- Email: [email protected]
Study of the road surface properties that control the acoustic performance of a rubberised asphalt mixture
Applied Acoustics 102 (2016) 33–39
Contents lists available at ScienceDirect
Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust
Study of the road surface properties that control the acoustic performance of a rubberised asphalt mixture V.F. Vázquez, S.E. Paje ⇑ University of Castilla-La Mancha (UCLM), Laboratory of Acoustics Applied to Civil Engineering (LA2IC), Avda. Camilo José Cela s/n, 13071 Ciudad Real, Spain
a r t i c l e
i n f o
Article history: Received 1 June 2015 Received in revised form 26 August 2015 Accepted 10 September 2015 Available online 22 September 2015 Keywords: Crumb rubber Tire/pavement noise Acoustic field evaluation Generation mechanisms
a b s t r a c t Simultaneously with the fact that vehicle industry has been able to lower the noise emission from driving vehicles, tire/pavement noise has become the most significant source of traffic noise. In order to reduce it, low noise surfaces are seen as a practical solution. One of these types of surfaces may be elaborated with bituminous mixtures with crumb tire rubber added to the binder in high content by a wet process. However, the generation mechanisms involved in the tire/pavement sound and the reasons of the noise attenuation achieved with these mixtures are not so clear. This study analyses the different generation mechanisms that take place in the tire/pavement sound generation in these crumb tire rubber pavements. The surface properties of the in-service pavement, which are most important in controlling the acoustic performance (texture, acoustic absorption and dynamic stiffness or mechanical impedance), have been measured for the characterization of a test track constructed with and without crumb tire rubber. After results correlation of these surface characteristics in a pavement with crumb tire rubber added by a wet process, it seems that the parameters of roughness and texture could have a relevant role in the global tire/pavement sound emission, whereas dynamic stiffness influence is relatively minor. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction For passenger vehicles at constant speed, tire/pavement interaction sound contributes significantly to the traffic noise at vehicle speeds greater than 40 km/h, and their generation mechanisms are influenced by pavement surface characteristics, such as sound absorption, texture depth and surface stiffness. Many tire/pavement sound studies are focused on the road texture, however, for a complete road surface characterization with respect to noise emission, the sound absorption and the dynamic stiffness (or mechanical impedance) of the surface should also be measured [1–12]. Low-noise road surfaces are seen as a good option to abate traffic noise. These surfaces include porous asphalt that, in contrast to dense asphalt surfaces, can absorb sound energy [13,14]. Porous road surfaces effect is focused in sound absorption mechanisms; nevertheless, these surfaces are not the only solutions in order to mitigate tire/pavement noise. For example, gap graded pavements with crumb rubber are another option in this respect [15]. Using crumb rubber modified asphalt, the mechanical
⇑ Corresponding author. Tel.: +34 902 204 100x3270. E-mail address: [email protected] (S.E. Paje). http://dx.doi.org/10.1016/j.apacoust.2015.09.008 0003-682X/Ó 2015 Elsevier Ltd. All rights reserved.
characteristics of the mixtures could be enhanced: crumb rubber increases pavement life and resistance to rutting and cracking [16]. Crumb rubber could be incorporated to the mixtures by dry and wet processes. In a dry process, crumb rubber is in the mixture replacing some of the solid fraction, as a part of the aggregates. On the other hand, in the wet process, it is added to bitumen before mixing it with the aggregates [17]. LA2IC has been contributing to the understanding of the acoustics and performance of asphalt pavements during the last few years [6,7,9–11,14,15,18]. Now, the aim of this study is to have a better understanding of the sound generation mechanisms of an in-service road pavement with crumb tire rubber, thus contributing to improve their acoustic design. To achieve this objective, characterization measurements of a gap graded mixture with crumb rubber added by wet process were carried out. The pavement characterization has been focused on the parameters that define the texture of the pavements: Mean Profile Depth (MPD) and International Roughness Index (IRI). The sound absorption and dynamic stiffness of the pavement surfaces were also assessed. All these parameters associated with the pavement surface are the most influential in the generation and propagation of tire/pavement noise. Results were correlated with sound generated by the tire/pavement interaction measured in the near field conditions.
34
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
dynamic stiffness (S) can be expressed in terms of complex numbers between the force (F) and the displacement (d) vectors of a tested surface:
2. Methodology 2.1. Experimental test track Three asphalt pavement sections located in Ciudad Real, Spain, are studied in this work. Two of them (experimental sections in Fig. 1) have a wearing course of a gap graded mixture (European denomination BBTM11A) containing crumb rubber added by wet process in a percentage of 20% by weight of bitumen (around 1.5% of the total weight of the mix); this is a high content for these types of road pavements. The last section (reference section) was constructed with the same gap graded mixture but without crumb rubber. The binder used for the experimental and reference test sections are respectively a high viscosity modified bitumen (BMAVC Spanish denomination) with crumb rubber and a conventional penetration grade bitumen. The characteristics of the binders employed for this research work are shown in Table 1. Crumb Rubber Modified Binder (CRMB) and conventional binder are quite different according to their characteristics. More detailed material properties of these pavements (grain size of the CRMB gap-graded mix, grain size of the crumb rubber used, etc.) are described elsewhere [15]. The measurement campaign for this work was performed after three years in service. 2.2. Dynamic stiffness Dynamic stiffness could play an important role in the tire/pavement sound generation, especially for surfaces with the same texture profile but different aggregate and bitumen content, or in surfaces aged by compaction [3,10,18]. Dynamic stiffness measurements have been achieved by means of the Non-Resonant Method, directly on the upper face of the surfaces studied [10,18]. The
S¼
F ½N d ½m
ð1Þ
The experimental set-up was composed of an impedance head that records the movement and force signals, a vibration exciter and an amplifier. Inset of Fig. 2 shows the impedance head and the exciter during a field measurement. A multianalyzer system was used to record the response and to produce the spectra of the dynamic stiffness. Sweep signals between 10 Hz and 7 kHz were used for the mixture excitation. 2.3. Sound absorption The acoustic absorption coefficient (a) is defined by the relation between the incident acoustic energy (Einc) and the absorbed acoustic energy (Eabs) by the bituminous mixture (without return).
a¼
Eabs Einc
ð2Þ
The acoustic absorption coefficient value depends on the one hand on the facility of the wave to enter the material pores and on the friction with the internal surface structure, which participates in the sound energy dissipation. Sound absorption coefficients of the mixtures studied have been evaluated by means of an impedance tube. Details of the measurement technique have been given elsewhere [6,11]. 2.4. Pavement surface profile The longitudinal profiles of the studied test sections (experimental and reference) were measured by means of the so-called LaserDynamicPG-LA2IC; a high-speed profiling laser device which allows measuring profiles of the surface course. The characteristics of this equipment are described elsewhere [15]. To characterize the surface texture of the sections studied, different parameters were used. The Mean Profile Depth (MPD) [22] was used to characterize the pavement macrotexture, whereas the International Roughness Index (IRI) [22] was used to characterize the pavement megatexture and roughness. The macrotexture corresponds to texture wavelength range from 0.5 mm to 50 mm, whereas the megatexture and roughness corresponds to texture wavelength above 50 mm up to 500 mm. The MPD influences
Fig. 1. Location with GPS coordinates of the test track section. Inset displays the equipment for the close proximity and longitudinal profile measurements of the road surface.
Table 1 Pavement’s specifications.
