Welcome to the new IOA website! Please reset your password to access your account.

Volume : 46

Part : 2

Proceedings of the Institute of Acoustics

 

Why A-weighting Network is inadequate for measurement of noise with high content of infrasound and low frequency noise (ILNF)?

 

Hadi Asady1, Mariana Alves-Pereira2, Mohammad Bakhtiari3, Adrian Fuente4, Siamak Pourabdian1, Farhad Forouharmajd1,*

1Department of Occupational Health Engineering, School of Health, Isfahan University of Medical Sciences, Isfahan, Iran.

2School of Sciences for Economics and Organizations, Lusofona University, Lisbon, Portugal.

3Behbahan Faculty of Medical Sciences, Behbahan, Iran.

4École d’orthophonie et d’audiology, Faculty of Medicine, Université de Montréal, Montreal, Quebec, Canada.

 

1 INTRODUCTION

 

Humans with normal hearing are able to detect sounds that fall within a frequency range from about 20 Hz to 20 kHz, with the upper limit dropping off somewhat in adulthood. Not all mammalian species are sensitive to the same range of frequencies, and most small mammals are sensitive to very high frequencies but not to low frequencies[1]. As can be seen in Figure 1 the human auditory system has threshold less than or equal to 10 dB for frequency range between 250-8000 Hz.

 

 

Figure 1: Audiogram showing the average human threshold for pure tones obtained in a sound field, hearing range is usually specified as the range of frequencies audible at a level of 60-dB SPL[2]

 

With decreasing of the frequency from the 250 Hz to lower frequencies the human hearing threshold increase and the sounds with level less than those thresholds are not auditable by most of humans. In the other words the sound level less than the human hearing threshold at the low frequencies (20<=f<=200 Hz) [3, 4] and infrasound (f<20 Hz) will not be detected and perception. This fact was reflected in the A-weighting network. The A-weighting filter was born with the work by Fletcher and Munson to determine loudness level contours for various sound levels. They selected 23 healthy, young males, whom were assumed to have good hearing. These subjects were exposed to a series of different, single, pure tones at different levels of loudness and asked to score the sounds for equal loudness. The tones, produced by valve oscillators and amplifiers, were presented to the listener by occluded (covering the ear) headphones. The basic range of 20 Hz to approximately 16 kHz was used because those were the limits of the equipment at the time[5]. Figure 2 shows the Fletcher and Munson loudness curves from their 1933 paper. Three years later, these curves were used in the first American standard for sound level meters developed by the Acoustical Society of America[6].

 

 

Figure 2: Loudness level counter[7]

 

Historically, the A weighting network on a sound level meter were derived as the inverse of the 40 dB Fletcher-Munson curves and used to determine sound level. Since A-weighting correction are too large for low frequencies[8] so we expect that carry out measurement of low frequency noise with using of this network underestimate the results too much. this problem was reported by some researchers. Alves-Pereira and et al. reported that the LFN components of noise-rich environments are most often unknown because noisy environments are usually only described by a dBA-level measurement[9]. Baliatsas and et al. showed that an issue of importance in terms of exposure characterization is the weighting method used to measure LFN. A-weighting is widely used in public health research. Adapted to the sensitivity of the average human ear, sound level meters set to the A-weighting network cannot efficiently evaluate the contribution of LFN components, since the human ear is less sensitive to very low-pitch or high-pitch noises[10]. The question is why A-weighting network is unappropriated for measurement of noises with high content of infrasound and low frequencies? Or approximately equal content of different frequencies including infrasound, low and high frequencies. We will try to answer to this question.

 

2 MATERIALS AND METHODS

 

At this part we used methods and equations which presented in the IEC 61672-1 as an international standard[11] for calculation of A, C and Z frequency-weighting networks.

 

2.1 Frequency-weighting C

 

For any frequency f in hertz, the C-weighting characteristic C(f) shall be calculated, in decibels, from:

 

 

Normalization constant 𝑪𝟏𝟎𝟎𝟎 represents the electrical gain, in decibels, needed to provide a frequency weighting of 0 dB at 1000 Hz. We can calculate 𝑪(𝒇) with using of formulas and values which presented in table 1 and 2. 

