Everything you need to learn about controlling noise.
General Noise Control Factors:
The various general noise factors are as follows:
ADVERTISEMENTS:
Energy Contained in a Sound Wave:
When work is carried out, a minute quantity of the work done by most engineers is not directly converted into useful work or into waste heat but is radiated at sound. It has been the minuteness of the quantity of energy required to produce a sound that is a major factor in the difficulty of noise reduction.
A machine produces a mere 10-6 to 10-9 of its total output as noise, but this minute fraction of sound may be representing a noise level of a shattering 100 dB (A) or more, such is the sensitivity of our ears. Not only that, but if the noise is having a particularly characteristic, such as a tone, or an impulsive character, or an intermittency, or if it carries information, the brain will rate the annoyance value of the noise at an even higher level than a straight forward measure of the sound level would indicate.
Thus, in trying to control sound we are considering after a minute fraction of the energy involved in the work producing process. It has been therefore difficult at the outset to redesign an existing device so that it would be inherently less noisy.
On the other hand, if a device has been badly made it can easily put out a much larger fraction of its power output as noise. In this case noise control has been much easier. However, there have been not many things that are built to be unnecessarily loud, unless it was the intention of the designer and the user.
It has been of course, pointless to class device which have been intended to produce noise such as the loudspeaker and the football rattle, has controllable noise sources their designers and operators and operators would generally wish them to be even louder if possible.
ADVERTISEMENTS:
Outdoor Sources:
As a first step, we should consider the type of noise source that must be controlled. Amongst outdoor noise, road traffic noise has been most widespread and usually the loudest. In recent years, the problem has become so intense that special measures, have been taken to limit the growth of this noise and may be even to reduce it.
There is a more localised form of noise, which produces levels far greater than that produced by road traffic, i.e., around airports. Peak levels well over 110 dB(A) can occur miles from the airport and whole communities can suffer continued exposure to extreme amounts of noise.
ADVERTISEMENTS:
A third important source of noise has been due to construction. A construction site, even though it is a temporary thing, is able to generate a great deal of annoyance during operation which can last up to three or four years. This annoyance gets exaggerated by the resentment that persons suffering from the noise feel about the change in their environment caused by the construction work and this tends to exacerbate.
Noise from factories has been a widespread source of complaint. Finally, there happens to be a domestic noise from one’s immediate neighbours which very often has no general solution, but has to be dealt with as individual cases.
Indoor Sources:
Mainly, these concern the control of noise within a factory and other places of work, and are receiving increased attention from the government. The control of indoor noise is important. Requirements of privacy and intelligibility for conversation in office mean that the quieter noises in offices must also be controlled. At the bottom end of the loudness scale, the stringent requirements for quietness that exist for dwelling houses imply that noise control of service systems needs attention.
The Approach to Noise Control:
ADVERTISEMENTS:
Noise control methods naturally break down into three parts- noise control at source, noise control along the path that the noise must take to reach the receiver, and noise control at the receiver, which includes his immediate acoustic environment.
Method of Measuring the Noise Level from Small and Large Sources:
Small sources of noise should be measured by placing sound level meters 1 metre from the source, at point around the source, spaced out. A large source should be measured at representative points around the source. In both cases, a measurement should be taken at any position where on operator’s ear has been likely to be for substantial lengths of time.
Noise Control at the Source:
There are at least three primary areas in which control of noise generated by the source may be initiated- proper design, proper equipment operation and equipment maintenance.
The proper design of equipment to minimise noise generation has been somewhat a complex engineering problem needing a strong background in the fundamental of vibrations, fluid mechanics, dynamics, materials, machine design. It should be apparent that in a treatment such as this, it has been impossible to develop the background needed to effectively design for noise control.
In a very general way, some of the things that may result from the good noise design are:
1. Using shock absorbing techniques for absorbing impact energy (for example, the use of non-metallic gears to reduce the noise generated by the metal-to-metal impact associated with metallic gears; the use of flexible mounts to support would planer knives to reduce the noise generated by the impact of the knives on the board).
2. Using efficient flow techniques for reducing noise associated with high-fluid velocities and turbulence (for example, the new quiet hydraulic pumps in which the flow paths have been redesigned to give less turbulence; the reduction in noise generated by control valves in both gas and liquid systems achieved by simply reducing system pressure to as low as possible).
