Technical aspect of road traffic noise mitigation measures

Transcription

1 Technical aspect of road traffic noise mitigation measures M+P.MVM revision 2 19 December 2005

2 title: Technical aspect of road traffic noise mitigation measures report number: M+P.MVM revision: 2 date: 19 December 2005 prepared for: Netherlands Ministry of Housing, Spatial planning and Environment. Local Environmental Quality and Transport Directorate The Hague purchase order: authors: Gijsjan van Blokland gijsjanvanblokland@mp.nl Erik de Graaff erikdegraaff@mp.nl Summary This report addresses technical, regulatory and economical aspects of noise control of road traffic. The issue is approached at all levels, ranging from control of the elements that defines the source, the usage of the source, the propagation and the reception of road traffic noise. Special attention is given to the control of the source of the rolling noise and the propulsion nois. The regulatory aspects are treated in more detail and the effects on both L den levels and on annoyance is given for three major European cities. Also the effects of the proposed noise control schemes on air quality are treated. Keywords: road traffic noise, regulations, measures, economy, air quality. 1 M+P Raadgevende ingenieurs bv Postbus CB Vught The Netherlands T F Vught@mp.nl Division of the Müller-BBM group 2005 / M+P Raadgevende ingenieurs bv. No part of this publication may be used for purposes other than agreed upon by client and M+P Raadgevende ingenieurs bv (DNR 2005 Art. 46). M+P.MVM03.2.1; revision 2 1

3 CONTENTS 1 INTRODUCTION 3 2 GENERAL SCHEME OF TRANSPORTATION NOISE MITIGATION 4 3 CHAIN APPROACH FOR ROAD TRANSPORT Source properties vehicles Tyres Road surfacing Usage of the source Infrastructure and propagation from source to receiver Road and town planning Building planning 22 4 EFFECT OF MITIGATION MEASURES 24 5 EFFECTS ON AIR QUALITY General Measures on the source Measures on the usage of the source Modal split Measures on propagation Measures on town planning and road planning Summary 27 6 REFERENCES 28 7 ANNEX 1: REGULATIONS AND LEGISLATIONS International regulation for road vehicles Motor vehicles with four or more wheels Motorcycles Regulation for tyres for road vehicles and its trailers Regulation for road surfaces 36 8 ANNEX 2: NOISE MAPS OF 3 EUROPEAN CITIES Introduction Noise emission model Noise exposure Noise maps Definition of scenarios Results Discussion and conclusions References 46 M+P.MVM03.2.1; revision 2 2

4 1 INTRODUCTION The EU working group on Health and Socio-Economic Aspects has been given the task to support and to evaluate the work of the Lärmkontor consortium working on an extensive study on the effectiveness of noise mitigation measures, referred to as the EffNoise study (ref. [1]). The Netherlands Ministry of Housing, Spatial planning and the Environment (VROM) has given M+P the task of evaluating the results of the EffNoise study and interpret them in a broader sense. It was concluded that the main challenge lies in the field of road traffic noise and therefore this study has focused to that item. This report is referenced in the Position Paper of the WG-HSEA on the subject of noise mitigation measures. This work has been performed parallel and in coordination with a study of KPMG on cost issues of measures on road traffic noise reported in [figure 147]. This task was interpreted in a broader sense as to evaluate community measures on noise, to formulate a position on this subject and to recommend high priority actions to the commission. This topic falls within the framework of the European Noise Directive, article 10, 1st part. The authors acknowledge the input from members of the WG-HSEA for this report.. M+P.MVM03.2.1; revision 2 3

5 2 GENERAL SCHEME OF TRANSPORTATION NOISE MITIGATION The topic of noise annoyance due to transportation activities is approached by investigating and mapping the total chain comprising source technology at one end and noise sensitivity of the human at the other end. The chain can then be broken down to its links and each link can then be studied with respect to its properties, the processes and parameters affecting the properties, the relation with adjoining links and possibly even relations with adjacent chains. A framework of this approach is given in the figure below (see figure 1). Shift between modes of transport Combined effect with noise from other sources Source properties usage of the source Propagation from source environment receiver 1.Noise abatement technology 2.Marketing aspects 3.Type approval regulations 4.Composition of vehicle fleet 5.Road surface choice 6.Tyre mounting 7.Local vehicle regulations 1.Number of vehicles 2.Speed 3.Driving behavior 4.Taxation 5.Curfews 1.Noise barriers 2.Building planning alongside infrastructure 1.Town planning 2.Housing density 3.Existence of quiet areas 1.Day/evening/night averaging 2.Frequency/time content 3. Existence of quiet areas 4.Dose-effect relations 5.Façade insulation Mitigation measures Non-acoustical effects of measures figure 1 scheme of chain approach to evaluate mitigation measures Mitigation measures can be grouped into technical measures and behavioral measures. The first group addresses the physics of generation and propagation of noise energy, the second group affects behavior of actors involved in decision on usage of vehicles/transportation means that emit noise. table I overview of technical and behavioural measures for noise mitigation. technical measures behavioral measures source technology management of source measures on propagation measures at receiver position legal instruments economic instruments social instruments Each of these measures can be or shall be activated by different administrative levels, such as there are the European Union (or supra-eu organizations such M+P.MVM03.2.1; revision 2 4

6 as UN), national, regional and local levels, by NGO s and last-but-not-least by manufacturers. In order to be able to evaluate technical and behavioral measures, a general concept of their impact is necessary. The subjective experience of noise (as nearly all sensory inputs) follows a logarithmic relation, which means that a difference is related to a relative change in the signal. A difference of + or - 1 db represents an increase or decrease in source strength energy with about 25%. A halving (or doubling) of number of energy leads to a level change of 3 db. For a subjective halving or doubling a 10 db difference (10 fold change in energy) is required. That leads to the following coarse relations:. 1. measures that control the intensity of noise sources have to be very effective to generate a noticeable change in the noise level. Traffic volume has to be reduced by more than 25% to obtain the just perceivable effect of 1 db and halving the perceived level implies an intensity reduction with a factor of 10; 2. A moving vehicle on a road, railway or along an aircraft fly path judged on base of its average noise level, The acoustic representation for this is a row of single sources together forming a line source. The noise lowering due to distance from a line source also follows the rule stated above, meaning that a 25% change in distance leads to a 1 db effect. 3. introduction of low noise technology in a population of vehicles becomes only effective after a significant fraction of the population is affected. For instance, when 25% of the tyre population is of a 3 db lower noise type, the average noise level drops with less then 0,5 db. 4. A similar mechanism generally imposes limits to the effect of noise due to interactions such as tyre/road and wheel/rail noise. Only when both components are of the low noise type, significant effects can be achieved. These fundamental relations have to be taken into account in evaluating measures and its effect on noise levels and already point into the direction of coordinated measures that act on the source. M+P.MVM03.2.1; revision 2 5

