Occupational Soundscapes – Part 13: Path Noise Control

     The discussion of noise control continues, pursuing the second priority (see Part 11):  “Abate transmission of sound along its path.”  Although the singular form is used in the priority statement, sound from a single source may travel along several paths, each requiring its own analysis and solutions.  This fact complicates noise control efforts and reinforces the pursuit of source noise control (see Part 12) as the first priority.
     For sound sources that cannot be sufficiently tamed, a number of potential path treatments are available.  The pursuit of path noise control begins with identification of transmission paths.  Assessment of each path’s contribution to, or potential for mitigation of, the soundscape informs the selection of appropriate measures to be implemented.  Finally, implementation and verification complete the process.

     The “blurred lines” in the hierarchy of controls were discussed in “Part 11:  Concepts in Noise Control” [20Mar2024].  What was not explicitly mentioned, but may have been noticed, in Part 12 is that there is also some “blurriness” between source and path noise control.  For instance, when discussing fan noise, treatments of ductwork were presented as source control because the sound had not yet become air-borne.  The transmission of sound, or vibration, along the duct also qualifies these treatments as path control.  However, the earlier the intervention, the better; thus the inclusion in the source control discussion.

Sound Fields
     Air-borne sound transmission paths are comprised of different regions, within which the behavior of sound differs.  These regions, called sound fields, are depicted in Exhibit 1, plotted as SPL vs distance from source.  The near field, as the name suggests, extends from the source to the distance at which the sound begins to act as a single, coherent, broadband noise.  Within the near field, SPL measurements vary with position, due to phase differences of the generated sound waves.  To minimize near field effects on SPL measurements, readings should be taken at a distance from the source greater than the wavelength at the frequency of interest.  For instance, to measure SPLs at typical audiometric frequencies (500, 1k, 2k, 4k, 8k Hz; see Part 5) at a single location, the sound level meter should be more than 0.686 m (2.25 ft) from the source [λ(500 Hz) = 0.686 m].

     Beyond the distance at which sound waves coalesce to propagate as a single sound is the far field region.  Within the far field is the region called the free field (or direct field), in which sound waves travel without interruption.  That is, the sound waves encounter no reflective or absorbent surfaces in the free field.  In many industrial settings, or other dense environments, the free field may extend for only a very short distance.
     Beyond the free field, reflections add to free field sound to create the reverberant field.  The transition from free field to reverberant field conditions varies by frequency and the characteristics of various materials upon which the sound waves impinge.

Directivity Factor and Index
     The preceding presentation of sound fields is applicable to transmission in any direction from a source.  However, the location of the source, relative to reflective surfaces, can change the nature of the sound field significantly.  The influence of reflective surfaces near a sound source is called its directivity factor (Q).  The formal definition of Q is somewhat wordy and esoteric; a more practical description is attempted here.
     As we know, sound is transmitted omnidirectionally.  When a source is located near a reflective surface, transmission in the direction of that surface is reduced, causing the sound to be more directional in nature.  The more reflective surfaces in close proximity to a source, the more directional the sound transmission becomes.  Numerically, the directivity factor is the ratio of the directional sound intensity to the nondirectional sound intensity; Q = Id/In.  The nondirectional sound intensity, In, is the intensity of sound that would exist at the same point if the reflective surfaces were not present.
     To visualize directivity in an occupational soundscape context, imagine the potential placements of equipment (i.e. sound sources) within a room.  If a source is suspended in the center of the room (e.g. a fan), the sound it generates propagates in all directions without interruption.  In this case, no reflective surfaces influence transmission of the sound; therefore, Id = In and Q = 1.
     A more common installation is the placement of a machine on the floor.  If located in the center of the room, only the floor effects directionality, limiting sound to hemispherical propagation.  In this case, Id = 2 In and Q = 2.  Extending this to placements adjacent to a wall and in a corner yields Q = 4 and Q = 8, respectively.  Stated another way, Q is the inverse of the fraction of the omnidirectional sound transmission sphere that remains in free field conditions, as depicted in the center column of Exhibit 2.

