READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE.
https://nrc-publications.canada.ca/eng/copyright
Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la
première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.
Questions? Contact the NRC Publications Archive team at
PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.
NRC Publications Archive
Archives des publications du CNRC
This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.
Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
Vibration response of floors and the effectiveness of toppings to control flanking transmission
Nightingale, T. R. T.; Halliwell, R. E.; Quirt, J. D.
https://publications-cnrc.canada.ca/fra/droits
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
NRC Publications Record / Notice d'Archives des publications de CNRC: https://nrc-publications.canada.ca/eng/view/object/?id=362bdfbf-7177-4d5c-958a-ea31d20166a2 https://publications-cnrc.canada.ca/fra/voir/objet/?id=362bdfbf-7177-4d5c-958a-ea31d20166a2
Vibration response of floors and the effectiveness of toppings to control flanking transmission
Nightingale, T.R.T. ; Halliwell, R.E.; Quirt, J.D.
NRCC-45377
A version of this document is published in / Une version de ce document se trouve dans Inter-Noise 2002, Dearborn, MI., Aug. 19-21, 2002, pp. 1-8
/i /i b
The 2002 International Congress and Exposition on Noise Control Engineering Dearborn, MI, USA. August 19-21, 2002
Vibration response of floors and the effectiveness of
toppings to control flanking transmission
TRT Nightingale, RE Halliwell, JD Quirt
National Research Council Canada, Institute for Research in Construction, M-27, Montreal Road, Ottawa, Ontario, K1A 0R6 CANADA
Abstract
A companion paper in INTER-NOISE 2002 [1] has shown that to achieve adequate sound insulation in many wood frame constructions, it is necessary to control flanking transmission involving the subfloor. This paper examines the effectiveness of toppings to control both airborne and impact flanking paths involving the floor surface. Factors such as surface density, hardness, presence and type of interlayer are systematically examined and found to be significant factors. The change in flanking sound insulation due to adding a topping is shown to be different for paths where energy propagates perpendicular to the joists, compared to those where energy propagates parallel to the joists. Floor vibration mappings reveal that a topping will change not only the power injected but also the propagation losses across the floor. The most effective toppings reduce input power and increase propagation losses relative to the bare floor. One type of topping exhibits significant improvement in the flanking sound insulation in one direction and a significant worsening in the other. The results are used to establish the basis for a semi-empirical model for flanking transmission in structures with very high internal losses.
1 Introduction and Evaluation Methods
This paper expands upon the results presented in a companion paper [1], examining the effect of adding a floor topping to the constructions shown in Figure 1 and Figure 2.
A B
C D
A B
C D
Figure 1: Sketch of the test specimen with joists perpendicular to the junction.
Figure 2: Sketch of the test specimen with joists parallel to the junction.
Effects due to changing the topping surface density or surface hardness, and adding an interlayer, are evaluated. The trends observed were similar for cases where the apparent sound insulation was dominated by flanking transmission (room pair AB) or by direct transmission (room pair AC). The magnitude of changes to the impact sound insulation for room pairs AB and AC are compared to show that in general the improvement for direct transmission will not match that for flanking transmission involving the floor surface. The change in impact sound insulation between room pair AB is examined for each of the toppings with the joists parallel to the junction (Figure 2), and with them perpendicular (Figure 1). The improvement for floor flanking paths can be strongly dependent on joist orientation.
The three toppings shown in Table 1 are a subset of those studied in this IRC/NRC study, but are sufficient to identify the more important trends associated with surface density, surface hardness, and method of attachment. These trends are examined for airborne and impact sources. The toppings were applied in both rooms A and B on the constructions shown in Figure 1 and Figure 2. The companion paper [1] gives construction details.
Material Nominal Thickness, mm Surface Density, kg/m3 Application Topping Effective
OSB Overlay 18 mm 11.7 23.4 Stapled 305 mm o.c.
Bonded Gypsum-concrete 25 mm 47.1 58.8 Bonded with agent
Floating Gypsum-concrete 38 mm 63.9 63.9 9 mm foam interlayer
Table 1: Properties of the topping layers. When the topping is bonded or mechanically fastened to the subfloor, it is convenient to introduce an effective density that is the sum of the surface densities of the subfloor and the topping.
The oriented strand board (OSB) overlay consisted of adding a second layer of the subfloor material. The OSB overlay was not continuous under the partition wall, stopped short of the sole plate of the wall, and did not contact the gypsum board. The bonded gypsum-concrete was applied according to the manufacturer’s recommended practice which is to apply a bonding agent to the subfloor and then pour the slurry in place allowing bonding to all surfaces contacted. This bonding includes the gypsum board of walls in rooms A and B. The final topping had a closed-cell foam interlayer completely covering the floor and the portion of the walls to a height equal to the thickness of the gypsum-concrete. Interlayer joints were sealed to stop the slurry leaking to the floor and walls.
