Railway Noise Reduction at the Source Only.

Is this Possible or Mandatory?

Markus Hecht 1 Fachgebiet Schienenfahrzeuge; Institut für Land- und Seeverkehr, Fakultät V - Verkehrs- und Maschinensysteme, Technische Universität Berlin, Sekr. SG 14, Salzufer 17-19, D-10587 Berlin

ABSTRACT Climate change urges CO2 reduction also in traffic. Electrified Railways are a powerful CO 2 free and energy and space efficient traffic mode already today. Modal shift is demanded but environmental noise of railway hinder rail traffic growth significantly. Noise tunnels are not only expensive but also reduce energy efficiency and increase interior noise. Noise barriers are very costly and optical obstacles. TSI-Noise values have many criticisms. They do not cover the most bothering situations, curve running and severe other critical operation conditions are missing. The noise limit values are very high. Many standard measures in all other technical fields, construction machines or road vehicles do not exist in the railway context. Even worse is the situation of the track noise. There are nearly no measures except the demand for smooth surfaces. But the noise radiation of rails and sleepers is not fix but can be reduced significantly. Even good railway practice in other countries does not push the noise reduction without legislative demands. Example are quiet high speed pantographs and tamping machines in Japan. With general measures and low noise barriers and skirts on each track side the immission limit even at densely operated lines of L DEN 55 dBA is possible.

1. INTRODUCTION

The environmental friendliness of electrified railways is well established. The only problems are airborne and structure borne noise emissions. Structure born noise emissions are limited to the immediate vicinity and shall not be covered here further. The European noise limits for vehicles, the so called NOI-TSI [1] only avoid extreme noisy vehicles. Therefore the quieter routes [2] valid from 8th December 2024 will not lead to sufficient relief. As in Germany and Switzerland also today there are no more cast iron braked freight wagons or other cast iron braked rolling stock. Most European countries see noise barrieres as the main instrument to protect people from noise behind walls. First the negative aspects of noise barrier and the deficiencies of NOI TSI are mentioned. Afterwards effective noise reduction measures at tracks and vehicles are shown in brief.

1 Markus.hecht@tu-berlin.de

2. Noise barriers and noise tunnels

For passengers noise barriers have two disadvantages. One is no visibility, see figure 1

Figure 1: Noise barrier on the Berlin-Munich High speed line 2019, passenger vision from ICE-train, landscape visibility was in the past, foto author The second is interior noise increase. Most times this is in the 2 dB region and therefore not so dominant. If instead of a noise barrier a noise protection tunnel is foreseen, then the noise increase will reach 4 to 7 dB inside the vehicle according to the tunnel construction and the vehicle design. In any case more noise will disturb the passengers. Also the running resistance of the trains increases and by this more propulsion energy is needed. Up to now mainly the higher investment costs for tunnels than for noise barriers are discussed but not the further disadvantages.

Another topic is reliability. Railways are accepted much more if they are a reliable mode of transport and not operated by chance. Even in the condition of few snow but strong wind noise barriers are trapping the snow mainly in between the two walls. So the high speed line between Berlin an Munich was not continous for five days in February 2021 because of snow removal failed.

Figure 2: Effect of snowdrift in noise barriers

3. NOI TSI deficiencies

NOI TSI first was introduced in 2006, amended in 2011 and now valid is the version from 2014

The first version did not have very ambiguous noise limit values and it was intended to make them stricter similar to the pollutants of combustion engines installed in non-road machinery [3], valid from 1998. Today a 13 dB reduction in NOx and particles is valid compared to the first issue. But at NOI TSI the tightening process was very weak, only few dB or nothing, more or less better specification and somehow simplification of the measurement process was the progress. There are huge differences in the specified vehicle types. For instance locomotives may be very noisy though there is much more freedom for noise reduction measures than with Electric Multiple Units (EMUs). Lutzenberger and Gutmann [4] proved that the NOI TSI values are significant above the state of the art values. But presumably the noise limits will not be lowered in next future as this would need an approval of all European countries. Also the idea of NOI TSI to define simply measurable conditions and not all critical conditions. So curve noise is completely omitted though most complaints of local residents are in curve running conditions, not only with trams but also with mainline rolling stock. Disturbance through wheel squeal is not the only problem but low frequency grumble with significant noise-level increase compared to straight track running is very dependent of running gear design. Also the power up and power down process is not limited. But NOI TSI gives just the upper limits and minimum operation conditions. Tighter limits and additional relevant operating conditions may be added for the tendering process for new vehicles. But industry is against this because of increased effort in acceptance tests is needed then. But compared to the benefits this additional effort is marginal. Table 1 shows several improvements. The variety of vehicle types is reduced. Fuel cell drive and battery driven vehicles will occur in significant numbers soon and would confuse the limits further. For residents it is not of interest which type of propulsion system is used. The limiting noise levels are reduced to values which are achievable with some effort, 1 to 2 % price increase is estimated. This is not to revise NOI TSI, that would be an attempt without any hope for success. But it is a recommendation admitted for tender process inside the actual valid NOI TSI.

