturbances is strikingly large in environments where the concentration level is high-demanded [4].

For the most part, human beings are exposed to man-made noise consisting of a combination of inaudible infrasound and audible noise components among which noise at low-frequency is especially problematic [1]. The most common sources of LFN in the work environment are larger ventilation systems, air conditioning, diesel engines, aircraft, compressors and blasting

(7)

2

[6]. Low frequencies occur at high levels, especially in transportation noise where LFN levels inside vehicles depends on the speed and size of the vehicle, road conditions including tires, and if windows are closed or not [1].

The disturbance experience is depended, not only on the sound pressure level, but also on the time of exposure, which in many cases judgments can’t be done solely by readings of

equivalent sound pressure levels [4].Consideration should therefore be given that the noise variances in time.

People exposed to community noise containing low-frequency components experience that the noise are more disturbing than noise without such components [1].

WHO

The World Health Organization (WHO) made a publication [34] where they recognize LFN as an environmental problem. Below is a selection of quotes from the publication:

 “The evidence on low-frequency noise is sufficiently strong to warrant immediate concern”

 “It should be noted that a large proportion of low-frequency components in noise may increase considerably the adverse effects on health”

 “Health effects due to low-frequency components in noise are estimated to be more severe than for community noises in general”

 “It should be noted that low-frequency noise, for example, from ventilation systems, can disturb rest and sleep even at low sound pressure levels”

 “For noise with a large proportion of low-frequency sound a still lower guideline value (than 30 dBA) is recommended”

 “When prominent low-frequency components are present, measures based on A-weighting are inappropriate.

 “Since A-weighting underestimates the sound pressure level of noise with low-frequency components, a better assessment of health effects would be to use C-weighting”

Low frequencies mean large wavelengths (see Table 1) and since absorption is very low, the low-frequency airborne sound can be propagated over great distances.

Frequency [Hz] 20 25 50 100 200 Wavelength [m] 17 13,6 6,8 3,4 1,7

Table 1. Frequency and wavelengths of low-frequency sound.

The large wavelengths make the LFN harder to be attenuated than noise of higher frequencies and therefore low-frequency sound can easily be propagated through walls, ceilings and floors

(8)

3

[33]. Like thunder that appears as a gunshot at close range and a rumble from distance, a sound that has been propagated over great distances consists only of lower frequencies [33]. The best way to minimize LFN is to control it at the source [1]. Examples of situations where the resulting sound may contain a high proportion of low frequencies are embedded control rooms, wheelhouses and cabs, and the use of conventional hearing protectors [6].

Infrasound

Infrasound is defined as sound with frequencies below 20 Hz [4, 20, 29-32]. Since a

perception threshold has been demonstrated, where the sound directly turns into audible sound with increased frequency, the human auditory organ may perceive infrasound if the sound pressure levels are sufficiently high[4-6, 16]. As for other type of sound, the perception threshold varies in sensitivity between individuals. Below 15 Hz, the tonal character of the sound ends, and the sound is being perceived rather as repetitive shocks or pressure waves [5]. Infrasound of moderate intensity is generated in processes or facilities where a large amount of air is pushed into movement including natural phenomena like; earthquakes, waterfall, volcanic eruptions, ocean waves, wind and thunder, and artificial sources such as diesel engines, compressors, jet aircraft, air-conditioning systems and other machinery [2]. Effects on the auditory organ, like temporary or permanent hearing damages or discomfort and pain sensations, only occurs at sound pressure levels above approximately 125-130 dB [5].

As a result of exposure to infrasound, effects like; fatigue, sleepiness, dizziness, experience of sea-sickness, diffuse uncomfortable feelings, disturbance effects and reactions, different kind of difficult experiences and more have been reported [4, 5]. The inconveniencies have, where appropriate, been reported to occur at sound pressure levels above the perception threshold which could possibly be explained that the whole inner ear, and not only the cochlea, including balance apparatus, is stimulated on exposure to infrasound [5].

Symptoms like post-exposure fatigue and headache, swaying sensations as if falling, lethargy and drowsiness, tinnitus, and respiratory difficulties have been shown by exposure to whole-body infrasound between 2-20 Hz; 100-125 dB SPL [1]. In a face-to-face situation, during whole-body infrasound exposures of high intensity, even though clear infrasound modulation of the speech signal occurred, subjects have little difficulty in understanding speech and this result could also be due to the visual information of the lip movement, but in understanding speech presented through earphones during whole-body infrasound exposure, minor

difficulties have been reported [1].

Infrasound below the perception threshold doesn’t seem to cause any kind of disturbance and it is therefore desirable to aim for an environment with infrasound that doesn’t exceed that threshold [4].

The pain threshold for infrasound appears to be about 140 dB around 20 Hz, increasing to about 165 dB at 2 Hz and to 175-180 dB for static pressure [16]. Calculated into dBA, the pain threshold for 20 Hz is around 90 dBA, which seems to be surprisingly low.

(9)

4

Infrasound consists of even lower frequencies than LFN (see Table 2), and consequently even larger wave-lengths, so it is therefore very difficult to attenuate by ground and atmospheric absorption.

Frequency [Hz] 1 3,15 6,3 12,5 20 Wavelength [m] 340 107,9 54 27,2 17

(10)

BACKGROUND R.STORM

5

Background

Weighting filters

Sound level meters contain weighting filters that, depended on frequencies, attenuate the microphone signal to provide the desired frequency responses. The frequency characteristics within the two commonly used filters, A and C filter, are showed in Figure 1.

The A-weighting filter gradually reduces the significance of frequencies below 1000 Hz, until at 10 Hz the attenuation is about 70 dB, which is rather close to the frequency response of the ear at low sound pressure levels [2]. It’s normal to use the sound level meter A-weighting filter when measuring environmental noise and assessing the risk of hearing damage. The C-weighting filter has an almost flat frequency response and is commonly used when measuring and assessing impulse sounds [5].

A quick and easy method to assess if a noise contains a concentration of sound energy in the low-frequency range is by using the difference between the dBC and the dBA reading of a sound level meter. A concentration of sound energy in the frequency range above 1 kHz is indicated by a small dBC-dBA difference, while a concentration in the low-frequency range is indicated by a large difference. The difference could be explained due to that the A-weighted filter response falls progressively below the C-weighted response. Disturbance due to LFN can be excluded if the measurements between the A-weighted and C-weighted filter differs by no more than 15 dB but if it differs even more, a third octave band analysis should be done to get a more accurate evaluation [4, 6]. WHO recommends performing a frequency analysis of the noise if the difference between the measurements is more than 10 dB [34]. However, Landström et al. [6] suggest that a synoptic control should only be done when the dBA level is higher than about 30 dBA.