7 3 CHAIN APPROACH FOR ROAD TRANSPORT 3.1 Source properties The total noise produced by a road vehicle consists of two main sources: 1. rolling noise, coming from the tyre/road interaction process; 2. power train noise, originating from the drive train of the vehicle. Aerodynamic noise is only relevant at extreme speeds and then probably only in the non A-weighting relevant- low and high frequency ranges. For nearly all normal occurring situation, this noise type can be neglected. The properties of the two sources of vehicle noise are defined by the following three components: 1. the tyre, 2. the road and 3. the power train Rolling noise Rolling noise is a direct result of the dynamic process occurring in the tyre/road contact patch. Due to the force variations working on the tyre belt, the tyre body is excited and the resulting bending waves cause noise radiation into the air. In this radiation process significant amplification is caused by the horn-like geometry found between the tyre belt and the road surface. In addition secondary processes such as air-pumping, stick-slip and stick-snap generate extra contributions, mainly on low-textured roads. The magnitude of the mechanical processes is defined by the rolling speed of the tyre and the geometry of irregularities in the contact patch between tyre and road, composed of the tread profile and the road surface texture. Variations in the mechanical properties of the tyre carcass are found to be less important. The propagation of the radiated sound, amplified by the horn-process, is significantly affected by the sound absorptive properties of the road surface Power train noise The power train excites noise through the gas exchange process in intake and exhaust, by the radiation of the sound due to the mechanical vibrations of the power train structure, that results from the mechanical processes in the drive line and by aerodynamic processes in the cooling fan of the engine and generator. Power train noise from Heavy Duty Vehicles is significantly louder than noise from passenger cars. This can be explained by the higher mechanical power, the lesser shielding and the less comfort oriented design of these vehicles. Within a certain vehicle class, the main parameters that affect the sound production is the engine speed. Secondary are engine load and engine type (Diesel or Otto). Consumer and legislative forces have resulted in significant reduction of these sources over the last twenty years and at this moment in frequently used driving conditions the rolling noise dominates over the power train noise in terms of A-weighted levels. Exceptions on this general rule are powered two wheelers, M+P.MVM03.2.1; revision 2 6

8 some type of sport cars, modified standard vehicles and HDV s under city driving conditions. The contribution of power train noise to the overall noise level of a vehicle is given in the figures below (see figure 2) L W [db(a)] Passenger Cars Heavy Duty Vehicles Medium Duty Vehicles Speed [v (km/h)] % of rolling noise Passenger Cars Heavy Duty Vehicles Speed [km/h] figure 2 Left: total sound power level of a typical passenger car and a typical medium and heavy duty vehicle as a function of speed, right: fraction of rolling noise in total level of vehicle. 3.2 vehicles Passenger Cars technology The technological properties of road vehicles are the combined result of general technological level of our society, the requirements put on the vehicles by regulating authorities and the marketing processes that is mainly steered by consumer demands. Application of advanced technologies is strongly influenced by costs and by reliability. Regulatory constraints are mainly seen as boundary conditions for design and in general are met just marginally. Since consumer demands generate a strong economical force, car design and engineering is to a large extend steered by them. Passenger cars are chosen and operated in general not very rational and emotional aspects become relevant, such as image, status, subjective quality, etcetera (see for instance ref. [5.] The present day passenger car noise characteristics are: Low interior noise levels. Preference for acoustic frequencies related to engine firing frequency. Comfort, durability and engine feedback are the key emotions related to interior noise. Exterior noise levels maximized by legislation, but mainly defined by the emotional positioning of the vehicle in the market (ref. [6]). Tyre noise optimized for its contribution to the interior level, exterior levels tuned to their contribution during type approval and then only for OEM tyres. M+P.MVM03.2.1; revision 2 7

9 economy The economical aspects of low noise vehicle technology are complicated. We distinguish two parts: 1 the cost of application of noise reducing technology 2 the cost/effectiveness of low noise vehicles. Low noise technology, effectuated by additional measures may add to manufacturers costs of the vehicle, because of the additional parts and treatments and due to the user costs because of additional weight. The overnight abatement costs (reducing the noise emission of an existing vehicle type) can be substantial. However once conceived in the engineering stage, the costs are estimated not to exceed 100 /vehicle/db for passenger cars (ref: [11, 12]). In fact, for cars, there exists no correlation between the exterior noise level (as measured under type approval conditions) and the selling price of the vehicle (ref. [7]). For the general noise emission experienced under normal traffic conditions, the tyre is the dominating part and thus vehicle technology only minor affects this part. The tyre will be addressed in paragraph 3.3. The complicating factor comes from the role that the acoustic emission plays within the imago of a product. Low noise technology interferes with this, since it implies constraints in the freedom of design, may affect marketing and therefore may lead to shifts in consumer choice. On a macroscopic scale this represents no cost factor, since it will not increase the total expenditures with respect to vehicles (it may even bring the consumer to buy lower priced cars) but on the level of an individual manufacturer it may affect sales. The cost/benefit ratio of vehicle measures to reduce the equivalent sound emission, based on current technology is positive on a macro-economic scale but can be significantly improved if combined with other measures regarding low noise tyres and/or low noise road surface. Several studies were made on this subject. See for instance references [10] and [11]. The effectiveness for reducing specific annoyance is larger then that for nonspecific annoyance [9]). regulation Cars are subjected to noise regulations by the EU (70/157) and the (identical) ECE (R51) that evaluate the car under a condition of moderate car speed and high engine speed and engine load. This condition represents a maximum normal usage of the vehicle, under which power train noise dominates. The limit values were reduced with about 8 db over the last 30 years. The effectiveness for the total noise emission of the vehicle is smaller since: general urban driving also comprises a large amount of rolling noise; the limit values were not technology forcing. Only the last sharpening has actually led to significant shift in the noise characteristics of the type approval values. At present a new type approval scheme is discussed within the EU and the ECE, in which the operating condition of the vehicle is shifted to lower engine M+P.MVM03.2.1; revision 2 8