Binder Penetration (0.1 mm, 25 °C) EN-1426 Ring and ball softening point (°C) EN-1427
CRMB (experimental)
B 50/70 (reference)
19 80
57 51
5.7%
6%
Air void content
Fig. 2. Dynamic stiffness measured on the pavement without CRMB (reference) and with CRMB (experimental section). An example of the coherence function is also shown.
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
surface characteristics including tire/pavement noise or skid resistance. It is well known that increasing texture amplitudes in the transition region between the macro and megatexture range increases the lower frequencies of the tire/pavement noise by enhancing tire vibrations. IRI index is related with megatexture range and roughness. It is the most commonly used index for evaluating the smoothness of road pavement surfaces, and it is an indicator of the ride quality. However, its relation with tire/pavement noise has not been sufficiently studied yet in crumb tire rubber pavements. In order to assess the character of the texture, the coefficient c is achieved from the MPD and the Root Mean Square Values (RMS) quotient. RMS is defined by:
2 1 x1 þ x22 þ þ x2n 2 RMS ¼ n
ð3Þ
where xn represents every MPD datum recorded along the sections, and n is the total number of data recorded. The coefficient c (c = MPD/RMS) gives information about the texture in the vertical plane: it can be positive regarding the plane of the road surface (c > 1.05), neutral (0.95 < c < 1.05) or negative (c < 0.95). A negative texture could lead to noise reduction at high frequencies, due to the attenuation of the sound wave energy. 2.5. Close proximity sound methodology In order to assess the influence on the sound generation mechanisms of an in-service road pavement with and without crumb tire rubber, a methodology for geo-referenced close proximity measurements developed by LA2IC was employed (see for example Ref. [7]). Trailer Tiresonic Mk4-LA2IC, to which a reference tire was assembled, was used in the close proximity sound measurements as a part of the test vehicle (see inset of Fig. 1). Two microphones are mounted very close to the wheel, in order to exclusively evaluate the acoustical performance of the asphalt mixtures. During the measurements, the vehicle speed was kept close to the reference speed selected for the in-service mixture assessment. Moreover, the evolution of the tire/pavement sound levels were corrected for speed variations around the reference speed, according to the expression [14]:
Lcorr ðtÞ ¼ Lmeas ðtÞ B log
v ðtÞ v ref
ð4Þ
35
shows experimental data and the associated error bars. The coherence function of the measurements is also shown in this figure. Coherence function informs about the relation between force and motion. Its value was close to unity between the frequencies studied, thus, a good relationship between the incoming signals was achieved with the measurement set-up. Inset of Fig. 2 shows the impedance head and the vibration exciter (shaker), which provides the excitation force, during an in-situ measurement on the mixture with crumb tire rubber. According to the results shown in this figure, the dynamic stiffness of the mixture with CRMB (experimental section) is higher than that of the reference mixture, between 150 Hz and 400 Hz. This agrees with recent studies carried out in mixtures sampled and compacted in the laboratory, where the use of crumb rubber modified binders lead in higher stiffness modulus than the same binder without crumb rubber [19–21]. On the other hand, the behavior of these mixtures agrees with the bitumen characteristics shown in Table 1: the softening value of binders is higher for reference binder (B 50/70), and the CRMB shows a better resistance to penetration than the binder without CR. According to these results, dynamic stiffness potentially could be responsible of higher sound emissions of a pavement with high content of CRMB added by wet process, i.e. the mixture used for the experimental section. Lower dynamic stiffness values would be necessary to achieve noise mitigation in the overall noise emissions, by this surface property. Fig. 3 shows the normal incidence acoustic absorption spectra of core samples, with and without crumb rubber with an air void content lower than 10%, see Table 1. These sample cores were extracted from each of the test sections studied by removing surface material in different locations three years after the mixtures were laid. The absorption coefficient has a similar behavior for all the mixtures studied. It is below 0.1 for most of the analyzed frequencies, and every value is below 0.25. According to this similar result, the sound absorption capacity will not be responsible for potential differences between the tire/pavement noise of the mixtures studied. This conclusion that pavements with an air void content lower than 15–20% are not able to generate significant sound absorption has also been reported previously for different asphalt pavements [3,6,11,15]. Field measurements to register the longitudinal profiles of the different test sections have enabled us to obtain, by numerical calculations, the MPD: the texture related with macrotexture range. The results of the MPD calculation from laser profiling are
where the speed constant B was obtained from the slope of the logarithmic regression of the sound pressure level and speed:
LCPtr ¼ A þ B logðVÞ
ð5Þ
On the other hand, since temperature variations between the different tests were insignificant, sound levels were free from the influence of temperature in the measurements performed; therefore, temperature corrections of sound levels, LCPtr, have not been carried out. In addition, it is important to note that the test equipment allows measuring close proximity tire/pavement sound and wearing course profile simultaneously, on the wheel-path followed by the reference tire in successive tests, and in a geo-referenced way. This aspect is crucial for the relationship between the different surface characteristics evaluated. 3. Analysis of measurements and discussion Dynamic stiffness of the mixtures was studied in in-situ conditions, on the wearing course of the experimental (with CRMB) and reference sections three years after the mixtures were laid. The dynamic stiffness results in N/m are shown in Fig. 2; the figure
Fig. 3. Normal incidence acoustic absorption spectra of core samples extracted from experimental and reference sections. The inset shows the core samples evaluated.
36
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
Table 2 Averaged texture parameters of the road after three years in service conditions. Surface course
MPD value (mm)
c
IRI (m/km)
Experimental section (CRMB) Reference section
1.09 1.17
1.00 0.99
0.78 1.18
presented in Table 2. As can be seen, a higher MPD over the reference section (mixture without crumb rubber) is observed. In addition, from longitudinal surface profiles, the index IRI was calculated, and their values are shown in Table 2. The coefficient c was also evaluated. This coefficient may be related with noise attenuation at high and low frequencies, since the coefficient defines the character (negative, neutral or positive) of the surface studied. As is shown in Table 2, significant differences between the section fabricated with a pavement incorporating CRMB and the reference section, with the same type of pavement but without crumb rubber, were found. These textures could produce different noise emissions levels due to tire/pavement interaction. In order to gain insight into the influence of crumb tire rubber on the surface properties that control the tire/ CRMB pavement noise of an in-service road pavement fabricated with a high content of CRMB, close proximity sound measurements were carried out on the test sections with and without crumb tire rubber, three years after the mixtures were laid. First, the dependence with speed of the sound emission from the interaction of the reference tire and the surface of the test sections was measured upon increasing speed from 20 km/h to 120 km/h. Thus, for the test section fabricated with the incorporation of crumb tire rubber, Fig. 4 shows the global sound levels LCPtr versus logarithm of speed. The linear regression of LCPtr and speed gives coefficient B = 35.4 dB (A), see Eq. (5). In addition, the relationship between LCPtr and speed for different frequency bands shows that coefficient B presents values of around 36 dB (A) for 1.25 kHz and 4 kHz, and 23 dB (A) for 400 Hz. These frequencies are attributed to a mechanism of sound generation related to the flow of air in and around the rolling tire and the pavement surface texture, and to a mechanism related to the excitation and vibrations of the tire tread elements during the contact with the pavement, respectively [14,23]. Fig. 4 indicates that the variation in the coefficient B increases nonlinearly with frequency. This was one of the main results of a previous study [14]. The coefficient B is greater at frequencies above around 1 kHz, implying
Fig. 4. Global sound levels and sound levels for different frequencies as a function of vehicle speed. Their corresponding linear regressions are also included.