(11)

BACKGROUND R.STORM

6

The A-weighting procedure is implied, by the results of a study made by Landström et al. [22], to lead to an overestimation of frequency tones and an underestimation of the low-frequency band noise with respect to tolerance levels during work. Further, it is indicated that dBA noise levels don’t predict any annoyance, resulting in the dBA weighting being a poor predictor of annoyance due to LFN [1, 8].

Broner and Leventhall made a study [14] where the psychophysical estimation technique, in which the subject quantifies his perception of a given physical attribute such as annoyance in terms of numerical estimates, was used. The technique was used to determine if a predictor of LFN annoyance, superior to the dBA, could be found. As possible candidates for best

predictor of higher level LFN, ten noise measures were chosen, including simple weighting networks as A, B, C, D1, D2 and E. Tentatively, it was concluded that the dBB noise measure could be used to predict the annoyance due to levels of LFN in the range 90-105 dB overall sound pressure levels.

(12)

HEALTH EFFECTS R.STORM

7

Health effects

Hearing loss

A common cause of hearing loss in our society is hearing damage caused by noise exposure. In a global perspective, hearing loss is the most common permanent work injury [33]. In order to induce hearing damage when exposed to LFN, the low-frequency levels have to be

extremely intense, or of very long duration according to Goldstein [1]. Other sources describe the noise in general without referring to any specific frequencies when they write about hearing damage [2, 35]. If no consideration is done concerning lower frequencies, it is

suggested that noise below 85 dBA doesn’t cause any hearing damage, but however, it should be noted that individual differences may vary [4].

Noise-induced hearing damage causes a threshold shift that usually appears at frequencies around 4000 Hz [2, 35]. Men are more often affected than women and there have been discussions whether it depends on differences in gender when it comes to the ear’s resistance due to noise, or if the difference solely depends on that men in general are exposed to more noise in their professional, military service and recreational activities than women [35]. With an aim to identify the LFN, especially the presence of infra sounds and pure tones in the noise spectrum of construction vehicles, a study was carried out by Jacobsson et al. [24]. They studied band-shafts, dumpers, excavators, wheel loaders, mobile cranes, tractors, trucks and motor graders. Sound was recorded at different moments of driving and working, and analyzed within 2-20 Hz and 0-500 Hz range. The results showed that a few vehicles, at the time of the research, exceed the determined level of noise exposure at 110 dB IL Pure tones exist in spectra of noise within one third of the studied vehicles at levels were it leads to restriction of exposure time. Based on the results, no indications have been found that machine operators are more affected by hearing loss than ordinary people.

Temporary hearing loss

High-level, short-term noise causes the ear to desensitize, especially at frequencies about 4000 Hz, but if the overload is removed soon enough, the threshold will shift back almost to its normal level [2, 4, 5, 35]. However, a temporary hearing loss may have arisen and the hair cells in the cochlea may still have gotten small, irreversible damages. Repeated exposure to high-level noise will most certainly lead to damage in many hair cells, resulting in a

permanent hearing loss.

Temporary threshold shifts varies in magnitude from a change in hearing sensitivity of a few decibels in a narrow band of frequencies to shifts so large that temporary deafness arises [2]. For hearing sensitivity to return to near-normal hearing levels, the time required can vary from a few hours up to three weeks [2].

However, it should be noticed though that LFN produces less temporary threshold shift than high-frequency noise does [2].

(13)

8 Permanent hearing loss

Long-term exposure to high-level noise damages the hair cells in the cochlea and causes a permanent hearing loss which cannot be cured [2, 4, 5, 35]. Noise containing impulse noise has a significantly greater effect on hearing impairment than constant noise with the same equivalent sound pressure level [4, 5, 35]. This could be explained due to that the muscles attached to the eardrum and the ossicles of the middle ear tighten the ossicular chain and pull the stirrup away from the oval window of the cochlea. This action is termed the acoustic reflex, but unfortunately the reflex doesn’t begin until 30 or 40 ms after the sound overload occurs, and full protection (up to 20 dB attenuation) doesn’t occur for about another 150 ms [2].

Hearing loss occurs mainly at frequencies between 2000 - 8000 Hz in the early stages and doesn’t interfere very much with ordinary speech, which is carried out mainly by the sounds between 300 - 3000 Hz, also called the speech band [2]. Loss of hearing of speech becomes more apparent as the exposure to noise continues.

Tinnitus

In addition to hearing loss, loud noise can give rise to more or less permanent tinnitus which is a common ailment among adults and often occurs in relation to a hearing impairment [4, 5, 33]. It is not itself a disease but a symptom resulting from a range of underlying causes such as hearing damages, deceases, side effects of medicines, malocclusion, stress, depression etc. [33]. Treatments to reduce the ailments exist but unfortunately there is no cure for tinnitus. After exposure to intense sounds, if sustained long enough, the first minute or so generally involves a “rushing noise” tinnitus, which doesn’t originate entirely in the cochlea but may involve higher auditory centers as well [2].

Annoyance

LFN has negative influences on annoyance, which is reported in a number of studies, and it also seems to lead to other effects such as fatigue, sleep disturbances and reduced social orientation [8].

A higher degree of annoyance has been reported when working under conditions of LFN exposure, and the annoyance is correlated to subjective estimations of symptoms like tiredness, dizziness, a feeling of pressure on the head and lack of concentration [11]. A review of LFN effects by Broner [16] indicates that the effects are similar, though to a different degree, to those of higher frequency noise. Further indications are given that the possible danger due to infrasound has been much over-rated. The high level of LFN in the 20-100 Hz range is of much more significance than infrasound at the same level in most

environments where LFN is a problem. It seems, on the basis of existing data, that the threshold for infrasonic effects is approximately 120 dB SPL. It is further concluded that the primary effect due to LFN appears to be annoyance.