     Practical application of knowledge of directionality often manifests in use of the directivity index (DI).  The directivity index represents the increase in SPL expected, given a value of the directivity factor, Q, and is calculated as DI = 10 log Q (dB).  Exhibit 2 provides a summary of Q and DI values and the shape of the sound field resulting from each sound source placement discussed.
     The directivity factor only accounts for the influence of major structures, i.e. floor and walls in rooms assumed to be rectangular.  Irregular shapes, other objects, and material properties are not considered.  Therefore, Q and DI values can only estimate the true characteristics of sound transmission.  However, these estimates typically suffice for evaluation of occupational soundscapes, particularly in preliminary assessments made for planning purposes.

Transmission Paths
     Sound is transmitted directly through air; it can also be first transmitted as vibration through solid materials.  A sound may be transmitted in several media simultaneously.  Examples of this “multimodal” transmission, from several sources, are depicted in Exhibit 3 – air-borne paths on the left and structure-borne paths on the right.  Though the root cause of a sound should be identified whenever possible, it need not be known to treat its transmission paths.  However, the spectral composition of the sound and characteristics of the room to be quieted are required to develop effective treatments.

     The multitude of possible transmission paths from a single sound source is represented by the example shown in Exhibit 4.  Some issues depicted in the example of a centrifugal fan and connected ductwork were addressed in Part 12; this installment provides guidance on other matters, such as flanking transmission, acoustic leaks, and variations in material properties.
     Flanking transmission occurs when sound travels around a barrier, rather than through it, taking the path of least resistance.  The example depicted in Exhibit 4 – that of a wall ending at a false, or suspended, ceiling, rather than extending to the top of the structure – is a common form of construction.  It is an “out of sight, out of mind” error; though a large area is available for air-borne sound transmission, it is not visible under normal circumstances.  Flanking is, therefore, often neglected in noise control efforts, yielding unsatisfactory results, particularly at sites where an SPL differential greater than 35 dB is needed.
     The guiding principle of acoustic leakage is “if air can pass, sound can pass.”  Leaks are common at the periphery of windows, doors, and access panels; they can also occur at any joint between materials.  When nonstructural partitions are built in a large space, attention must be paid to connections to structural members and other partitions to prevent leaks.  Many times, the flanking described above and leaks share a root cause – insufficient sealing of nonstructural partitions to structural members.  Constructing a wall can visually separate two areas while they remain acoustically linked.
     Sealing acoustic leaks is often straightforward.  Cracks and joints can be caulked; larger gaps can be filled with absorbent material.  Installing compression seals or acoustic gaskets can effectively reduce leakage around doors and access panels.
     A sound wave that impinges on a barrier divides its energy three unequal ways, as depicted in Exhibit 5.  A portion of the incident sound energy is reflected by the barrier.  The remainder continues through, where it is further reduced by absorption within the barrier material.  The final remainder is transmitted, or reradiated, from the opposite side of the barrier.  The difference between the incident sound energy and the transmitted sound energy is the transmission loss (TL) of the barrier.  TL is frequency-dependent and is an important property of noise control constructions.  Analysis and treatment of transmission paths, properties of noise control materials and their use in combination are discussed further in the following sections.

Barriers and Partial Enclosures
     A barrier is any wall or other physical obstruction to line-of-sight transmission.  A partial enclosure is a set of barriers that is open in one or more directions (e.g. sides or top, if floor-mounted).  Noise control may be the express purpose for constructing a barrier, or it may be one of several objectives.  Additional objectives could include protection from moving parts, contamination control, aesthetic improvement, and others.  Additional objectives may influence design decisions regarding barrier size, placement, materials, etc., but they must not be allowed to compromise proper management of the soundscape.
     As shown in Exhibit 5, the first influence a barrier exerts on a sound wave is to reflect a portion of its energy.  These reflections establish or intensify the reverberant field, shown in Exhibit 1.  Rigid materials with hard, nonporous surfaces are the most effective sound reflectors.  Sealing surface porosity can increase a barrier’s reflectivity.  Depending on the location of the barrier relative to sources and receivers, this could be beneficial to noise control or detrimental to soundscape management.  A common example is a block wall that is painted for aesthetic purposes or to facilitate cleaning.  Increased reverberation results, potentially hindering speech communication or increasing overall SPL in the area, requiring other measures to counteract the effect.

