Noise from Railway Crossings

Transcription

1 Noise from Railway Crossings - Inter noise paper 2007 Danish Road Institute Report Ministry of Transport and Energy

2 Road Directorate Guldalderen 12 DK-2640 Hedehusene Denmark Telephone Telefax Roadinstitute.dk Title Noise from Railway Crossings - Inter noise paper 2007 Dated September 2007 Author Hans Bendtsen, Sigurd Thomsen Published by Road Directorate, Danish Road Institute Copyright Road Directorate, All rights reserved Photo Hans Bendtsen Print Electronic ISSN electronic ISBN electronic Reports published by the Danish Road Directorate can be requested from the bookshop: Telephone Telefax schultz@schultz.dk

3 Noise from Railway Crossings - Inter noise paper 2007 Hans Bendtsen Sigurd Thomsen Danish Road Institute Report Ministry of Transport and Energy

4 xx

5 Contents Preface... 5 Forord Introduction The Project Noise Results Measurement method and conditions Frequency spectra Statistical uncertainty Conclusion Acknowledgements References

6 4

7 Preface Paper presented at Inter-noise The design of the pavement near the rails as well as the joints between the pavement and the rails are important. Six railway crossings have been selected for noise measurements. The results showed that the angle at which the vehicles crossed the rails was of importance to the noise levels. Rail crossings at an angle of approximately 45 relatively to the road generally were less noisy than rail crossings at 90. The least noisy and the most noisy rail crossing gave rise to a vehicle noise level of 70.9 db and 78.5 db, respectively, a difference of 7.6 db. This highlights that the design of rail/tram crossings significantly affects the noise from road traffic. The surface texture at the rail crossings was also measured but no unique relation with the measured noise levels could be seen. 5

8 Forord Paper præsenteret på Inter.noise Konstruktionen af vejbelægningen ved en jernbaneoverskæring samt forbindelsen mellem vejbelægning og jernbanen er af betydning. Vejtrafikstøjen ved seks udvalgte jernbaneoverskæringer er blevet målt. Resultaterne viste at vinkelen som køretøjerne krydsede jernbaneoverskæringen havde betydning for vejtrafik støjniveauet. Jernbaneoverskæringer med en vinkel på omtrent 45 i forhold vejen var generelt mindre støjende end jernbaneoverskæringer med en vinkel på 90 i forhold til vejen. Vejtrafik støjniveauet ved den mindst og mest støjende jernbaneoverskæring blev målt til henholdsvis 70,9 db og 78,5 db, hvilket fremhæver at konstruktionen ved jernbaneoverskæringen har en betydelig indflydelse på vejtrafikstøjen. Overflade teksturen ved jernbaneoverskæringerne blev også målt, men ingen enestående sammenhæng mellem de målte støjniveauer og overflade tekstur kunne konstateres. 6

9 1. Introduction There are four main sources of noise where a tram, light-rail or railway crosses a road: The noise from the train or tram passing by. The noise from acoustical horns onboard trains to warn the vehicles and people travelling on the road that a train will soon be passing. The noise from the vehicles passing by on the road. The noise from acoustical warning systems (if there are any) prior to and during trains or trams crossing the road. The total environmental noise exposure next to railway crossings is the sum of the contributions from these four main noise sources. In this project the noise from the road traffic is in focus. When a vehicle approaches a railway crossing the driver might reduce speed before the crossing and accelerate after the passage of the crossing. This will have some effect on the road traffic noise. A B C B A G E D F F D E Figure 1. Different factors (marked by arrows in different collars) that are important for the generation of rolling noise when a vehicle passes a railway crossing. This is an example with rubber plates on both sides of the rails [3]. 7

