Industrial Noise
Noise for health and safety
Sonomètres Pulsar 
Lärmmessgeräte Pulsar Industrial Noise
Despite the existence of Noise at Work regulations since 1989, “How do we measure it ?” is a common question noise professionals hear, so perhaps some light can be shed on what seems a complex subject.
Noise of course is simply the effect of changes in air pressure impinging on the ear. The basis of the regulations is that if the accumulation of this pressure change is too great over a period, damage will result. The range of pressures a human can hear is from about 0,001 Pa (Pascal) up to about 200 Pa. Such a range is hard to deal with so we first ‘A-weight’ the sound to mimic a notional ears response, then take the logarithm of the pressure and scale the results from 0 to 140 to get “Sound Level” in decibel; very simple in concept but it makes calculations hard.
More data on decibels and the various noise metrics are on this site
What we measure.
In industry two different noise metrics are used to determine safe exposure. One effectively representing the energy of the noise as an average level and another representing the absolute highest pressure occurring, called the PEAK. The ‘energy’ metric in the UK from 2006 is to be in terms of LEX, the average level over an 8-hour daily period, this is currently called LEP,d but only the name is being changed for the new regulations, the metric is exactly the same. Formally Lex is the Leq or average level normalised to an 8 hour working day. Different actions need to be taken at different levels, but from 2006, there will be an absolute Limit Value of 85dBA for the ‘average’ level.
Today there is only one PEAK ‘Limit Value’ of 140 dB(C) above which no worker must be exposed (200Pa in pressure), but from Feb 2006, there will also be two ‘action’ levels of 135dBC, 137dBC and the same ‘limit value’ of 140dBC peak that exists today.
How to protect the workers and meet the regulations?
First you have to measure the noise levels and there are two basic ways to do this, either by measuring the actual EXPOSURE of each individual worker with a Noise Dosemeter, now usually called a Personal Sound Exposure Meter (PSEM), or measuring the noise in the whole area and deducing the exposure of each worker using a Sound Level Meter (SLM). Traditionally, professional acousticians have considered the SLM to be the more accurate method, mainly because it was considered that workers could modify dosemeter data. As well, classes of SLM exist that are more accurate in the laboratory than PSEM, even if in the field this accuracy has little or no advantage, especially as untrained users can easily make errors when using an SLM. Today, most logical people think that in the real world, measuring at the ear of the exposed worker is likely to be more relevant to the situation and as a result PSEM are used.
USA industry being more practical, has always favoured the use of dosemeters simply because they measure the actual noise affecting the worker. However, now a new generation of ultra-miniature dosemeters is available, such as the Noise DoseBadge [page link] and this gives huge improvements in accuracy and ease of use and allows the Health and Safety regulations to be met at far lower cost.
The shoulder mounted noise doseBadge is so small that it does not affect the worker in any practical way and it also allows many workers to be monitored relatively automatically. Further, as the badges have no user controls and no obvious read-out, the wearer is far less tempted to ‘modify’ the data; thus these new noise badges have revolutionised factory noise data collection. Traditional PSEM, which were belt-mounted with a cable to a microphone on the shoulder such as models 66 & 67, are still available and some older users prefer these as have many more features, although few of the added features are needed to meet the EU Directive but may be useful for data analysis.
As an aside, a few genuine noise experts have very firm and long-held opinions as to which method – SLM or PSEM – is the better and will defend their position as though it was a religious or political argument. Such non-scientific, dogmatic, thinking does not help the cause of noise control and can confuse those not so expert. SLM and PSEM methods both allow the regulations to be met and providing the instruments used have some form of ‘official’ performance approval; either is acceptable.
Sound level Meters
While for most routine noise measurements, the PSEM is the method of choice; there are still some measurements, where an SLM may be required. For example, if the overall level in dBA is too high, some form of hearing protection must be provided to bring the level at the ear below 85dBA. For the greatest noise reduction, the chosen ear protection should match the SPECTRUM of the noise and to do this the frequency content of the noise must be measured. The two main ways are to divide the frequency spectrum into OCTAVE or THIRD OCTAVE bands. There are typically 9 octaves to be measured and 27 third octaves, so octaves are easier to manage and for this application are almost always adequate. An SLM which measures octaves exists in two forms, a REAL TIME ANALYSER or RTA, such as Model 30 where each octave is measured simultaneously, or the manual method where each octave is measured one at a time, Model 73 is an example. Clearly as noise is changing all the time, the manual method may not be measuring the same noise in each band, so with manual instruments either an average must be taken or the noise must be very steady. Like the PSEM versus SLM argument, some experts will argue for one over the other, often passionately. However, most of the historical noise data was taken using manual instruments and in the hands of an expert, they really have little if any disadvantage, except where single impulses are found.
How to do it?
The first action is to obtain and read your national regulations and any advice leaflet, in the UK this is in the form of an HSE booklet, which gives excellent and accurate advice as well as the nomogram needed where the noise varies. Then a survey must be done to identify all persons who may be exposed to noise at or above the first action level. Logic suggests that the first thing to do is to list all areas and then do a simple survey using a conventional sound level meter. Now categorise each area, for example
| Level in decibel A-frequency-weighted | Action required to meet the regulations |
| Areas always below 75dB | Report only – no other action |
| Areas between 75 & 85 dB for a proportion of the time |
Initial survey using an Leq meter or PSEM |
| Areas where levels above 85dB are found |
A full survey with clear identification of levels & action |
A Full Survey will include full details of all the noisy processes, in other words, mapping the various noise sources, determining the people exposed and the various levels; all obvious matters. The subsequent report should include immediate actions needed – such as ear defenders - as well as longer term noise reduction strategies. It is clear that in any noisy space, the loudest noise must be used for everyone in the vicinity, unless a separate measurement is taken at each worker’s location.