Absorbers
     Sound waves that are not reflected traverse the barrier, losing energy to absorption.  Effective reflectors tend to be poor absorbers and vice versa.  Therefore, a highly-reflective barrier of homogeneous construction typically dissipates little sound energy by absorption.  To achieve significant sound level reductions, barriers are often lined with a layer of absorbent material.
     Absorbent material placed on the incident side of a reflective barrier reduces reverberation and the amount of energy available for transmission through the barrier.  If the absorbent is susceptible to abrasion, chemical attack, or other damage in the incident environment, it can be placed on the opposite side of the barrier.  This arrangement has no effect on the reverberant field on the incident side of the barrier; it reduces transmitted noise, but to a lesser degree than incident-side mounting.
     Effective absorbent materials tend to be lightweight (i.e. low density), porous, and cellular (e.g. foam) or fibrous (e.g. carpet).  Other types of absorbers may also be used to dissipate sound energy.
     Reactive absorbers, or Helmholtz resonators, were mentioned in connection to mufflers installed on internal combustion engine exhausts or pneumatic equipment in Part 12.  A Helmholtz resonator consists of a cavity connected to the surrounding atmosphere via a small, necked opening.  The mass of air in the cavity acts as a spring, oscillating in response to passing sound waves, absorbing energy.  This type of absorber is effective in a very narrow frequency range; it must be designed (i.e. “tuned”) to closely match a troublesome frequency.
     The frequency range for which a Helmholtz resonator is effective can be extended by adding absorbent material in the cavity.  While unlined resonators are typically most effective at low frequencies, absorbent materials are better absorbers of high-frequency sound energy.  Fabricated panels are commercially available to exploit the benefits of combination in a straightforward installation.
     A diaphragmatic absorber is a panel that oscillates at a frequency matching the incident sound wave or one of its harmonics.  As with the resonators discussed above, inducing oscillations in a mass transfers energy to that mass, where it is dissipated.  The sound attenuation provided by this type of absorber is difficult to predict, though it is usually most effective at lower frequencies.  It may require less effort and time to properly specify other noise control measures to target frequencies of concern.
     An absorber’s effectiveness is quantified by its sound absorption coefficient, α, the proportion of incident sound energy that is not reflected or transmitted.  To be “industrially useful,” according to NIOSH, material should have a coefficient of α > 0.60 at 500 Hz and higher frequencies.  At lower values of α, the amount of material required to achieve necessary SPL reductions could be prohibitive.
     Though theoretically impossible, highly-absorptive materials may be catalogued with α > 1.0.  Many values used to develop noise controls are estimates; actual performance may vary from that calculated or anticipated.  Awareness of potential deviations allows inclusion of a margin of error in specifications and prevention of overly-optimistic expectations.
     Absorbers may also be catalogued by their noise reduction coefficients (NRC).  An absorber’s NRC is simply the arithmetic average of its sound absorption coefficients at 250, 500, 1000, and 2000 Hz, rounded to the nearest 0.05.  Owens Corning considers materials with NRC > 0.40 to be “sound absorbers;” this is a significantly lower threshold than the NIOSH guideline cited above.  NRC may be useful for initial at-a-glance comparisons of materials, but the frequency dependence of absorption requires that individual α values be analyzed to ensure appropriate specification of materials.  Coefficients of example materials are compiled in Exhibit 6 for comparison.

     An alternative specification of the absorption capability of a barrier is the sabin.  A sabin is the equivalent of the sound absorption provided by one square foot (1 sq ft) of a “perfect” absorber (i.e. α = 1.0).  It may be called a “standard” or English sabin, when using “English” units, to explicitly differentiate it from a metric sabin.  A metric sabin is equivalent to one square meter (1 sq m) of perfect absorption and is equal to 10.76 standard sabins.
     To determine the total absorption (A) provided by a panel or other treatment, multiply the area of the treatment and its sound absorption coefficient:  A = S α.  The total absorption of a composite panel, wall, etc. is found by summing the absorptions of its constituents:  A = ∑(Si αi).  Be certain that the system of units is known to all who use this information; a factor-of-ten error can be difficult to overcome!  Also, the symbols used may not be intuitive, due to inconsistency with other contexts.  However, this notation is consistent with existing literature on acoustics and noise control; to avoid further confusion, it will be maintained.