A research, which was part of an environmental research program dealing with annoyance due to noise, was made by Broner and Leventhall [20]. 20 subjects with an average age of 31

(14)

9

years participated and all of them reported themselves as having good hearing. They carried out a psychophysical magnitude estimation task and rated the annoyance and loudness of higher-level LFN stimuli. In order to allow comparisons of annoyance response across frequencies, the noise stimuli consisted of seven 10 Hz bandwidths in the range 20-90 Hz. The result showed that the attributes of annoyance and loudness were judged similarly when considering the group of subjects as a whole. However, some sex differences did occur within the group. It was found that males had a significantly different rate of annoyance growth when compared to females, and it was also found that the loudness response by males was significantly higher than for females. Unacceptability was in the study found to generally increase as frequency increased, and as level increased for a given stimulus frequency range. No real evidence was discovered to indicate that any particular frequencies were responsible for the annoyance due to higher-level LFN.

Based on four papers, Bengtsson [26] made a thesis where the aims were to evaluate the influence of LFN on performance, annoyance, other subjective effects, cortisol levels and subjective stress. A further aim of this thesis was to evaluate whether the frequency balance and modulation frequency in a LFN influenced a subject’s perception of pleasantness. All studies were laboratory experiments. To create realistic conditions considering the

environment in the exposure room, exposure time, sound pressure level and the characteristics of the work, was of great importance. During exposure to LFN and a reference noise at an A-weighted sound pressure level of 40 dB, the experiment reported in papers I and II comprised 32 subjects who worked under high workload for two hours with four performance tasks. The experiment reported in paper III comprised 38 subjects who worked for four hours with six performance tasks under low workload during exposure to LFN or a reference noise at 45 dBA. The experiment reported in paper IV comprised 30 subjects who varied the level of the sound characteristic’s frequency balance and modulation frequency in the LFN to make the noise more pleasant. The results showed that exposure to LFN during work can lead to subjective annoyance and increased cortisol levels, even at moderate sound pressure levels, and it also seems to impair the performance. The effects were influenced by workload and noise sensitivity, and performance effects are hypothesized to be mediated by impaired learning and reduced attention. The noise should contain no or little perceivable modulation and a lower relative content of low frequencies to achieve a more pleasant LFN.

With an aim among others to evaluate the prevalence of annoyance in relation to noise

exposure, a study was made by Persson Waye and Rylander [8]. In the study a cross-sectional questionnaire and noise measurement survey was undertaken among 279 randomly chosen subjects exposed to noise in their homes from heat pump and ventilation installations. 108 of them were exposed to LFN and as controls, 171 persons living in similar residential areas but exposed to mid-frequency noise, where chosen. The results showed that among the persons exposed to LFN, the prevalence of annoyance was significantly higher as compared to controls.

The effects of noise and whole body vibration were analyzed in a laboratory study by

Landström et al. [27], based on noise and vibrations exposures via loudspeaker and a shaker. During three different types of exposure, comprised by 24 subjects, the effects on fatigue, annoyance and performance were analyzed. In the first condition, the exposure consisted of noise at 50 Hz modulated at 2 Hz (85 dBlin). In the second condition, exposure to whole body vibration at 2 Hz (0.56 m/s2) were made and in the third condition, combined exposure to these noise and vibration signals were made. The highest rated fatigue and annoyance, lowest

(15)

10

pulse activity, and lowest performance, was correlated to the exposure of the combined stimuli.

To examine the sources of diesel locomotive noise that give rise to annoyance, an experiment was made by Kantarelis and Walker [15]. The experiment was designed to study the influence on response of frequencies below 250 Hz. The noise was presented from three trains in 24 conditions at four levels, with and without the low frequencies and the response was recorded on a 10-point scale. The amplitude modulation present in diesel engine noise has been shown to have importance in determining annoyance. In determining response, the low-frequency content of the noise was important. The frequency and the depth of the modulation are related to the exhaust manifold design, the fundamental firing frequency and the firing frequency of the engine. To reduce the adverse subjective response, changes in the manifold design may be of importance.

Concentration, rest and sleep

Subjective reports of abnormal fatigue are common in case studies and if continuous LFN gives rise to that kind of reaction, it may affect people on their work performance by lowering the attention and concentration. Symptoms like tiredness, fatigue, disturbed sleep and rest, lack of concentration and attention, and effects on mood have been reported by exposures to LFN [4, 8-12, 26]. Even at moderate levels, monotonous sound can increase sleepiness, especially if the sound is dominant in the low frequencies [4].

Intense sounds, if sustained long enough, will produce a physiological fatigue consisting of a threshold shift that persists up to 16 hours [2]. Temporary threshold shift that persists for even longer time is called pathological fatigue and is produced by more severe exposures, and as much as three weeks can be required for this delayed recovery to be complete [2].

The reported disturbance of rest and concentration in a study made by Persson Waye and Rylander [8], was significantly higher in a group exposed to LFN as compared to a control group. 15-20 % reported presence of disturbance in areas exposed to LFN and in control areas, only 3-4 % reported disturbance.

A study by Persson Waye et al. [12] comprised a larger number of subjects in an extended period of acclimatization nights. 26 normal hearing subjects with an average age of 26 years were recruited to sleep for five consecutive nights in a sleep laboratory. On the 4th night, half of the subjects were exposed to LFN and on the 5th night, they had their reference night with no exposure, while the reverse conditions were present for the other half of the group.

Immediately after wake up, and after 15, 30 and 45 minutes, saliva samples for cortisol determination were taken. Using questionnaires in the morning and in the evening, subjective evaluations of sleep and mood were obtained. The LFN was a wide-band ventilation noise which had an A-weighted sound pressure level of 40 dB. The results showed that subjects felt less socially orientated and were more tired in the morning after nights with exposure to LFN. On the cortisol secretion, no effect of noise condition was found.

Landström et al. [22] made two experiments with a total of four groups of 24 subjects each. They were exposed to either 100 or 1000 Hz tones or broadband noise in a sound chamber specially constructed for noise exposures in a wide frequency range. The subjects were engaged in a simple reaction time test or a rather difficult grammatical reasoning task during exposure. They adjusted the noise to two annoyance levels, defined by their interference with

(16)

11

task performance for each exposure. The results showed that significantly higher sound pressure levels were accepted (6 dB) during the simple reaction time test than during the reasoning test. Much lower levels (22 dB) were accepted of the 1000 Hz tone than of the broadband noise. For the 100 Hz tone, mean tolerance levels were 33 dB higher than that for the 1000 Hz tone and the difference is 14 dB larger than would be expected from

A-weighting, and thus suggests that this frequency weighting overestimates the influence of the low-frequency sound. One reason for these results may be that subjects’ settings were based on experienced effects of noise on task performance.