10 It is mainly the tyre-road noise that will be influenced by the passage of a railway crossing because of changes in the pavement texture and structure (see Figure 1). The following is of importance: The type of pavement and its surface characteristics on the road crossing the railway. The type of material used on the outer sides of the rail (if any special material is used) and its surface characteristics. The type of material used between the rails (if any special material is used) and its surface characteristics. The joint between the pavement and the material used on the outer sides of the rail (if any special material is used). The outside joint between road material and the rail. Width, height differences and depth can influence the generation of rolling noise. The inside joint between road material and the rail. Width, height differences and depth can influence the generation of rolling noise. The angle between the joint and the driving direction of the vehicles can also influence the noise. As it can be seen a series of factors related to the design and construction of a railway crossing has an importance for the generation of rolling noise from the vehicles passing the railway crossing on the road. Increased noise levels may occur when the tyres roll across the rails. Tyre vibrations will be transmitted to the surrounding air as noise. Such vibration generated noise will typically be in the frequency range below Hz. Wear and tear of the pavement might also influence the noise. The foundation of a road construction and the foundation of a railway construction are often built up in different ways. At the same time the load by trains on the rails and by vehicles on the road are different. This can lead to different effects of tear and wear on the change of the height of the pavement and the rails over time. A typical result is that a difference in the height of the rail and the surrounding pavement develops over time and these height differences increases the rolling noise of the tyres as they pass such height differences. Therefore frequent maintenance of the railway crossings is of great importance in order to optimize the reduction of road traffic noise. In some cases the joints between the pavement material and the rails might cause increased noise levels every time a tyre hits the joint. This might generate impulse like noise which can cause a higher level of annoyance for the neighbours than the maximum noise levels reflects. In the case of impulse like noise it can be discussed whether to add a noise penalty to the measured noise levels in order to compensate for the increased annoyance. In the Danish administration of noise from factories etc. 5 db is added to the measured noise levels if the noise has an impulsive character. 8

11 2. The Project Different types and design of rail- and tramway crossings are used on urban roads. When cars are crossing the rail- or tramway the tyre-road noise may increase due to height discontinuities and joints between the pavement and the rail. The EU project SILENCE [1] started in February 2005 includes a Task F1 focusing on this subject. The objective was: To produce a ranking in relation to road traffic noise of commonly used existing types of rail/tramway crossings. To find solutions which are as silent as possible. In order to do so nearly all crossings between the lines of the North Zeeland local railways and roads have been inspected. On this background 6 railway crossings have been selected for noise measurements (See Figure 2): One railway crossing with asphalt concrete on both sides of the rails (Asphalt /asphalt-old-90 ). The pavement was very uneven and with holes caused by wear and tear. One railway crossing with concrete blocks between the rails (Old asphalt/ concrete-45 ). The concrete was worn down. Two railway crossings with asphalt concrete beside the rails and rubber plates between the rails. One crosses the road at an angle of 90 (New asphalt/rubber-90 ) the other at around 45 (New asphalt/rubber-45 ). Two railway crossings with rubber plates beside the rails and rubber plates between the rails. One crosses the road at an angle of 90 (New rubber/rubber-90 ) the other at around 45 (Rubber/rubber-new-45 ). The two first crossings were worn. Between the rails denoted Asphalt/asphalt-old-90 the surface was dense asphalt concrete (DAC8) stretching about 75 cm on both sides. The old crossing with concrete was constructed of 0.65 by 0.65 m concrete slabs. The remainder of crossings was constructed using rubber slabs (a product called STRAIL) placed between the rails and at two crossings before and after the track as well. An angle of 90 results in both wheels hitting the rail at approximately the same time while this is not the case at crossings with an angle of around 45. The roads leading to the crossings were (judged by the measurement team) paved with different types of dense asphalt concrete or the like with a maximum aggregate size of 8 to 11 mm. The pavements were more than two years old and they were all in a good condition. On this background it is estimated that the noise emission from vehicles on these pavements are very much alike, +/- 1 db. 9

12 Asphalt/asphalt-old-90 Rubber/rubber-new-45 Rubber/rubber-new-90 Asphalt/rubber-new-45 Asphalt/rubber-new-90 Asphalt/concrete-old-45 Figure 2. Photographs showing the measurement sites [3]. 10

13 3. Noise Results 3.1 Measurement method and conditions The road traffic noise has been measured by the Statistical Pass-By method (SPB) [2]. The maximum noise level LB pafmaxb as well as the third octave band spectrum and the vehicle speed have been measured. Only vehicles passing with a constant speed were accepted. Vehicles with defect silencers or the like that would influence the measured noise levels were omitted. The average air temperatures during measurements were 14 to 21 ºC. As the temperature variation is very little and as all the temperatures are close to 20 ºC no temperature correction has been carried out. The measured vehicle speed was corrected for the influence of the angle between radar and direction of travel. Table 1. Number of vehicles measured [3]. Rail crossing Passenger cars Dual-axle trucks/busses Multi-axle trucks/busses Asphalt/asphalt-old Asphalt/concrete-old Rubber/rubber-new Rubber/rubber-new Asphalt/rubber-new Asphalt/rubber-new Table 1 shows the number of vehicles that were included in each measurement series. The number of dual-axle trucks and busses included is limited and way below the requirements of at least 30 in the SPB standard [2]. Therefore results for this vehicle group have been included only for the two roads with at least 10 vehicle pass-bys, and these results are only indicative. There were so few multi-axle trucks and busses that these results have been omitted. Linear regression analyses of the measured maximum A-weighted noise levels on the logarithm to the base 10 of the corrected vehicle speeds have been carried out. The resulting regression lines for passenger cars have been combined in Figure 3 while the vehicle noise levels LB vehb at the reference speed are given in Table 2 The regression line representing the old asphalt/asphalt-90 railway crossing has been extended slightly outside the range of validity specified in [2] in order to estimate the LB vehb value at the reference speed of 50 km/h. 11