Table 3 Slope of the linear regressions, B, of the studied test track sections. High and low frequencies are also shown. Surface course
Experimental section with CRMB Reference section without CRMB
Slope (coefficient B) Eq. (5) Global
High frequencies (4 kHz)
Low frequencies (400 Hz)
35.4
36
23
36.5
31
26
that at high frequencies, the sound from the tire/pavement interaction increases with speed at a greater rate than sound at lower frequencies. Table 3 shows the slope of the linear regressions, B, of both test sections with and without crumb tire rubber. As can be seen, global sound levels generated by the interaction of the tire and the pavement with high content of crumb tire rubber increases linearly proportional to 35 log(V). This result of the sound-speed slope, similar for both reference and CRMB sections, is usually observed on dense and gap-graded asphalt concrete surfaces or porous asphalt surfaces [6,7,14,15], depending on the conservation state of the surface and/or its texture. In Table 3, the relative importance of low frequencies on the sound generation as a function of speed, for both types of mixtures (slope relatively low), and their similar behavior of the global sound levels can be seen. Additional analysis of the speed influence on the sound frequencies could be useful; thus, Fig. 5 shows the relationship between sound spectrum and speed at different speeds between 30 km/h and 120 km/h, for the section with CRMB. This figure shows that the increment of the sound levels is about 14 dB (A) at low frequencies (400 Hz) between 30 km/h and 120 km/h, whereas the increment at medium and high frequencies (1.25 and 4 kHz respectively) is about 20 dB (A). As can be seen, the coefficient B increases nonlinearly with frequency in mixtures with CRMB. A similar behavior was also reported for dense asphalt [14]. As is observed, the significance of low frequencies, attributed to a mechanism related to vibration induced by the pavement surface, on the sound levels as a function of speed is small compared with higher frequencies, attributed to a mechanism related to aerodynamic effects near the tire/pavement contact patch.
Fig. 5. Sound spectra of the experimental section with CRMB measured in close proximity to the tire/pavement contact patch at different speeds.
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
Fig. 6. Frequency spectrum map (50 km/h) of the in-service surfaces, with and without crumb rubber. A color scale in dB (A) was used for sound levels. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The acoustical influence of the incorporation of crumb rubber in high content to a gap graded bituminous mix, three years after the mixture with CRMB was laid, is reflected in Fig. 6. This figure shows the LCPtr frequency spectrum mapping. It was performed at 50 km/h, between the test sections with and without CRMB in a length of around 650 m. The reference section (without crumb rubber), with a length of 200 m, is included between points 800 m and 1000 m distance in this figure. As can be observed in this figure, sections fabricated with high content of CRMB present, after three years, an important capacity to reduce the rolling noise. The mean values of the global sound levels associated to each test section, measured in close proximity to the tire/pavement contact patch at reference speed of 50 km/h, are shown in Table 4. Averaged values shown in this table were calculated from measurements along six stretches of at least 100 m, before and after the reference section, using different runs. Additional analyses are necessary to comprehend the mechanisms responsible for the above-mentioned results shown in Fig. 6. Thus, Fig. 7 shows the averaging of the close proximity sound spectra between 300 Hz and 5 kHz in the one-third octave bands. Spectra, speed corrected, are recorded rolling at 50 km/h over a representative length (at least 100 m) of the test sections. The figure highlights the different behavior of the reference and experimental (CRMB) sections as a function of frequency. As it is shown in Fig. 7, the sound reduction achieved with the incorporation of high content of crumb rubber added by wet process is located at medium frequencies (0.8–1.25 kHz), which have the greatest effect in the global LCPtr sound levels shown in Fig. 6. This sound reduction observed in a mixture with high content of CRMB at medium frequencies is not linked to a higher mechanical stiffness presented by the mixture with crumb rubber added by wet processes (see Fig. 2). In addition, as above mentioned, the absorption of the CRMB is not responsible for the better acoustical behavior of the gap-graded pavement with high content of CR. On the other hand, at frequencies above 2 kHz the behavior is the opposite; the experimental section with CRMB presents higher sound levels than the reference section without crumb tire rubber. Nevertheless, these frequencies have a minor contribution in the
37
Fig. 7. Sound spectra measured in close proximity to the tire/pavement contact patch at 50 km/h three years after the mixtures were laid.
global sound levels. At low frequencies (300–700 Hz), no great differences were found between the sound levels of the mixtures studied. To increase the insight into the tire/CRMB pavement sound generation mechanisms responsible for its better acoustical behavior shown in Fig. 6, a study of the close proximity sound levels, related with some of the characteristic bands has been done along the test sections. The bands considered in this study were 400 Hz, 800 Hz and 4 kHz, which correspond to low, medium and high frequencies. Fig. 8 shows the close proximity sound levels that correspond to every band studied and the global sound levels. One of the aims of this work is to find out the existence of a good correlation between close proximity tire/pavement sound levels and the other surface characteristics in an in-service mixture with high content of crumb rubber. In this sense, Fig. 9 shows an example of the relationships between LCPtr levels and texture indexes (MPD and IRI), recorded in the measurements conducted along a stretch of 650 m. As is shown, the profile of texture index in the macrotexture range (MPD) is not correlated along the stretch, at least with tire/pavement global sound levels. However, at the reference section, but not only at this section, also in the sections with CRMB after the reference section (at 1100 m distance), it appears high noise levels on high texture amplitudes, which may be well correlated. On the other hand, the index IRI, which is related with the megatexture and the roughness of the pavement surface, seems
Table 4 Averaged LCPtr after three years in service conditions. Surface course
LCPtr (dB (A)) (50 km/h)
Experimental section (CRMB) Reference section
88.3 89.7
Fig. 8. LCPtr global sound levels and LCPtr sound levels for low, medium and high frequencies along the test section with and without CRMB.
38
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
Fig. 9. Correlation between LCPtr global sound levels and texture indexes (MPD and IRI).
to have a better relationship with the global sound levels LCPtr measured along the total stretch shown in Fig. 9. In accordance with Fig. 8 (800 Hz) and Fig. 9 (IRI), profiles of texture in the megatexture range (IRI), and not only in this range, may be highly correlated with medium frequencies: this is concluded from the fact that higher IRI values observed in the reference section has led to higher sound levels around 800 Hz. Therefore, in an in-service road pavement fabricated with a high content of crumb tire rubber added by wet process, texture seems to be the main mechanism responsible for the noise attenuation compared with the same pavement without CRMB. Fig. 10 shows the relation between the tire/ CRMB pavement sound emissions at high, medium and low frequencies and the different parameters of the surface texture (MPD, IRI and parameter c). The figure shows profiles with mean values every 100 m. As is shown in this figure, the high frequencies of the sound from the interaction of a rolling tire and the pavement surface (with and without CRMB) are related with the coefficient c, which defines the positive or negative character of the surface texture above the plane of the pavement surface. Every 100 m section tested has a neutral texture, however, the reference section has a texture more negative than the experimental section with CRMB, thus, the high frequency sound could be attenuated inside the voids between particles whose upper surfaces form a nearly flat plane. This is not properly an absorption mechanism, but a sound reduction due to dispersion and multiple reflections of the tire/ pavement sound emitted. Medium frequencies (around 800 Hz) seem to be strongly correlated to the megatexture and the roughness of the surface studied, as is shown in Fig. 10. As can be observed, the IRI of the reference section is higher than that recorded from the experimental CRMB mixture. Moreover, medium frequencies are the most relevant frequencies in the global tire/pavement noise, thus, the IRI may be an important texture index in order to achieve an acoustic characterization of wearing courses, at least, in gap graded bituminous mixtures as those presented in this work. On the other hand, the profile of the texture amplitudes at wavelengths in the range of macrotexture (MPD), which is related with impact and vibration sound generation mechanisms, influences tire/pavement sound generation at low frequencies (around 400 Hz), as expected: as texture amplitude increases, the vibration levels increase causing higher levels of noise, particularly at these frequencies. The MPD of the reference section (between 800 and 1000 m distance) is generally higher than those of the experimental section, however, high levels of MPD were also recorded in the reference section (between 1100 and 1200 m). These levels produce high noise levels at low frequencies, as is shown in Fig. 10.