——> wind

Table 1: Recommendation for noise limits of Multiple Units in a tender process in comparison to the actual NOI TSI noise limits for Diesel Multiple Units (DMU) and Electric Multiple Units (EMU)

Also two very relevant other conditions are added: maximum level for power up an power down, as this is done during night time in depots which are often very close to residential areas. Additional the level increase for curve running is limited for tighter and medium tight curves. A level increase for curve running is inevitable because of wheel-rail interaction but can be limited by design measures to the low values stated in table 1. 4. Track Noise Contributions

1 up to 560 kW Power, above 83 dB

Another very neglected field of noise mitigation at the source is track noise, mainly noise radiated from rail and sleeper. Most railway infrastructure undertakings (RIU) do not have work instruction documents how to provide low noise situations. Rolling noise and impact noise are fields to be concerned. Just roughness limit control is done in some limited districts [5].

Regarding rolling noise the EN 3095 [6] from 2005 to 2013 in defined parameter levels for minimum and maximum rolling noise levels in Table E.1. Most times parameters for maximal rolling noise are used in practice, for instance the rail type or the pad stiffness or pad loss factors for the pad between rail and sleeper. Adding all the differences logarithmically gives a sum of 11,6 dB. For sure not always the worst situation occurs, but nevertheless this gives an idea of a huge potential for noise mitigation at the source. Impact noise very often occurs on the crossing nose of switches, figure 1 and can have many reasons.

DMU EMU — Recommendation Standstill LyaurldB) 7265 38 ‘Standstill + Compressor LyaigrldB] 76 68 60 Standstill + Air Release Lyarmn [4B] 8585 65 Power up / Power down Lyarmas (4B) - : 70 Starting + Braking Lysranc(@B] 82! 80 2 Pass by (S0km/h) Lyaxgr lB] 8180 14 Curve Running R270-350m + AL[aB] : 4 Curve Running R 380-500m + AL{@B] - 2

Figure 3: Possible impact sources at the crossing nose; 1 weld- or fishplate connections, 2 deteriorated nose, 3 impact by hollow worn wheels, 4 lost gravel support by crumbled stones A basic improvement are switches with movable crossing noses, Figure 3. They are much more expensive in investment, but have some benefit in maintenance and reliability, not only impact noise is reduced. But up to now switches with movable nose are only installed if high speeds on the diverging route are required, not because of noise reasons.

1 3 4 3 |

Figure 4 Switch with movable crossing nose at Deutsche Bahn, foto author 5. solutions

For light and medium loaded train lines the above mentioned measures will be enough.

For very densely operated lines further measures are needed. This are skirts on the vehicles and low noise barriers at each side of each track, figure 5. Because of the European structure gauge in the

lower part of the vehicle skirts cannot be as low as in the American profile an just help up to 2 dB reduction. Low noise barriers alone just give reductions in the 4 dB area. But together 15 dB can be achieved. Of course with the other measures on tracks and vehicles a L DEN 55 dBA is possible even at densely operated lines. With integrated railways it is hard to bring the vehicle and the track side together. But if there are clear interface descriptions in modern railway operation it should be possible. Noise related track prices with sufficient impact could support the process of noise reduction at the source.

Figure 5: 55 cm low-noise barrier and skirts at bogies of railfreight cars and locomotive SBB 460 at a test track in Brannenburg in Bavaria left, vehicles alone right, foto author 6. REFERENCES

1. N.N.; COMMISSION REGULATION (EU) No 1304/2014 of 26 November 2014, on the

technical specification for interoperability relating to the subsystem ‘rolling stock — noise’ 2. https://www.era.europa.eu/content/noise-tsi-quieter-routes_en 3. Directive 97/68/EC of the European Parliament and of the Council of 16 December 1997 on the

approximation of the laws of the Member States relating to measures against the emission of gaseous and particulate pollutants from internal combustion engines to be installed in non-road mobile machinery 4. Lutzenberger,S; Gutmann, Ch.; Ermittlung des Standes der Technik der Geräuschemissionen

europäischer Schienenfahrzeuge und deren Lärmminderungspotenziale mit Darstellung von Best- Practice-Beispielen, Umweltbundesamt, Dessau, UBA Texte Nr 12/2013 5. Rothhämel, J.; Schröder, S.; Koch, B.; Akustischer Fahrflächenzustand im netz der DB Netz AG;

ZEV Rail 2015 (Jahrgang 139), Ausgabe Sonderheft 2015 6. N.N.; Railway applications - Acoustics - Measurement of noise emitted by railbound vehicles

(ISO 3095:2005); German version EN ISO 3095:2005 7. Schindler, Ch.; Dellmann, T.; Hecht, M.; et al; Handbuch Schienenfahrzeuge - Entwicklung

Produktion Instandhaltung , 2014, DVV-Media Verlag, Hamburg, Akustik, p. 154-181 8. Frauenknecht, Christian R.; Erfolgsgeschichte einer Lokomotive, Appenzeller Verlag, Herisau,

2000