10 speed and load. Although being more representative for the actual driving condition, it lacks regulatory power since the noise characteristics of the tyres, that become more relevant for the type approval result, are not controlled for that vehicle in the after market, the same way as silencers are. The tyre noise directive will not be adequate to fill this gap. The expected effect of the new developments in UN/ECE must therefore not be over-estimated (see chapter 7 for more background information on regulations) Heavy Duty Vehicles Technology and economy Driveline noise of HDV s contributes significant to the overall noise level in urban areas. Consequently reductions do result in lowering of L den values and even more in reduction of specific annoyance. This is corroborated by the investigation of noise levels in three major European cities. (see chapters 4 and 8 for a summary). Technological applications for HDV s are solely steered by economic ratios and by requirements of regulations. Since the regulations for the control of drive train noise emission of HDV s has been effective in reducing power train noise over the last 20 years (see chapter 7) already several measures have been applied. Initially as additional measures, in a next stage incorporated in the design of the vehicle. Apart from technology forcing regulations, the effectiveness can also be explained by the fact that noise reduction is a spin-off of measures on combustion processes to reduce NO X (such as pilot fuel injection and turbo-charging) and to improve fuel-efficiency (turbo-charging and low engine speed/high engine torque concepts). Because vehicle technology is constantly improving, one expects that additional noise reducing potential has become available since the last directive change (92/97) of Illustration for this is the removing of underside shielding of HDV s as a consequence of low noise engine concepts introduced in the vehicles since then. Re-installing of underside shielding will result in additional reduction of the driveline sound production. The costs of this is estimated to be about 1000 /vehicle. The Eff-Noise study presented figures for 6 db noise reduction. Such a decrease requires severe measures and will therefore be considerably more expensive. regulation HDV type approval schemes test the vehicle under conditions of moderate vehicle speed and high engine speed and engine load, conditions under which the power train in general dominates the noise emission. Because at the present situation the vehicle is unloaded, high powered types operate in very transient conditions, that are not stable and furthermore produce excessive stick-slip noise in the tyre contact area. A new test was therefore developed within WG42 of ISO TC43 and TC22 that comprises a more stable and a more M+P.MVM03.2.1; revision 2 9

11 representative test situation. (see chapter 7 for more information on regulations) Powered Two-Wheelers A special case are mopeds and motorcycles. Noise reduction of these vehicles is not related to the costs per vehicle. On the contrary, the noisy exhausts are non-standard and a noisy vehicle is more expensive then a silent one, leading to a very good cost/effect ratio when the wide application of these devices is prevented. In this case the costs involved relate to the enforcement. It is estimated (see ref. [11]) that effective control of IRESS (illegal replacement exhaust silencing systems) could reduce the overall sound production with about 5 db and specific annoyance most probably even more. Effective enforcement requires a few minor adaptations of the vehicle (electronic device to control engine speed) and clear and unambiguous E-marking of the OEM silencers. The road side check can then be done easily and only requires a measuring system and capacity of enforcement officers. The investment per checked vehicle is estimated to be less then 5 and is estimated to require about 0,1 man hour (NL experiences ref [14].). The cost/abatement ratio will be further improved by the fact that a higher level of enforcement leads to a drastic refitting of silencers by original ones. A clear example can be found in Greece, presented in the figure below. Percentage of non permissible emissions Control check months (2-year period : April March 1998) figure 3 Monthly variations of the Athens Noise Control Roadside Check results (percentage of motorcycles with non permissible noise emissions in the total sample (ref. [15]). 3.3 Tyres technical aspects The paradoxical situation exists that, although the noise characteristics of tyres dominate the total noise production of the road vehicle, the interest from both legislation institutions and consumer to this aspect is very small. International type approval schemes exhibit ineffective regulations, both with respect to the present limit values and with respect to the tightening that is scheduled (see paragraph 7.4.4). M+P.MVM03.2.1; revision 2 10

12 The consumer, not being informed about the external noise properties of the tyre, normally makes his choice on price, design and possibly rolling resistance information. Strong lead here is the tyre type originally mounted on the vehicle. This explains the interest of tyre manufacturers to become OEM supplier. The present marketing trends towards wider tyres, run flat capabilities and higher speed indices are counter productive to low noise and possibly also other properties such as aquaplaning and comfort. Two extensive studies have demonstrated that there exists no correlation between rolling resistance, wet braking capabilities and noise production of car tyres (ref.16). This would imply that environmental and safety aspects are not in jeopardy when tyres are noise optimized, at least for another 2 to 3 db s. Its obvious that design of extremely low noise tyres will imply another balance between all the properties a tyre has to fulfill. rolling resistance 1,4 1,2 1 0, L [db(a)] braking distance [m] L [db(a)] figure 4 relation between rolling noise level and safety and economy features of passenger car tyres. Source [16]. Major steps are possible when break through technology, such as the composite wheel-design becomes available. It has been demonstrated that such tyres exhibit a significant lower levels (ref [17] and [18] report 10 db lower values compared to market average levels). The figure below gives a recent example of such a tyre. figure 5 Break through technology for tyres (source Such airless systems have been proven to be extremely low noise (ref. [17, 18]). M+P.MVM03.2.1; revision 2 11

13 3.3.2 Economy The tyre manufacturing business operates in a very competitive environment. The design of the tyre is therefore mainly steered by marketing issues. Design constraints come from regulations on safety and noise and from manufacturers liability. In this balancing of properties, exterior rolling noise properties are defined by OEM requirements and by the present regulation. The cost of reducing tyre noise can in a first approach be set to zero on a macro scale (corroborated by the lack of correlation between price and rolling noise level of tyres), although an individual manufacturer may suffer from lower sales. For HDV tyres, it might be that improving noise characteristics lead to a slight reduction of durability, but any relation between noise and mileage cannot be based on existing evidence, especially since external noise has only very recently become one of the properties to be looked at (ref. [16]) In terms of cost/abatement low noise tyres also rank high (see ref. [10], [11] and [7]). 3.4 Road surfacing Technical aspects table II The road surface is in many of the cases the dominating factor in the process of noise generation by road vehicles. Optimized texture of the surface suppresses vibrational excitation of the tyre belt, while maintaining good grip between tyre and surface. Additionally acoustic absorption reduces the noise amplification due to the horn effect and lowers the propagation from propulsion noise from the vehicle into the environment. Three classes of low noise surfacings can be distinguished. Each type affects tyre/road noise generating mechanisms in a different way and therefore exhibits different effects for different vehicle categories in different speed ranges. Also the application field of the different technologies can be very specific, due to the surface properties in the field of skidding resistance and durability to cornering heavy duty vehicles (one of the most wear affecting factors for porous surfaces). overview of low noise road surface types and the effect for cars and HDV s at three speed ranges. road surface type effect for small cars effect for HDV s smooth textured dense surface 0 to -2 db 0 to +2 db smooth textured thin semi porous surfacing -2 to -4 db 0 to -2 db medium textured, thick porous surfacing -3 to -6 db -3 to -6 db M+P.MVM03.2.1; revision 2 12