Transmission Loss
     Transmission loss (TL) can be defined a few different ways.  It was introduced conceptually in the previous section as simply “the difference between the incident sound energy and the transmitted sound energy.”  As depicted in Exhibit 5, this difference is comprised of the energy reflected by a barrier and that absorbed within it.
     The TL discussed here is actually the “apparent” transmission loss, known to some as the apparent sound reduction index.  This formulation, with “apparent” dropped for convenience, is used because it is most useful in practical applications, a category in which occupational soundscapes naturally belong.  This is so because the apparent TL allows for flanking transmission that is likely to occur outside strictly-controlled laboratory environments.  The present discussion considers stand-alone barriers and partial enclosures, where the occurrence of flanking is certain.  In “real-world” conditions, apparent TL provides a better estimate of relevant noise control metrics.
     The transmission loss of a panel is determined experimentally by placing it between two reverberant rooms and measuring the difference in SPL between the source and receiving rooms.  The panel’s TL is then calculated using the following equation:
  TL = SPLs – SPLr + 10 log (S/A) = SPLs – SPLr + 10 log S – 10 log A (dB),
where SPLs is sound pressure level in source room (dB), SPLr is sound pressure level in receiving room (dB), S is surface area of panel (sq ft or sq m), and A is total absorption in receiving room [sabins (sq ft) or metric sabins (sq m)].  Practically, it may be more useful to rearrange the expression to estimate SPL differentials that can be achieved with various barrier or enclosure configurations.
     The frequency dependence of a barrier’s transmission loss is divided into four regions, or zones, as shown in Exhibit 7.  In each zone, TL is controlled by a different characteristic of the barrier or panel.

     At very low frequencies, a panel’s stiffness controls sound transmission, with decreasing effectiveness as the frequency increases.  Once the resonance zone is reached, TL begins to appear erratic, though it trends higher with increasing frequency.  The addition of damping material to a panel smooths the TL curve in this zone.  At approximately twice the lowest resonant frequency (i.e. one octave higher), damping effects disappear and sound transmission becomes mass controlled.
     In the mass controlled zone, TL increases linearly at a rate of approximately 6 dB/octave according to the Mass Law, expressed as TL = 20 log (f M) – 47 dB, where f is the incident sound frequency (Hz) and M is the surface density of the barrier.  Surface density is the mass per unit area of incident surface (kg/sq m).  From the Mass Law expression, it can be seen that doubling the barrier thickness (doubling surface density or mass) achieves the same 6-dB increase in TL as a one-octave increase (doubling) in incident sound frequency.  Large increases in barrier thickness can quickly become cumbersome, however.
     The mass controlled zone ends at the barrier’s critical frequency (fc).  The critical frequency is the lowest at which coincidence occurs.  Coincidence, or coincidence effect, occurs when the speed of sound in air and in the barrier are equal.  Sound at the critical frequency must be at grazing incidence (θ = 90°) for coincidence to occur.  At higher frequencies, coincidence occurs at progressively lower angles of incidence (θ), where direct (i.e. perpendicular) incidence is defined as θ = 0°.
     Coincidence causes a barrier to become a more-effective transmitter of sound, as the large drop in TL, called the coincidence dip, above the critical frequency shows.  In the coincidence controlled zone, damping effects return, reducing the severity of the dip.  However, mass law performance should be considered an asymptotic limit in this zone.
     A few points to which the preceding discussion alludes are worth making explicitly, including:

• The vertical axis (ordinate) in Exhibit 7 is labeled “sound reduction index,” an alternate term for transmission loss, as noted above.
• There are no values of TL or frequency given in Exhibit 7 because the amount of sound reduction and the transitions between zones will vary by material and construction details of barriers.  For a homogeneous barrier, the critical frequency can be calculated:
  

where c is the speed of sound (m/s), t is barrier thickness (m), P is barrier mass density (kg/m^3), and E is the elastic modulus of barrier material (kg/m s^2).  Heterogeneous barriers are often constructed to join reflective and absorptive materials and to reduce coincidence effects below a level of significance.

• There is no coincidence effect unless the air-borne wavelength (λ) is less than the structure-borne wavelength (λB) of the sound.  Coincidence occurs when sin θ = λ/λB.  For direct impingement, sin 0° = 0; coincidence does not occur.  This has little practical effect, however, as the context of discussion is a reverberant room, where sound impinges at all angles.
  
• Once again, the spectral composition of incident sound is critical information needed to apply noise control measures in the appropriate zones of transmission loss.

