Testing and Noise Assessment of Two Types of Bridge Expansion Joints: Case Study

Abstract

Expansion joints mounted on the edges of bridges can cause excessive noise and environmental nuisance. Currently, there are no standardized European methods for assessing the noise of these devices. This article presents the results of investigations of two expansion joint devices: a modular one used for small displacements and a finger one used for large displacements. The method proposed in the Austrian standard was used to evaluate the acoustic effects. The exposure levels of the sound were compared after analyzing 100 reliable car passes through each device. Acoustic signals were recorded and analyzed at three points. In the case of the modular device, the average exposure sound level above the device was 2.6 dB higher than the noise above one outside the device. For a finger device, the difference was 1.2 dB. The latter device can be considered “low-noise”. The amplitude–frequency characteristics of the recorded phenomena were also analyzed to show which frequencies are responsible for excessive noise. The dependence of sound emissions on the speed of cars was also determined. The conducted research has shown that the adopted method can be successfully used for the acoustic evaluation of expansion joint devices.

1. Introduction

Expansion joints are installed on the edges of bridge spans to fill gaps caused by temperature changes, rheological processes, and loads.

Cars driving through these devices often generate noise that can exceed acceptable levels in the environment. Bridge structures, which usually improve traffic conditions, can often incite criticism or even complaints in such cases. The sounds generated when driving through expansion joints are different from the sounds coming from traffic on a flat, even surface, as these are mainly impulse sounds, distinguished from other ones by their high volume, which results in them being more bothersome. They depend primarily on the type of device and its technical condition. In the assessment of expansion joints, it is important to determine to what extent the noise from an expansion joint exceeds the noise caused by road traffic outside the device.

This problem is recognized in many regulations. For example, the standard in [1] recommends that when preparing an investment, specific requirements for noise emissions should be included in the technical specifications for expansion joints. However, these requirements have not been specified.

One of the documents used in the design of expansion joints in Poland contains recommendations for the selection of bridge expansion joints as well as their installation and acceptance [2]. These recommendations indicate the need to assess the noise generated by expansion joint devices; it is noted that such an assessment is very subjective, but the criteria for this assessment have not been specified.

In the guidelines in [3], which present the typical elements and equipment of road engineering structures, it is stated that the impact of noise emitted by vehicles should be taken into account when selecting the expansion joint, particularly in residential zones and recreational areas. The guidelines for the design of environmental protection elements and devices on road engineering structures [4] provide general recommendations for the selection of devices that take low noise emissions into account. None of these documents provide a method for assessing expansion joints in terms of acoustics, nor are there any acoustic criteria for the selection, acceptance, or replacement of these devices.

Analyzing the literature, we can find many examples of noise tests of expansion joints, with many different methods used. Measurements were made at different points, and different evaluation parameters and signal analysis methods were adopted. The results of such studies cannot be compared. For example, in paper [5] the results of measurements of noise peaks for expansion joints are presented. Measurements were carried out only at the axis of the devices, 0.3 m above the surface. No measurements were made at the reference point/section. The measurement results were closely related to passing cars, and the acoustic parameter adopted made it difficult to objectively assess the devices. A similar problem was discussed in paper [6], where in addition to noise, vibrations of the expansion joint device were analyzed. It has been shown that in some cases the increase in noise at the device is caused by vibrations of the elements of the expansion joint. The mechanism of sound generation was also analyzed in [7,8,9]. Devices with low noise emissions have been presented, such as those in [10].

In paper [11], it was noted that various test methods were used, including the use of a different measurement distance from the road or the height of the microphone. In this paper, the noise was measured at a distance of 7.5 m from the axis of the measured lane and at a height of 5.0 m from the road surface.

Particularly noteworthy is the test method presented in paper [12] and developed on this basis by the Austrian standard [13]. The method adopted in these studies consists of measuring the exposure sound level, L_AE, at a distance of 3.75 m from the lane axis, 1.2 m above the device, and at the reference point, i.e., at the same distance from the lane axis and at the same height, but at a distance of at least 30 m from the device. In addition, it was recommended to take measurements under the device. The sound exposure level was adopted for the acoustic assessment of expansion joint devices. It should be noted that the adopted parameter is used in acoustics to describe individual sound events and effectively reflects the value of the energy of an impulse sound event that arises, for example, when a car drives over a bridge expansion joint. This parameter is independent of time and traffic intensity, which allows for the better comparability of individual sound events and the determination of the dependence of the sound level on the speed of cars. The standard in [13] allows for testing in two equivalent ways: A—measurements during the passage of a research vehicle with precisely defined parameters and B—measurements under traffic, taking into account 100 journeys of passenger cars and vans. An acoustic classification of the devices was also proposed.