14 3.4.2 economy Road surface measures are very effective in terms of road noise reduction and exhibit in most cases better cost/benefit ratios traditional measures such as noise screens and sound insulation (refer to section 3.6.1). The costs involved are generally attributed to the higher maintenance costs which are required to preserve a good quality of both the surface texture and the acoustic absorption of the surface. The cost/benefit ratio of low noise road surfacings is made in relation with other measures on screens and façade insulation and will therefore be treated in the section on infrastructure (see section 3.6). 3.5 Usage of the road vehicle Speed reduction Speed reduction has the advantage of being a simple way to reduce noise (in general with about 2 db per 10 km/h speed change) but requires enforcement and street design changes in order to make it durable. Additionally indirect costs are involved in the increase of traveling time. Compared to the in general very high costs of additional façade insulation, it will exhibit a positive cost/benefit ratio. There may be differences between urban areas and nonurban areas Drive style figure 6 Driving states of a road vehicle in urban areas While design and technical state of the vehicle defines its acoustic characteristics, the actual usage of a road vehicle determines its noise production. The main parameter in this is the vehicle speed and acceleration. Speed and noise levels of tyres and power train recorded during an urban drive cycle. (vehicle VW Golf IV, ref. [20] velocity 100 V [km/h] time [s] Lp7.5m [db(a)] (calculated) tyre engine total time [s] M+P.MVM03.2.1; revision 2 13

15 Speed defines the level of rolling noise. Speed together with acceleration determines the level of the engine noise. Acceleration because of the increase in engine load and because of the down shifting behavior of the gearbox. The figure 6 presents the noise level of a vehicle during urban driving. The noise map of passenger cars (presented in 22 and an example given in figure below, see figure 7), corrobarates that speed is the dominating parameter for the total noise production and therefore the traffic situation and road type determines the vehicle emission. Acceleration, that affects engine speed and engine load, accounts for the vertical spread in the graph. figure 7 Noise map of a passenger car when driving in an urban environment. The figure presents the chance of occurance of noise/speed events. Only the situations with positive acceleration lasting longer then 2 s are taken (ref [22]). Both the noise production of the engine and the tyre follow a roughly similar relation of 35 log(v). The mechanical (torque verse engine speed) characteristic of the engine in relation to the gear box and the vehicle weight defines the shifting pattern and therefore the engine speed/driving speed relation. Low torque to mass vehicles will run at relative high engine speeds, while high torque (especially when this is available at low engine speed) to mass ratio vehicles operate mainly in the lower engine speed range. % of nominal engine speed normal accelerated urban driving at 50 km/h engine speed at single events Power to Mass ratio [KW/T] figure 8 typical engine speeds during accelerating conditions found at urban driving around 50 and during single events as a function of the vehicle power to mass ratio. Source: [22]. This is indicated in the figure 8 in which measurement results of several passenger cars in urban traffic were analyzed in terms of engine speed during M+P.MVM03.2.1; revision 2 14

16 the accelerating phase at around 50 km/h. At lower speeds the engine speed curve will be higher. Additionally a curve is given for single events meaning acceleration events that occur less often. Control of the passenger car sound production in urban situations therefore lies mainly in control of the speed and limiting maximal acceleration. The first can be controlled by speed limits (and enforcement), the latter is more a drivers attitude issue and reflects cultural differences. For instance Japanese drivers operate at lower engine speeds and lower accelerations then European drivers. At present there is renewed interest in fuel-saving driving that mainly addresses control of vehicle speed and engine speed. Such driving will also imply lower noise production Traffic situations, traffic flow and traffic intensity The speed and acceleration values are mainly defined by the traffic situation and in a lesser, but still relevant degree, by the driver attitude (see part ). table III average speeds in typical EU street situations. (ref. [1]). Road type / traffic situation speed [km/h] Standstill [%] Residential street < 30 km./h 29 9 Residential street <50 km/h 39 6 City centre Very busy main streets with traffic lights <50 km/h Busy main streets with traffic lights <50 km/h Normal main streets with traffic lights <50 km/h Main streets with right of way <50 km/h 46 3 Main streets with right of way >50 km/h 56 5 Rural streets <70 km/h 63 1 Rural streets >70 km/h 81 3 Urban motorway 93 0 Traffic calming has an effect when it affects the average speed and is able to reduce peak speeds. Unfortunately traffic calming improves also the flow of traffic that leads to higher speeds and thus jeopardizes the noise effect. It will however surely be beneficial for air quality (see section 5.3). The spectral composition of the noise however will also change considerably, with a higher contribution of low frequency, exhaust oriented, noise, which has better propagation into neighboring areas and inside dwellings (see part 3.8) The effect of control of traffic intensity is small. In order to achieve a 3 db reduction, intensity must be lowered with 50%. Limiting the number of heavy vehicles can however be very effective since the same effect of -3 db can be obtained by phasing out the 5% share of heavy vehicles in a normal main street with traffic lights. M+P.MVM03.2.1; revision 2 15

17 3.5.4 Drivers attitude The driver attitude influences this behavior by its own gear shifting strategy, the setting of the automatic gearbox and the usage of the accelerator that can cause down shift. One can distinguish between quiet disciplined driving up to aggressive and obtrusive driving. This attitude will not significantly affect the average speed, since this is mainly defined by the traffic situation, but will affect the average engine speed and peaks in the vehicle speed and even more important, the vehicle acceleration. table IV Typology of car drivers in France [4]. Driving style Percentage of drivers Quiet, disciplined and fuel saving drivers 26% Rather quiet and slightly fuel saving drivers 23% Fast and anticipating drivers 23% Sportive and no fuel saving drivers 15% Aggressive and obtrusive drivers 8% Others 5% In another study the effect of driving attitude on the maximal acceleration was studied (see figure 9). acceleration [m/s2] 3 2,5 2 1,5 1 0,5 0 normal urban hectical urban Power to Mass ratio [KW/t] figure 9 acceleration as a function of power to mass ratio in two driving styles (see [4]). Technical aspects of drive style changes The type of usage of the road vehicle affects the noise production significantly. Since both engine speed and vehicle are decisive parameters in the noise production of the vehicle, fast driving and fast acceleration may increase the equivalent level with 3 to 10 db. This is illustrated in figure 10 that gives noise vs. speed curves for both normal and aggressive driving for HDV, small cars and motor cycles (source IMMA, ref.[19]). The increase, normalized to the vehicle speed is mainly caused by the higher level of propulsion noise. In addition to this comes the effect of higher vehicle speeds, experienced during aggressive driving, which can add another 2 to 3 db to the noise production level. M+P.MVM03.2.1; revision 2 16

18 figure 10 effect of driving style to noise production. noise level, in terems of L Amax at 7.5 m vs. vehicle speed for PTW (Powered Two Wheelers), small car and HDV s (Heavy Duty Vehicles) for both normal driving (line) and aggressive driving (dots), ref [19]. Economical effects of drive style changes The economical effects of a noise minimizing style of driving are clearly positive. A careful adaptive driving style saves fuel, reduces engine and tyre wear, reduces exhaust emission and leads to less accidents. All are very relevant cost factors in vehicle usage. The counter-effect of longer traveling time, was several times demonstrated to be surprisingly minor. This has for instance lead to the installation of speed reducing devices in commercial vehicles long before they became mandatory. Apart from in-vehicle control and changes of the drivers attitude, also external speed control will reduce vehicle speed. Since the majority of the drivers meet the speed limits, the effect on the equivalent level is generally limited, but it reduces peaks in vehicle speed, and therefore in the number of single events Modal shift Freight transport Modal shift from road to rail does in general not improve the environmental situation. The noise production per unit freight is about 5 db higher (barely compensated by the railway bonus) and due to the lower requirements on exhaust emission from diesel locomotives also air quality is not improved. CO 2 emission will be smaller due to the lower rolling and air resistance. Furthermore through roads are planned around cities while rail lines frequently cross urban areas resulting in higher exposure of the population. Shift to transport over water leads to very positive effects, since the noise production by a unit of freight is more then 20 db lower then by road. The cost transfer involved in modal shift of freight however can be significant. Cost elements are: M+P.MVM03.2.1; revision 2 17