3. Reducing fluid jet velocities as jet noise is proportional to the eight power of jet velocity. Often the air jets from pneumatic control valve exhaust ports are a major source of industrial noise. By reducing such jet velocities much quieter operation could be attained.
4. Reducing sound-radiating areas. For example, as much as 5 dB A reduction has been attained at the operator of industrial sewing machines by detaching the metal cloth table from the machine and mounting the table separately on a wooden workbench. Also, in some situations it has been possible to divide a large radiating surface into smaller areas, thus changing both the frequency and magnitude of the noise emitted.
5. Reducing peak accelerations by using the maximum time to produce required velocity or displacement changes. This is critical in the design of cams and cam followers. In many industrial machines, such as those used in making straight pins, the major noise sources have been the cams and followers.
6. Significantly separating operating speeds and resonant speeds. As large amplitude vibrations are associated with resonant speeds, noise generation could be reduced by keeping operating speeds well removed from the resonant speeds to prevent excitation to resonance.
Although this list is incomplete, one general conclusion can be drawn the optimum time to incorporate engineering design considerations relating to noise has been the initial design phase. Many of these considerations offer potential for abating the noise generated by existing sources; however, they would usually need extensive and expensive redesign and modification.
The second primary area has been proper operation. All equipment should be operated at the design conditions. Operating equipment at design pressure and speed should result in minimum noise generation.
Several areas relating to proper operation are:
1. If acoustic guards, covers, or enclosures have been used, make sure that they are in place and that all openings are acoustically sealed. Noise is a lot like water -it will rush out through any cracks or openings.
2. If the equipment has been equipped with a muffler system, makes sure that it is working according to design. Since many mufflers depend upon absorption within the case, significant changes in insertion loss can occur with age and use.
3. We should install machinery on adequate mountings and foundations to reduce structure borne sound and vibrations. Many good isolation pads are commercially available. Make sure your choice isolates both sound and mechanical vibrations. An isolator that is good for mechanical vibrations may pass the higher frequency sound energy without significant reduction.
4. We should use proper cutting speeds and feed rates to control tool chatter in machining operations.
5. We should provide good support for tool bits and work-piece to reduce vibrations. Often, an increase in constraint will decrease these vibrations and eliminate the noise generation.
6. We should apply additional sound control devices. Inlet and discharge silencers or mufflers can often be added with a small investment. Effective acoustic enclosures have been often easy to design and construct.
The important point to be made here has been that the manner in which a piece of equipment is operated can significantly influence the magnitude of the noise generated.
The third primary area has been equipment maintenance. Many maintenance techniques are known that can effect noise. Some of these could just as well have been classified under the operation area.
The following list proves potential maintenance activities:
1. We should maintain good dynamic balance. It decreases rotating forces and normally reduces noise generated by secondary sources, such as shaking guards, shells, of vibrating enclosures.
2. When purchasing replacement components, like gears, motors, and pumps, we should not ignore noise specifications. Most vendors can quote sound power data for their products. Using these data we can accurately predict the noise impact.
3. When installing fluid pipes or lines, we must use gradual rather than abrupt transitions from one part of flow passage to another.
4. We should improve lubrication. Inadequate lubrication due to poor design, degradation of due age, or incorrect lubricants have been often cause of bearing noise and also cutting noise associated with machining.
5. We should install bearings correctly. Improper installation has been often the reason for bearing- noise problems.
6. We should reduce mechanical run out of shafts. It is also able to improve both initial static balance and dynamic balance. By reducing this source of vibration excitation, many components bearing, gears, and cams may generate less noise and have greater life.
It should be apparent that maintenance is an important element in controlling the noise generated by equipment. Maintenance department should be instructed that noise is a very important consideration in all of their activities.
However, the various methods for noise control at source have been described as follows:
(a) Containment:
The control of noise at source is considered to be the cheapest and most reliable method of noise control. Once sound energy is moved away from its immediate source it becomes an all pervading problem. The area of the surface of a sphere gets increased as the source of its radius, and this means that the nearer we move to the source the smaller the area has been which we have to deal with in order to contain the sound. This idea of containment or isolation, then, has been the main principle of noise control.
