Noise dosimeter
From Wikipedia, the free encyclopedia
A noise dosimeter (American) or noise dosemeter (British) is a specialised sound level meter intended specifically to measure the noise exposure of a person integrated over a period of time; usually to comply with Health and Safety regulations such as the EU Directive 2003/10/EC, or the equivalent American OSHA rules.
Contents |
[edit] History
The first dosimeters were designed in about 1969 as a result of the American Walsh-Healey Act (50-204.10). The regulations resulting from this legislation set legal limits on the "amount" of noise to which a worker could be exposed. This act was followed by the OSHA - Occupational Safety and Health Act of 1970, which was passed by the U.S. Congress and became effective in April, 1971. The full text of this act can be seen in the Code of Federal Regulations, Title 29, Chapter XVII, Part 1910.
It was recognized that the cumulative effect of noise exposure is a prime cause of industrial hearing damage. While it was known that the damage was proportional to a combination of sound pressure and time, the precise relationship between noise level, exposure time and the resulting hearing damage was not then well understood.
The American OSHA noise exposure limits, with the benefit of hindsight, are now widely discredited outside the USA, but at the time they were the best that could be done and were the essential first step in protection against hearing damage risk due to noise. It was a revolutionary regulation and reflected the then best current acoustic knowledge.
The act and subsequent Occupational Safety and Health Administration (OSHA) regulations set the permitted noise dose by a trade-off of time and level following the data in table 1.
| Exposure duration per day - in hours | Sound level in dB(A) |
|---|---|
| 8 | 90 |
| 6 | 92 |
| 4 | 95 |
| 3 | 97 |
| 2 | 100 |
| 1.5 | 102 |
| 1 | 105 |
| 1/2 | 110 |
| 1/4 or less | 115 |
Table 1, the OSHA permitted noise exposure levels
When this regulation was introduced, measuring sound level was poorly understood and only a handful of manufacturers offered suitable equipment world wide. The act led to a huge expansion in suppliers of sound level meters and dosimeters, about 30 different brands being introduced, mostly in the United States, although a few United Kingdom and other companies were also started. Today, most of these have ceased trading, once again leaving just a handful of companies world-wide manufacturing noise dosemeters.
One of the classic meters from the early 1970s is shown in figure 1, the Dupont dosimeter. This stored the data on a chemical cell.
[edit] Measuring industrial noise
In many factory noise situations, a sound level meter is a perfectly suitable device to check compliance with OSHA or EU requirements. If a Class 2 sound level meter is used, or what used to be called a "Type 2," with due allowance made for the measurement uncertainty, and in any part of particular premises there is no noise level above say 80 dB(A)S, (80 decibel A-frequency and S-time-weighted), it is clear that every worker will be below a limit set at say 85 dB(A) and there is no need to use more complex measurement devices. However, if it is suspected or known that sound levels above 85 dB(A)S are found, a dosimeter was traditionally the usual instrument required - at least in the USA and its sphere of influence.
Work mainly in the UK and Germany after about 1970, demonstrated that the '5 dB doubling' rule in Table 1 above and used in the USA did not correspond very well with hearing damage risk and the International Organisation for Standardization (ISO) recommended an 'equal energy rule', where a 3 dB increase in level halved the permitted exposure time; 3 dB being a doubling of the energy although a 6 dB increase is a doubling of the sound pressure. The ISO document started at the same point of 90 dB(A) for 8 hours as the criterion value, but then the exchange of time and level followed Table 2.
| Permitted Exposure time | Level in dB(A) |
|---|---|
| 8 hr | 90 |
| 4 hr | 93 |
| 2 | 96 |
| 1 hr | 99 |
| 30 min | 102 |
| 15 min | 105 |
| 7.5 min | 108 |
| 225 sec | 111 |
| 112.5 sec | 114 |
As can be seen, a worker in the USA could be exposed to 110 dB for 30 minutes, whereas under International rules the maximum time was about 3.8 minutes - a very significant difference; the American worker being less protected .
[edit] Other differences in exposure description
There was also a difference in some countries how levels below the criterion level of 90 dB(A) were handled. Some authorities thought that there should be a threshold at 90 dB(A), so that a level of 87 dB(A) counted as zero and the exposure time for levels under 90 dB(A) could be infinite; others said that the 3 dB exchange should continue down, so that at 87 dB(A) the exposure should be limited to 16 hours. Even others suggested a threshold that was not the same as the criterion level having perhaps the criterion at 90 dB(A) but a threshold at say 87 dB(A).