14 79 77 L pafmax, 7.5 m [db re 20µPa] Speed [km/h] Asphalt/asphalt-old-90 Rubber/rubber-new-90 Asphalt/rubber-new-90 Rubber/rubber-new-45 Asphalt/concrete-old-45 Asphalt/rubber-new-45 Figure 3. Maximum A-weighted noise levels (passenger cars) with time weighting F, at a distance of 7.5 m, as a function of speed. Reference speed 50 km/h [3]. By visual assessment of the crossings in Figure 2 one would expect higher noise levels at the old crossing asphalt/asphalt-90 and old asphalt/concrete-45, because of their condition. But this is only half-way true. The old crossing asphalt/asphalt-90 was the noisiest, but the noise levels at the old asphalt/concrete-45 were in the same range as at the new asphalt/rubber-45. These two crossings yielded the lowest noise levels. The three noisiest crossings had a track at an angle of approximately 90 relatively to the road vehicle direction of motion, whilst the three least noisy crossings had tracks at an angle of 53 to 55 (all denoted -45 crossings here). The main reason for the noise level difference is probably the oblique angle leading to a softer approach of the tyre to the discontinuity than at 90. The difference between the measured average maximum pass-by noise level at the noisiest and the least noisy crossing was 7.6 db. At the most silent rail crossing the noise level was 70.9 db which is nearly the same as at a 2 years old dense asphalt concrete with a maximum aggregate size of 11 mm. Table 2. LB pafmax 7.5 mb noise levels at 50 km/h, passenger cars [3]. Rail crossing Asphalt/asphalt- Rubber/rubbernew-90 Asphalt/rubber Rubber/rubber- Asphalt/concrete- Asphalt/rubber- old-90 -new-90 new-45 old-45 new-45 Noise level LB AFmaxB [db]

15 85 83 L pafmax, 7.5 m [db re 20µPa] Rubber/rubber-new-45 Rubber/rubber-new Speed [km/h] Figure 4. Maximum A-weighted noise levels (dual-axle vehicles) with time weighting F, at a distance of 7.5 m, as a function of speed. Reference speed 45 km/h [3]. Regression lines for dual-axle vehicles can be seen in Figure 4 and Table 3 shows noise levels at the reference speed of 45 km/h for dual-axle vehicles at two railway crossings. These results are indicative. The difference was 3.2 db and an impact of 90 was the noisiest as with the results for passenger cars. Table 3. LB pafmax 7.5 mb noise levels at 45 km/h, dual-axle vehicles [3]. Rail crossing Rubber/rubber-new-90 Rubber/rubber-new-45 Noise level L AFmax [db] Frequency spectra Results of 1/3-octave band frequency analysis are shown in Figure 5. The differences between different railway crossings were seen over the whole frequency range, but especially in the lower frequency regions indicating that the vibration noise is increased. The crossings with an angle of 90 also have a higher level in the high frequency domain over 1000 Hz. 13

16 [db(a)] Frequency [Hz] Asphalt/asphalt-old-90 Rubber/rubber-new-90 Asphalt/rubber-new-90 Rubber/rubber-new-45 Asphalt/concrete-old-45 Asphalt/rubber-new Figure 5. 1/3-octave band A-weighted average frequency spectra at 50 km/h [3]. 3.3 Statistical uncertainty The statistical uncertainties are expressed as half the width of a 95 % confidence interval and can be seen in Table 4 for passenger cars and Table 5 for dual-axle vehicles. Rail crossing Asphalt/asphaltold-90 Table 4. Statistical uncertainties passenger cars [3]. Asphalt/rubber -new-90 Asphalt/concreteold-45 Rubber/rubbernew-90 Rubber/rubbernew-45 Asphalt/rubbernew-45 Cars [db] Table 5. Statistical uncertainties dual-axle vehicles [3]. Rail crossing Rubber/rubber-new-90 Rubber/rubber-new-45 Dual-axle [db]