The impact sound pressure levels were measured with the ISO hammer box located at the same four positions near the center of the floor in room A. The hammers impart power only into the floor surface, which is involved in all transmission paths to rooms B and C. Hence, the change in impact sound pressure level measured in rooms B and C with and without a topping indicates its effectiveness for controlling flanking and direct transmission of impact sound. The procedures of ISO 140-7 were followed; these are essentially the same as ASTM E1007 (although E1007 does not allow reporting of data when the rooms do not share a portion of a floor ceiling assembly, as in the case of room pairs AB and AD).
There are four primary airborne transmission paths from room A to room C. The paths between rooms A and C that do not involve the floor (i.e., wall in room A to the wall and ceiling in room C) have significantly greater sound insulation than paths involving the floor
and may be safely ignored. The remaining two paths involve the floor with the flanking path from the floor surface to the lower wall (in room C) being less important than the direct for the cases considered here. This allows direct comparison of the apparent sound reduction index (ISO 140-4) or apparent transmission loss (ASTM E336 applied without suppression of flanking paths) with and without the topping to assess the effectiveness of the topping to control direct transmission.
For room pair AB, the sound insulation of the partition wall was not sufficient to allow the direct comparison method to be used to assess the effectiveness of toppings to control floor flanking paths. For assessing floor paths, the apparent sound insulation of the partition wall was increased by shielding in front of the wall in room B. Measurements between rooms A and B made with the shielding in place are estimates of the flanking transmission loss for paths involving the floor in room B.
2 Effect of Floor Surface Density
Figure 3 shows that doubling the effective surface density from 11.7 to 23.4 kg/m2 by fastening an 18 mm OSB overlay results in about a 10 dB increase in the apparent transmission loss for frequencies above about 160 Hz. This improvement is very significant but does not sufficiently suppress paths involving the floor in room B to allow the apparent sound insulation to be determined only by direct transmission through the partition wall.
10 20 30 40 50 60 70 80 63 125 250 500 1k 2k 4k Frequency, Hz Measured Apparent without topping R'w 38 (effective density 11.7 kg/m2) Bonded 25mm gypsum concrete R'w 57 (effective density 58.8 kg/m2) Fastened 18mm OSB Overlay
R'w 48
(effective density 23.4 kg/m2)
Partition Wall Potential Rw 50
A
B
C
D
Figure 3: Apparent airborne sound insulation of flanking paths involving the floor in room B showing the effect of increasing the surface density of bonded or mechanically attached topping. Included is the potential transmission loss for the partition wall measured in a lab. The bonded gypsum-concrete topping, having an effective density more than five times that of the bare floor, does suppress floor paths enough so that the apparent sound insulation is no longer limited by flanking. The exception is the frequency range around 1 kHz, where the apparent transmission loss of the floor path exhibits a dip. This is associated with the critical frequency of the gypsum-concrete topping bonded to the OSB subfloor.
Figure 4 shows the impact sound insulation between room pair AB for the same two toppings. With the OSB overlay it is evident that above about 250 Hz the trends in the two floors with OSB surface are very similar, but with the levels being about 10 dB lower when the OSB overlay is applied.
30 40 50 60 70 80 63 125 250 500 1k 2k 4k Frequency, Hz N o rm al iz ed Impac t S ound P res s u re Lev el , dB Bonded 25mm gypsum concrete L'n 69 No Topping
Bare OSB Floor L'n 73 Fastened 18mm OSB overlay L'n 62 Im pr ov ing I m pa c t S ound I n s u la ti on
A
B
C
D
Figure 4: Impact sound pressure levels measured in room B, with the ISO hammer box in room A, for bonded or mechanically fastened toppings of widely varying surface density. Below about 160 Hz the OSB overlay is considerably less effective. The much greater surface density of the bonded gypsum-concrete causes more reduction of the impact sound pressure levels below about 800 Hz. However, above 1kHz the levels increase relative to the OSB overlay - and at 2kHz relative to the bare floor - because the power injected by the ISO hammer box depends on the impedance of both the impact device and the surface being struck. At high frequencies, more power is injected by the steel hammers of ISO hammer box into the much higher surface impedance of the gypsum-concrete topping than into the more compliant OSB subfloor. This highlights potential difficulties with impact sources that do not correctly mimic the effective impedance of a human foot and shoe.
3 Interlayer Under Gypsum-Concrete Toppings
Figure 5 shows measured apparent airborne sound insulation for flanking paths that involve the floor in B, when a gypsum-concrete topping is applied with or without an interlayer.