Fig. 10. Correlation between tire/pavement sound at high, medium and low frequencies with coefficient c, IRI and MPD respectively. Reference section without CRMB is the shaded area.
4. Conclusions To analyze the influence of crumb tire rubber on the surface properties that control the acoustic performance of an in-service road pavement, different sections of a road rehabilitated with a bituminous mixture with high content of crumb rubber incorporated by a wet process were used. The main conclusions that can be derived from the laboratory and field measurements at this stage are as follows: – Dynamic stiffness and sound absorption are not relevant characteristics in the noise mitigation achieved with the experimental section with a high content of crumb rubber added by wet process; mixture with 1.5% content of CRMB of the total weight of the mixture, and an air void content less than 10%. The dynamic stiffness of the experimental section was slightly higher than those of the reference section. Lower dynamic stiffness values would be necessary to achieve noise mitigation in the overall noise emissions, by this surface property. Absorption coefficients were very low for every test section, below 0.1 in the most of the analyzed frequencies. – Texture seems to be the main responsible for the differences found between the tire/pavement noise emissions of the experimental and reference sections. The parameters calculated from the texture profile: MPD, IRI and coefficient c, show relevant differences between the experimental and reference sections. Adding crumb tire rubber may have some effect on surface texture; since mean values of experimental and reference sections are rather different. – IRI values (megatexture and roughness) of the reference sections seem to be the main responsible for the tire/pavement
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
sound levels achieved in medium frequencies, which are the most relevant frequencies in the global tire/pavement noise emissions of the analyzed test sections. According to these results, IRI is an important surface characteristic that should be taken into account when pavement acoustic studies are carried out. Further research is needed to gain an understanding of the mechanisms involved in the generation and propagation of the tire/pavement noise of experimental sections with a high content of crumb rubber added by wet process, related to road surface properties. Acknowledgements This work was partially supported by the Spanish Ministry of Economy and Competiveness with European Regional Development Funds, in the project BIA 2012-32177 within the framework of the National Plan for Scientific Research and by the Regional Council of Science and Technology of Castilla-La Mancha, project PPII-2014-012-A. The authors also wish to acknowledge J. Palomares and M.R. López from the Laboratory of Public Works of Ciudad Real for their valuable assistance. References [1] Wik TR, Miller RF. Mechanisms of tire sound generation. Research Center, B.F. Goodrich Tire Co. SAE no. 720924; 1972. p. 2633–43. [2] Heckl M. Tire noise generation. Wear 1986;113:157–70. [3] Sandberg U. Road traffic noise – the influence of the road surface and its characterization. Appl Acoust 1987;21:97–118. [4] Meiarashi S, Ishida M, Fujiwara T, Hasebe M, Nakatsuji T. Noise reduction characteristics of porous elastic road surfaces. Appl Acoust 1996;47:239–50. [5] Bennert T, Hanson D, Maher A, Vitillo N. Influence of pavement surface type on tire/pavement generated noise. J Test Eval 2005;33:94–100. [6] Paje SE, Bueno M, Terán F, Viñuela U, Luong J. Assessment of asphalt concrete acoustic performance in urban streets. J Acoust Soc Am 2008;123:1439–45. [7] Paje SE, Bueno M, Terán F, Miró R, Pérez-Jiménez F, Martínez AH. Acoustic field evaluation of asphalt mixtures with crumb rubber. Appl Acoust 2010;71:578–82.
39
[8] Mak KL, Hung WT, Lee SH. Exploring the impacts of road surface texture on tyre/road noise – a case study in Hong Kong. Transport Res Part D 2012;17:104–7. [9] Bueno M, Luong J, Terán F, Viñuela U, Paje SE. Macrotexture influence on vibrational mechanisms of the tyre-road noise of an asphalt rubber pavement. Int J Pav Eng 2014;14(7):606–13. [10] Vázquez VF, Paje SE. Dynamic stiffness of bituminous mixtures. Measurement technique and influence on tire/road noise. In: Proceedings of the inter-noise 2013 congress, (Innsbruck, 2013). [11] Luong J, Bueno M, Vázquez VF, Paje SE. Ultrathin porous pavement made with high viscosity asphalt rubber binder: a better acoustic absorption? Appl Acoust 2014;79:117–23. [12] Liao G, Sakhaeifar MS, Heitzman M, West R, Waller B, Wang S, et al. The effects of pavement surface characteristics on tire/pavement noise. Appl Acoust 2014;76:14–23. [13] Fujiwara T, Meiarashi S, Namikawa Y, Hasebe M. Reduction of equivalent continuous A-weighted sound pressure levels by porous elastic road surfaces. Appl Acoust 2005;66:766–78. [14] Paje SE, Bueno M, Viñuela U, Terán F. Toward the acoustical characterization of asphalt pavements: analysis of the tire/road sound from a porous surface (L). J Acoust Soc A 2009;125:5–7. [15] Paje SE, Luong J, Vázquez VF, Bueno M, Miró R. Road pavement rehabilitation using a binder with a high content of crumb rubber: influence on noise reduction. Construct Build Mater 2013;47:789–98. [16] Huang Y, Bird RN, Heidrich O. A review of the use of recycled solid waste materials in asphalt pavements. Resour, Conserv Recy 2007;52:58–73. [17] Licitra G, Cerchiai M, Teti L, Ascari E, Fredianelli L. Durability and variability of the acoustical performance of rubberized road surfaces. Appl Acoust 2015;94:20–8. [18] Vázquez VF, Luong J, Paje SE. Non-destructive technique for dynamic stiffness and acoustic assessment: study of gap-grades mixtures with different content in bitumen. In: International conference on noise and vibration engineeringISMA2014, 15–17 (Leuven, Belgium, 2014). [19] Arabani M, Mirabdolazimi SM, Sasani AR. The effect of waste tire thread mesh on the dynamic behavior of asphalt mixtures. Construct Build Mater 2010;24:1060–8. [20] Navarro FM, Rubio MC. Influence of crumb rubber on the indirect tensile strength and stiffness modulus of hot bituminous mixes. J Mater Civ Eng 2012;24(6):715–24. [21] Lee S, Akisetty CK, Amirkhanian SN. The effect of crumb rubber modifier (CRM) on the performance properties of rubberized binders in HMA pavements. Construct Build Mater 2008;22:1368–76. [22] PIARC 1987. Report of the committee on surface characteristics. In: Proceedings of XVIII word road congress, (Brussels, Belgium, 1987). [23] Sandberg U, Ejsmont JA. Tyre/road noise reference book. Informex; 2002.