Because the early regulations - both American and International - had 90 dB(A) for 8 hours as their criterion, this was counted as "100% dose" and as a result, many early dosimeters were scaled in terms of 'percentage dose'. This while being simple to understand was very misleading indeed. Clearly '100% dose' in the USA and the rest of the world was different, except in the special case of a steady level of 90 dB(A), but it took some time before it was realised that there was no way of converting a '%dose' taken under OSHA rules to the ISO equal energy rules. Indeed as late as 1974, an American paper at the International Congress on Acoustics in London tried to show how to do a conversion, but it contained a significant error in the maths.
One further added complication was that the US Air Force, to try and improve their health protection, had yet another doubling rule where the time was halved for every 4 dB increase, but still starting at 90 dB(A).
The tables 1 and 2 show the permitted exposure times for different levels, both based on A-weighting, but there are other more complex differences. The OSHA rules assumed an exponentially averaging sound level meter with S-time-weighting (originally called Slow Time Constant) such as was given by the dc output of a classical sound level meter; while others based their rules on a linearly averaged pressure squared metric, for example the noise exposure in Pa2h. These two systems gave radically different outputs from the detector, the difference depending on the form of the noise and they are incompatible, although instrument tolerances may hide the difference in some cases.
Finally, the American ANSI standards demanded that the microphone be calibrated by a random incidence wave - that is sound coming from all directions, whereas the International standards called for a plane wave coming from a single direction; again these two systems are incompatible.
The one constant was that all regulations used A-frequency-weighting, even though even this was specified slightly differently in the USA.
[edit] Variations in legal requirements
This meant that the user had to specify under what legal regime he wanted to use his dosemeter, as each country could - and did have their own local laws for such measurements. In each region, sometimes down to regions with tiny populations such as Western Australia, only one combination was usually legally accepted and that could be very different to an adjoining region. While these decision may have seemed sensible at the time, users were in general not adequately informed of these problems and this resulted in many measurement errors.
The five main parameters that were commonly different in different political entities were:
- Threshold
- Criterion level
- Exchange rate
- Exponential or linear integration
- 'Head-on' or random incidence calibration
There were of course others, such as the single event maximum level, the peak value etc.
To reduce the variations of instrument required to be manufactured, some commercial companies made complex "Universal" units where all these things could be selected by the user. As measuring noise was not well understood, such complexity clearly militated against accurate results and many anecdotal stories give examples of huge errors resulting; as very few unqualified users understood the complex issues involved.
[edit] International standards
The international body that specifies the technical requirements of such instruments as sound level meters and dosemeters is the International Electro-technical Commission (IEC) based in Geneva; whereas the method of their use is normally given in an ISO publication. However, in any particular political region, local laws apply as the IEC and ISO publications only have the status of "recommendations" and so countries could - and did - have their own sets of rules - many of which were technically flawed and in some cases scientifically impossible. Every new regulation thus made the concept of "%dose" more meaningless. The "100%" dose was different in different countries, but many users did not understand this and would buy low cost USA built dosemeters where the American "100%" was not correct for their local regulations and usually very much under-estimated the noise exposure.
[edit] The permitted levels reduced
During the 1980s and 1990s many workers - led by Scandinavia - determined that the '90dBA for 8 hours' limit was far too high and an unacceptable number of workers would be damaged at these levels, so a level of 85 dB(A) for 8 hours was felt to be a better criterion. Even later the EU reduced the limits still further to the 80 dB(A) we have today, as given in the UK's "The Control of Noise at Work Regulations 2005.". These regulations follow closely the EU Directive 2003/10/EC, normally called the Physical Agents Directive.
A further complication for the sound level meter designer was that it was realised that a single very high noise peak could instantaneously damage hearing, so a limit was originally set by the then European Community so that no worker should ever be exposed to an rms acoustic pressure of more than 200 Pa - equating to 140 dB re 20 μPa - and that this should be measured using an instrument with no frequency weighting. This while a good idea, was patent nonsense as 200 Pa could be generated by trains going through tunnels, closing a door, in fact many everyday things could cause such a pressure wave below the frequencies that could be heard or could cause hearing damage. Accordingly C-frequency-weighting was specified to measure the peak level as this has a flat frequency response between 31 Hz and 8 kHz. However, this missed a significant amount of important energy and a new frequency weighting of 'Z' (zero) weighting was specified in IEC 61672 : 2003 that has a flat response from at least 20 Hz to10 kHz.