     Transmission loss of a material may also be characterized by its sound transmission class (STC) or noise isolation class (NIC).  STC and NIC are single-number indices that represent the frequency-dependent TL and insertion loss, respectively.  These indices are analogous to the use of NRC for absorption coefficients.
     A material’s STC and NIC are determined using 1/3 octave band measurements from 125 to 4000 Hz (16 total; see Part 6).  The data are plotted and compared to standard curves while applying a set of rules that ultimately determine the ratings assigned.  Details of the procedures are not provided here because they are of limited value in the context of industrial settings.
     An STC rating is equivalent to the sound intensity reduction, in dB, that a material is expected to provide.  It is typically used to assess the prevention of transmission of intelligible speech; thus, it may be more applicable to a government building than a factory.  NIC compares sound levels with and without a panel in place.
     The standard caveat applies to STC and NIC:  theory and practice differ; variables present in application are not represented in laboratory tests.  The best results can be obtained by using constituent octave band data.  However, single-number ratings may be useful to narrow the field of candidate materials for testing; examples of material TL and STC ratings are tabulated in Exhibit 8 and Exhibit 9, respectively.

Path Length
     As mentioned above, standalone barriers and partial enclosures cannot force sound to impinge on attenuating material; flanking is assured.  To maximize the benefit provided by a barrier, the length of the flanking path between source and receiver must be considered.  Recall that doubling the distance from a source reduces sound intensity by 6 dB at the receiver.
     In Part 10, low-frequency communication signals were recommended when the direct path (i.e. line of sight) to listeners is obstructed.  This is because low frequencies experience greater diffraction around obstacles, resulting in a smaller shadow zone, as depicted in Exhibit 10.  It is the Fresnel theory of diffraction that allows us to predict sound level reductions attainable with a barrier of certain size and placement.  To do so, consider the diagram, in Exhibit 11, of Fresnel zones created by placing a barrier between a source and a receiver.  Although the description references sound flanking the top of a barrier, the method is the same when applied to flanking around the side of a tall barrier.

     The information needed to predict a barrier’s performance are:

• A – distance from sound source to top of barrier (m).
• B – distance from top of barrier to receiver (m).
• D – direct path distance from source to receiver (i.e. no barrier present) (m).
• δ – path length difference (m); δ = A + B – D.
• N – Fresnel number (dimensionless); N = ± 2 δ/λ.  N > 0 in the shadow zone (i.e. attenuation zone) and N < -0.2 in the bright zone (i.e. free field).  Greater attenuation is achieved at higher Fresnel numbers, attained with greater path length difference and/or higher frequency.

     To estimate the attenuation provided by a barrier, the Fresnel number is used to locate a corresponding point on a standard curve.  A simplified method is to calculate it as follows:  ΔSPLf = 10 log (3 + 0.12 f δ), where

• ΔSPLf is the barrier attenuation at frequency f (i.e. difference in SPL at receiver with and without barrier present) (dB).
• f is the frequency of incident sound (Hz).
• δ is the path length difference (m).

Summing the SPL differentials (see Part 4) provides an estimated overall SPL reduction due to the barrier.  The practical limit of sound level reduction by use of a barrier is approximately 20 dB; greater reductions require additional noise control measures.
     There are several design and placement decisions to be considered in order to maximize the attenuation potential of a barrier, including:

• Place the barrier as close as is practical to the sound source or the receiver; both should be in the direct field to attain significant level reductions.
• The barrier should be as tall as is practical, given other constraints.  Its effective height (H in Exhibit 12) is the vertical distance from the source to the top; its total height is less important.
• Each side of the barrier should be at least twice as far from the source as is the top of the barrier (i.e. width > 4 H).
• To prevent transmission (i.e. to “force” flanking), construct a barrier with surface density greater than 2 lb/sq ft (9.76 kg/sq m).
• To minimize the impact of ceiling reflection, ceiling height should be at least 150% of the source-to-receiver distance (D); see Exhibit 13.
• To achieve 10 dB or greater attenuation, the angle between the lines defining (1) incidence at the top of the barrier and (2) the direct path from the top of the barrier to the receiver should be greater than 30°.  This is the angle θ in Exhibit 12; it is also shown in Exhibit 13.

     Adding absorbent material to the ceiling, hanging baffles, or other similar treatment can also reduce reflections of sound reaching a receiver.  The more effective the barrier or enclosure, the more noticeable ceiling reflections become.