Contents lists available at ScienceDirect
Applied Acoustics journal homepage: www.elsevier.com/locate/apacoust
Study of the road surface properties that control the acoustic performance of a rubberised asphalt mixture V.F. Vázquez, S.E. Paje ⇑ University of Castilla-La Mancha (UCLM), Laboratory of Acoustics Applied to Civil Engineering (LA2IC), Avda. Camilo José Cela s/n, 13071 Ciudad Real, Spain
a r t i c l e
i n f o
Article history: Received 1 June 2015 Received in revised form 26 August 2015 Accepted 10 September 2015 Available online 22 September 2015 Keywords: Crumb rubber Tire/pavement noise Acoustic field evaluation Generation mechanisms
a b s t r a c t Simultaneously with the fact that vehicle industry has been able to lower the noise emission from driving vehicles, tire/pavement noise has become the most significant source of traffic noise. In order to reduce it, low noise surfaces are seen as a practical solution. One of these types of surfaces may be elaborated with bituminous mixtures with crumb tire rubber added to the binder in high content by a wet process. However, the generation mechanisms involved in the tire/pavement sound and the reasons of the noise attenuation achieved with these mixtures are not so clear. This study analyses the different generation mechanisms that take place in the tire/pavement sound generation in these crumb tire rubber pavements. The surface properties of the in-service pavement, which are most important in controlling the acoustic performance (texture, acoustic absorption and dynamic stiffness or mechanical impedance), have been measured for the characterization of a test track constructed with and without crumb tire rubber. After results correlation of these surface characteristics in a pavement with crumb tire rubber added by a wet process, it seems that the parameters of roughness and texture could have a relevant role in the global tire/pavement sound emission, whereas dynamic stiffness influence is relatively minor. Ó 2015 Elsevier Ltd. All rights reserved.
1. Introduction For passenger vehicles at constant speed, tire/pavement interaction sound contributes significantly to the traffic noise at vehicle speeds greater than 40 km/h, and their generation mechanisms are influenced by pavement surface characteristics, such as sound absorption, texture depth and surface stiffness. Many tire/pavement sound studies are focused on the road texture, however, for a complete road surface characterization with respect to noise emission, the sound absorption and the dynamic stiffness (or mechanical impedance) of the surface should also be measured [1–12]. Low-noise road surfaces are seen as a good option to abate traffic noise. These surfaces include porous asphalt that, in contrast to dense asphalt surfaces, can absorb sound energy [13,14]. Porous road surfaces effect is focused in sound absorption mechanisms; nevertheless, these surfaces are not the only solutions in order to mitigate tire/pavement noise. For example, gap graded pavements with crumb rubber are another option in this respect [15]. Using crumb rubber modified asphalt, the mechanical
⇑ Corresponding author. Tel.: +34 902 204 100x3270. E-mail address: [email protected] (S.E. Paje). http://dx.doi.org/10.1016/j.apacoust.2015.09.008 0003-682X/Ó 2015 Elsevier Ltd. All rights reserved.
characteristics of the mixtures could be enhanced: crumb rubber increases pavement life and resistance to rutting and cracking [16]. Crumb rubber could be incorporated to the mixtures by dry and wet processes. In a dry process, crumb rubber is in the mixture replacing some of the solid fraction, as a part of the aggregates. On the other hand, in the wet process, it is added to bitumen before mixing it with the aggregates [17]. LA2IC has been contributing to the understanding of the acoustics and performance of asphalt pavements during the last few years [6,7,9–11,14,15,18]. Now, the aim of this study is to have a better understanding of the sound generation mechanisms of an in-service road pavement with crumb tire rubber, thus contributing to improve their acoustic design. To achieve this objective, characterization measurements of a gap graded mixture with crumb rubber added by wet process were carried out. The pavement characterization has been focused on the parameters that define the texture of the pavements: Mean Profile Depth (MPD) and International Roughness Index (IRI). The sound absorption and dynamic stiffness of the pavement surfaces were also assessed. All these parameters associated with the pavement surface are the most influential in the generation and propagation of tire/pavement noise. Results were correlated with sound generated by the tire/pavement interaction measured in the near field conditions.
34
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
dynamic stiffness (S) can be expressed in terms of complex numbers between the force (F) and the displacement (d) vectors of a tested surface:
2. Methodology 2.1. Experimental test track Three asphalt pavement sections located in Ciudad Real, Spain, are studied in this work. Two of them (experimental sections in Fig. 1) have a wearing course of a gap graded mixture (European denomination BBTM11A) containing crumb rubber added by wet process in a percentage of 20% by weight of bitumen (around 1.5% of the total weight of the mix); this is a high content for these types of road pavements. The last section (reference section) was constructed with the same gap graded mixture but without crumb rubber. The binder used for the experimental and reference test sections are respectively a high viscosity modified bitumen (BMAVC Spanish denomination) with crumb rubber and a conventional penetration grade bitumen. The characteristics of the binders employed for this research work are shown in Table 1. Crumb Rubber Modified Binder (CRMB) and conventional binder are quite different according to their characteristics. More detailed material properties of these pavements (grain size of the CRMB gap-graded mix, grain size of the crumb rubber used, etc.) are described elsewhere [15]. The measurement campaign for this work was performed after three years in service. 2.2. Dynamic stiffness Dynamic stiffness could play an important role in the tire/pavement sound generation, especially for surfaces with the same texture profile but different aggregate and bitumen content, or in surfaces aged by compaction [3,10,18]. Dynamic stiffness measurements have been achieved by means of the Non-Resonant Method, directly on the upper face of the surfaces studied [10,18]. The
S¼
F ½N d ½m
ð1Þ
The experimental set-up was composed of an impedance head that records the movement and force signals, a vibration exciter and an amplifier. Inset of Fig. 2 shows the impedance head and the exciter during a field measurement. A multianalyzer system was used to record the response and to produce the spectra of the dynamic stiffness. Sweep signals between 10 Hz and 7 kHz were used for the mixture excitation. 2.3. Sound absorption The acoustic absorption coefficient (a) is defined by the relation between the incident acoustic energy (Einc) and the absorbed acoustic energy (Eabs) by the bituminous mixture (without return).
a¼
Eabs Einc
ð2Þ
The acoustic absorption coefficient value depends on the one hand on the facility of the wave to enter the material pores and on the friction with the internal surface structure, which participates in the sound energy dissipation. Sound absorption coefficients of the mixtures studied have been evaluated by means of an impedance tube. Details of the measurement technique have been given elsewhere [6,11]. 2.4. Pavement surface profile The longitudinal profiles of the studied test sections (experimental and reference) were measured by means of the so-called LaserDynamicPG-LA2IC; a high-speed profiling laser device which allows measuring profiles of the surface course. The characteristics of this equipment are described elsewhere [15]. To characterize the surface texture of the sections studied, different parameters were used. The Mean Profile Depth (MPD) [22] was used to characterize the pavement macrotexture, whereas the International Roughness Index (IRI) [22] was used to characterize the pavement megatexture and roughness. The macrotexture corresponds to texture wavelength range from 0.5 mm to 50 mm, whereas the megatexture and roughness corresponds to texture wavelength above 50 mm up to 500 mm. The MPD influences
Fig. 1. Location with GPS coordinates of the test track section. Inset displays the equipment for the close proximity and longitudinal profile measurements of the road surface.
Table 1 Pavement’s specifications.