Paper [14] presents the results of tests of two devices during the passage of several vehicles. The assessment was based on the measurement of the exposure sound level at a distance of 3.75 m from the road axis and at a height of 1.2 m—above and outside the device. Attention was drawn to the lack of a unified methodology for noise measurements of expansion joint devices.

Paper [15] presents results of tests of expansion joint devices, where the microphones were located in the axis of the devices, at the edge of the road lane, 0.3 m above the expansion joints, and at the same height, but about 30 m from the tested devices. The authors presented the measurement procedure and the results of tests of four types of expansion joint devices: finger joint, single-module, single-module with a soundproofing overlay, and multi-module. The devices were located on six different bridge structures. The assessment was based on the equivalent sound level recorded within 15 or 30 min or 24 h. Paper [16] presents the results of noise tests of modular devices and the assessment of the impact of noise shields installed under the device on noise levels. The assessment was based on the measured equivalent sound level recorded before and after the installation of the shields. Paper [17] reports the results of a study on the noise of a modular device without and with a silencing overlay during the passage of two types of vehicles, i.e., passenger cars and trucks. The increase in noise at the devices caused by these vehicles was compared. The measurements were carried out at the axis of the device and at a distance of 30 m from this axis. The exposure sound level was analyzed, and it was shown that a passenger vehicle has a greater impact on the increase in noise next to the device. In paper [18], the sound pressure level in the vicinity of two expansion joints was investigated—one device was mounted perpendicular to the road axis, with a second at an angle of approx. 52°. The sound pressure level values with and without correction were compared.

The literature analysis showed a very diverse approach to the noisiness testing of expansion joints and the lack of a unified, harmonized European method that would be reliable for their assessment. The Austrian method [13] was adopted for the research discussed in this article, because, above all, this method best describes the impulsive nature of acoustic phenomena, allows for the assessment of the change in noise over time, determines the effect of travel speed on noise, and allows for the classification of devices in terms of acoustics. The article will present the results of research and the assessment of the noise of two expansion joints using this method (one of the devices was the subject of complaints from local residents). In addition, the tests are aimed at confirming the usefulness of the method used for the assessment of the devices in operation.

2. Characteristics of the Tested Expansion Joints

2.1. Modular Device Characteristics

Modular expansion joints are one of the most popular devices. They can be single-module or multi-module. This article discusses the research of a single-module device. For these devices, the range of compensated displacements is 80 (±40) [3]. The advantages of modular expansion joints include low maintenance requirements and a long service life. The disadvantages of modular expansion joints include difficulties in assembly and the lack of the possibility of renovation when vehicles are moving on-site.

The tested device was mounted on a motorway bridge, where the maximum permissible speed of vehicles was 140 km/h. Its total width was 155 mm, and the width of the gap was 40 mm. Top and bottom views of the device are shown in Figure 1.

Local residents complained about the noise of this device. The device was in good technical condition but was mounted 5 to 8 mm lower in relation to the bituminous surface in the immediate vicinity. It should be noted that according to [2], the bituminous surface layer should be laid from 0 to 3 mm above the expansion joint. The inaccuracy of the device in the vertical position in relation to the surface should not exceed ±2 mm. Therefore, it can be concluded that the device does not meet the requirements of [2].

2.2. Finger Device Characteristics

Finger devices are used in medium- and large-span bridges. For these devices, the range of compensated displacements is from 100 to 800 (from ±50 to ±400) [3].

The advantages of these expansion joints include the ease of passage for motor vehicles and the possibility of installation and repair when individually closing single lanes on a bridge structure, which is very difficult or impossible to organize when using other types of devices. The disadvantages of finger expansion joints are their high price, the need to clean the apron on which rubbish proceeds through the gaps between the finger elements, and the inability to allow bicycle traffic. The longitudinal gaps between the “fingers” are about 40 mm wide, and the bicycle wheel can fall into them and become wedged. Therefore, it is recommended to use these devices on facilities intended only for car traffic.

The tested device, like the previous ones, is located along a highway. The total width of the device during the tests is 1000 mm, and the width of the gap is 250 mm. The device was in good technical condition; it was lowered by 1 to 4 mm in relation to the bituminous surface. On the bottom, the device was equipped with a plastic apron, which had dehydration and anti-noise functions. Top and bottom views of the device are shown in Figure 2.

3. Research Methodology

The method used for the discussed research was similar to the one described in the standard in [13]. This method has many advantages, including the following:

• Effectively describes the impulse nature of the phenomena occurring at expansion joints;
• Reflects the acoustic characteristics of devices;
• Evaluation of the devices does not depend on the traffic volume and the day of measurements;
• Allows for the comparison of different devices;
• Allows for the acoustic classification of the devices;
• Allows for the determination of the impact of the technical condition of devices on sound emissions and for assessing changes in their noise levels over time;
• Allows for the assessment of the dependence of sound emission levels on the speed of cars (each passage is analyzed separately).

