Surveying and Assessing Noise
Prevent noise exposure to protect workers’ hearing
Methodologies have evolved that allow noise exposure databases to facilitate risk assessments (ASCC, 2008 and Hohmann, 2008). In addition, there are guidance notes, checklists and noise risk assessment tools which help to determine when noise is likely to be problematical (EU-OSHA, 2014).
While risk assessments can be undertaken without the need for noise measurements (HSE, 2005), whenever any significant risk exists it would be difficult to justify not using site-specific measurement data.
Noise risk assessment – legal requirements
In any assessment the key requirement is to assess risk using appropriate ‘criteria’. Throughout Europe specific legal obligations arise at the upper (UEAV) and lower exposure action value (LEAV). One of the legal obligations on employers is that they must make a ‘suitable and appropriate assessment of risk’, in consultation with the employees and/or their representatives. An appropriate assessment of the risk arising from exposure must be undertaken by a competent assessor. The primary purpose of the assessment is to clarify what needs to be done to protect the health and safety of all employees who are exposed to noise.
When carrying out a risk assessment, particular attention should be given to the following:
- The level, type and duration of exposure, including any exposure to impulsive noise
- The exposure limit values and the exposure action values
- The effects of exposure to noise on employees whose safety or health is at particular risk from such exposure (e.g. particularly vulnerable employees)
- As far as technically possible, any effects on employees’ safety and health resulting from interactions between noise and work-related ototoxic substances, and between noise and vibrations
- Any indirect effects on employees’ safety or health resulting from interactions between noise and warning signals or other sounds that need to be observed in order to reduce the risk of accidents
- Information on noise emissions provided by the manufacturers of work equipment
- The availability of alternative equipment designed to reduce noise emission
- The extension of exposure to noise beyond normal working hours under the employer’s responsibility
- Appropriate information obtained from health surveillance including, where possible, published information
- The availability of hearing protectors with adequate attenuation characteristics
The above constitute the minimum European legal requirements and they need to be interpreted relative to the likely magnitude of risk.
Recommended survey/assessment strategies
A combination of well-planned, reliable, representative measurements with good interpretation is essential to establish the level of risk arising from exposure to noise and to assess controls.
There are two principal types of instrumentation/survey methodologies to quantify noise exposure (dosimetry and static or SLM-based measurements). Initially a ‘walk-through’ survey of a facility can establish a list of all areas and workstations where the sound pressure level (SPL) is likely to approach or exceed 80 dBA (LAeq).
In planning the assessment it is recommended (ISO 9612:2009) that a logical strategy, as follows, is adhered to:
- Work analysis
- Selection of measurement strategy
- Conduct measurements
- Error handling and uncertainty evaluations
- Calculations, and presentation of results
The requirements of ISO 9612 do not normally apply to ‘routine assessments’ but more specifically they apply to the ‘determination of noise exposure to engineering grade; e.g. for detailed noise exposure studies or epidemiological studies of hearing damage or other adverse effects’. Nonetheless, certain aspects of the standard can be utilised to add reliability to the findings of routine assessments.
In all assessments, it is important that consultations are held with managers, supervisors, health and safety committees and employee representatives in scoping out the survey/assessment. This will help to ensure its representativeness and its acceptability. Furthermore, it is recommended that a Class 1 sound level meter (SLM) is used to determine LAeq values at the position(s) the employees occupy throughout their shift and at representative ‘static’ positons. The LEX, 8h can then be calculated based on the combination of LAeq values and the time spent in each area. Alternatively, if a worker moves throughout a limited area, we can determine the LAeq for the entire area (spatial average SPL) by moving the microphone to all those areas that s/he occupies (i.e., measure at positions that their ears would occupy). This can be done by ‘shadowing’ an employee for a representative period. Another option is to use the ‘worst case’ and the maximum LAeq measured at the noisiest position and assume that the worker spends all of the time in that position (worst-case assumption).
The choice of instrument influences the uncertainty of the measurements and Class 1 instrumentation is recommended as it is more precise than Class 2. In addition, the specified tolerance limits for Class 1 instruments are applied for the temperature range from -10°C to +50°C. This is important if exposure arises in refrigerated environments (e.g. cold stores). For Class 2 instrumentation the influence of variations in air temperature on the measured SPL is only specified over the range from 0°C to +40°C (IEC 61672-1:2002 and IEC 61252).
The ‘SLM or static-based’ methodology is particularly useful when workers operate at fixed positions and where the measurement results describe the noise at the operator’s ear, attributable to a specific tool, workstation or task. In some cases, it may be advisable that this is supplemented with a ‘second method of quantifying exposure’ – by using dosimeters which are worn by the workers throughout the working day or for a representative period.
Dosimeters generally display the total ‘noise dose’ over the sample period and/or report the results as a percentage LEX, 8h (in Europe, usually 80 or 85 dBA), and often include LAeq and other data. However, as the microphone is worn on the body (shoulder or lapel), there is very likely to be an additional uncertainty to dosimeter measurements because of localised noise disturbance. Despite this, the International Standards Organisation previously claimed them to be the preferred method for quantifying exposure (ISO, 1997). However, the ISO standard for workplace noise assessment was updated in 2009 and ISO, like many authorities, are now somewhat wary of dosimeters.