Binder Penetration (0.1 mm, 25 °C) EN-1426 Ring and ball softening point (°C) EN-1427
CRMB (experimental)
B 50/70 (reference)
19 80
57 51
5.7%
6%
Air void content
Fig. 2. Dynamic stiffness measured on the pavement without CRMB (reference) and with CRMB (experimental section). An example of the coherence function is also shown.
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
surface characteristics including tire/pavement noise or skid resistance. It is well known that increasing texture amplitudes in the transition region between the macro and megatexture range increases the lower frequencies of the tire/pavement noise by enhancing tire vibrations. IRI index is related with megatexture range and roughness. It is the most commonly used index for evaluating the smoothness of road pavement surfaces, and it is an indicator of the ride quality. However, its relation with tire/pavement noise has not been sufficiently studied yet in crumb tire rubber pavements. In order to assess the character of the texture, the coefficient c is achieved from the MPD and the Root Mean Square Values (RMS) quotient. RMS is defined by:
2 1 x1 þ x22 þ þ x2n 2 RMS ¼ n
ð3Þ
where xn represents every MPD datum recorded along the sections, and n is the total number of data recorded. The coefficient c (c = MPD/RMS) gives information about the texture in the vertical plane: it can be positive regarding the plane of the road surface (c > 1.05), neutral (0.95 < c < 1.05) or negative (c < 0.95). A negative texture could lead to noise reduction at high frequencies, due to the attenuation of the sound wave energy. 2.5. Close proximity sound methodology In order to assess the influence on the sound generation mechanisms of an in-service road pavement with and without crumb tire rubber, a methodology for geo-referenced close proximity measurements developed by LA2IC was employed (see for example Ref. [7]). Trailer Tiresonic Mk4-LA2IC, to which a reference tire was assembled, was used in the close proximity sound measurements as a part of the test vehicle (see inset of Fig. 1). Two microphones are mounted very close to the wheel, in order to exclusively evaluate the acoustical performance of the asphalt mixtures. During the measurements, the vehicle speed was kept close to the reference speed selected for the in-service mixture assessment. Moreover, the evolution of the tire/pavement sound levels were corrected for speed variations around the reference speed, according to the expression [14]:
Lcorr ðtÞ ¼ Lmeas ðtÞ B log
v ðtÞ v ref
ð4Þ
35
shows experimental data and the associated error bars. The coherence function of the measurements is also shown in this figure. Coherence function informs about the relation between force and motion. Its value was close to unity between the frequencies studied, thus, a good relationship between the incoming signals was achieved with the measurement set-up. Inset of Fig. 2 shows the impedance head and the vibration exciter (shaker), which provides the excitation force, during an in-situ measurement on the mixture with crumb tire rubber. According to the results shown in this figure, the dynamic stiffness of the mixture with CRMB (experimental section) is higher than that of the reference mixture, between 150 Hz and 400 Hz. This agrees with recent studies carried out in mixtures sampled and compacted in the laboratory, where the use of crumb rubber modified binders lead in higher stiffness modulus than the same binder without crumb rubber [19–21]. On the other hand, the behavior of these mixtures agrees with the bitumen characteristics shown in Table 1: the softening value of binders is higher for reference binder (B 50/70), and the CRMB shows a better resistance to penetration than the binder without CR. According to these results, dynamic stiffness potentially could be responsible of higher sound emissions of a pavement with high content of CRMB added by wet process, i.e. the mixture used for the experimental section. Lower dynamic stiffness values would be necessary to achieve noise mitigation in the overall noise emissions, by this surface property. Fig. 3 shows the normal incidence acoustic absorption spectra of core samples, with and without crumb rubber with an air void content lower than 10%, see Table 1. These sample cores were extracted from each of the test sections studied by removing surface material in different locations three years after the mixtures were laid. The absorption coefficient has a similar behavior for all the mixtures studied. It is below 0.1 for most of the analyzed frequencies, and every value is below 0.25. According to this similar result, the sound absorption capacity will not be responsible for potential differences between the tire/pavement noise of the mixtures studied. This conclusion that pavements with an air void content lower than 15–20% are not able to generate significant sound absorption has also been reported previously for different asphalt pavements [3,6,11,15]. Field measurements to register the longitudinal profiles of the different test sections have enabled us to obtain, by numerical calculations, the MPD: the texture related with macrotexture range. The results of the MPD calculation from laser profiling are
where the speed constant B was obtained from the slope of the logarithmic regression of the sound pressure level and speed:
LCPtr ¼ A þ B logðVÞ
ð5Þ
On the other hand, since temperature variations between the different tests were insignificant, sound levels were free from the influence of temperature in the measurements performed; therefore, temperature corrections of sound levels, LCPtr, have not been carried out. In addition, it is important to note that the test equipment allows measuring close proximity tire/pavement sound and wearing course profile simultaneously, on the wheel-path followed by the reference tire in successive tests, and in a geo-referenced way. This aspect is crucial for the relationship between the different surface characteristics evaluated. 3. Analysis of measurements and discussion Dynamic stiffness of the mixtures was studied in in-situ conditions, on the wearing course of the experimental (with CRMB) and reference sections three years after the mixtures were laid. The dynamic stiffness results in N/m are shown in Fig. 2; the figure
Fig. 3. Normal incidence acoustic absorption spectra of core samples extracted from experimental and reference sections. The inset shows the core samples evaluated.
36
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
Table 2 Averaged texture parameters of the road after three years in service conditions. Surface course
MPD value (mm)
c
IRI (m/km)
Experimental section (CRMB) Reference section
1.09 1.17
1.00 0.99
0.78 1.18
presented in Table 2. As can be seen, a higher MPD over the reference section (mixture without crumb rubber) is observed. In addition, from longitudinal surface profiles, the index IRI was calculated, and their values are shown in Table 2. The coefficient c was also evaluated. This coefficient may be related with noise attenuation at high and low frequencies, since the coefficient defines the character (negative, neutral or positive) of the surface studied. As is shown in Table 2, significant differences between the section fabricated with a pavement incorporating CRMB and the reference section, with the same type of pavement but without crumb rubber, were found. These textures could produce different noise emissions levels due to tire/pavement interaction. In order to gain insight into the influence of crumb tire rubber on the surface properties that control the tire/ CRMB pavement noise of an in-service road pavement fabricated with a high content of CRMB, close proximity sound measurements were carried out on the test sections with and without crumb tire rubber, three years after the mixtures were laid. First, the dependence with speed of the sound emission from the interaction of the reference tire and the surface of the test sections was measured upon increasing speed from 20 km/h to 120 km/h. Thus, for the test section fabricated with the incorporation of crumb tire rubber, Fig. 4 shows the global sound levels LCPtr versus logarithm of speed. The linear regression of LCPtr and speed gives coefficient B = 35.4 dB (A), see Eq. (5). In addition, the relationship between LCPtr and speed for different frequency bands shows that coefficient B presents values of around 36 dB (A) for 1.25 kHz and 4 kHz, and 23 dB (A) for 400 Hz. These frequencies are attributed to a mechanism of sound generation related to the flow of air in and around the rolling tire and the pavement surface texture, and to a mechanism related to the excitation and vibrations of the tire tread elements during the contact with the pavement, respectively [14,23]. Fig. 4 indicates that the variation in the coefficient B increases nonlinearly with frequency. This was one of the main results of a previous study [14]. The coefficient B is greater at frequencies above around 1 kHz, implying
Fig. 4. Global sound levels and sound levels for different frequencies as a function of vehicle speed. Their corresponding linear regressions are also included.