19 Extra costs of handling in multi-modal terminals; Extra costs of time delay because of either choice of the slower means of transportation, or the time involved in modal shifting; In case of transport by rail and water, savings are obtained by decreased staff costs (a small crew can handle up to 300 TEU s in a ship and about 50 TEU in a train while a road vehicle carries on average 1,5 TEU), and by the lower specific fuel consumption. figure 11 Inland shipping system carrying about 100 TEU (equivalent to about 60 HDV s). The cost/benefit ratio is very much affected by the costs of time delay and by the cost of extra handling for situations where the destination cannot be reached by water or rail. passenger transport Although the noise production level of a passenger transported by road is about the same as that for rail, modal shift to rail is seen as an improvement due to the lower annoyance rating of railway noise. Traveling by boat is not an alternative, while traveling by plane is incomparable due to the distance effect. At small distance the annoyance rating is higher due to the relative larger contribution of take-off and landing noise. Long distance flights benefit from the fact that most of the traveling is done at altitudes where noise exposure of the population is nearly non-existent. The beneficial effects by shifts from car to bicycle, walking and subway are beyond discussion, since these means of transportation are more or less noiseless (Subway: as long as the system is underground, systems above the ground can be extremely noisy due to the light supporting structure that vibrates easily). A shift to city busses will improve noise (and air quality) only when modern types are used. Older types can be very noisy and exhibit low quality exhaust air emissions. The cost/benefit ratio cannot be clearly established. Relevant are the actual traveling time loss, the usage of the duration of the train journey and the assumed positive effects of biking and walking. Additional costs for redesigning transport routes in urban areas must at the other hand be taken into account (bicycle paths, pedestrian areas, etc.). M+P.MVM03.2.1; revision 2 18

20 3.6 Infrastructure and propagation from source to receiver Barriers technology Screening is the traditional method for controlling environmental noise. Its effect lies in the fact that noise only partly bends around a subject (referred to as diffraction). The efficiency of this process is affected by two parameters: the sound reduction is defined by the path length difference between the undisturbed path and the diffracted path, relative to the wave length of the sound. This means that high frequency sound is shielded very efficient, while low frequencies bend easy and will therefore be relative unaffected, meteorological conditions of the lowest layer of the atmosphere such as wind speed gradients and temperature gradients cause bending of sound rays, that reduce the effective height of the barrier and thus reduces shielding efficiency; economy If aesthetics is not of importance and cheap materials and reduced maintenance can be applied, the cost/benefit ratio can be rather positive. Negative cost factors are: the necessity to apply sound absorbing cladding when applied on both sides of the road; improving the aesthetic quality and giving them transparency, as is often done when applied close to living areas; complicated barrier designs that are required in case of entrance and exit roads. general Barriers are in general effective, can be built overnight and requires a limited amount of floor space. Constraints in the application lies in reduced efficiency when: sources, barrier and receivers are wide apart (then the diffracted ray path, lies close to the undisturbed ray path) as is more and more the case with multi lane motor ways; in the case of high rise apartment buildings, when barriers either have to be very high, or partial covering constructions are applied; its limited ability to modify them when increased reduction is required. for both the users of the road and the people living in the vicinity the barriers are considered as obstacles that block view and therefore affect the environmental quality in a negative way; M+P.MVM03.2.1; revision 2 19

21 due to the spectral shift of the noise caused by the diffraction process, the efficiency of barriers is reduced when applied in combination with housing façade insulation. All these factors cause a considerable drop of the barriers on the cost/abatement ranking list as is corroborated by several studies(see: [7], [10] and [26]). Road surfaces. The technical details are presented in paragraph 3.4. The noise mappings of three major European cities, presented in chapter 4, demonstrate the effectivity of application of low noise road surfaces over measures on vehicles and tyres. Although the road surface is integral part of the noise generation process it has to be considered part of the infrastructure with respect to planning and application. The effect of road surface variations exceeds the effect of vehicle and tyre as is illustrated in the figure below. It can be seen clearly that the most noisy passage on the silent surfaces are far below the most silent passages on the noisy surface. This is especially the case for medium and high speeds, but also at urban speeds there is a distinct difference. S.P.B. level [in db(a)] Brushed concrete drain asphalt 0/8, 80 mm thick v [in km/h, log scale] figure 12 cruise-by data (based on maximal A-weighted levels at 7.5 m distance) of about 2000 vehicles on transversely brushed cement concrete and double layered porous asphalt concrete. The level difference between a very noisy and very silent surface can amount up to 15 db, which is equivalent to a reduction in vehicle intensity with a factor of 30 or a reduction in speed of about 4 (e.g. from 120 Km/h to 30 Km/h). A variation up to 6 to 8 db can be observed quite frequently. Since such level differences are directly transferred to the housing façades, application of a low noise road surface is among one of the most effective measures to control traffic noise, not only in absolute effect, but also in cost effectiveness as several studies have proved ((see: [7], [10] and [26]). Although very effective in suppressing rolling noise its constraint lies in the limited durability when applied wrongly. Therefore much research nowadays is dedicated to improving the durability (see ref[28]). M+P.MVM03.2.1; revision 2 20