10 20 30 40 50 60 70 80 90 63 125 250 500 1k 2k 4k Frequency, Hz 9mm Foam Interlayer with 38 mm gypsum concrete R'w 59 Bonded 25 mm gypsum concrete R'w 57
Partition Wall Potential R'w 50 Measured Apparent without topping R'w 38
A
B
C
D
Figure 5: Apparent airborne sound insulation of flanking paths involving the floor in room B showing the effect of an interlayer between the subfloor and gypsum-concrete topping. Included is the potential transmission loss for the partition wall measured in a lab.
The small difference in effective surface density between the two toppings means that differences greater than about 2 dB should be attributable to the interlayer. Applying this criterion to the results of Figure 5 suggests the interlayer does not appreciably improve the airborne flanking sound insulation for frequencies up to about 200 Hz. A significant improvement is observed above 800 Hz, the critical frequency of the 38 mm gypsum-concrete.
The impact sound insulation shown in Figure 6 indicates that below about 160 Hz there is little benefit of using an interlayer. However, significant improvements are observed at most frequencies above 160 Hz, and above 800 Hz there is a trend of increasing benefit with increasing frequency.
The data of Figure 5 and Figure 6 indicate that the benefit of an interlayer under a cementitious topping is considerably greater for impact than airborne sound insulation. While this study uses 9 mm foam to demonstrate the effectiveness of an interlayer, it should be noted that significant improvements can also be obtained using much thinner and less compliant materials as long as they prevent bonding to the subfloor. Complete discussion of interlayers is beyond the scope of this paper.
30 40 50 60 70 80 63 125 250 500 1k 2k 4k Frequency, Hz Bonded 25mm gypsum concrete L'n 69 No Topping Bare Floor L'n 73 Interlayer and 38 mm gypsum concrete L'n 62
A
B
C
D
Figure 6: Impact sound pressure levels measured in room B as a result of the ISO hammer box in room A for a gypsum-concrete topping with and without an interlayer. Shown for comparison are the impact sound pressure levels when the floor was bare.
4 Effect of Topping on Direct and Flanking Transmission
Since the impact sound pressure levels for room pair AC are controlled by direct transmission, the change in AC sound insulation can be compared to the change for AB to assess the effectiveness of a topping to control direct and flanking transmission, respectively. Each topping was applied to the constructions of Figure 1 and Figure 2 to determine if the orientation of the joists is an important factor.
Figure 7 shows the change in the receiver room impact sound pressure level due adding an 18 mm OSB overlay. The figure indicates that the overlay reduced the levels (improved sound insulation) for both direct and flanking transmission. For frequencies above 200 Hz, regardless of the orientation of the joists, there is a greater reduction in sound pressure level for room pair AB than AC. The OSB topping is more effective at controlling flanking transmission involving the floor surface than direct transmission.
Fastened 18mm OSB topping -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 63 125 250 500 1k 2k 4k Frequency, Hz AC
Joists perpendicular to junction
AB
Joists perpendicular to junction AB
Joists parallel to junction
AC
Joists parallel to junction
A
B
C
D
Figure 7: Change in impact sound pressure level due to applying the 18 mm OSB overlay as a function of the orientation of the joists shown in Figure 1 and Figure 2.
Figure 8 shows that for both direct and flanking transmission, adding the gypsum-concrete topping without an interlayer will increase the impact sound pressure level relative to the bare floor in the high frequencies. The most important feature however is the relation of curves. Unlike Figure 7, the change to the direct transmission (room Pair AC) is bounded above and below by the changes to the flanking transmission (room pair AB). For one joist orientation, the topping is more effective at controlling flanking transmission (room pair AB) than direct transmission (room pair AC) while for the other joist orientation the same topping is more effective at controlling direct transmission than flanking.
Bonded 25mm gypsum-concrete topping
-25 -20 -15 -10 -5 0 5 10 15 20 63 125 250 500 1k 2k 4k Frequency, Hz AC
Joists perpendicular to junction AB
Joists perpendicular to junction AB
Joists parallel to junction AC
Joists parallel to junction
A
B
C
D
Figure 8: Change in impact sound pressure level due to applying the bonded 25 mm gypsum-concrete topping as a function of the orientation of the joists shown in Figure 1 and Figure 2.