The Guide requires an instrument that can measure with the following minimum characteristics:
| A-frequency weighted Leq |
| C-frequency weighted peak around 140dB |
| At least Type 2 accuracy class |
The Pulsar Instruments, models 71 and 72 manual units, or 63, 64 data logging, or the dedicated new Assessor all have these characteristics. Class 1 instruments such as models 71 are more accurate – IN THE LABORATORY – but in a factory with echoes off the walls, multiple noise sources, a lower cost Class 2 is acceptable to most people.
It also suggest the following additional features if ear protection is needed.
| C- weighted Leq and/or octave-band analysis |
Models 73 and 74 (manual) and 30, 30-2, 33 and 33-2 (Automatic) all provide these functions.
It is required that the meter is calibrated BEFORE & AFTER each set of measurements. If it is correct before and after the readings, you have an excellent chance it was correct DURING them and has not been accidentally damaged. Pulsar Acoustic Calibrator model 100 is the correct unit for all these sound level meters and a version of this has been fully Pattern Approved by the PTB in Germany. Many users are careless about such routine calibration, but it is essential if your noise records are to be believed. Always note that you have done it in the report..
A full survey using a PSEM
The problem is approached differently. Mapping the various noise sources is not so important, but it is still done ‘for the record’. Instead the main data produced is a database of each employee, giving all the relevant noise data and perhaps other pollutants. The data will include the various departments or even individual machines where the employee is located, even though when using a PSEM the physical location of an individual exposed person is less important. It is the actual exposure of each person that is logged and recorded.
In some ways many acoustic problems disappear when a PSEM is used. The many reflections from the various walls and machines are not directly relevant, as the sum total of noise at the workers ear is what is measured – and that is what we are trying to control. The important matters are that the PSEM is correctly calibrated and that the worker wears it all the time and that these facts are recorded. Clearly, for most people, using a PSEM, in particular one like the doseBadge, is a quicker and easier method than the tedious plotting of each and every noise source.
To use a PSEM, many precautions are the same as for an SLM. The batteries must be charged and the unit must be calibrated before the measurement. The PSEM is then fixed to the worker as near to the ear as convenient and set running. In the case of the doseBadge, simply pointing the READER at the badge and setting it going with the infrared link does this.
At the end of the working shift, the badge is stopped by the same method and removed. Back in your office, each badge is now downloaded by the Reader – up to 100 badges may be done, but ten are more typical – and the data is now available to read. Of course most people will now transfer all the data to a PC using the software provided when it is automatically inserted into the noise database in the correct place. It is clear that one of the main reasons to use a PSEM is the simplicity of the process.
Further action.
However the data has been taken, the following is now done.
L EP,d or Lex below 85dB(A)
Below the first action level, you must ‘consider if anything can be done to further reduce the noise’. This is a platitude and not a sensible legal requirement, but nevertheless a wise suggestion.
L EP,d or Lex between 85 & 90 dB(A)
Between the first and second action levels, the employer must ‘on request’ provide suitable and efficient hearing protection. To do this, you need to know the frequency content of the noise and select the ear defenders to match this noise. To do this properly, an Octave Band sound level meter is used and the Pulsar models 30, 73 and 74 are very suitable. All these have Leq and Peak as well as octave bands. The software for the 73 and 74 provides the hearing defender selection for you, based on the manufacturers data.
L EP,d on Lex above 90dB(A)
Above the second action level OR if the Peak level is exceeded, the worker must be removed from the noise. Ideally, this should be by reducing the noise, but the employer may be allowed to ensure that GOOD AND EFFECTIVE ear protection is worn. The criteria for selection of such ear defenders are given above. Above the second action level, protection is mandatory and the ear defenders MUST get the noise level at the ear below the second and/or peak action levels. Regulation 11 requires the employer to provide information on the ear protection and how to use it. All zones where the levels are above the 2nd or peak action levels shall have a warning sign and all tools which create such noise shall be labelled warning employees to wear protection. As well, specific tasks, which are in levels above action level 2, must be identified. Cleary, the best thing to do is to reduce the noise AT SOURCE, by screening machines, slowing them down, changing technology or any of the many ways available
Who should measure?
In the UK, a ‘competent person’ should do it. One way to become a competent person is to attend a short course – typically four days - such as those approved by the Institute of Acoustics [link]. These are excellent and after a short examination, offer formal certification. Universities and other educational organisations, such as colleges of Further Education, mainly run these.
However, Pulsar Instruments offer a short “One – Day” course written and presented by members of the Institute that give an insight into the basics of noise and the use of SLM and PSEM to measure it. While such a 0ne-day course does not substitute for a longer formal course, it does allow a non-scientific user a ‘start’ in the art and science of noise measurement. The course offers as little theory as possible – consistent with understanding – but concentrates on the practical use of the instruments and the “does and don’ts” of noise. In motoring terms, it is a course for drivers, not mechanics.
Very copious notes are provided and these are not only useful on a day-to-day basis but provide a very valuable reference source, including “all the theory you will ever want” in an appendix. This course is available to anyone, but purchasers of Pulsar Instruments are offered a discount.