22 3.7 Road and town planning general The planning of the urban areas defines to a great extent the choice and usage of transport mode and the geometries of source to receiver propagation. Urban areas that separate transport and living will exhibit lower levels of noise exposure than urban areas in which both functionalities are mixed. There are several issues to be addressed in the planning of urban areas and infrastructure. They will be treated in general in the next paragraphs. More detailed analysis are made on base of noise mapping of three major European cities that differ considerably in town and road planning zoning A general approach to control the noise impact is to increase the distance from source to receiver. It can be demonstrated that the cost/benefit ratio of zoning is poor. With houses at 10 m distance from the centre of the road, an additional 10 m reduces the noise levels with less then 3 db, at an estimated cost of 40 M /km (ground costs of 200 /m2). The next 3 db costs about double (80 M /km), while the next 3 db is achieved by again doubling the costs. In case of high rise buildings at both sides of the street, the reflected contribution will partly destroy the reduction effect, leading to even more negative cost/benefit figures. Furthermore, urban sprawl increases traffic. It can be deduced that both effects cancel out and thus the general exposure of population in a densely urbanized area is equal to that of an area with wide spread housing. This is corroborated by [27] who compared several urban areas with widely varying population density and concluded that the sound production per unit area remained the same, even if such different cities as Los Angeles and Hong-Kong were compared Road planning The planning of a circular road, replacing main arteries through a city, will improve the noise situation in that city. However, moving traffic to the outside of the city causes limitations in the future expansion of the city and exciting noise problems then. Also, higher speeds result in increased sound levels along side the road (on average about 2 db per 10 km/h increase). Its beneficial effects lies in the reduction of traveling time and the improvement of air quality. The emission from vehicles is optimal under smooth driving conditions at around 80 km/h, stop-and-go traffic exhibits high emission of fine particles and NO X. The concentration of vehicles on major arteries and ring roads further improves cost/benefit figures of noise reducing measures such as low noise surfacings and barriers. The comparison of noise figures between Munich (no ring-road) and Amsterdam (ring-road) illustrates the beneficial effects of a ring road combined with barriers and low noise surfacings. The assumption is of course that the other characteristics are comparable. M+P.MVM03.2.1; revision 2 21

23 table V comparison of noise exposure of city with and without ring road system. n.b. the Amsterdam situation, the ring road system is combined with noise barriers and low noise surfacings city % of population above 65 db(a) Lden Amsterdam (with ring road) 15 Munich (no ring road) 25 See chapter4 and 8 for more information. 3.8 Building planning Noise spread in urban areas cannot be explained by straightforward sound propagation formulas. The process of reflection and diffraction causes a complex sound field in which very silent areas are found close to noisy spots. The sound comes either from individual traffic events in the close vicinity or from the big city background hum. Urban building design and city planning carries many opportunities for improving the sound-scape. Noise shielding from main arteries can be combined with enhancement of local natural sounds so not only the annoying background noise is suppressed, but also the acoustical environment is improved by giving the sounds, characteristic for that environment, a chance. In general, urban sound-scape improvement is not only db(a) reduction, but involves distinction between annoying and natural sounds. Nevertheless, some basic principles can be extracted, based on statistical evidence. 1. The closed façade building geometry, found in densely populated areas results in high sound shielding from this local noise. Especially when oriented in block like structures, the inner-block space can exhibit very low levels, although the background hum will also be noticeable there. Such housing areas can combine high sound exposure levels at the street side, with low exposure at the rear. 2. Open or row type of building will not improve the sound-scape, but on the contrary, due to multiple reflections, sound exposure may become higher. This is worsened in case of absence of a clear traffic plan and flow is allowed through all streets. 3. Applying noise absorptive materials, for road surfaces, but also for building facades will suppress the reverberation field in city streets, especially between closed facades, and will also reduce propagation through the urban area. Though no wide scale design is known, these ideas are put forward in several occasions. (see for instance Ref. [29]). 4. A considerable effect can be obtained by careful planning of noise sensitive objects and noise producing infrastructure, using the screening effect of noise insensitive objects. In a modern active city living, transporting, working and recreational activities are mingled. M+P.MVM03.2.1; revision 2 22

24 Concentrating transportation (be it rail or road) on a limited number of axis and surrounding these axis with building rows of office and recreational buildings generates large areas that, apart from the omnipresent city hum, can be quite quiet. The cost/benefit figure of such a planning can be very positive, since placing offices next to transportation axis, results in high exposure of these buildings which is in general regarded positive and valuable, leads to good accessibility with private and public transport, while such areas from a housing point of view are either less desirable (and therefore less valuable) or requires large investments in façade measures. For the more quiet back areas housing activities are preferred over business applications. Many cities already exhibit this type of planning, but one may assume that more careful and active pursuing of this principle is possible Quiet areas/quiet facades Average façade exposure does not completely explain the subjective evaluation of environmental noise. The presence of quiet areas has a beneficial effect on the perception of it. This again can be acquired by careful town planning and using such principles as shielding by building objects (see preceding paragraph) or block wise design of apartments buildings with back yards/gardens with surround shielding. Cost/benefit ratios are hard to define since the effect is mainly subjective and furthermore, changing a present situation into the desired situation will be extremely expensive. At the other hand, in new situations it will not require significant extra investments (see also preceding paragraph). M+P.MVM03.2.1; revision 2 23

25 4 EFFECT OF MITIGATION MEASURES table VI The effect of noise reducing measures on the power train, on the tyres and on road surfaces is assessed on base of available noise mapping models of three major European cities, Amsterdam, Munich and Madrid (chapter 8, ref [5). These models were modified as to allow independent changes on the power train, the tyre and the road surface for both light and heavy vehicles, under free rolling and accelerating conditions. The calculations are carried out for three scenarios: - Scenario I: 2010: no extra effort - Scenario II: 2010: extra effort - Scenario III: 2010: much extra effort And for each scenario the number of persons in 1 db wide exposure classes were calculated. These figures were interpreted in terms of annoyed and highly annoyed fractions of the populations. The application and the models are described in Annex B. Here we present the overall results. Noise annoyance averaged over three major EU cities Scenario I Scenario II Scenario III Annoyed 28,1 % 25,8 % 21,5 % 19,0 % Highly annoyed 13,2 % 11,8 % 9,2 % 7,9 % It can be seen that a 20 to 30% reduction in annoyance is estimated when extra effort is given to noise reduction. An additional 10 to 20% can be gained when much extra effort is given to reducing noise. The detailed analysis in Annex B shows that the Road surface is the most effective source component to work on, but that the power train is about as relevant as he tyres. Of the total effect about half originates from the road surface, 1/4 from the tyre and ¼ from the power train. M+P.MVM03.2.1; revision 2 24

26 5 EFFECTS ON AIR QUALITY 5.1 General The local environmental noise quality of road transport cannot be regarded separate from the aspects on air quality (AQ). At this moment the EU focuses on PM 10 and NO X. Mitigation measures for noise that cause counter productive effects for AQ have a small chance of implementation while at the other hand attractivety of noise mitigation measures, that are also beneficial for AQ will be large Many of the mitigation measures described in this report have an effect on the local air quality, especially the components PM 10 and NO X. Road traffic is the dominant source for NO X and contributes with about 1/3 to PM 10. The largest part in this contribution originates from Diesel engines. Additional improvements with respect to CO 2 emission do affect total vehicle concepts. 5.2 Measures on the source On the level of measures on the technology of the source, the relation between further suppressing the source strength for sound and the quality of the exhaust emissions is strong. For instance suppression of the ignition delay and smoothing of the combustion process is beneficial for both improving the air quality with reduction of the NO X and for reducing the typical combustion noise of diesel engines. With respect to PM 10, there exists a clear trade-off between NO X and PM 10, both cannot be optimized at the same time. A special measure on the source is the replacement of Diesel engines by Otto engines running on LPG or natural gas. Due to the lower and smoother combustion pressures these engines are less noisy. For Otto type engines the beneficial effect is the additional sound damping effect of the catalyst converter in the exhaust silencing system. For both engine types the trend towards higher engine torque at low engine speeds, required for reducing specific mileage of the vehicle is also very beneficial for noise, since it allows the driver to operate the engine at lower engine speeds and thus lower noise levels. A strong example of correlated improvement of exhaust emission, mileage and noise is the hybrid engine technology, that allows running on electric power in cities, especially in frequent stop-and-go traffic. 5.3 Measures on the usage of the source On the level of measures on the technology of the source, the relation between noise sensitive driving and clean driving is ambiguous. Smoothly flowing traffic at moderate speeds (about 60 km/h) exhibits minimal particle and NO X emission. Increasing the speed slightly increases the gaseous emission as well as noise (ref [30]). Decreasing the speed will decrease noise, but increases gaseous emission, partly because the engine operates outside their optimal speed/load area and partly because transient conditions are introduced for which the engine cannot adapt itself quickly enough. M+P.MVM03.2.1; revision 2 25