(b) Sound Insulation and Sound Absorption:
These are two different but complementary means of noise control are often confused. Sound insulation means the reduction of sound as it passes through a wall or barrier or in the case of vibration, through isolators or discontinuities. Here the main contribution to the reduction in the sound level is the fact that by far the largest fraction of the sound energy gets reflected back from the barrier.
For example, if a wall is having a transmission loss of 30 dB, it implies that only a thousandth part of the energy falling on the wall gets transmitted but upto 999/1000 parts may get reflected back. Sound absorption, on the other hand, takes place when sound is partially reflected on a wall. Sound is absorbed in many ways at a boundary and the reflected sound is reduced because of these absorbing processes.
With good sound absorbers 90 per cent of the incident sound energy can be absorbed. Perhaps the most significant difference between insulation and absorption has been the fact that absorption involves a degradation of sound energy into heat whereas insulation can be attained without any such degradation at all.
(c) Enclosures:
It has been desired to reduce the noise from a particular source, this can often be carried out by means of an enclosure. However, the mere surrounding of the source by massive partitions have been not adequate, because the volume enclosed with the source becomes a reservoir for noise energy.
What happens then has been that sound levels inside the enclosure increase until the rate of loss of energy within the enclosure has been equal to the rate of production of sound energy from the source. If the only way in which sound can escape from the enclosure has been through the wall, it is immediately obvious that the sound energy will build up until the rate of radiation of the sound from the enclosure has been equal to the rate of production of sound from the source and the situation will be as if the enclosure did not exist at all (except for directionally effects).
It, therefore, becomes necessary to include within the enclosure an alternative means by which the sound energy can be converted into heat. This is absorbed by including sound-absorbing material within the enclosure and when this is carried out most of the sound energy is lost into the sound absorber and the insulation becomes effective.
A critical problem with barriers has been that of permitting access to the machinery or noise source that is enclosed. Assess would be needed for a number of reasons- for monitoring, adjustment of control, the supply of raw materials and the removal of finished products and also for the removal of waste heat.
With a little ingenuity it is possible to overcome most of the control problems by providing a small window and access hatch, preferably double glazed, in the barrier. Gauges could be read without opening the barrier and controls that only have to be adjusted occasionally can be left within the barrier.
Controls which have to continually adjusted need special treatment and this can often means re-designing the machine so that the barrier becomes an integral part of the whole device. It must be dealt with at the time of specification of the machine, which we will discuss below, in the case of an existing machine it has been often possible to provide extensions to controls, or in the case of electrical equipment to remove switches and rheostats to a remote position.
For continuous process machinery, the through put of material has been a more demanding condition. It has been often necessary to locate the immediate operator of the machine inside the barrier area, in which case the operator must either bear a noise greater than that recommended, which has been involving the use of ear defenders more or less permanently, or if the operator is a member of a team carrying out the work which involves less noisy operations, the team can get cycled through the working day so that no individual operator exceeds the noise dose limit.
The final reason for providing access has been the matter of cooling. Unfortunately, all materials having good sound absorbing properties also tend to have good thermal insulating properties, which implies that a temperature sensitive process cannot be enclosed in sound insulating barriers.
However, it becomes possible to provide access for cooling air by means of chimney-like constructions. A duct, although it gives a direct air path to the surrounding area, does provide a certain amount of attenuation, particularly if the duct has been lined with sound absorbing material.
Handbooks are available from the duct manufacturers which are quoting the value of sound attenuation that can be expected from a given duct cross-section and length. These should be used to calculate the length of chimney needed to provide adequate sound reduction as well as ventilation.
Machinery Design and Specifications:
When new machinery has to be installed in a factory, it would be prudent to regard the noise that will be produced by the machinery so as to ensure an acceptable working environment for employees. The writing in of noise specifications would be straight-forward, but the main difficulty has been due to the interaction of the machine with the environment where it is being installed and the difficulty of measuring the noise produced by the machine in the environment where it is being manufactured.
It would be important to remember that the machinery when installed in the factory should produce a noise level not exceeding XdB (A). However, this makes the supplier of the machine in a difficult situation. He is not having any knowledge what acoustical environment exists within the factory, but he can measure the noise level from the machine within his own environment.