Experience of relief when the source is turned off

Compared to high-frequency noise, LFN is not immediately distracting and a common reaction to LFN and especially ventilation noise is a sense of relief when the source is turned off, even if you previously haven’t been aware of the noise presence [6, 33]. With a view to examine the importance of this experience on performance, a study was made by Kjellberg and Wide [7] where the subjects weren’t aware of that a broadband noise (15-1000 Hz, with the dominating energy in the lower frequency bands) of 51 dBA was part of the experiment. Two groups were compared in the study, a group who worked in silence and where the noise was turned on after 20 minutes (silence-noise), and a group who worked in noise and where the noise was turned off after 20 minutes (noise-silence). The results show that both the noise and the interruption of noise seem to have affected the performance. The results also show that the response times were longer for the noise-silence group, which indicates that the noise delayed the learning of the task. The response time were somewhat insignificantly prolonged when the noise was turned off, whereas the error rate dropped sharply and significantly more than in the silence-noise group. The slower responses could partly explain the lowered error rate but the difference between the groups, when error rates were corrected for the changes of response times, remained significant. The fact that the effect differed between the two groups shows that the direction of change is of great importance.

Performance

There have been a large number of studies on effects on performance due to exposure of LFN and most of them come to the same conclusion that LFN impairs work performance.

With a special interest to study objective and subjective effects over time, a pilot study by Persson Waye and Rylander [9] was carried out to assess method evaluating effects of LFN on performance. A ventilation noise of a predominantly mid-frequency character and one of predominantly low-frequency character were used. 14 randomly selected healthy subjects with an average age of 25 years performed three computerized cognitive tests alternatively in the mid-frequency noise or the LFN. After the session, questionnaires were used to evaluate symptoms, effects on mood and estimated interference with the test results caused by temperature, light and noise. As compared to the mid-frequency noise exposure, the results showed that the subjective estimations of noise interference with performance were higher for the LFN. Data from the study indicates that response times were longer for subjects exposed to LFN. Further indications were given that cognitive demands were less well coped with under LFN condition, and it could be hypothesized that LFN exposure was more difficult to habituate to. A significant relation between reduced activity and performance time indicates that increased fatigue is of importance concerning effects of LFN exposure.

(17)

12

In a study by Bengtsson et al. [10], the aim was to evaluate effects of moderate levels of LFN on attention, tiredness and motivation in a low demanding work situation. To create

conditions representative of a normal workday in a control room environment, the A-weighted sound pressure level, performance tasks and instructions to the subjects were chosen. Two ventilation noises were used: one of a low-frequency character and one reference noise of a flat-frequency character. Both of them had the same A-weighted sound pressure level (45 dB). 38 subjects with an average age of about 25 years were recruited to work with six

performances tasks, in the noises in a between subject design, for four hours. The tasks, chosen to be sensitive to tiredness and motivation, were of a primarily monotonous routine-type character. Cortisol levels were measured in saliva and subjective reports were collected using questionnaires. The result showed that LFN negatively influenced performance on two tasks sensitive to reduced attention and on a proof-reading task, but performances of tasks aimed at evaluating motivation were not significantly affected. The results of the subjective reports and the saliva samples confirmed that the aim of creating a low demanding work situation was attained. There were also indications that subjects in the LFN condition, whether they were correct or erroneous, needed a longer response time and that they gave a greater number of erroneous answers. LFN can thereby be hypothesized to make the subjects less attentive, which then negatively affects their performance.

Persson Waye et al. [11] studied the possible interference of LFN on performance and annoyance. For the study, 32 subjects with an average age of about 23 years were recruited. Each person underwent a hearing test and only persons with normal hearing were allowed to participate. They performed a series of performance tasks in a noise environment. The subjects were categorized, based upon responses to questionnaires, as having a high or low sensitivity to noise in general and LFN in particular. By using questionnaires, their subjective reactions to the test session were recorded. Two ventilation noises were used as exposure noises, one with predominantly low-frequency content (LFN) and the other with

flat-frequency noise content (reference noise), both at a level of 40 dBA. Four performance tasks were used in the experiment and they were chosen in order to involve different levels of mental processing. By instructing the subjects to work as rapidly and accurately as possible, a high workload was generated. The effects were evaluated based upon terms of changes in performance and subjective reactions. As compared to the LFN condition, the results showed that there was a larger improvement of response time over time during work with a verbal grammatical reasoning task in the reference noise. It is further indicated that LFN interfered with a proof-reading task by lowering the number of marks made per line read. When working under conditions of LFN, the subjects reported a higher degree of annoyance and impaired working capacity. It is suggested that the quality of work performance and perceived annoyance may be influenced by exposure to LFN at commonly occurring noise levels.

Social orientation

Effects to LFN may appear over time and can result in lower social orientation such as more disagreeable, irritated, ill-tempered and less cooperative and helpful [8-10, 12]. Subjects, in a study made by Persson Waye et al. [12], were found to feel less socially oriented in the morning, when exposed to LFN during night.

(18)

13

Pleasantness

A tendency to lower pleasantness such as more bothered, depressed and less content has been reported after exposure to LFN [9].

People high-sensitive to LFN

For subjects rated as high-sensitive to LFN, the effects are more pronounced, while partly different results are obtained for subjects rated as high-sensitive to noise in general and it is suggested that subjects categorized as high-sensitive to LFN may be at highest risk [11].

Speech intelligibility

Speech perception involving intelligibility of phonetically balanced words is affected by LFN and speech intelligibility is significantly influenced by frequencies as low as 20 Hz [1].

Medical symptoms

Besides the direct effects due to LFN exposure, symptoms like feeling of pressure on the eardrum, dizziness and feeling of pressure on the head have been reported [9-11].

Among people reported themselves as annoyed by LFN, a proportion of 22 % of a study by Persson Waye and Rylander [8] reported headaches when exposed to LFN. It was suggested that the medical and psycho-social symptoms, including headache, are related to general annoyance or may be a result of disturbed sleep and rest caused by exposure to LFN. Vestibular Effects

Common measures of activation of the vestibular system are visual nystagmus and posture equilibrium, and exposure to low-frequency sound may create such ailments as disorientation, nausea and balance disturbance [1].

(19)

LIMITS AND CRITERIA R.STORM

14

Limits and criteria

Exposure time

The greater the noise levels the shorter time is needed for a hearing damage to occur. Permanent and prolonged exposure to noise with an A-weighted sound pressure level exceeding about 85 dB presents a risk of hearing damage [4]. However, the same source suggests that individual differences may vary greatly so that people high-sensitive to noise may be at risk of hearing damage after prolonged exposure to noise, even with A-weighted sound pressure levels about 75-80 dB.