Rooms and Total Enclosures
     Rooms and total enclosures are structures that are closed in all directions.  Common usage of the terms often differentiates them by the nature of construction or intended use.  A “room” typically refers to a space occupied by humans and of robust construction, possibly utilizing structural members of a building.  “Enclosure,” in contrast, often refers to a structure surrounding equipment, of a temporary nature, constructed, for example, of reconfigurable panels.
     While the connotations of common usage may be useful shortcuts in conversation, use of the terms in this way is not strictly accurate; parameters must be established at the outset to avoid misunderstanding.  Many design decisions, operational procedures, etc. depend on proper understanding of noise control requirements; all assumptions must be validated.
     Much of the preceding discussion of partial enclosures also applies to total enclosures; e.g. the transmission loss of an enclosure is the sum of its constituent barrier TLs.  Completing an enclosure, however, adds points of analysis needed to fully characterize its performance.
 
Room Constant and Reverberation
     The total absorption of an enclosure can be characterized by its room constant, R.  Room constant is calculated at each frequency as

where

is the average absorption coefficient of all barriers comprising the enclosure, and S is the total surface area of the enclosure (sq ft or sq m).  A larger enclosure, of identical construction, has a higher room constant which corresponds to a less-intense reverberant field within the enclosure.
     Knowing the room constant allows us to estimate the location of the transition from free-field to reverberant conditions (see Exhibit 1).  The distance from the source at which this transition takes place, r’, known as the transition zone, is estimated for each frequency as

     An enclosure’s room constant is also effected by its contents, including equipment, furniture, people, and so on.  The contribution to R of a room’s contents is much more difficult to calculate, however.  A rule of thumb for industrial environments is to increase R by 25% or more.  The sound intensity calculations are relatively insensitive to this adjustment; therefore, the exact value used is noncritical.
     For other settings, such as conference rooms, performance halls, etc., tabulated data are available from various sources to aid in estimating the effect of a room’s contents on R.  A sample of the type of data available is provided in Exhibit 14.

     A related measure is a room’s reverberation time, TR.  Reverberation time is defined as the time required for sound intensity to decrease 60 dB from its steady-state value upon silencing the source.  A room’s reverberation time is estimated as TR = 0.05 V/(S α_bar) (s), where V is the volume of the room (ft^3) and S is surface area (sq ft) or TR = 0.16 V/(S α_bar) (s), where V is the volume of the room (m^3) and S is surface area (sq m).  (S α_bar) is the total absorption of the room in sabins or metric sabins, respectively.
     The curves in Exhibit 15 represent mean reverberation times for various types of venue; actual values are highly variable.  As can be seen from the curves, longer TR is desirable for music, while speech intelligibility requires relatively short reverberation times.  The “ideal” TR in an industrial setting can be generalized “as short as possible,” as the vast majority of sound is unwanted (i.e. noise) and often at objectionable intensity.

Build-Up
     When a sound source is completely enclosed, the energy trapped inside causes “sound build-up,” as intensity increases until a steady-state condition is reached.  The average sound pressure level inside the enclosure (SPLE) is found as follows:  SPLE = PWL + 10 log (4/RE) (dB), where PWL is the sound power level of the enclosed source (dB) and RE is the enclosure’s room constant (sq m).  The level at the exterior of the enclosure (SPLEXT) is determined as follows:  SPLEXT =  SPLE – TLE – 6 dB, where TLE is the average transmission loss of the enclosure panels (dB).
     Sound energy is absorbed by converting it to thermal energy.  Without proper ventilation, “heat build-up” inside an enclosure can reach dangerous levels.  Performance of temperature-sensitive equipment can be altered and, in extreme cases, a fire hazard can be created.
     The amount of air flow needed to maintain a stable temperature inside an enclosure is determined as follows:  Qv = 1.76 W/ΔT, where Qv is the ventilating air flow required (ft^3/min or cfm) (assumed to be at ambient temperature), W is the amount of heat generated in the enclosure (W), and ΔT is the allowable temperature rise (°F).  Air flow through an enclosure should be as uniform as possible to prevent hot spots.  Treating the calculated value as a minimum is advisable, though a balance must be sought with the additional noise concerns associated with fans and ductwork, as discussed in Part 12.