Table 3 Slope of the linear regressions, B, of the studied test track sections. High and low frequencies are also shown. Surface course
Experimental section with CRMB Reference section without CRMB
Slope (coefficient B) Eq. (5) Global
High frequencies (4 kHz)
Low frequencies (400 Hz)
35.4
36
23
36.5
31
26
that at high frequencies, the sound from the tire/pavement interaction increases with speed at a greater rate than sound at lower frequencies. Table 3 shows the slope of the linear regressions, B, of both test sections with and without crumb tire rubber. As can be seen, global sound levels generated by the interaction of the tire and the pavement with high content of crumb tire rubber increases linearly proportional to 35 log(V). This result of the sound-speed slope, similar for both reference and CRMB sections, is usually observed on dense and gap-graded asphalt concrete surfaces or porous asphalt surfaces [6,7,14,15], depending on the conservation state of the surface and/or its texture. In Table 3, the relative importance of low frequencies on the sound generation as a function of speed, for both types of mixtures (slope relatively low), and their similar behavior of the global sound levels can be seen. Additional analysis of the speed influence on the sound frequencies could be useful; thus, Fig. 5 shows the relationship between sound spectrum and speed at different speeds between 30 km/h and 120 km/h, for the section with CRMB. This figure shows that the increment of the sound levels is about 14 dB (A) at low frequencies (400 Hz) between 30 km/h and 120 km/h, whereas the increment at medium and high frequencies (1.25 and 4 kHz respectively) is about 20 dB (A). As can be seen, the coefficient B increases nonlinearly with frequency in mixtures with CRMB. A similar behavior was also reported for dense asphalt [14]. As is observed, the significance of low frequencies, attributed to a mechanism related to vibration induced by the pavement surface, on the sound levels as a function of speed is small compared with higher frequencies, attributed to a mechanism related to aerodynamic effects near the tire/pavement contact patch.
Fig. 5. Sound spectra of the experimental section with CRMB measured in close proximity to the tire/pavement contact patch at different speeds.
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
Fig. 6. Frequency spectrum map (50 km/h) of the in-service surfaces, with and without crumb rubber. A color scale in dB (A) was used for sound levels. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
The acoustical influence of the incorporation of crumb rubber in high content to a gap graded bituminous mix, three years after the mixture with CRMB was laid, is reflected in Fig. 6. This figure shows the LCPtr frequency spectrum mapping. It was performed at 50 km/h, between the test sections with and without CRMB in a length of around 650 m. The reference section (without crumb rubber), with a length of 200 m, is included between points 800 m and 1000 m distance in this figure. As can be observed in this figure, sections fabricated with high content of CRMB present, after three years, an important capacity to reduce the rolling noise. The mean values of the global sound levels associated to each test section, measured in close proximity to the tire/pavement contact patch at reference speed of 50 km/h, are shown in Table 4. Averaged values shown in this table were calculated from measurements along six stretches of at least 100 m, before and after the reference section, using different runs. Additional analyses are necessary to comprehend the mechanisms responsible for the above-mentioned results shown in Fig. 6. Thus, Fig. 7 shows the averaging of the close proximity sound spectra between 300 Hz and 5 kHz in the one-third octave bands. Spectra, speed corrected, are recorded rolling at 50 km/h over a representative length (at least 100 m) of the test sections. The figure highlights the different behavior of the reference and experimental (CRMB) sections as a function of frequency. As it is shown in Fig. 7, the sound reduction achieved with the incorporation of high content of crumb rubber added by wet process is located at medium frequencies (0.8–1.25 kHz), which have the greatest effect in the global LCPtr sound levels shown in Fig. 6. This sound reduction observed in a mixture with high content of CRMB at medium frequencies is not linked to a higher mechanical stiffness presented by the mixture with crumb rubber added by wet processes (see Fig. 2). In addition, as above mentioned, the absorption of the CRMB is not responsible for the better acoustical behavior of the gap-graded pavement with high content of CR. On the other hand, at frequencies above 2 kHz the behavior is the opposite; the experimental section with CRMB presents higher sound levels than the reference section without crumb tire rubber. Nevertheless, these frequencies have a minor contribution in the
37
Fig. 7. Sound spectra measured in close proximity to the tire/pavement contact patch at 50 km/h three years after the mixtures were laid.
global sound levels. At low frequencies (300–700 Hz), no great differences were found between the sound levels of the mixtures studied. To increase the insight into the tire/CRMB pavement sound generation mechanisms responsible for its better acoustical behavior shown in Fig. 6, a study of the close proximity sound levels, related with some of the characteristic bands has been done along the test sections. The bands considered in this study were 400 Hz, 800 Hz and 4 kHz, which correspond to low, medium and high frequencies. Fig. 8 shows the close proximity sound levels that correspond to every band studied and the global sound levels. One of the aims of this work is to find out the existence of a good correlation between close proximity tire/pavement sound levels and the other surface characteristics in an in-service mixture with high content of crumb rubber. In this sense, Fig. 9 shows an example of the relationships between LCPtr levels and texture indexes (MPD and IRI), recorded in the measurements conducted along a stretch of 650 m. As is shown, the profile of texture index in the macrotexture range (MPD) is not correlated along the stretch, at least with tire/pavement global sound levels. However, at the reference section, but not only at this section, also in the sections with CRMB after the reference section (at 1100 m distance), it appears high noise levels on high texture amplitudes, which may be well correlated. On the other hand, the index IRI, which is related with the megatexture and the roughness of the pavement surface, seems
Table 4 Averaged LCPtr after three years in service conditions. Surface course
LCPtr (dB (A)) (50 km/h)
Experimental section (CRMB) Reference section
88.3 89.7
Fig. 8. LCPtr global sound levels and LCPtr sound levels for low, medium and high frequencies along the test section with and without CRMB.
38
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
Fig. 9. Correlation between LCPtr global sound levels and texture indexes (MPD and IRI).
to have a better relationship with the global sound levels LCPtr measured along the total stretch shown in Fig. 9. In accordance with Fig. 8 (800 Hz) and Fig. 9 (IRI), profiles of texture in the megatexture range (IRI), and not only in this range, may be highly correlated with medium frequencies: this is concluded from the fact that higher IRI values observed in the reference section has led to higher sound levels around 800 Hz. Therefore, in an in-service road pavement fabricated with a high content of crumb tire rubber added by wet process, texture seems to be the main mechanism responsible for the noise attenuation compared with the same pavement without CRMB. Fig. 10 shows the relation between the tire/ CRMB pavement sound emissions at high, medium and low frequencies and the different parameters of the surface texture (MPD, IRI and parameter c). The figure shows profiles with mean values every 100 m. As is shown in this figure, the high frequencies of the sound from the interaction of a rolling tire and the pavement surface (with and without CRMB) are related with the coefficient c, which defines the positive or negative character of the surface texture above the plane of the pavement surface. Every 100 m section tested has a neutral texture, however, the reference section has a texture more negative than the experimental section with CRMB, thus, the high frequency sound could be attenuated inside the voids between particles whose upper surfaces form a nearly flat plane. This is not properly an absorption mechanism, but a sound reduction due to dispersion and multiple reflections of the tire/ pavement sound emitted. Medium frequencies (around 800 Hz) seem to be strongly correlated to the megatexture and the roughness of the surface studied, as is shown in Fig. 10. As can be observed, the IRI of the reference section is higher than that recorded from the experimental CRMB mixture. Moreover, medium frequencies are the most relevant frequencies in the global tire/pavement noise, thus, the IRI may be an important texture index in order to achieve an acoustic characterization of wearing courses, at least, in gap graded bituminous mixtures as those presented in this work. On the other hand, the profile of the texture amplitudes at wavelengths in the range of macrotexture (MPD), which is related with impact and vibration sound generation mechanisms, influences tire/pavement sound generation at low frequencies (around 400 Hz), as expected: as texture amplitude increases, the vibration levels increase causing higher levels of noise, particularly at these frequencies. The MPD of the reference section (between 800 and 1000 m distance) is generally higher than those of the experimental section, however, high levels of MPD were also recorded in the reference section (between 1100 and 1200 m). These levels produce high noise levels at low frequencies, as is shown in Fig. 10.