The daily noise exposure levels, within working environment, are suggested by Swedish regulations to not exceed 85 dB within a working day of 8 hours [4].

Calculating daily noise exposure levels

If the working day differs from 8 hours, the daily noise exposure level LEX,8h, expressed in dB,

can be calculated, according to this formula [4]

where

LpAeq,Te = the measured equivalent A-weighted sound pressure level under the time Te

Te = the daily exposure time, expressed in hours, for noise with equivalent

A-weighted sound pressure level LpAeq,Te

T0 = 8 hours

To calculate Te knowing LpAeq,Te, the formula can be modified to

and when inserting the known variables the modified formula for calculating Te is

When inserting values into a diagram, it is obvious that the exposure time is halved for every increase of 3 dB (see Figure 2).

(20)

15

Figure 2. From 85-130 dB, the exposure time is halved for every increase of 3 dB from 8 hours down to 1 s.

When exposed to an equivalent sound pressure level of 85 dBA, the exposure time of a working day is set to 8 hours and if the sound pressure level would be as much as 112 dBA, the exposure time of the noise mustn’t exceed more than 1 minute.

LFN limits

For high-demanding working conditions where concentration and work performance is of big importance, Figure 3 can be applicable as guideline according to AFS 2005:16 [4].

It should be noted that the levels are expressed in equivalent sound pressure level and not including any weighting network. Further, it’s just a recommendation and not a law, and the

(21)

16

values are developed for a global perspective. Consideration should be given to the nature of the work, especially when it comes to economic point of view.

For indoor rooms in permanent dwellings and holiday homes, Swedish Welfare has proposed general advices on indoor room noise exposure (see Table 3), where indoor rooms considers room for sleep and rest, daily life and dining room used as a bedroom [40]. The general advices also include room intended for education, care, and bedrooms in temporary dwellings.

Frequency [Hz] 31,5 40 50 63 80 100 125 160 200 Sound pressure level [dB] 56 49 43 41,5 40 38 36 34 32

Table 3. Third octave band levels of proposed LFN limits in homes.

As can be seen in the table, the sound pressure levels are generally 5 dB below the criteria for high-demanding working conditions. Sound pressure levels of these general advices should be used as a guideline concerning health perspective.

SEN 590111

A Swedish Standard, SEN 590111 [45], with risk assessment of noise-induced hearing loss, including limits and criteria by noise exposure, was published in 1972 and repealed in 1998. The standard doesn’t seem to have the same approach to health requirements as today since it for instance is leaving 10 % of the people in the risk zone of hearing damage [46]. The maximum permissible time of noise exposure during a working day is shown in the diagram below [46] (see Figure 4).

For example, if the noise is 100 dB at 250 Hz, the time of exposure must not exceed more than 2 hours within a working day.

(22)

HEARING PROTECTION R.STORM

17

Hearing protection

The most effective and economic way to control noise is at its source but sometimes the noise may still consists such high levels that the hearing may be at risk. By the proper use of

hearing protectors, the hearing can be protected even in noisy environments. It is important though that the hearing protectors make an airtight seal with the ear to get maximum

protection. To ensure maximum protection and the best comfort of earmuffs, sealing rings and muffler pads should been replaced twice a year [41].

Noise would still reach the inner ear by bone conduction even if hearing protectors were complete effective in blocking airborne sound. Sound transmitted by bone conduction is generally 40-50 dB below that transmitted through an open ear canal so it doesn’t really compromise the effect of hearing protectors [2].

When noise levels exceed 100 dBA for an 8-hour period, the use of wearing double hearing protection is necessary [39]. When the noise is being attenuated by both plugs and muffs, hearing protection normally can’t be used to limit the effects of disturbance due to the poor attenuation of lower frequencies [4].

Effects of spectacles

A study was carried out by Lemstad and Kluge [36] to investigate how much spectacles influence the attenuation of earmuffs. Using broad-band stationary noise on human subjects, measurements was performed with four different safety spectacles. Spectacles are concluded to introduce a very significant reduction in the attenuation and the effect can be from

moderate to severe depending on the subject and spectacles. The attenuation was reduced as much as 29 dB (at 200 Hz) for one of the subjects when wearing a pair of spectacles with 4 mm thick sidebars in combination with the muff, as compared to wearing the muff only. Spectacles reduce the attenuation at low frequencies by introducing a leakage that creates a Helmholtz resonator effect. Depending on muff volume, the resonance frequency is in the 200-300 Hz range, but the attenuation is reduced in a much broader frequency range than that. Also at 3-5 kHz the opening may produce a resonance which can reduce the attenuation significantly. The attenuation can even be negative at resonance and the noise can be

amplified by as much as 8 dB (at 250 Hz) if the fitting is poor. For optimal attenuation, a good fit and thin side bars are of great importance but unfortunately, many safety spectacles have rather thick bars which may produce poor muff attenuation. However, helmets with integrated eye protection exist which possibly avoid the need for side bars crossing the muff seal

altogether.

Attenuation

To demonstrate the attenuation of some commonly used hearing protectors, data has been recorded from manufacturers’ brochures and plotted into a diagram (see Figure 5) [41-44]. The levels shown for each device are the Assumed Protection Value (APV), which is the difference between the mean attenuation and the standard deviation. Two earmuffs from Peltor (Optime I and III) and three plugs from different manufacturers are shown and, as can be seen, no attenuation data for the muffs at 63 Hz has been stated from the Peltor brochure [41]. However, it is still applicable in comparison with the other devices.

(23)

18

Figure 5. The attenuation of some commonly used plugs (dashed lines) and muffs (non-dashed lines).

By the hearing protection devices shown in the diagram, the earplugs (dashed lines) seem to have greater attenuation at lower frequencies as compared to the earmuffs (non-dashed lines). With levels from 31 dB (at 63 Hz) to 42 dB (at 4000 Hz), it could be concluded that the frequency response of the Uvex X-Fit is relatively flat, approximately 11 dB. This brings better quality to the attenuated sound as compared to the other hearing protectors where the attenuation levels varies as much as 28 dB (Peltor Optime III). If we take a look at the muffs, the lower frequencies is far away as much attenuated as the higher, which will lead to a noise dominated by low frequencies, which is experienced to be more disturbing than noise with a flat frequency character [1].