17 4. Conclusion On the background of this study the following tentative recommendations for the design of rail- and tramway crossings with as low a noise emission from passenger cars as possible can be derived: The noise is reduced by around 4 db when the angle between the road and the rails is around 45 instead of 90. Using rubber plates next to the rails reduces the road traffic noise in relation to using asphalt or cement concrete. Using rubber plates only between the rails seem to reduce the noise level by 1 or 2 db in relation to using rubber on both sides of the rails. This is difficult to explain and may be caused by features not noted. The noise levels from dual-axle heavy vehicles at two crossings, a new rubber/rubber at 90 and a new rubber/rubber at 45, showed the same trends as the noise levels from passenger cars. However, no conclusions can be drawn on heavy vehicle noise because at most of the investigated railway crossings too few truck pass-by noise levels were recorded. The surface texture at the rail crossings was also measured but no unique relation with the measured noise levels could be seen. The least noisy railway crossing (a new asphalt/rubber at 45 ) had the lowest peak to peak range in surface texture. 15

18 5. Acknowledgements This paper has been produced by the Danish Road Institute/Road Directorate (DRI) as part of the work [4] carried out by the Forum of European Highway Research Laboratories (FEHRL) in Task F1 of the EU project SILENCE in cooperation with Ulf Sandberg and Michael Ögren from Swedish National Road and Transport Research Institute (VTI). The Maintenance Division of the Danish Road Directorate has helped by suggesting roads with different types of railway crossings. 16

19 6. References [1] Silence homepage: [2] ISO : "Acoustics Method for measuring the influence of road surfaces on road traffic noise Part 1: The statistical Pass-by-method" [3] Railway crossings Road traffic noise measurements. Road Directorate/ Danish Road Institute. Technical Note 49, ( [4] Reduction of noise from vehicles crossing discontinuities on urban streets. Deliverable F.D2 SILENCE, EU sixth framework programme. 17

20 xx

21 Rapport / Report Nr. Titel/Title/Shortcut No. 136 European co-operation in COST Bringing ALT activities closer together Forfatter/Author Gregers Hildebrand Michael E. Nunn 137 Traffic management and noise reducing pavements - Recommendations on additional noise reducing measures 138 Mechanistic Design of Semi-Rigid Pavements - An Incremental Approach Hans Bendtsen Jürgen Haberl, Johan Litzka Ernst Pucher Ulf Sandberg Greg Watts Finn Thøgersen Christian Busch Anders Henrichsen 139 Holdbarhed af Drænasfalt Carsten Bredahl Nielsen Jørn Raaberg Erik Nielsen 140 Indbygning af skrivekridt - Et fuldskalaforsøg 141 Noise reducing pavements - State of the art in Denmark 142 Economic assessment of traffic noise in planning Danish experiences 143 Organising urban noise abatement - New ideas 144 Two-layer porous asphalt - for urban roads 145 Thin noise reducing pavements - Experiences Sten Thorsen Poul Panduro Knud A. Pihl Hans Bendtsen Bent Andersen Lars Ellebjerg Hans Bendtsen Hans Bendtsen Lene Nøhr Michelsen Brian Kristensen Hans Bendtsen Bent Andersen Jørn Råberg Lars Ellebjerg Jørgen Kragh Hans Bendtsen Bent Andersen 146 Cost-benefit analysis on noise-reducing pavements Lars Ellebjerg 147 Traffic management and noise - INTER-NOISE 2006 Hans Bendtsen Lars Ellebjerg 148 Noise reducing thin layers for highways - INTER-NOISE 2006 Hans Bendtsen Sigurd N. Thomsen 149 Noise reducing thin pavements urban roads Sigurd N. Thomsen Hans Bendtsen Bent Andersen 150 Integration of noise in PM Systems - Pavement Management and noise 151 Noise Control through Traffic Flow Measures - Effects and Benefits Hans Bendtsen Bjarne Shmidt Lars Ellebjerg 152 Noise from Railway Crossings Inter Noise Paper 2007 Hans Bendtsen Sigurd N. Thomsen

22 Road Directorate Niels Juels Gade 13 P.O. Box 9018 DK-1022 Copenhagen K Denmark Telephone Telefax Road Directorate Guldalderen 12 DK-2640 Hedehusene Denmark Telephone Telefax Road Directorate Thomas Helsteds Vej 11 P.O. Box 529 DK-8660 Skanderborg Denmark Telephone Telefax vd@vd.dk Roadinstitute.dk