The method employed has been usually to specify that the noise level measured at a distance from the machine should not exceed XdB (A), and to compute the resulting noise field from a simple estimate of the acoustics of the room in which the machine is to be installed. In most cases, the noise level in the room would not exceed the level measured close to the machine.
Exceptions take place when the machine is kept by itself in a small, highly damped chamber. Under these conditions, the direct field gets dominant and the noise level must be calculated from inverse square law and directionality.
The design of machines for quietness has been technical and specialised. A few general rules can be given here, but individual problems would have to be tackled on their own merits with the help of more technical information.
The general principles have been as follows:
1. Impacting parts of machines must be enclosed.
2. An enclosure should have sound-absorbing material which should be suitable for the environment within the enclosure.
3. Manufacturing tolerances should be kept as low as possible; rotating machinery must be dynamically balanced.
4. All rotating or impacting machines must be based on antivibration mountings.
5. All rigid connections of the machine in the way of electricity, water, air, gas, etc., should be having vibration decouplers around the machine and water ducts should have pressure release chambers.
6. Internal combustion engines should be properly silenced.
7. Machines would not be sited in small reverberant chambers, but kept within absorbent enclosures sited in large work areas. The over-all layout of a building or factory should be designed so that noise critical areas (offices, show-rooms etc.) should be kept away from noisy areas by using buffer non- noise critical areas, such as kitchens, canteens, lavatories, corridors, etc.
Vibration, Isolation and Damping:
Machines and other sources of impact noise (principally footsteps) have been providing acoustic energy into the frame of a building in the form of vibration which would then run throughout the building and, by re-radiation from walls, would be making noise problems at quite remote parts of the building.
It has been now known that under circumstances where a room is having a resonant frequency at the presence of some remote vibration source, such as a pump, the noise level in the room would become high due to tuned resonance effects. Adjacent rooms without tuned resonances can get unaffected by the noise.
Such anomalous forms of high sound level can also be brought about by central heating system where the large area of the radiators can serve as efficient acoustic radiators and can couple the sound within the pipes to the room, generating a high noise level.
The main cure for these problems has been to provide discontinuities. If we construct a building in a single steel frame, then the vibration energy can run through the frame virtually unhindered and give rise to a noise problem wherever high coupling or resonance effects can occur.
It is possible to introduce the discontinuities into the system by bolting the framework together with loss pads between adjacent members, thus attenuating the vibration energy as it passes through the structure. In direct analogy with the airborne case, however, it has been found to be most effective to cut off the vibration energy of source.
We must first attempt to distinguish between load- bearing connections from the vibration source and non-load- bearing connections. The weight of the machine must be supported somehow and this has to be carried out by providing a strong suspension for the mass associated with the machine.
The principal problem here has been that the degree of freedom introduced by such suspension gives rise to a resonant frequency. It becomes necessary to keep this resonant frequency down in the region where little vibration energy exists and to fit override stops to disallow occasional high vibration amplitude during, for example, run up on a machine when high low-frequency amplitudes will exist for a short period of time. Figure 2 shows the typical natural frequencies which are obtained with various common methods of suspension.
Noise Reduction at Receiver:
Noise has been received by people and more exceptionally by delicate instrumentation and it is often necessary to control the level of the noise received. This is normally achieved by treating the room or area within which the receiver is situated and we therefore have to study the acoustics of this situation.
There exists little opportunity for noise control at the receiver. Normally, the permissible noise level have been set for the receiver, and engineering techniques must be used on the source and/or paths in order to limit the exposure of the receiver. Personal hearing protection devices are in reality a treatment of the transmission path from source to receiver but, because they are worn by the receiver, it makes some sense to treat them as part of the receiver.
One other aspect relating to the receiver gets associated with controlling noise exposure by so-called administrative techniques. In other words, it is possible to control both the exposure time and level of workers by arranging their work pattern in a way which would limit the amount of time that they spend in certain noise level environments.
Theoretically, it is a powerful concept but in practice it is having limited potential. Due to workers still requirements and union job descriptions, it has been often very difficult to arrange an individual’s work pattern so that it involves activities in significantly different noise level environments.
Another complication has been the difficulty of assessing the cost. We would normally accept some loss in productivity when the range of required skills gets increased. Presently, there appears to be no basis for assessing the cost of this type of noise control even when the work pattern could be arranged to limit exposure.