The earmuffs and earplugs shown in the diagram can be seen below (see Figure 6 and 7). Measured data from an Atlas Copco machine without cabin, and the attenuation data (APV) from Peltor Optime III hearing protectors, have been collected to see the effects on

attenuation by properly wearing a pair of earmuffs (see Figure 8) [41].

(24)

19

Figure 8. Sound pressure levels of an Atlas Copco machine in an underground mine with full stall, with and without hearing protectors.

The measurement of the machine at convertor stall was carried out in a mine, and as can be seen, wearing hearing protection effectively attenuates the noise. The total sound pressure levels have been calculated to 96.9 dBA for the machine only and 73.4 dBA when wearing hearing protection. The difference is approximately 23 dB, which is a great improvement and the hearing may no longer be at risk since the A-weighted sound pressure level doesn’t exceed 85 dB and, when considering the individual differences, neither exceed 75 dB [4].

Earmuffs and noise helmets

In a survey by Pääkkönen and Tikkanen [13], the ability of earmuffs and helmets to reduce LFN was tested. First without and then with earmuffs, noise was measured near the opening of the auditory canal. As sound sources, steady noise and LFN impulses were used. In the frequency range of 4-250 Hz, the results showed that earmuffs attenuated noise levels of 1-20 dB, and available prototype noise helmets attenuated noise levels of 5-39 dB. Generally, the attenuation improves with increased frequency.

Attenuation of wearing double hearing protection

Several studies have been made considering the attenuation of wearing both earplugs and earmuffs in combination. To calculate the attenuation of wearing both earplugs and earmuffs, it is not only to add together the attenuation data, instead a formula needs to be used. In a study by Damongeot et al. [25], data were collected from different laboratories and completed by their own tests. The aim was to look at the possibility of finding a formula for predicting the attenuation given by any pair of protectors from the attenuation given by each one, not in frequency bands, but on the basis of a global attenuation index.

(25)

20

An empirical formula was finally found, which allows the assessment of the global attenuation given by any combination of protectors

where

P = The global attenuation of the earplug alone M = The global attenuation of the earmuff alone

PM = Estimated attenuation of plugs and muffs in combination

When considering all the protectors, the accuracy of the assessments made by this formula is around ± 4 dB with a confidence interval of 90 % and when considering those protectors having attenuation higher than 15 dB, the accuracy is around ± 3 dB [25].

A research by Abel and Armstrong [3] was conducted to measure the attenuation that may be achieved by wearing an earplug and an earmuff in combination. An aim of the research was to delineate the relative contributions of each type of device to the outcome. A group of 16 subjects ranging in age 23-50 years, and with no hearing loss nor a history of ear diseases, participated in the experiment. Two plugs and two muffs were tested both singly and in the four possible combinations of muff/plug. With the ears non-blocked, hearing thresholds were measured and fitted with each of the eight protector alternatives at one-third octave bands ranging between 500-8000 Hz. Attenuation scores for each device or combination of devices were obtained by subtracting the non-blocked outcome from the protected outcome within frequency for each subject. The results showed that the combined attenuation of both a muff and a plug falls short of the sum of the attenuations observed for the individual components. For frequencies at and beyond 2000 Hz, the observed mean attenuation values were close to published bone conduction limits. Below 2000 Hz, the combined attenuation was broadly determined by the earplugs, and remarkably there was a significant gain in wearing double protection. However, the amount of benefit was determined exclusively by the choice of plug (see Figure 9).

(26)

21

Figure 9. The attenuation observed for two earmuffs and two earplugs worn singly and in combination (EP: E-A-R foam plug; BP: Bilsom

Soft plug; EM: E-A-R 3000 muff; BM: Bilsom 2315 muff). (Figure is copied from Abel and Armstrong’s research).

Another case were they investigated the protection provided by double hearing protection was carried out in a study by Tubb et al. [17], where a number of experiments were devised to measure the levels of sound attention afforded by these devices. The experiments involved a passive and an Active Noise Reduction (ANR) headset and a set of passive and ANR

earplugs. A total number of four individual hearing protection devices. Two independent experiments were made, the first one was based on an objective microphone method and the second one was based on a subjective method. Both experiments were carried out to calculate the attenuation of the hearing protection worn both singly and in combination with both earplugs and headset. The results showed that the active attenuation of an ANR headset appears to offer little advantage when used over earplugs in a double protection system. However, when worn in combination with either active or passive headsets, ANR earplugs have been shown to increase the low-frequency attenuation. The results also showed that the headset suffers no degradation of performance when worn over earplugs, which means that there is likely to be some mechanical coupling between the headset and earplug that reduces the overall attenuation attained.

A study by Pääkkönen et al. [19] evaluated the noise attenuation of earplugs and earmuffs, and their combined use against heavy weapon noise in field conditions for military personnel. A miniature microphone inserted into the ear canal measured the noise attenuation. Thirteen volunteers were tested against pink noise, the noise of explosions, and against the noise from weapons of different kinds. The results showed that the attenuation was 16-23 dB for earplugs and 10-20 dB for earmuffs. The combined use of both earplugs and earmuffs gave attenuation values of 24-34 dB, which were smaller values than expected.

(27)

22

Speech intelligibility

In an experiment by Brungart et al. [23], a group consisted of six normal hearing listeners wearing no hearing protection, earmuffs, or earplugs and earmuffs were asked to perform a speech intelligibility task. The task required them to segregate two simultaneous talkers who were either presented from the same loudspeaker or spatially separated by 90º. The subjects were also asked to determine the location of the target talker in each trial in a large anechoic chamber were the experiment was conducted. The results showed that listeners who can reliably localize sound sources while wearing earmuffs cannot do so when they are wearing both earmuffs and earplugs, but they are still able to benefit from the spatial separation of the competing talkers. The use of double hearing protection is suggested to cause spatially

separated sound sources to be heard at locations that are inaccurate, but still distinct enough to enhance the segregation of speech.