Fig. 10. Correlation between tire/pavement sound at high, medium and low frequencies with coefficient c, IRI and MPD respectively. Reference section without CRMB is the shaded area.
4. Conclusions To analyze the influence of crumb tire rubber on the surface properties that control the acoustic performance of an in-service road pavement, different sections of a road rehabilitated with a bituminous mixture with high content of crumb rubber incorporated by a wet process were used. The main conclusions that can be derived from the laboratory and field measurements at this stage are as follows: – Dynamic stiffness and sound absorption are not relevant characteristics in the noise mitigation achieved with the experimental section with a high content of crumb rubber added by wet process; mixture with 1.5% content of CRMB of the total weight of the mixture, and an air void content less than 10%. The dynamic stiffness of the experimental section was slightly higher than those of the reference section. Lower dynamic stiffness values would be necessary to achieve noise mitigation in the overall noise emissions, by this surface property. Absorption coefficients were very low for every test section, below 0.1 in the most of the analyzed frequencies. – Texture seems to be the main responsible for the differences found between the tire/pavement noise emissions of the experimental and reference sections. The parameters calculated from the texture profile: MPD, IRI and coefficient c, show relevant differences between the experimental and reference sections. Adding crumb tire rubber may have some effect on surface texture; since mean values of experimental and reference sections are rather different. – IRI values (megatexture and roughness) of the reference sections seem to be the main responsible for the tire/pavement
V.F. Vázquez, S.E. Paje / Applied Acoustics 102 (2016) 33–39
sound levels achieved in medium frequencies, which are the most relevant frequencies in the global tire/pavement noise emissions of the analyzed test sections. According to these results, IRI is an important surface characteristic that should be taken into account when pavement acoustic studies are carried out. Further research is needed to gain an understanding of the mechanisms involved in the generation and propagation of the tire/pavement noise of experimental sections with a high content of crumb rubber added by wet process, related to road surface properties. Acknowledgements This work was partially supported by the Spanish Ministry of Economy and Competiveness with European Regional Development Funds, in the project BIA 2012-32177 within the framework of the National Plan for Scientific Research and by the Regional Council of Science and Technology of Castilla-La Mancha, project PPII-2014-012-A. The authors also wish to acknowledge J. Palomares and M.R. López from the Laboratory of Public Works of Ciudad Real for their valuable assistance. References [1] Wik TR, Miller RF. Mechanisms of tire sound generation. Research Center, B.F. Goodrich Tire Co. SAE no. 720924; 1972. p. 2633–43. [2] Heckl M. Tire noise generation. Wear 1986;113:157–70. [3] Sandberg U. Road traffic noise – the influence of the road surface and its characterization. Appl Acoust 1987;21:97–118. [4] Meiarashi S, Ishida M, Fujiwara T, Hasebe M, Nakatsuji T. Noise reduction characteristics of porous elastic road surfaces. Appl Acoust 1996;47:239–50. [5] Bennert T, Hanson D, Maher A, Vitillo N. Influence of pavement surface type on tire/pavement generated noise. J Test Eval 2005;33:94–100. [6] Paje SE, Bueno M, Terán F, Viñuela U, Luong J. Assessment of asphalt concrete acoustic performance in urban streets. J Acoust Soc Am 2008;123:1439–45. [7] Paje SE, Bueno M, Terán F, Miró R, Pérez-Jiménez F, Martínez AH. Acoustic field evaluation of asphalt mixtures with crumb rubber. Appl Acoust 2010;71:578–82.
39
[8] Mak KL, Hung WT, Lee SH. Exploring the impacts of road surface texture on tyre/road noise – a case study in Hong Kong. Transport Res Part D 2012;17:104–7. [9] Bueno M, Luong J, Terán F, Viñuela U, Paje SE. Macrotexture influence on vibrational mechanisms of the tyre-road noise of an asphalt rubber pavement. Int J Pav Eng 2014;14(7):606–13. [10] Vázquez VF, Paje SE. Dynamic stiffness of bituminous mixtures. Measurement technique and influence on tire/road noise. In: Proceedings of the inter-noise 2013 congress, (Innsbruck, 2013). [11] Luong J, Bueno M, Vázquez VF, Paje SE. Ultrathin porous pavement made with high viscosity asphalt rubber binder: a better acoustic absorption? Appl Acoust 2014;79:117–23. [12] Liao G, Sakhaeifar MS, Heitzman M, West R, Waller B, Wang S, et al. The effects of pavement surface characteristics on tire/pavement noise. Appl Acoust 2014;76:14–23. [13] Fujiwara T, Meiarashi S, Namikawa Y, Hasebe M. Reduction of equivalent continuous A-weighted sound pressure levels by porous elastic road surfaces. Appl Acoust 2005;66:766–78. [14] Paje SE, Bueno M, Viñuela U, Terán F. Toward the acoustical characterization of asphalt pavements: analysis of the tire/road sound from a porous surface (L). J Acoust Soc A 2009;125:5–7. [15] Paje SE, Luong J, Vázquez VF, Bueno M, Miró R. Road pavement rehabilitation using a binder with a high content of crumb rubber: influence on noise reduction. Construct Build Mater 2013;47:789–98. [16] Huang Y, Bird RN, Heidrich O. A review of the use of recycled solid waste materials in asphalt pavements. Resour, Conserv Recy 2007;52:58–73. [17] Licitra G, Cerchiai M, Teti L, Ascari E, Fredianelli L. Durability and variability of the acoustical performance of rubberized road surfaces. Appl Acoust 2015;94:20–8. [18] Vázquez VF, Luong J, Paje SE. Non-destructive technique for dynamic stiffness and acoustic assessment: study of gap-grades mixtures with different content in bitumen. In: International conference on noise and vibration engineeringISMA2014, 15–17 (Leuven, Belgium, 2014). [19] Arabani M, Mirabdolazimi SM, Sasani AR. The effect of waste tire thread mesh on the dynamic behavior of asphalt mixtures. Construct Build Mater 2010;24:1060–8. [20] Navarro FM, Rubio MC. Influence of crumb rubber on the indirect tensile strength and stiffness modulus of hot bituminous mixes. J Mater Civ Eng 2012;24(6):715–24. [21] Lee S, Akisetty CK, Amirkhanian SN. The effect of crumb rubber modifier (CRM) on the performance properties of rubberized binders in HMA pavements. Construct Build Mater 2008;22:1368–76. [22] PIARC 1987. Report of the committee on surface characteristics. In: Proceedings of XVIII word road congress, (Brussels, Belgium, 1987). [23] Sandberg U, Ejsmont JA. Tyre/road noise reference book. Informex; 2002.