 

Table 1:

 

 

Table 2:

 

 

2.2 Frequency-weighting A

 

The A-weighting characteristic 𝑨(𝒇) shall be calculated, in decibels, from:

 

 

Normalization constant 𝑨𝟏𝟎𝟎𝟎 represents the electrical gain, in decibels, needed to provide a frequency weighting of 0 dB at 1000 Hz. We can calculate 𝑨(𝒇) with using of formulas and values which presented in table 2 and 3.

 

Table 3:

 

 

2.3 Frequency-weighting Z

 

For any frequency in the range of a sound level meter, the Z-weighting characteristic 𝒁(𝒇) shall be given, in decibels, by

 

                                                            𝒁(𝒇) = 𝟎 𝒅𝑩

                                                            Equation 3

 

2.4 Converting formula

 

The formula for converting octave-band sound pressure levels into sound levels on the X-weighting network is:

 

 

LX-weighting (dB): Sound level on the X-weighting network, X being C, A or Z.

Lpi (dB) = Sound pressure level for the ith octave band,

F(dB) = Correction factor,

 

3 RESULTS

 

With using of formulas 1, 2,3, weight values related to the three different weighting network including C, A, Z were calculated at different frequencies (related to the one-third octave bands) (see table 4). As you can see in table 4 highest weight was related to the frequency 10(Hz) for both A and C weighting network. The weighing increase with increasing of frequency from 10 to 5000 Hz and 10 to 160 Hz in the A and C weighing network respectively and after those frequencies starting to drop. For C-weighing network weight for nine frequencies including 200, 250, 315, 400, 500, 630, 800, 1000, 1250(Hz) are zero (0 dB) while in the A-weighing network just frequency 1000 Hz have zero (0 dB) weight. Weighing for Z networks was zero (0 dB) for all frequencies.

 

Table 4: Sound level conversion chart from flat response(Z) to A and C weightings.

 

 

Now with using of weighing values that was presented in the table 4 we want to calculate the sound total pressure of an artificial data (assumption this data was gathered with using of the sound level meter in Z weighing mode, actually this data set shows sound pressure levels at the different one- third octave frequencies in an imaginary place) with using of the equation 4. In this data set the sound pressure levels at the infrasound and low frequencies are higher than high frequencies. We can see same kind of data in some public and occupational settings[9, 12]. As you can see in table 5, for noise with high content of infrasound and low frequency noises, A-weighing network highly drop down the total sound pressure in comparison to the C and especially Z weighting networks.

 

Table 5:

 

 

3.1 Discussion

 

American Conference of Governmental Industrial Hygienists recommend 85 dB(A) for 8-hour exposure with noise in the occupational settings[13]. As you can see in the table 5 if we have a place (occupational or non-occupational) which the most of the noise energy transferred by the infrasound and low frequencies, we will not detect it if we use of A-weighing network for the noise level measurements. Since the calculated total sound pressure level with using of A-weighting network (73.68 dB) was more less than occupational noise exposure standards (85 and 90 dB(A) recommended by ACGIH and OSHA respectively for 8 hours' exposure) so planning for noise control is not needed in such situations. In such condition, we can expect that hearing system damage not be occurs and other demonstrated health effects of high-level of noise exposure such as cardiovascular diseases, annoyance, cognitive development and hyperactivity in vulnerable population groups such as children, anxiety, angiocardiopathy, and impaired hormone secretion not be seen[10, 14].

 

However, some of studies have shown than exposure of people with noise in occupational or non- occupational settings also under the recommended standards levels could increase some health and Well-being problems between them. Many studies tried to show that these kinds of problems are attributed to the infrasound or low frequency noise exposure[15, 16]. KP Waye reported that having exposure with low frequency noise could cause sleep disturbance, reduced Wakefulness/greater fatigue, reducing of the work performance and also hearing loss, annoyance[17]. Also, JA ALVES and et al. reported than low-frequency noise is an agent that interferes with the performance of work tasks[18]. In addition to these kinds of effects some studies reported that exposure with low frequency noise at sufficiently high SPLs (Sound Pressure Levels) induce vibrations mainly in the chest and stomach and may cause a multi symptom disease called vibro-acoustic-disease (VAD) in long term exposure[17, 19]. With looking of these studies specially about VAD we can realize that the SPL has an important role in causing of those problems but with using of A-wreathing network the SPL was underestimated and also, we could not recognize the places with high content of infrasound and low frequency noise, at this situation so we will not have preventive infrasound and low frequency exposure control strategy.