Table 2 indicates the permissible noise exposures currently associated with most occupational safety and health standards.
When the daily noise exposure is composed of two or more periods of different levels, the combined effect should be considered, rather than the individual effect of each. That is, if the sum of the following fractions exceeds unity (1), then the mixed exposure is considered to exceed the limit value.
We know,
Where,
Cn = total time of exposure at a specified noise level.
Tn = permissible item at that exposure.
An example will illustrate how to determine compliance. Suppose an employee works for 6 hours in a noise level of 90 dBA, 1 hour in a level of 100 dBA, and 0.5 hour in a level of 105 dBA.
Let us calculate the cumulative exposure for this working day:
As the cumulative exposure fraction exceeds unity, these employees sound exposure is not compliance. Now, if by administrative control we could arrange this employee’s work pattern so that he spends 2 hours at 90 dBA, and 0.5 hour at a 100 dBA, 0.5 hour at 105 dBA, and the remainder of the work shift at less than 90 dBA, then the employees would be in compliances as-
In summary, the opportunities for noise control at the receiver have been worthy of consideration but are usually very limited in practice. The real challenge in noise control have been associated with interrupting the path and abating sound at the source. We will now discuss some methods of noise reduction.
The subject can be understood by considering a source of sound located within a room radiating sound energy. At most frequencies the sound energy is travelling away from the source and gets multiply reflected from the boundaries of the room, and a uniformly diffuse sound field (which is called the reverberant field) builds up within the room.
First there has been the near field within which we are so close to the source that individual characteristics of the source are able to control the high sound level. Further away from the source there has been a free field in which the source can be treated substantially as a point source at which the noise levels fall off by 6 dB for each doubling of the distance from the source.
This free field can lead, but does not always, into the reverberant field which fills the rest of the room space. The reverberant field has been generally uniform within the room except in the region of the source or near to highly absorbing areas.
i. The Near and Free Field:
In the near and free field the sound level has been controlled by the sound being directly radiated from the source and has been independent of the room acoustics. Any control of the sound level therefore must be by means of direct reduction of the source power or by using a barrier between the source and the receiver.
ii. Free Field Radius:
The free field radius may be defined as the distance from the sound source at which the free field gets merged into the reverberant field. If the source is having directional characteristics this radius would be dependent on orientation and it can also be influenced by the presence of other sound sources within the room.
It has been possible for the free field radius to exceed the typical dimensions of the room in which case an effective reverberant field does not exist within the room and free field noise control techniques only can be used.
iii. Reverberant Field:
It is possible to control the reverberant field level by the power of the source, the size of the room and the amount of sound absorbing material within the room. Within the reverberation field it has been possible to control the sound level by controlling the power from the source, for example with enclosures, or by using sound absorbers within the room, but not be using sound barriers (as opposed to enclosure) near to the source.
By increasing the sound absorption within the room, the reverberant field level could be reduced but this is not an efficient process because it is necessary to double the area of sound absorbers so as to reduce the reverberant field level by 3 dB.
Most room already have a fair amount of sound absorbers so that the cost benefit ratio for this procedure quickly becomes uneconomic and indeed impractical. Another effect has been that when the absorption in the room gets increased.
2. Noise Reduction along Sound Paths:
Once sound has left a source and gets established in the surrounding medium, either air or the structure of the building, it would travel some distance before reaching the point at which the noise will take place. When this situation arises, certain steps have to be taken to reduce the transmission of noise as it travels down the path.
Common paths for noise have been airborne paths like ducts and corridors, and it has been usual to include in this category walls that break up airbornes sound paths, and not be regard walls as independent vibration paths.
Alternatively, vibration energy may transmit through building structures directly and may arise from direct excitation from the source, or indirect excitation through a sound field produced in the room having the source. Where a noise source has been directly coupled to conducting paths like pipes of air ducts, the paths are able to carry sound energy.
In order to control noise along the path, we must consider how sound gets transmitted through ducts, corridors directly through walls and along pipes and through the structure of buildings.
3. Transmission of Sound along Ducts:
The manufacturers of fans and ducting systems include those who have attempted to solve the problems of noise. It is desirable to contact them directly at the design stages of building and they will provide sound levels and sound insulation values together with prices and delivery information for any reasonable noise requirement.