[edit] The PSEM arrives
To try and simplify the situation, in the 1980s Working Group 4 of IEC Technical Committee 1 started to write a new standard for a dosemeter, but decided that for many good reasons, a new name was called for and the long - but more correct name - of Personal Sound Exposure Meter (PSEM) was used. WG4 was mainly made up of design engineers from the International Sound Level meter companies together with scientists from various national acoustic laboratories and a few academics. The result of their efforts became IEC 61252 :1993, the current PSEM standard. This has tolerance limits based on a Class 2 - what was a Type 2 - sound level meter, but because it is intended to be worn on the body, it has relaxed directional characteristics.
The favoured metric for many scientists was simply the sound exposure in pressure-squared-time, for example Pa2h and this was used for the PSEM standard and at a stroke this removed all the various options for measurement. However for health and safety legislation in Europe for legal purposes the metric chosen was the daily personal noise exposure level, LEP,d, which corresponds to LEX,8h as defined in international standard ISO 1999: 1990 clause 3.6, and is expressed in decibels A-frequency-weighted [dB(A)]. In simple terms this is the normalised sound exposure expressed in decibels.
[edit] Use of dosimeters
The original dosimeters were designed to be belt worn with a microphone connected to the body of the dosemeter and mounted on the shoulder as near to the ear as practicable. These devices were worn for the full work shift and at the end would give a readout initially in percentage dose, or in some other exposure metric. These were the most common way of making measurements to meet legislation in the USA, but in Europe the conventional sound level meter was favoured. There were many reasons for this, but in general in Europe the dosemeter was distrusted for several reasons, some being.
- The cable was considered dangerous as it could catch on rotating machinery
- The dosemeter could tell you the level had been exceeded, but it did not say when this happened
- Workers could falsify the data very easily
- The device was big enough to affect the work pattern
In the USA - where most of the early devices were manufactured, these reasons did not seem to matter so much.
To remove these European objections, dosemeters became smaller and started to include a data store where the Time History of the noise, usually in the form of Short Leq was stored. This data could be transferred to a personal computer and the exact pattern of the noise exposure minute by minute plotted. The usual method used was to store data in the form of Short Leq, a French concept that helped to bring computers into acoustics. As well, dosemeters started to incorporate a second C-frequency-weighted channel that allowed the true peak to be indicated. By the time the PSEM standard was published, many major sound level meter companies - in both Europe and the USA had a dosemeter in their range. A typical noise dosemeter is shown in figure 2
[edit] The 21st century devices
The next technology breakthrough came when in the 1990s the United Kingdom Department of Trade and Industry awarded a SMART grant to Cirrus Research to design an ultra-miniature dosemeter. It was to be so small and light that it would not affect the worker and as well was to have no microphone cable. The resulting device (Fig. 3) - the first true dosebadge - was a twin channel device able to meet all the requirements of the European Directive and also the market need for data storage. The device had no internal display nor any controls, so workers would not be tempted to try and 'modify' the readings; instead the data was transmitted and the device controlled by an infra-red link .
Today, such devices are available from several manufacturers - at least one with a full Intrinsic Safety certificate for use in hazardous atmospheres. Some sophisticated ones have extra channels to store data on the state of the battery, any 'out of range' signals and some are able to be used on the USA OSHA as well as EU equal energy rules. Others have a "Time History" store, where the exposure minute by minute is stored for the full working shift, allowing Health and Safety Officers to pinpoint the exact time of any high energy noise and assist in determining the cause.
The most recent design innovation for noise dosemeters offers a multiple-point approach to assess workers daily noise exposure. This approach relies on the deployment of a set of dosemeters that have wireless (Bluetooth) communication with a handheld device such as PDA (fig. 4).
[edit] References
- EU Directive 2003/10/EC, normally called the Physical Agents Directive.
- Occupational Safety and Health Act. Code of Federal Regs, Title 29, Chapter XVII, Part 1910.
- IEC 61252 :1993 - Acoustics - Personal Sound Exposure Meter
- Wallis A.D. & Krug R.W. "A data storing dosimeter" Proc IOA Vol 11 part 9 101-106 Nov 1989.
[edit] External links
The following major manufacturers are among those who offer current technology PSEM, claimed to comply with IEC 61252 : 1993. and who were involved in the preparation of the PSEM standard and appointed the named engineers - among others - to the international working group.
- Denmark Bruel & Kjaer P. Hedegaard
- United Kingdom Casella CEL Ltd. R.Tyler
- United Kingdom Cirrus Research plc R.W. Krug
- United Kingdom Pulsar Instruments plc A.D.Wallis
- USA Quest Technologies Inc. E. Kuemmel
The following other major manufacturer produces instruments that claim to comply with the requirements of the standard but was not involved in the preparation of it.
- France 01dB-Metravib