During our own research, a set of microphones of the same type, windscreens, appropriate measuring cassettes, and B&K PULSE version 15.1.0 software were used; the measuring tracks were calibrated. The microphones, similar to those in [13], were located in three points:

• At a distance of 3.75 m from the lane axis at the axis of the expansion joint, 1.2 m above the road (M1);
• At the same distance from the lane axis and at a distance of a min. of 30 m from the device (reference point) and 1.2 m above the road (M2);
• Under the expansion joint device in the lane axis (M3).

The scheme location of the measurement points is shown in Figure 3 and Figure 4.

In order to assess the acoustic characteristics of the expansion joints, the sound exposure level was determined as follows:

$$L_{AE}=10\mathrm{l}\mathrm{o}\mathrm{g}\left[{\displaystyle \frac{1}{t_0}}{\int }_{t1}^{t2}{\displaystyle \frac{p_A^2}{p_0^2}}dt\right], \mathrm{d}\mathrm{B}$$

(1)

where t₁ and t₂—the beginning and end of the evaluated event (observation time); t₀—reference time equal to 1 s; p_A—measured sound pressure, corrected by the frequency response A; and p₀—reference pressure value equal to 20 μPa.

The sound exposure level, L_AE, for a reference time of 1 s is used to describe single sound events. Unlike the equivalent sound level, which is determined in time, T, it is independent of time, which allows for the better comparability of individual sound events.

The L_AE reflects the total sound energy of a single event that is emitted during the passage, both at the reference section and above the expansion joint, and can be converted into an equivalent sound level. For this reason, the L_AE parameter was adopted for the assessment of expansion joints. In accordance with [13], a two-stage classification of devices in terms of acoustics has been proposed—an expansion joint device is considered “low-noise” if the L_AE values above the device are of a max. of 2 dB greater than the value at the reference point.

According to [13], at least 100 authoritative crossings of passenger cars and vans were registered for each device (method “B”). Each pass was recorded with a camera set in the axis of the expansion joint at a speed of 240 frames per second and was analyzed individually. In order to determine the speed of a car, the time between the first and second axle of the car was determined—Figure 5. Then, the type of vehicle was identified and the wheelbase was read from its technical data. The vehicle model was identified using AI technology—Google Lens [19]. The speed of cars was calculated using the following formula:

$$v= {\displaystyle \frac{d}{t_2-t_1}}$$

(2)

where t₁—time for the first axle pass through the device [s], t₂—time for the second axle pass through the device [s], and d—the wheelbase of the vehicle [m].

The use of a camera with a speed of 240 frames per second made it possible to estimate the speed of cars with an accuracy of no less than 2.5 km/h, which can be considered sufficient for the analyses conducted.

During the research, the surroundings of the device were monitored through direct observation and the additional recording of movement with a wide-angle camera. Only reliable crossings, i.e., without interference from other sources, were selected for the analysis.

The analysis only took into account when there were no other vehicles nearby or in the remaining lanes. Only those crossings were considered authoritative, the maxima of which were more than 6 dB higher than the neighboring minima at the measurement point. This requirement allows for the elimination of possible interference from the rest of the traffic. If the difference between pass-by noise and ambient noise is more than 10 dB, any accompanying effects that may occur (additional sources unrelated to the analyzed vehicle such as wind, reflections, etc.) can be disregarded as they do not have a significant impact on the determined exposure level.

Simultaneously with the recording of sound pressure, meteorological parameters such as temperature, wind speed, and relative humidity were measured by a mobile weather station.

4. Results

4.1. Modular Device

The tests were carried out at an air temperature of 17.0 to 22.0 °C, with an average wind speed of 0.5 m/s and a humidity of 75%. The measurements were carried out during the normal operation of the facility, but the emergency lane and adjacent lane were closed to traffic for the duration of the measurements. The scheme of the location of measurement points is shown in Figure 3 and Figure 4. The roadway, measurement points, and weather station are shown in Figure 6. Only passenger cars and vans were taken into account.

Figure 7 shows the values of the exposure sound level at points M1 and M3 depending on the values at reference point M2 for 100 crossings over the device. A diagonal line indicates values 2 dB greater than the values at reference point M2. It can be seen that most of the passages through the expansion joints (65%) generate a sound level at the M1 point that is more than 2 dB higher than that at the reference point. According to the adopted criterion, the device cannot be considered “low-noise”.

Table 1. Average exposure sound level at points M1, M2, and M3.

Parameter	Measuring Point			Differences
	M1	M2	M3	M1 − M2	M3 − M2	M3 − M1
Arithmetic mean LAE [dB]	87.9	85.3	80.9	2.6	−4.4	−7.0
Standard deviation [dB]	2.75	2.34	3.57	-	-	-

shows the average values of the exposure sound level. It can be seen that for 100 recorded rides, the exposure sound level above the device is on average 2.6 dB higher than the value at the reference point. The level below the device is on average 4.4 dB from the value at the reference point and 7.0 dB lower than the level above the device.