Before talking to such people, however, it has advisable to have an. understanding of what has been a ‘reasonable noise requirement’ and what problems need to be regarded. We should not expect total silence from a system but it is reasonable to expect sound levels below ambient noise levels when the building is in use. The noise generated by the system should be graded to the existing noise climate of the building and will be related to the power of the driving fan.
A second point has been that ventilating systems are forming a link between the inside and the outside of building and thus noise can get transmitted through the systems and can get generated produced outside the building by the system itself.
It is not usual for such noise to make trouble except in the case of large industrial plants where the use of giant centrifugal fans can make discrete frequency noises (at the frequency at which the blades of fan pass the outlet), or in the case of discotheque noise in built-up areas. Conversely if a ventilating system has been installed in a noise-critical area such as near an airport the noise transmitted into the building from outside must be regarded.
In all cases, therefore, the three factors to consider have been the noise produced by the system inside and outside the building, and the noise transmitted through the system. Noise from fans can be known from reference to manufacturer’s information or by empirical formulae derived by Beranek.
It is in the form of a “spectrum”. A spectrum is a chart which depicts the variation of loudness as a function of frequency. The coupling of a fan to its associated duct does not greatly influence the noise produced by centrifugal fans, but axial fans have been sensitive to obstruction and bends in the ducting for a length upstream and downstream of twice fan diameter.
It is important to remember that sound radiation from a fan unfortunately travels equally well upstream and downstream. This is as the speed of sound (343m) (sec) has been in excess of any air speed used in practical systems. It has been found that circular ducting has been inherently more rigid than rectangular ducting and thus circular ducting transmits sound along the duct to cause a problem at the far end, whereas rectangular ducting tends to permit sound to pass through walls of the duct, possibly causing problems near the fan.
Commercially, duct attenuating systems are known but it is necessary to remember that noise could be transmitted down a duct in two ways. Firstly, it gets transmitted along the air path within the duct. To counteract this, the provision of sound-absorbing material within the duct has been necessary.
The achievable attenuation is dependent upon the sound absorption coefficient of the living. The absorption coefficient will depend upon frequency and upon the type of sound absorber employed. Secondly, noise gets transmitted along the rigid walls of the duct.
To attenuate this noise, it becomes necessary to incorporate a non-rigid bellows section to d- couple the duct from the vibration produced by the fan. Bends within the ducts will be able to attenuate noise travelling along the duct but can cause noise in their own right if turbulence is generated at the bend.
The provisions of aerodynamic guide vanes within the duct are able to reduce such noise. At the end of a duct the sound must pass from the duct into the air space of the room A useful effect occurs here- that of end reflection. This phenomenon is taking place at the open end of an organ pipe and helps it to resonate.
In the case of general noise travelling along the duct, it has been possible that as much as 90 per cent of the sound can get reflected from the end. It is to overcome this effect that the mouths of wind instruments are belled, permitting the efficient coupling between sound within the tube and the surrounding air.
The aim of the designer of ventilating systems has been to achieve the opposite effect by providing low coupling to the room while avoiding localised noise source, which are generally caused by turbulence produced within constricting grills. Much more detailed information on this extensive subject can be attained by referring the literature produced by duct manufacturers.
4. Factory Controlling Sound Insulation:
i. Transmission Loss or Sound Reduction Index:
The transmission loss of a panel is used to refer to the ability of the panel to resist the transmission of energy from one side of the panel to the other. When sound energy is incident on a panel most reflected, but a small fraction gets transmitted to form a sound wave beyond the panel.
The fraction of energy transmitted is called the “transmission coefficient” and the transmission loss (TL) is related to by:
TL = 10 log10 (1/τ)
Transmission loss has been a property of a panel and is only one of the number of the factors that go to make up sound insulation. This is because sound insulation has been a property of adjacent room and includes the effects of room acoustics and flanking transmission. A number of factors are able to control the transmission loss of a wall- these include the mass of the panel, the number of the layers it contains and its structural stiffness.
ii. Mass Law:
The mass law is regarded as the basic rule of thumb for determining the transmission loss of a panel.
