Noise barriers

CAV.To.Mi. campaign (2005)

Introduction

The new high speed railway line Torino-Novara, Italy, was opened in February 2006. About 100 000 m2 of acoustic barriers have been installed along the 86 km of the line, to protect the environment from the noise emitted by trains.

Barrier type Installed quantity Percentage
Metallic cassettes filled with glass wool 25 000 m2 25%
Timber and metal cassettes filled with glass wool 25 000 m2 25%
Concrete panels with a porous side 30 000 m2 30%
Acrylic sheets and concrete panels (reflective) 15 000 m2 15%
Others   5 000 m2   5%

In September 2005 the noise screens were almost all installed and the DIENCA Acoustics group was asked to verify their acoustic intrinsic characteristics: this means to measure the airborne sound insulation and reflection characteristics (which are different from the insertion loss) on site. The task was very demanding; in fact, it was requested to carry out the measurements:
- according to high quality technical standards;
- on site during the final phase of the construction work;
- ending in 7 days;
- working only in selected time slots (Saturday, Sunday, overnight after 22); for the rest of the time a special test train was active on the line.
The DIENCA Acoustics group decided to accept the challenge with some further specifications:
- the measurement method shall conform to the European technical specification CEN/TS 1793-5 [1], the sole technical guideline applicable on site;
- the test shall be applied to sample sections of the noise barriers, selected at random;
- the customer shall give the logistic support to move quickly inside and outside the railway line, to provide light spots during the nights, etc.

View of the railway line from inside    View of the railway line from inside    Overnight measurements

Procedure easyness to use

The equipment, conforming to CEN/TS 1793-5, proved to be easy to be moved inside the railway line. The logistic support of the customer was of great help to respect the tight time schedule: it was possible to place the equipment also close to barrier sections very difficult to be reached.

Placing the microphone grid for airborne sound insulation measurements outside a viaduct. Placing the microphone grid for airborne sound insulation measurements outside a concrete barrier.

According to CEN/TS 1793-5 the in situ value of the sound insulation index (SI) should be obtained averaging the values subsequently measured at nine points in front of the sample under test displacing the same microphone [1]. Due to the short time available for the measurements, it was decided to work with a 9-microphone grid in place of a single microphone. For the sound reflection index (RI) measurements the usual procedure was applied [1].

Microphone grid overnight. Microphone grid outside a viaduct. Reflection index measurements.

Results - Metallic cassettes barriers

The figure shows the curves of the Sabine’s reflection coefficient (measured in the laboratory) and the reflection index (measured in situ) obtained during the 2005 measurement campaign compared with the same curves obtained for metal/mineral wool cassettes from the same manufacturer in 2000.

Graph.

Reflection coefficient (in the laboratory) and reflection index (in situ) of the metallic cassettes filled with mineral wool barrier type. (a) in the laboratory, filled with rock wool; (b) in the laboratory, filled with glass wool; (c) in situ, filled with rock wool; (d) in situ, filled with glass wool.

The cassettes are identical except that in the year 2005 they were filled with glass wool while in the year 2000 they were filled with rock wool. As can be seen, the laboratory curves are very similar and get the same single number rating: 20 dB (maximum allowable). The curves measured in situ with a 5-year time interval show a reasonable resemblance above the 250 Hz 1/3 octave band and get a 1 dB difference between their single number ratings: 7 dB rock wool (2000) , 6 dB glass wool (2005). As discussed elsewhere [2,3], the in situ results are different from the laboratory results because of the differences in the test sound fields (diffuse in the laboratory, directional in situ), in the definition of the relevant quantity (the reflection index should not be confused with the reflection coefficient) and of the post processing methods used. It should also be noticed that the reflection coefficient in the laboratory, rS, has been obtained complementing to one the Sabine’s sound absorption coefficient αS; when, as a result of the Sabine’s approximation, αS has values exceeding 1, rS has been set to zero.

Formula.

According to CEN/TS 1793-5, the in situ measurements of the sound insulation index must be repeated in front of the acoustic elements and in front of a post. The figure shows the sound insulation index curves measured in front of the cassettes with a 5-year time interval: on average they are similar and get the same single number rating (31 dB).

Graph.