 

3.2 Conclusion

 

It is obvious that measurement of noise level in all kind of places (occupational or non-occupational) with using of A-weighting networks underestimated the total sound pressure level. Using of this network will not show us the places with high content of infrasound and low frequency noise so we will under noise effect without we could recognize it. So, at these situations using of Z-weighting network will show the real sound levels and different frequencies including infrasound and low frequencies.

 

3.3 Acknowledgement

 

This research has been supported by Isfahan University of Medical Sciences grant (ID number: 3401350). This fund was used for data analysis. Conflict of Interest: The authors report no potential conflicts of interest.

 

4 REFERENCES

 

  1. Purves D: Neuroscience, Sixth edition edn. New York: Oxford University Press New York; 2018.
  2. Heffner HE, Heffner RS: Hearing ranges of laboratory animals. J Am Assoc Lab Anim Sci 2007, 46(1):20-22.
  3. Leventhall H: Low frequency noise and annoyance. Noise and Health 2004, 6(23):59-72.
  4. Eggermont JJ: Chapter 10 - Nonauditory Effects of Noise. In: Noise and the Brain. edn. Edited by Eggermont JJ. San Diego: Academic Press; 2014: 266-300.
  5. Rapley B, alves-pereira M, Bakker H: The inadequacy of the A-frequency weighting for the assessment of adverse effects on human populations; 2017.
  6. Pierre RLS, Maguire DJ: The Impact of A-weighting Sound Pressure Level Measurements during the Evaluation of Noise Exposure. In : 2004 ; 2004.
  7. Fletcher H, Munson WA: Loudness, Its Definition, Measurement and Calculation*. Bell System Technical Journal 1933, 12(4):377-430.
  8. Roberts C: ECOACCESS GUIDELINE FOR THE ASSESSMENT OF LOW FREQUENCY NOISE. 2008, 33.
  9. Pereira MA, João JdM, Castelo Branco NAA: PUBLIC TRANSPORTATION AND LOW FREQUENCY NOISE: A HEALTH HAZARD? 2004.
  10. Baliatsas C, van Kamp I, van Poll R, Yzermans J: Health effects from low-frequency noise and infrasoundsound in the general population: Is it time to listen? A systematic review of observational studies. Sci Total Environ 2016, 557-558:163-169.
  11. Commission IE: Electroacoustics—Sound level meters—Part 1: Specifications (IEC 61672- 1). Geneva, Switzerland 2013.
  12. SHEHAP AM, SHAWKY HA, EL-BASHEER TM: Study and Assessment of Low Frequency Noise in Occupational Settings, vol. 41; 2015.
  13. ACGIH: TLVs and BEIs based on the documentation of the threshold limit values for chemical substances and physical agents & biological exposure indices. In . : ACGIH Cincinnati, OH; 2019.
  14. Li X, Dong Q, Wang B, Song H, Wang S, Zhu B: The Influence of Occupational Noise Exposure on Cardiovascular and Hearing Conditions among Industrial Workers. Sci Rep 2019, 9(1):11524.
  15. Chiu C-H, Lung S-CC, Chen N, Hwang J-S, Tsou M-CM: Effects of low-frequency noise from wind turbines on heart rate variability in healthy individuals. Scientific Reports 2021, 11(1):17817.
  16. Schmidt JH, Klokker M: Health Effects Related to Wind Turbine Noise Exposure: A Systematic Review. PLOS ONE 2014, 9(12):e114183.
  17. Persson Waye K: Noise and Health - Effects of Low Frequency Noise and Vibrations: Environmental and Occupational Perspectives. In . Volume 4 , edn.; 2011: 240-253.
  18. Araújo Alves J, Neto Paiva F, Torres Silva L, Remoaldo P: Low-Frequency Noise and Its Main Effects on Human Health—A Review of the Literature between 2016 and 2019. Applied Sciences 2020, 10(15):5205.
  19. Alves-Pereira M, Castelo Branco NAA: Vibroacoustic disease: Biological effects of infrasoundsound and low-frequency noise explained by mechanotransduction cellular signalling. Progress in Biophysics and Molecular Biology 2007, 93(1):256-279.