Enclosure Access Points
     Normal operation and maintenance requires equipment to remain accessible.  Small enclosures, or covers, can be removed to provide access for maintenance tasks, but this is usually infeasible for larger enclosures.  The four key access types are doors, windows, tunnels, and utilities.
     A door allows a person to fully enter an enclosure.  An access panel may be used when all required tasks can be safely performed from the exterior (e.g. within arm’s reach, with a tool, etc.).  Access panels are treated as a type of door for this discussion for two key reasons:

1. The methods of sealing and materials of construction, for a given enclosure, are the same.
2. Both present the risk of being left open, severely limiting the enclosure’s noise reduction capability.

      Windows are used when visual monitoring of the interior is required, but physical access is not.  The types in use range from a single sheet of clear polycarbonate to residential-style multiple-pane windows.  The choice of window depends on the construction of the enclosure and the attenuation required.
     Beyond the cost considerations, the number and size of doors and windows in an enclosure significantly impact its ability to reduce sound levels.  As seen in the curves of Exhibit 16, an enclosure’s attenuation capability is reduced as a function of the difference between the access point and wall TLs and the proportion of the wall comprised of the access.  If multiple accesses with different TLs reside in a single wall (e.g. door and window), calculating the effective TL of the wall, as presented above, is more accurate.

     Tunnels are openings in enclosures used to pass material into or out of an enclosure, typically on a conveyor.  The effect on sound levels of a typical three-sided tunnel are shown in Exhibit 17.  The effect of fully enclosing the conveyor (i.e. adding a fourth side to tunnel, below conveyor) depends, in large part, on the type of conveyor installed.  For example, a full tunnel is likely to provide greater noise reduction when an open roller-type conveyor is used than when one with a heavy rubber belt is installed.  In any case, it is recommended that any tunnel be constructed with a length at least twice the largest cross-sectional dimension (i.e. width or height) of the opening in the enclosure.
     Additional noise reduction can be achieved by closing the tunnel opening between material transfers.  The greatest advantage is gained from this approach if closure is automated to ensure use and cycle times are long relative to transfer times.  Frequent transfers reduce the potential advantage; at or near continuous material flow, no reduction in noise exposure can be achieved by this method.
     Utility access points can easily be overlooked, as they are often located in inconspicuous areas of enclosures.  This group includes any passage through an enclosure of electrical wiring, pneumatic or hydraulic plumbing, ventilation ducts, exhaust ports, process fluids, and so on.  The necessity of these passages, combined with a lack of awareness of their impact on noise control efforts can lead to carelessness in installation.  Proper sealing is paramount; this fact must be conveyed to installers and maintainers to ensure noise-reduction targets are achieved.
     Irrespective of the cause of leaks – door gaps, construction joints, unsealed utility pass-throughs, etc. – the combined impact can be enormous.  An opening as small as 0.1% of total area can halve the noise reduction capability of a high-TL enclosure!  Leaks also cause attainable attenuation to rapidly plateau; a high-TL enclosure with leaks is no more effective than one with a modest TL rating.  Curves representing attenuation loss due to acoustic leaks are shown in Exhibit 18.

Methods of Isolation
     The discussion of barriers and enclosures has been biased toward isolation of the source, consistent with the priorities established in the source-path-receiver (SPR) model (see Part 11).  An alternative approach is to isolate the receiver(s), creating another blurred line between path and receiver noise control.
     Isolation of receivers is achieved by housing them in an enclosure – or room – when it is not possible or practical to enclose the source.  Examples include mills, foundries, and construction sites, where the equipment or work area is too large to enclose.  A receiver enclosure may be a “control room,” where equipment is remotely operated, or simply a refuge from high-intensity surroundings used when tasks permit or during break-times.
     When an enclosure is used to house personnel, the noise control challenges can become greater than when enclosing equipment.  The number and size of windows and doors is typically higher, the ventilation requirements are more stringent, numerous electrical and data cables are often required, and plumbing may be needed.  Fortunately, analysis and solution of these issues remain as presented above for equipment enclosures.

     An example noise control “journey” is presented in Exhibit 19; panel A depicts the initial condition.  Noise is a known problem; personnel are issued hearing protection devices (HPDs) as a reactive, interim solution.  In panel B, an installed partition provides some protection from direct sound exposure, but the intensity of the reverberant field requires continued HPD use.  The addition of absorptive material along the ceiling, shown in panel C, reduces the reverberant field intensity such that HPD use is no longer required for those engaged in “quiet” operations.  Finally, in panel D, the source of noise is fully enclosed, reducing noise levels sufficiently to cease HPD use for all operators.