If the know the mass per unit area is kilograms per square metre (M), then the transmission loss at a frequency f would be given by:
The sound insulation will get increased by between 5 and 6 dB per octave (one octave represents a doubling of frequency) and by a similar amount for the doubling of the mass of the wall.
Now subjectively, a difference of 5 dB in the sound level of a noise implies that only a small difference in the level is obtained. Thus, it is often difficult to get a significant increase in the sound insulation of an existing wall by merely increasing the total weight of the wall.
A domestic party wall is having a double layer of brick which can weigh up to 400 kilos per square metre (90 lbs per square foot). It is clearly not practical to add, say, 10 dB to the mass law insulations of such a wall by increasing the weight by a factor of 4.
The position has been more hopeful with lightweight partitions, but even so it has been better to proceed on a basis of understanding of the overall problem than merely to increase the weight of the partition.
iii. The Coincidence Effect:
It is possible to deform a panel in many different ways without disturbing the edge position (boundary conditions) by keeping different uneven loads on the panel. Sound waves falling on the panel are having the same effect except that because the pressure due to sound patterns (modes) in the room are time-dependent.
It is possible to excite a panel into a high response in two ways. A panel mode will respond well to a room mode when either the shape of the panel mode has been the same as the shape of room mode, or the frequency of the panel mode is the same as the frequency of the room mode.
Normally the two effects take place separately, and it has been now known that at low frequencies the models that have resonant frequencies equal to the room mode frequency (the resonant modes) response well and control the vibration amplitude of the panel, but these modes do not radiate sound well.
The modes having a similar shape to the forcing room mode respond well (but not as well as the resonant modes), but it has been these panel modes which are responding above their resonant frequencies, and hence in a mass controlled manner, which get coupled well the space beyond the panel and carry the sound energy through the panel.
It can be shown that this gives rise to the same mass law as was originally derived by considering infinite panels. The modes in the two-dimensional panel, however, behave differently as a function of frequency, than do the three-dimensional room modes, and as frequency increases, the resonant frequency of the shape coupled modes becomes equal to the resonant frequency of the room modes. Thus, we are having a coincidence of resonant and non-resonant coupling. This coincidence gives rise to a sudden high coupling of the room/panel/room system and a consequent drop in the sound insulation.
This coincidence behaviour has been found to be quite common in light-weight panels has been generally plasterboard, asbestos, glass and wood, and undertaken a limiting factor in the insulation that can be obtained with these materials. Thus, control of the coincidence effect can give rise to a significant improvement in sound insulation.
Much research work has been undertaken by the manufacturers of these materials (particularly by the manufacturers of glass and plasterboard). These manufacturers will readily advise on the choice of suitable configurations so as to avoid coincidence troubles.
iv. Multiple Layers:
An efficient method of increasing the sound insulation of a wall has been to construct the wall of a number of separate layers. As it is the discontinuity at the boundary between two different materials that makes reflection and therefore reduction of the transmitted sound, it would appear reasonable that the more discontinuities there have been the less sound will get transmitted.
However, a number of theoretical and practical difficulties arise. From a theoretical point of view, sound which gets reflected from one layer will travel back to the previous layer and then be reflected again. This sets up a multiple reflection successive layers of the multi-layer wall and the overall effect has been to reduce the insulation below that which might be expected from the sum of the individual insulations of the individual layers in the wall.
The introduction of gaps in a wall also allows the presence of two sorts of resonance. At low frequencies the masses of the two walls react against springiness of the trapped air and this makes a mass spring mass resonance which in a typical multilayer wall has a frequency between 50 and 150 Hz.
At much higher frequencies, it has been possible that the thickness of the air cavity could become equal to one-half of the wavelength of the sound. When this happens, standing wave resonance takes place in the air cavity. Both of these effects can reduce the beneficial increase of insulation derived from multi-layer panels.
Upon studying measured values of the sound insulation of such walls, it has been found that the effect of the low frequency mass spring mass resonance has been to hold down the sound insulation at low frequencies so that however, many layers are incorporated in the wall, the low frequency sound insulation is not improved.
However, the gradient of the insulation versus frequency curve has been found to increase in proportion to the number of layers in the panel. Experimental measurements have been observed indicating that the mass law increase of 5 dB per octave for a single panel gets increased to about 10 dB per octave for a double layer panel and further to 15 dB per octave for a three layer panel.