In the frequency range 160 – 2000 Hz with the joists perpendicular to the junction, the topping attenuates floor flanking paths more than direct transmission though the floor. The opposite is true when the joists are parallel to the junction since in the frequency range 400 – 2000 Hz the topping is better at controlling direct transmission than flanking transmission. Differences in the changes in impact sound pressure levels may be explained by recognizing that adding the topping not only changes the power injected by the ISO hammer box but also
changes propagation losses in the floor surface. A change to the injected power should affect direct and flanking transmission similarly. However, for the highly damped floor structures considered here, changes to propagation losses will be more important for flanking transmission than for direct transmission, because propagation losses determine the incident structural power at the flanking junction.
85 95 105 115 125 135 145 0 1 2 3 4 5 6 7 8
Distance from ISO Hammer Box, m
S u rf ac e V e lo c ity , dB ( a rb ) Room A No Topping
Joists parallel to junction
No Topping
Joists perpendicular to junction
18mm OSB Overlay Joists parallel to junction
18mm OSB Overlay Joists perpendicular to junction
Room B junction
Figure 9: Surface vibration levels at 1 kHz measured along two orthogonal lines from the source with and without the fastened 18 mm OSB overlay.
85 95 105 115 125 135 145 0 1 2 3 4 5 6 7 8
Distance from ISO Hammer Box, m
S u rf ac e V e lo c ity , dB ( a rb ) Room A No Topping
Joists parallel to junction
No Topping
Joists perpendicular to junction
Bonded 25mm gypsum-concrete Joists parallel to junction
Bonded 25mm gypsum-concrete Joists perpendicular to junction
Room B junction
Figure 10: Surface vibration levels at 1 kHz measured along two orthogonal lines from the source with and without the bonded 25 mm gypsum-concrete.
Estimates of the change in propagation losses can be obtained from velocity levels measured along two orthogonal lines from the source. One line is parallel to the joist orientation, while the other is perpendicular to the joist orientation. Results for the bare floor and the 18 mm OSB overlay, shown in Figure 9, indicate that for both joist orientations the topping increased propagation losses in the exposed surface (i.e., there is a greater level difference between source and junction with the topping than without).
Figure 10 shows that adding the bonded gypsum-concrete topping applied to the same bare floor increases propagation losses when the joists are perpendicular to the junction. Here the vibration energy propagates parallel to the joists to reach the junction. The opposite is true when the joists are parallel to the junction, since to reach the junction the vibration energy must propagate perpendicular to the joists. Thus, the bonded topping will be less effective when applied to flanking paths where the joists are parallel to the junction. This is consistent with the trends in the impact sound pressure levels shown in Figure 8.
The vibration data in Figure 9 highlight the mechanisms involved in flanking transmission between the floor surface in room A and the floor surface in room B. Vibration energy is attenuated with distance, and the rate of attenuation depends on the direction relative to the joists. With the joists parallel to the direction of propagation, the levels decrease reasonably monotonically with distance up to the joint, indicating that the field is not diffuse. (This was confirmed by a detailed mapping of the complete floor.) As a first order approximation only the direct field need be considered. Thus, the product of the injected power and an attenuation factor will give the effective power at the junction. Junction attenuation should be estimated using the difference in vibration levels close to the junction rather than space average levels because of the strong vibration gradient in the source surface. Using this criterion, Figure 9 shows that adding the OSB overlay may slightly reduce junction attenuation. The receiver room vibration levels exhibit very little attenuation with distance when compared to those of the source plate. The receiver surface will have two radiation mechanisms, one due to the reverberant field and other due to the near field of forcing points. The relative importance of these mechanisms has not been examined for these floors. Qualitatively, the draw-away curves illustrate the behaviours that would have to be included in a prediction model for such assemblies.
5 Discussion and Conclusions
A floor topping can be very effective for controlling flanking paths involving continuous structural elements (OSB subfloor and/or joists) in floor systems. High surface density is desirable for both airborne and impact sound insulation. The power injected by an impact source depends on the impedance of the exposed surface. Hence cementitious toppings should be finished with a compliant covering, and/or an interlayer should be used.
A topping that increases propagation losses relative to the bare floor will be more effective in controlling flanking transmission than direct transmission. This has important implications: results for direct transmission under laboratory conditions (ISO 140-5) can not be applied to predict changes in flanking paths when the added topping significantly changes propagation attenuation. These data suggest that for wood frame structures the error may exceed 10 dB. It should be recognised that the high structural attenuation in the floor surface will make the impact sound pressure level extremely sensitive to the location of the hammer box when there is significant flanking involving the floor surface. A fair comparison of results can only be made if the source is located a uniform distance from the flanking junction.
6 Acknowledgements and References
This work was supported by a consortium that included Canada Mortgage and Housing, Forintek Canada, USG, Owens Corning, Marriott International, and Trus Joist MacMillan. 1). R. E. Halliwell, J. D. Quirt, and T. R. T. Nightingale, “Construction details affecting flanking transmission