Sound insulation index measured in situ in front of three types of cassettes: (a) full metal filled with rock wool;(b) full metal filled with glass wool; (c) mixed timber/metal filled with glass wool.

The following figure shows the sound insulation index curves measured in front of a post: there is a clear shift in the 1/3 frequency bands values between 100 Hz and 1250 Hz; the single number rating difference is as large as 4 dB. This could be due to a poor sealing at the junction between panels and posts.

Graph.

Sound insulation index measured in situ in front of a post where the junction between acoustic and structural elements has a great influence: (a) full metal filled with rock wool;(b) full metal filled with glass wool; (c) mixed timber/metal filled with glass wool.

This assertion is supported by the results obtained for a different test section, where the cassettes, made by a another manufacturer, have a different design at the junction. The single number ratings are: 32 dB (2000) and 32 dB (2005) for the measurements in front of the acoustic elements; 33 dB (2000) and 32 dB (2005) for the measurements in front of a post.

Graph.

Sound insulation index values measured at a different site, where the metallic cassettes of a different manufacturer are installed, compared with the values obtained for the same products five years before.

Conclusions

The in situ verification of the acoustic intrinsic characteristics of the noise barriers installed along the high speed railway line Torino-Novara gave the opportunity to check the CEN/TS 1793-5 procedure on a large construction site, where the operating conditions are very different from the laboratory ones and when many measurements have to be done in a short time. The successful conclusion of the task is a strong argument in support to CEN/TS 1793-5.
The reliability of the measurements is supported by the comparison with the results of laboratory and in situ tests done five-six years before on similar noise barriers from the same manufacturers. The CEN/TS 1793-5 procedure for sound reflection revealed to be sensitive to small changes in the composition of the acoustic elements (e.g. rock wool replaced with glass wool) while the laboratory tests got the same single number ratings for metallic cassettes filled with rock wool, with glass wool and partly made with timber.
Airborne sound insulation measurements according to CEN/TS 1793-5 confirms once more the sensitivity of the method to detect small faults of the product design or the workmanship. It is important to distinguish between the differences in sound insulation due to the typical variance of construction works and the reproducibility of the measurement method. Doing this, the presented results suggest a difference between similar in situ results of 1 dB or less in the single number rating, a performance equal or better than that of the corresponding laboratory measurements. These conclusions holds for the whole set of tests, of which it was possible here to show only a small part.
Finally, the gained experience suggests that it is possible the repeated application of the CEN/TS 1793-5 measurement method over long time intervals in order to check the acoustic durability of traffic noise reducing devices, as recommended in prEN 14389-1.

Acknowledgments

The authors would like to acknowledge the effective logistic support received from the CAV.To.Mi. consortium during the measurement campaign.

Basic references

  1. European Technical Specification CEN/TS 1793-5:2003 Road traffic noise reducing devices – Test method for determining the acoustic performance — Part 5: Intrinsic characteristics - In situ values of sound reflection and airborne sound insulation, European Committee for Standardization, Brussels.
  2. Adrienne Research Team (1998) Test method for the acoustic performance of road traffic noise reducing devices, Final Report to the European Commission, DG XII – SMT Project MAT1-CT94049.
  3. GARAI M., GUIDORZI P. (2000) Experimental verification of the European methodology for testing noise barriers in situ: sound reflection, Proc. Inter-Noise 2000 (invited paper), Nice. (PDF, 158 KB)
  4. GARAI M., GUIDORZI P. (2000) Experimental verification of the European methodology for testing noise barriers in situ: airborne sound insulation, Proc. Inter-Noise 2000 (invited paper), Nice. (PDF, 99 KB)
  5. GARAI M., GUIDORZI P. (2000) "European methodology for testing the airborne sound insulation characteristics of noise barriers in situ: Experimental verification and comparison with laboratory data", J. Acoust. Soc. Am., 108(3), 1054-1067. (PDF, 446 KB)
  6. GARAI M., GUIDORZI P. (2008) "In situ measurements of the intrinsic characteristics of the acoustic barriers installed along a new high speed railway line", Noise Control Eng. J., 56(5), 342-355. (PDF, 940 KB)
  7. GARAI M. (2008) "Barriere acustiche", in: Manuale di acustica applicata, De Agostini Città Studi Ed., Torino.