Proper wearing of hearing protection

With an aim to determine whether the noise attenuation with the use of earplugs can be

improved by teaching the proper insertion of earplugs to users, a study was made by Toivonen et al. [21]. A total number of 54 randomly selected male subjects aged 18-25 years were divided into two groups, where one group (25 persons) acted as a control group and didn’t undergo training before earplug insertion, and the other group (29 persons) was given a lecture on the use and insertion of earplugs. The latter group also practiced inserting the earplugs into their ear canals under the supervision of an occupational health nurse. The subjects inserted the earplugs into their ear canals and the success of the training was measured by the Microphone In Real Ear (MIRE) and Real Ear At Threshold (REAT) methods, and also by visual evaluation. The results showed that the averaged A-weighted noise attenuation was 21 dB for the untrained group and 31 dB for the trained group,

according to the MIRE method. The attenuation at 1000 Hz was 24 dB for the untrained group and 30 dB for the trained group, with the REAT method. The training improved the visual insertion quality from 1.9 to 2.6 (on a scale 0-3) for the men tested. As summary, training in earplug insertion is important for good attenuation and for diminishing poor attenuation to a minimum.

National Institute of Occupational Safety and Health (NIOSH) researches have observed that hearing protection devices are often misused. As a result, workers may be exposed to high levels of noise without realizing it because they think they are wearing their hearing protection properly, believing they are protected. Foam earplugs and earmuffs are the two most common types of hearing protection devices used in the drilling industry. Improper use of both of them causes reduced hearing protection and may be harmful to the hearing. To teach workers about the correct way to use hearing protection devices, NIOSH [39] has made an instruction guide for proper insertion of earplugs (see Figure 10 for illustration):

1. Before inserting foam earplugs, they should be rolled into a very thin cylinder with no creases. You should begin squeezing the earplug lightly as you roll it between your forefinger and thumb in order to get the diameter of the cylinder as small as possible and at the same time crease free. Then, as the plug becomes more tightly compressed, you gradually apply progressively greater pressure.

(28)

23

Figure 10. Three simple steps to get properly insertion of earplugs. (Figure is copied from NIOSH’s instruction guide).

2. In order for properly insertion, you must reach one hand around the back of your head and pull up and back on your outer ear to straighten the ear canal. If the ear canal isn’t straightened out, the earplug cannot slide in far enough. This is due to the natural curve inside your ear. When the earplug is in far enough, to provide a good seal, you should be able to feel it.

3. To allow the foam earplugs to expand, they must be held by your finger in your ear canal for about 10-20 seconds. To ensure the best fit possible, you should then release and push again for another 5 seconds. The foam earplug will then be positioned entirely within your ear canal.

Comfort

Hsu et al. [18] made a study that evaluates and recommends improvements for the comfort of hearing protection. To understand workers’ perceptions and actual comfort needs when wearing hearing protection was the main purpose of the study. To identify the critical factors for comfort of hearing protection a questionnaire was designed. To measure the comfort indices, an earmuff comfort tester was designed. Also an experiment was conducted to

measure workers’ perceived comfort into quantitative data. The range of these comfort indices in which workers will feel comfortable was determined from the data. The results were

summarized as guidelines to improve the design of current hearing protection. It shows that the most important reason why workers perceive that hearing protection is uncomfortable is that there are difficulties in having conversations. Also pressure sensation was of great importance to discomfort. Important design factors that influence the comfort of earmuffs are air-tightness, weight, heat-sinking ability, texture, and headband force. Adjustability and wearing convenience were rated less important. It was also concluded that an earmuff with larger interior space tends to have lower inner temperature.

(29)

SOUND PROPAGATION IN MINES R.STORM

24

Sound propagation in mines

Mine workers, due to the reflection of machine-generated noise that would otherwise dissipate in an aboveground situation, are exposed to additional noise levels underground.

Determination of the sound-absorption coefficients is essential when modeling the sound field in an underground mining environment. Mining machines operates in an acoustic

environment that is a critical factor which is affecting the sound pressure level exposure for mining machine operators. For determining underground absorption coefficients, due to the brittle composition of materials such as coal and slate, the impedance tube testing is an inappropriate method. Classic absorption coefficients estimation using T60 measurements, will not work well in an underground environment since the theory assumes a finite room, a diffuse field and a relatively uniform absorption. None of these assumptions are unfortunately true in an open-ended mine entry.

In order to develop a measurement method for determining absorption coefficients for

underground mines, a method is presented by Kovalchik et al. [28]. The method is based upon using a ray-tracing technique to determine absorption coefficients for underground mines. Absorption coefficients are determined and presented for octave bands from 63 Hz to 8 kHz and they are essential for determining and predicting potential overexposure to machine operators in different mine environments. Through computer-based modeling, determining the sound absorption coefficients is necessary when describing the sound field in an underground mining environment. The sound level exposure for the operator can be predicted based on the acoustics characteristics of the environment, once the absorption coefficients are known. The technique presented in this study provides a viable method for determining the octave band sound absorption coefficients in underground mines. Once the data is filtered using an A-weighting algorithm, the difference in measured and calculated sound pressure levels of the model provides a good fit and a viable method for determining the octave band sound absorption coefficients in not only underground coal mines, but also in all kind of underground mines.

To calculate the noise characteristics of a given source and room configuration, a ray-tracing based technique is used by a core noise modeling software called Raynoise [38]. The specific information used for input into the Raynoise package consists of [28]:

1. The X, Y, and Z coordinates of the calibrated noise source

2. The height above the floor of the bottom and top measurement points (see Figure 11) 3. The total length of the crosscut being examined

4. The crosscut height and width at each measurement plane location

(30)

25

Figure 11. Measurement scheme used in determining the absorption coefficients in an underground mine environment [28]. (Figure is

copied from the paper made by Kovalchik et al. [28]).

The difference in measured and calculated sound pressure levels of the model provides a good fit and the final absorption coefficients determined from the model are shown in Table 4.

Octave band [Hz] 63 125 250 500 1000 2000 4000 8000 Absorption coefficient 0.03 0.04 0.20 0.14 0.15 0.19 0.28 0.45

Table 4. Final absorption coefficients determined from model runs [28].

It should be noted that the confidence in the absorption coefficients for the lower frequencies, from 63 Hz to 250 Hz, is not very high since the background noise levels being so close to the noise source output. Especially at the 250 Hz frequency, it doesn’t seem to follow the general trend.