27 Driving style, influenced by drivers attitude, affects gaseous emission directly. Aggressive driving that implies high engine loads and engine speeds and which was shown to cause higher noise emissions (see 3.5.2) causes higher fuel usage and therefore higher emissions, enhanced by the fact that the transient conditions encountered in this driving style leads to higher emissions. A style aimed at fuel saving by driving in low engine speeds, leads directly to lower emissions of both gaseous type and of noise. 5.4 Modal split Transfer of passenger transport from road to rail or freight transport from road to ship or rail is beneficial for the air quality. Both CO 2 and NO X emission per unit decreases. A negative exception is the transfer to bus/tram in which the unit reduction is hampered by the low occupancy of these modes of transport. Below are given result from a study of the Netherlands Statistical Agency (CBS), Ref [31]. figure 13 CO 2-emission (left) and NO X emission (right) per 1000 passenger-kilometers, From left to right: rail, bus/tram/metro, airplane, car. figure 14 CO 2-emission per 1000 tonkilometers, Left: road-transport, mid: ship, right: rail. 5.5 Measures on propagation Barriers are known to have also beneficial effects on local air quality. The gaseous emissions are forced upwards and will then be diffused by the wind currents. M+P.MVM03.2.1; revision 2 26

28 Façade insulation in general is combined with noise damping ventilation devices. These devices can be equipped with filters removing particles from the ventilation air. 5.6 Measures on town planning and road planning Removing through transport from the living areas to a ring road situated outside the city will result in local improvements of the air quality and sound exposure of the population, but the regional air quality will not automatically improve since both the driving distances and the speeds increases. Town planning involving concentration of traffic in main arteries and surround them with high rise office buildings will shield both noise and gaseous emissions from the living areas behind them. 5.7 Summary table VII The effects of mitigation measures on both noise and AQ are given in the table below. effect of mitigation measures on noise and on air quality. group Source related measures Usage of source propagatio n Road/town planning Mitigation measure measure Combustion optimization Diesel engine Combustion optimization Otto engine Diesel/Otto Hybrid/electric Effect on air quality Effect on noise Diesel Otto (LPG/NG) ++ + Speed reduction - to + + Fuel saving style + + Modal split 0 to - 0 to + Barriers + ++ Façade insulation + filterering/damping ventilation systems Ring road 0 to + + Traffic arteries and boundary of commercial buildings remarks v> 60 kmh: + v< 60 kmh: effect only indoors + ++ local AQ: ++ regional AQ: - M+P.MVM03.2.1; revision 2 27

29 6 REFERENCES 1. Lärmkontor Gmbh et al., EffNoise-service contract relating the effectiveness of noise mitigation measures, EC projectnr B4-3040/2002/346290/MAR/C1, 1st draft Working Group Health & Socio-Economic Aspects, Position paper on the effectiveness of noise measures, January Morgan, Nelson and Steven, Integrated Assessment of noise reduction measures in the road transport sector, TRL and RWTÜV report prepared for EC contract ETD/FIF , 1st draft for comment, July EU Road Traffic Noise Policy in the Past, Present and Future, van Beek (TNO), 1st draft for comment, G.Rau, Akustik- und Schwingungsoptimirung an Getriebe trägern, ATZ, 107, pp (july 2005); 6. P.Zeller et..al., Future approaches for vehicle acoustics and community noise, proceedings of 3rd Styrian Noise, Vibration and Harshness Congress, June 2005, Graz, Austria. 7. KPMG, cost-effectiveness of noise measures, Report for the Netherlands Ministry of Housing, Spatial Planning and the Environment, February G.J. van Blokland, Integral approach to reduce traffic noise emission, an introduction, proceedings Euro-Noise 98, pp , October J.E.F. van Dongen, Prevalence of acceleration at low speed (=50 km/h) and driving at constant low speed (=50 km/h) and the influence on community annoyance, TNO-PG, Leiden, January 2001; GRB informal document 8 Feb COWI, Strategy to limit road traffic noise, part 3: cost effectiveness analysis of noise measures, Miljøstyrelsen, C. Treleven, Regulatory Impact Assessment of New Vehicle Noise Test, TRL September Bayerische Landesambt für Umweltschutz, Fachliche Grundlagen für bayerisches Konzept zur Minderung der Strassenverkehrsgeräusche mit technischen Schwerpunkt, 2002, not published. 13. International Motorcycle Manufacturers Association (IMMA), Motorcycle noise, the curious silence, Geneva, June B.Kortbeek, NL Environmental Ministry, personal communications, G. van Blokland, M.S. Roovers, A.H.W.M. Kuijpers, Transportation noise in Europe: an overview, M+P report EEA , prepared for EEA inn Copenhagen, October R.Stenschke et.al. Tyre/Road noise emission, rolling resistance and wet braking behaviour of modern tyres for heavy duty vehicles, Proceedings of Inter-noise 2001, pp , 17. H.E.Hansson, Design of a composite wheel, Proceedings of the International Tyre/Road noise conference, INTROC 90, pp , M+P.MVM03.2.1; revision 2 28