Many practitioners and researchers have first-hand experience of the dubious nature of dosimeter data, but this has apparently been overlooked for their ease of use, despite Earshen’s warnings as early as 1980. Nonetheless, current HSE guidance includes a note of caution and states that a dosimeter ‘tends to increase the potential false contributions to measurements and thereby the measured sound pressure level’ (HSE, 2005, p. 96).
In 2004 a landmark publication described major difficulties with dosimeter data and highlighted some of the pitfalls associated with dosimeter-based surveys. Kardous and Wilson investigated various noise measuring instruments at firing ranges, and while emphasising the impulsive nature of the noise environment, they concluded that dosimeter readings were ‘nearly always in error’. Their results highlighted limitations exhibited by dosimeters, which include peak pressure clipping, unreliable dose-response relationship, and overall lack of capability to record signal parameters that may better describe and quantify the hearing damage risk from exposure to impulse noise (Kardous and Wislon, 2004).
Many assessors have experience of elevated dosimeter SPLs which defied explanation and a thorough risk assessment should seek to explain where, when and how elevated SPLs arose. In some cases static measurements will verify or disprove the dosimeter data. However, preference should always be given to precision SLM data. According to the ISO:
“Any high peak sound levels recorded by the instrument which were not validated by observation shall be investigated and commented on in the report.” (ISO: 9612, 2009)
Among the dosimeter’s drawbacks is that an instrument will provide a limited number of ‘useful’ measurements per day, despite the availability of data logging units. Generally the dosimeter will be attached to a worker who for practical reasons cannot be constantly observed. Thus, dosimeter data is not generally ‘attended’ by the assessor and the assessor cannot honestly testify to its validity, unlike a measurement which was personally supervised in its entirety.
For dosimeters, the size of the windscreen is usually limited. However, by using a hand-held or tripod-mounted SLM with a larger windscreen, the potential effect of airflow-induced noise (e.g. ventilation) can be minimised. ISO, 2009 states:
“Contributions from wind and airflows depend on the wind speed and the size of the windscreen. A-weighted sound pressure levels around 80 dB are usually not significantly influenced by airflow speeds up to 10 m/s, provided the windscreen is of 60 mm diameter or more.”
Thus, dosimeter data that is sourced in impulsive, refrigerated and/or well-ventilated environments is likely to conceal potentially significant error. When we consider the use of SLM and dosimeter options along with various strategies for assessing noise, the measurement of noise from particular tasks with an SLM can be the most reliable method for various scenarios. Coupled with information on the duration for each activity, a range of exposures can be easily estimated. However, dosimeters are useful where access issues arise (e.g. workers in machine cabs) and when employees are particularly mobile.
Thus, workplace noise measurements can be effectively made on an ‘activity, job or task basis’ by using an SLM to log a series of LAeq and peak (dBC) SPL readings over selected times. The selected monitoring periods must be representative of operational cycles when normal or elevated noise levels are likely to be recorded (thus, the need for scoping and consultation).
The instrumentation must be calibrated on-site directly before and after each series of measurements and should be subject to a traceable calibration by an accredited laboratory (preferably within one year). The SLM should be generally held at arm’s length or mounted on a tripod. The measurement locations should be generally located close to the worker’s ear position (c. 15 cm from the head) in an orientation which generates the maximum sound pressure readout. Photographs of survey positions and floor plans help clarify matters and add certainty to the data.
The duration of the sampling interval will be determined by the nature of the sound source and should always be long enough to obtain a representative measurement. If the noise is steady, a short term sample will be sufficient (e.g. 60 seconds), although ISO (2009) generally advocates a minimum sampling period of five minutes. If the noise level changes rapidly and varies, we must wait for the LAeq value to ‘flatten off’ and stabilise to within 1 dB. If the noise is from a cyclic operation we must sample over a number of cyclic activities/operations (a minimum of three). Thus, the measurement duration should be tailored to the variations in the workplace noise and this will be largely dependent upon the nature of the work and the characteristics of the noise. We should also ensure that any short-duration high level noise exposures are also included as these can have a significant effect on the Leq values.
Observations on activity levels and work patterns should be recorded and as well as logging noise levels, a range of data should be collected, such as variations in employee work patterns, use of portable hand tools, machine speeds or loads, rate of work, variations in production and/or raw materials which may affect the sound levels. Very often localised factors will determine what sampling strategy should be adopted and sometimes a preliminary walkthrough and scoping visit is recommended to assist the planning and scoping of the assessment.
It is recommended that similar exposure groups (SEGs) are identified and that representative samples and noise measurements are undertaken for all pertinent SEGs. At all times, notes should be taken of the operational conditions (machine speeds, products etc.). In addition, consideration should be given to maintenance and cleaning activities, when different exposure scenarios may arise.