By data from Mine Safety and Health Administration (MSHA), the roof bolting machine is indicated to be third among all equipment and second among equipment in underground coal mining whose operators exceed the MSHA-PEL (MSHA-Permissible Exposure Limit). The National Institute for Occupational Safety and Health (NIOSH) has, in response to this, conducted a study to reduce overexposures of noise to operators of roof bolting machines. An important segment of a research made by Matetic et al. [37] was to determine, characterize and measure sound power levels radiated by a roof bolting machine during different drilling configurations. The determined sound power levels generated during the drilling cycle represents the overall sound power generated by the machine and is therefore of major

(31)

26

assess the effectiveness of proposed controls for reducing the noise exposure to roof bolting machine operators. A method for predicting sound pressure levels to roof bolting machine operators, for differing drilling types and parameters, could be determined by using the sound power level results obtained from the laboratory and a commercially available acoustical modeling package. A method for predicting sound pressure levels at the operator’s position of a roof bolting machine in an underground coal mine, using sound power levels determined in the laboratory, is provided in the study. If the octave sound power levels, octave absorption coefficients, room or entry dimensions, and measurement locations in x, y and z coordinates are known, a commercially available software package, Raynoise [38], can be used to show how the sound power relates to sound pressure level at a given point with. Once the sound power level of the roof bolting machine is determined, the characteristics of the specific roof bolting machine are necessary for sound level prediction. Underground coal mine data was collected to compare predicted and measured sound pressure levels to determine validity. The underground measured results were the same as the results of the predicted sound pressure level data, which means that the method is valid for predicting a roof bolting machine operator’s noise dosage underground, given any type of drilling configuration or drilling method utilized from laboratory testing.

T60 measurement in an enclosed area of a mine

To measure how much time it takes for a signal to diminish 60 dB below the original sound, a measurement of reverberation time was done in an enclosed area of a mine in Kvarntorp, Sweden, by me and my mentor at Atlas Copco. The enclosed area had the dimensions 81.9x11.5x4.9 m (LxWxH). As sound generating source, we had a balloon that was inflated almost to a maximum and then punctured with a needle. The sound level meter used for the measurement was a B&K 2260 Investigator.

For calculating the reverberation time, Sabines formula can be used [2]

where

c = a constant (0.161 s/m)

V = volume of the room (81.9 • 11.5 • 4.9 = 4615 m3)

S = room’s surface area (2 • 4.9 • 11.5 + 2 • 4.9 • 81.9 + 2 • 11.5 • 81.9 = 2799 m2) a = absorption coefficient

(32)

27

and when inserting the known variables, the absorption coefficient a can be calculated for each frequency

Calculated results are shown in Table 5.

Third octave band [Hz] 50 63 80 100 125 160 200 Reverberation time [s] 8.93 6.31 6.34 4.28 3.08 2.54 2.17 Absorption coefficient 0.03 0.04 0.04 0.06 0.09 0.10 0.12 Table 5. Absorption coefficients calculated from formula knowing the reverberation times for the different frequencies.

(33)

CONCLUSIONS R.STORM

28

Conclusions

The primary effect due to LFN appears to be annoyance and the noise should contain no or little perceivable modulation and a lower relative content of low frequencies to achieve a more pleasant LFN. Combined stimuli of both LFN and vibration seem to cause more annoyance than exposure to the noise only.

People high-sensitive to LFN seem to be at highest risk since the effects are more

pronounced. Further indications are given that people who are annoyed by noise are at greater risks to be more affected by the ailments than people who don’t get bothered by it at all. Example of symptoms that have been reported by exposure to LFN:

Annoyance, lower performance, irritation, depression, fatigue, tiredness, tinnitus, concentration difficulties, lower attention, sleepiness, less cooperative, disturbed sleep, experience of relief when the source is turned off, nausea, headache, dizziness, feeling of pressure on the eardrum and head,

disorientation, balance disturbance.

The A-weighting network is implied to lead to an overestimation of low-frequency tone and an underestimation of low-frequency band noise with respect to tolerance levels during work. Further, it is indicated that the A-weighting network is a bad predictor of annoyance due to LFN since the dBA noise levels don’t predict any annoyance. The B-weighting network could be used to predict the annoyance due to LFN in the range of 90-105 dB overall sound pressure levels.

The greater the noise levels the shorter time is needed for a hearing damage to occur. Machine operators aren’t more affected by hearing loss than ordinary people. Noise containing impulse noise has a significantly greater effect on hearing impairment than constant noise with the same equivalent sound pressure level. The time required for hearing sensitivity to return to near-normal hearing levels after a temporary threshold shift can vary from a few hours up to three weeks. However, LFN produces less temporary threshold shift than high-frequency noise does. Repeated exposure to high-level noise will most certainly lead to a permanent hearing loss since it damages the hair cells in the cochlea.

There should be no risk of hearing damage if the A-weighted sound pressure level is below 75 dB during an 8 hour working day.

The best way to control noise is at its source. The hearing can be protected by the proper use of hearing protectors, even in noisy environments. Earplugs seem to have greater attenuation at lower frequencies as compared to earmuffs. The attenuation of earmuffs generally improves with increased frequency. When wearing a pair of spectacles, the attenuation can be reduced as much as 29 dB as compared to wearing the muff only. The use of wearing double hearing protection is necessary when noise levels exceed 100 dBA within an 8-hour period.

(34)

CONCLUSIONS R.STORM

29

A formula can be used to predict the attenuation of wearing both earplugs and earmuffs

where

P = The global attenuation of the earplug alone M = The global attenuation of the earmuff alone

PM = Estimated attenuation of plugs and muffs in combination

Mine workers are exposed to additional noise levels underground since the reflection of machine-generated noise doesn’t dissipate as in an aboveground situation. It is essential to determine the sound-absorption coefficients when modeling the sound field in an underground mining environment. The software Raynoise, based on a ray-tracing technique, can be used to calculate the noise characteristics of a given sound and room configuration.

HEALTH RISKS DUE TO EXPOSURE OF LOW-FREQUENCY NOISE : HÄLSORISKER VID EXPONERING AV LÅGFREKVENT BULLER

Examensarbete 15 högskolepoäng C-nivå

HEALTH RISKS DUE TO EXPOSURE OF

LOW-FREQUENCY NOISE

Summary

Acknowledgements

Contents

Introduction

Aim

Noise

Low-frequency noise

Infrasound

Background

Weighting filters

Health effects

Hearing loss

Tinnitus

Annoyance

Concentration, rest and sleep

Experience of relief when the source is turned off

Performance

Social orientation

Pleasantness

People high-sensitive to LFN

Speech intelligibility

Medical symptoms

Limits and criteria

Exposure time

LFN limits

SEN 590111

Hearing protection

Effects of spectacles

Attenuation

Speech intelligibility

Proper wearing of hearing protection

Comfort

Sound propagation in mines

T60 measurement in an enclosed area of a mine

Conclusions