     Key take-aways from the Exhibit 19 example, in brief:

• HPD use is a default short-term response to noise exposure, but is the last resort as a long-term solution.  This point is discussed further in Part 14.
• Progressing through insufficient solutions can be avoided with up-front analysis and planning.
• None of the treatments depicted address the sound power of the source; reducing it may render less-extensive path treatments sufficient.
• High-intensity exposure for some personnel should not be considered inevitable.  Had the source enclosure (i.e. protection of the machine operator) been considered first in the example, all personnel would have reaped maximum benefit without intervening measures unnecessarily consuming resources.
• Though it is a hypothetical example, the type of progression presented is far too common.  The urge to “do something” simple, visible, and ineffective must be suppressed in favor of purposeful analysis and true solutions.

And So Much More
     This installment of the “Occupational Soundscapes” series pushes the limits of the “easily consumable” objective to simply be a reasonable overview or introduction to path noise control.  There are myriad recommendations for design and construction details of enclosures, material alternatives, and so on that address practical applications of noise control.  On the theoretical side, discussion of additional performance indices and material ratings have been omitted.  The goal of this, and all entries on “The Third Degree,” is to provide a foundation on which readers can build knowledge while putting the basics into practice.  If it inspires investigation beyond the limitations of this medium that lead to improved soundscapes, it will be deemed a tremendous success.

     For additional guidance or assistance with Safety, Health, and Environmental (SHE) issues, or other Operations challenges, feel free to leave a comment, contact JayWink Solutions, or schedule an appointment.

     For a directory of “Occupational Soundscapes” volumes on “The Third Degree,” see Part 1: An Introduction to Noise-Induced Hearing Loss (26Jul2023).

References
[Link] An Introduction to Acoustics.  Robert H. Randall.  Addison-Wesley; 1951.
[Link] The Effects of Noise on Man.  Karl D. Kryter.  Academic Press; 1970.
[Link] Human Engineering Guide to Equipment Design (Revised Edition).  Harold P. Van Cott and Robert G. Kinkade (Eds).  American Institutes for Research; 1972.
[Link] Architectural Acoustics, 2ed.  K.B. Ginn.  Brüel & Kjaer; November 1978.
[Link] Industrial Noise Control Manual (Revised Edition).  National Institute for Occupational Safety and Health (NIOSH); December 1978.
[Link] Compendium of Materials for Noise Control.  National Institute for Occupational Safety and Health (NIOSH); 1980.
[Link] Handbook for Industrial Noise Control.  National Aeronautics and Space Administration; 1981.
[Link] Noise Control in Industry – A Practical Guide.  Nicholas P. Cheremisinoff.  Noyes Publications; 1996.
[Link] Fundamentals of Industrial Ergonomics, 2ed.  B. Mustafa Pulat.  Waveland Press; 1997.
[Link] “Hearing Protection.”  Laborers-AGC Education and Training Fund; July 2000.
[Link] “Noise and Vibration.”  Evan Davies in Plant Engineer’s Reference Book, 2ed.  Dennis A. Snow, ed.  Reed Educational and Professional Publishing Ltd.; 2002.
[Link] “Noise Control Design Guide.” Owens Corning; 2004.
[Link] “Engineering Controls for Reducing Workplace Noise.”  Robert D. Bruce.  The Bridge; Fall 2007.
[Link] Engineering Noise Control – Theory and Practice, 4ed.  David A. Bies and Colin H. Hansen.  Taylor & Francis; 2009.
[Link] “Noise – Measurement And Its Effects.”  Student Manual, Occupational Hygiene Training Association; January 2009.
[Link] “Controlling Noise at Work.”  (UK) Health and Safety Executive (L108- 3ed); 2021.
[Link] The Noise Manual, 6ed.  D.K. Meinke, E.H. Berger, R.L. Neitzel, D.P. Driscoll, and K. Bright, eds.  The American Industrial Hygiene Association (AIHA); 2022.
[Link] “Technical Guide for:  Noise Control – Engineering Controls, Work Practices, & Administrative Controls.”  Georgia Tech; May 2023.
[Link] “Noise control.”  Wikipedia.

Jody W. Phelps, MSc, PMP®, MBA
Principal Consultant
JayWink Solutions, LLC
[email protected]