The survey is only one part of the process and the determination of daily personal exposure involves a series of calculations which take account of the sound level at specific positions (or during jobs/tasks) and the time spent at these positions/tasks during a ‘normal’ working day.
The daily noise exposure level of workers can be determined using the following formula:
LEX, 8h = Leq + 10 Log (T/8)
where Leq is the measured noise level and T is the exposure time in hours. This formula is used when only one form of activity or exposure arises over the working shift (or when the measured noise captures any variations). In situations where a range of tasks, activities and/or noise levels arise, the following formula is used:
LEX, 8h =10 Log [(1/8) (t1_100.l L1 + t2_100.l L2…+tn_100.l Ln)], where,
Ln is the SPL to which the employee is exposed (Leq for the nth measurement period) in dBA and tn is the duration of the exposure, i.e. the corresponding exposure time – nth measurement hours.
An estimate of exposure time is fundamental to the calculation of LEX, 8h and in each case consultations should be held with the individual workers and their supervisors in order to determine the appropriate time. As a form of quality control, it is recommended that a combination of calculation methods is used. HSE’s exposure calculators and ready-reckoners are available at: http://www.hse.gov.uk/noise/calculator.htm.
Under certain conditions, the Physical Agents Noise Directive (2003/10/EC) permits the use of weekly noise exposure levels for activities where the daily noise exposures vary markedly from one working day to the next. The use of weekly exposure levels is only likely to be appropriate when daily noise exposure on one or two working days in a week is at least 5 dB higher than the other days, or the working week comprises three or fewer days of exposure.
The weekly personal noise exposure, LEP,w can be determined using the following formula :
LEP, w =10 Log [(1/5) (100.l LEP,d1
+ 100.l LEP,d2 + 100.l LEP,d3 + 100.l LEP,d4 + 100.l LEP,d5 + 100.l LEP,d6 + 100.l LEP,d7)]
In the above formula LEP,d1 is the LEX, 8h (LEP,d) for working day one, d2 for day two etc. It is important to remember that when we use the LEP, w formula, we should always use a reference time of five days, i.e., the nominal working week is five days.
Reporting and Uncertainty
Article 4 (5) of Directive 2003/10/EC expressly states that the assessment of the measurement results shall take into account the measurement inaccuracies determined in accordance with metrological practice. Therefore, a thorough risk assessment will include a comprehensive dataset and a calculation of the expanded uncertainty (JCGM, 2008 and ISO 2008 - Guides to the expression of uncertainty).
There have been a number of recent updates in the standards relating to instrumentation with BS EN 61672-1:2003 being replaced by BS EN 61672-1:2013. However, the limitations of the instrument forms only part of the uncertainty budget (Payne, 2004).
Some of the key elements to be addressed in the report include:
- Details of all measurement results, monitoring conditions and survey data
- Calculations or estimates of employees' exposures, assumptions used, and uncertainty
- Comparison of the employee exposures with the statutory action values and exposure limit values
- Consideration of variations in SEGs
- Identification of all workstations/activities where there may be a risk from noise and who is likely to be affected
- An assessment of exactly when and where hearing protection will be required
- An assessment of the suitability and likely performance of hearing protection under site-specific conditions
- Details of all areas and activities where noise-control measures are required
- Preliminary recommendations on the types of control measures which are likely to succeed
- Identification of all employees who need to be provided with health surveillance and training
- Identification of workstations and or activities for which noise control measures should be prioritised
The findings of the risk assessment must be carefully recorded and an action plan should be developed to identify and document the steps taken to meet the requirements of the law.
Quantifying exposure is just one small part of managing the risk of noise-induced hearing loss (NIHL). It is essential that individual workers personally engage in Hearing Conservation Programmes (HCP) if successful outcomes are to be achieved (Royster and Royster, 1998). Thus, high quality training and awareness programmes are a pre-requisite for all relevant workers.
Where significant exposure exists, it is considered good practice that an annual or biennial review of the noise assessment is undertaken. However, in the event of a case of NIHL, any changes in work practices/equipment, or if there is another reason to believe that it is no longer valid, then a sooner review of the noise risk assessment should be undertaken.
Worldwide noise-induced hearing impairment is the most prevalent irreversible occupational hazard. Noise is the most common cause of tinnitus, and according to WHO (1999) and leading authorities it is also the foremost harmful physical agent in the workplace (Soalheiro et al, 2012). The merits of HCPs have been well established and a comprehensive structured and coordinated approach is required to manage the risk of NIHL. Undertaking surveys and providing hearing protection will not protect workers’ hearing. Nor will any amount of safety signage or vague commitments to ‘purchase quiet equipment’. Moreover, periodic thorough assessments, effective noise avoidance and control measures, training and awareness programmes, properly managed hearing protection schemes and coordinated audiometric assessments must be all be provided. Thus, all of these critical elements need to be applied if we are to make significant inroads in reducing the alarming and intolerable incidence of NIHL.