However, this can be extended to a rather large 20 dB per octave for a four-layer panel. At high frequencies around 3000 HZ and over, the effect of the standing wave resonances could be seen. This is not a marked effect and does not even cause a reduction in sound insulation like occurs with the coincidence effect, but generally gives rise to a flattering off of the sound insulation versus frequency curves which nevertheless continue upwards.
The beneficial effect of multi-layering can generally be obviated by internal vibration bridges within the panel. For structural reasons it has been often impossible to build a double panel without some form of cross bracing between panels either on the edge or over the face of the panel.
This gives rise to vibration bridging which are able to transmit sound through the panel cancelling out the beneficial effect of air gaps and other discontinuities. A case in point concerns an experiment that was undertaken with steel plates containing, a vacuum. The idea was undertaken with steel plates containing a vacuum.
The idea of this was that a vacuum should not be able to conduct any sound and this multi-layer panel, in theory, should give perfect insulation. In practice, to sustain a vacuum between two plates a large number of internal struts had been necessary, keeping the two plates apart.
The sound insulation was measured before and after the air was taken out from between the panels and it was reported that the sound insulation of the panel with a vacuum was less than without. It was because the pressure of the atmosphere caused an improvement in the coupling of the two plates via bridges which then transmitted sound energy with greater efficiency.
In practice, it has been reported that the effect of flanking has been found to overcome almost completely the beneficial effect of layering so that a double brick wall has a performance little better than it would have if it were a single layer acting according to the mass law.
v. Sound Absorbing Material:
Sound absorbing material plays a vital role in controlling insulation that can be obtained between two rooms. It is assumed that by placing sound absorbers inside the cavity, the resonance mentioned above can get reduced and so the sound insulation of the panel increased.
In practice, the results have been not as good as may be expected for two reasons- in the first case, the mass spring mass resonance takes place at a very low frequency and sound absorbers are not having good sound absorbing properties in this range. It is then found that sound absorption is having little effect on mass spring mass resonance.
The small amount of sound absorption that may be present already in the cavity has been sufficient of drown this resonance very strongly. Hence the addition of further absorbers has little effect. Sound absorbers may also be kept over the face of the panel if the panel is light-weight and has small insulation, quite significant improvements in sound insulation have been observed.
As much as 15 dB could be obtained at high frequencies, especially where there has been a coincidence affect, as the presence of the sound absorber cancels out the increase in transmission obtained by this effect. Sound absorbers by themselves have very little inherent sound insulation. A four inch layer of polyurethane foam will be having a transmission loss of only 5 to 8 dB.
vi. Sound Insulation and Transmission Loss:
The quoted transmission loss of a panel has been not necessarily the sound insulation that could be obtained between two rooms separated by that panel. Other factors come into play. Principally, these have been the exposure area (S) and the effective sound absorption within the room receiving the noise (A).
Sound insulation (I) and transmission loss could be related by the equation:
I = TL-10 log S/A
From this equation it follows- that the larger the exposure area between the two rooms, the smaller will be the insulation. However, if the absorption in the receiving room gets increased this will increase the insulation.
Insulation obtained in practice is thus generally related to the area of sound absorber, but we must note that it has been inefficient to try and improve sound insulation by increasing sound absorption since normal rooms have a fair amount of sound absorption anyway, and that according to the equation, even if this amount of sound absorption were doubled, the insulation would only increase by 3 dB, which has been hardly a significant increase in insulation. The quantity ‘S’ could only be controlled at the planning stage and reveals the importance of minimum areas of contact between noisy and quiet areas.
5. Noise Control in Large Rooms with Many Sources:
In a large room having many sources, the model of near, free and reverberant field still applies, but there have been complications. The level of the reverberant field has been controlled by the total sound power that has been emitted into the room and this level can exceed the level in the near field of some of the quieter noise sources, so that zero free field radius will apply to these sources.
The free field radius of a dominant source will substantially remain the same as if that source alone was radiating into the room, but if a number of sources of roughly equal strength are dominant then the free field radius will be reduced by the increased reverberant field due to the other sources. Fortunately, it has been quite easy to trace the mixture because of the long time required to establish a new equilibrium at the low temperature.