Figure 8 presents examples of amplitude–frequency characteristics of the sound level. It can be seen that the sound pressure level above the device (M1) increases in the low- and medium-frequency range, i.e., from 50 to 1000 Hz; it does not change in the range of higher frequencies. Under the expansion joint device (M3), we can observe high attenuation of sounds with a frequency above 600 Hz.

Figure 9 shows the L_AE values depending on the vehicle speed at points M1, M2, and M3. A linear trendline and R-squared coefficient trend are also presented. As the speed of the cars increases, the sound exposure levels increase almost equally above the device (M1) and above the road in the reference cross-section (M2). When changing the speed from 70 km/h to 140 km/h, the sound exposure level increases by an average of 34 dB at the M1 point, by 3.1 dB at the M2 point, and 4.9 dB at the M3 point. Therefore, it can be assumed that with a twofold increase in speed, the volume of the tested device increases at point M1 by 0.3 dB and by 1.6 dB at point M3.

4.2. Finger Device

The tests were carried out at an air temperature of 18.0–22.8 °C, with an average wind speed of 1.3 m/s and an average air humidity of 53.6%. As with the previous device, the measurements were carried out during the normal operation of the facility; the emergency lane and adjacent lane were closed. Measurement points are shown in Figure 3 and Figure 4. Only passenger cars and vans were taken into account.

Figure 10 shows the values of the sound exposure level at points M1 and M3 depending on the values at point M2. Similarly to the previous lane, the diagonal line indicates values 2 dB greater than the values at reference point M2. Almost all crossings (98%) generate noise at the M1 point (above the expansion joint) that is not much higher than the level at reference point M2. Only two passes resulted in an increase of more than 2 dB. Therefore, the device can be considered “low-noise”.

Table 2. Average exposure sound level at points M1, M2, and M3 for the finger device.

Parameter	Measuring Point			Differences
	M1	M2	M3	M1 − M2	M3 − M2	M3 − M1
Arithmetic mean LAE [dB]	89.1	87.9	73.0	1.2	−14.9	−16.1
Standard deviation [dB]	1.98	1.73	2.73	-	-	-

shows the average values of the sound exposure level for the finger device. It can be seen that, for 100 analyzed passes, the average L_AE above the device is 1.2 dB higher than the sound level at the reference cross-section. The level below the device is 14.9 dB lower than the level at the reference point and 16.1 dB lower than the level above the device.

Figure 11 presents examples of the amplitude–frequency characteristics of the sound level recorded during the passage of a passenger car.

It can be seen that the sound pressure level above the expansion joint (M1) is slightly higher in the range of 100 to 800 Hz and is close in the remaining range to the level at the reference point. Under the device, we can observe high sound attenuation practically in the entire range. Figure 12 shows the L_AE values depending on the speed of cars (it should be noted that several drivers exceed the speed limit of 40 km/h). Analyzing, as before, the range of 70 km/h to 140 km/h, it can be stated that the sound exposure level increases on average by about 4.1 dB at the M1 point, 3.2 dB at the M2 point, and 5.3 dB at the M3 point. As the speed increases, the volume of the device in this case decreases in relation to the L_AE of the road surface. It should also be noted that large L_AE values can occur randomly at lower speeds (e.g., 90 km/h). This occurs when the engine noise exceeds the noise generated at the contact point between the wheels on the surface and the expansion joint.

5. Conclusions

This article presents a methodology for testing the noise of bridge expansion joints and the results of the tests of two devices. The tests were carried out on the basis of the standard in [13], analyzing 100 passenger cars and vans for each device. Sound exposure levels for each crossing were measured, and the impact of expansion joints on noise was determined. The amplitude–frequency characteristics of the sample rides were presented, which allowed for the determination of the frequency range in which changes in sound characteristics occur. The impact of car speed on the noise of devices was also analyzed.

For the adopted criteria for the evaluation of devices, one of them (the finger device) was classified as “low-noise”. The average sound exposure level above the device was only 1.2 dB higher than the sound level at the reference cross-section. This device can be successfully used in urban areas. However, the modular device presented in this article should not be installed in such areas. The average sound exposure level above this device was 2.6 dB higher than the sound level at the reference cross-section. It should be noted that the presented results refer to specific devices, their technical condition, and the specified accuracy of the installation.

The discussed studies confirm that the evaluation method used can be employed to evaluate, compare, and verify expansion joint devices. This method can be the basis for the acceptance of new devices or the replacement of devices that do not meet requirements. Studies have shown that a small modular device (without soundproofing pads) can be significantly noisier than a finger device designed for large bridges. The research also showed that the speed limit will slightly reduce the nuisance associated with the passage of cars through expansion joint devices.