30 18. U.Sandberg, tyre/road noise from an experimental composite wheel, Proceedings of the International Tyre/Road noise conference, INTROC 90, pp , ACEM, European Motorcycle Industry, Striving against traffic noise, how powered two-wheelers can contribute, 20. E. de Graaff, Level of driveline noise and tyre/road noise of passenger cars in urban and sub-urban driving, M+P.MVM , Roovers et al; Transportation Noise in Europe: An Overview, M+P report EEA project state of the environment, 2nd draft version -, October ISO/DIS 362-1, Measurement of noise emitted by accelerating vehicles Engineering method Part 1: M and N categories, H. Steven; Investigations on Improving the method of noise measurement for powered vehicles, FIGE report , December 1998; 24. Calm network, Community Noise Research Strategy Plan, Noise Technology Status report, November 17, Proceedings 2nd Styrian Noise, Vibration and harshness Congress, Acoustic Optimisation in the vehicle development process of the future, new demands - new solutions, AVL, August KPMG/M+P/RIVM, Costs-effectiveness of quieter road surfaces, T. Kihlman, City traffic noise a local or global problem, Proceedings Inter- Noise 1999, pp The Netherlands Noise Innovation Program : G R Watts and P A Morgan, Predicting the additional benefits of porous asphalt within street canyons and confined spaces,, proceedings INTERNOISE DEFRA: Vehicle emission factor database v S. Schenau, Trein het minst milieubelastend,26 september 2005: ISO 10844; Specification of test tracks for the purpose of measuring noise emitted by road vehicles. 33. G.J. van Blokland, D.F. de Graaff; Effect of tyre noise limits on traffic noise; report M+P MVM 94.4, June M.S. Roovers; Measurement results passenger car tyres; GRB informal document no 5 September M.S. Roovers; IPG lijst Stille banden: onderbouwing eerste versie; report M+P NOV.5.1 May G.J. van Blokland; Integral approach to reduce traffic noise emission, an introduction Euronoise Munich 1998 M+P.MVM03.2.1; revision 2 29

31 7 ANNEX 1: REGULATIONS AND LEGISLATIONS 7.1 International regulation for road vehicles The road vehicle and its tyre is internationally considered commodity that is presented on the market and because of its implication on health, safety and environment must meet certain technical qualities. Therefore exterior sound produced by motor vehicles is subjected to several regulations and directives The requirements stated by single countries are more and more harmonized in supra-national regulations. Within the EU such a process has started more than 30 years ago and recently a world wide harmonization has been approved between the major trade blocks, the so called GTR s (Global Technical Regulations). These are developed under the hood of Working Party 29 of the UNECE (Economical Commission of Europe) and its objectives are laid down in the 1998 AGREEMENT. The EU is contracting party in this agreement. In both the UNECE and the EC the approach towards legislation is oriented from an economical point of view. The main objective is to facilitate and encourage international trade, by removing trade barriers and within this frame work, national technical regulations are trade barriers, since they hinder the easy access of foreign products. This explains the strong role of DG Enterprise in the development of type approval systems. Also within the UNECE the technical regulations are mainly discussed over by national bodies for vehicle regulations and with vehicle industry as non-formal discussion partner. With respect to noise related vehicle regulations, the position of the environmental institutions such as DG environment or the UNECE Committee on Environmental Policy in these discussions is weak. This is clearly reflected in the technology following nature of the technical requirements that prefer common agreement over strong positions. This position is totally different from a comparable vehicle regulation on exhaust emission. Here the role of environmental and health issues is much stronger and the resulting regulations force new technology of vehicles to be developed. The three major parts that this vehicle regulation consists of are 1. the description of the type of test the vehicle has to be subjected to; 2. the requirements that are stated with respect to the test result; 3. the administrative rulings with respect to introduction, registration and exceptions. These issues all have to be studied in order to be able to evaluate the quality and effectiveness of this regulation. 7.2 Motor vehicles with four or more wheels This category of vehicles is described in directive 70/157/EC and regulation UNECE R51. M+P.MVM03.2.1; revision 2 30

32 7.2.1 Test method vehicles passenger cars and light duty vehicles The current test method for passenger cars consists of a wide open throttle acceleration of the vehicle at a speed of about 50 km/h in 2nd and 3rd gear (some types only 3rd) for manual gearboxes and in drive for automatic transmissions. Such a test condition evaluates the sound characteristics of the vehicle in a condition of medium vehicle speed and medium to high engine speed and high engine load. The total test result on average is explained for about 40% by the rolling process and for about 60% by the propulsion process. Heavy duty vehicles The HDV test procedure differs from the one for cars, since the HDV test is directed to testing at the gear ratio that leads to the highest test result, in most cases characterized by a condition of rated engine speed and full engine load. Although the test procedure focuses to the propulsion process, the unloaded condition can give rise to high stick-slip noise of the driven wheels. This could be avoided by mounting worn qualifier tyres, which is allowed in the current procedure, to avoid too high tyre noise contributions. New test procedure At present in several groups, ISO, UNECE and EU, a new test method is discussed with the aim of improving the representativity of the test method for urban driving conditions. In all cases realistic (full tread depth) tyres have to be mounted. For passenger cars, this development will result in a vehicle driving state that, relative to the present procedure, operates the vehicle at higher gear ratios and lower accelerations. This method better reflects the noise source distribution in urban traffic, where rolling noise dominates over propulsion noise, but consequently is less discriminating between noisy and silent drive trains. HDV s wil be tested with payload on the driven axle and at a maximized engine speed. This method avoids the stick slip noise of tyres and better represents the actual contribution of rolling noise and propulsion noise in urban traffic Technical requirements The requirement on the test result is mainly that a certain limit value may not be exceeded by the vehicle under test. This limit value is at present 74 db(a) for passenger cars, 76 to 77 db(a) for light duty vehicles and 78 to 80 db (A) for heavy duty vehicles and busses. M+P.MVM03.2.1; revision 2 31

33 7.2.3 Administrative rulings and allowances The process of determining the test result, in order to check compliance with the technical requirements, carries uncertainties which is accounted for by subtracting 1 db from the measurement result. This reflects the position of the regulatory body as one who has to prove that a certain vehicle does not meet the requirements. Standard statistical procedure is that, when a certain quality has to be proven, beyond a certain failure level, the uncertainty is added to the measurement result. Within the COP (Conformity Of Production) procedure it is stated that vehicle from the production line must meet the requirements, but that to account for inter vehicle spread, a 1 db margin is used. This is sensible since vehicle production will then target to the required level. A second aspect that affects the compliance check is the allowances that are given due to the fact that the technology used in the vehicle is assumed to make it harder to reach such limits. The application of direct injection for diesel engines (DI) gives an allowance of 1 db and special purpose vehicles such as vehicles with off-road capabilities get an extra allowance of 1 to 2 db. Furthermore sports cars (defined by min power and exceeding a certain power to mass ratio) receive an allowance of 1 db (in addition to the beneficial test condition of only 3rd gear). A special case are the light duty vehicles that, although largely based on passenger car technology, exhibit significantly higher limit values of 76 to 77 db(a) and added to that there is a DI-diesel allowance (while 95% of these vehicles nowadays are equipped with this engine type). For HDV s no such allowance exists Discussion The present vehicle regulation, in force since 1970, has resulted in a significant improvement of the noise characteristics of HDV s. This is the combined effect of representative measurement methods, limit values that present state-of-theart technology and lack of special rulings. The regulation has been less effective for passenger cars, which can be explained by the same factors, but working in opposite direction, i.e. a less representative, measurement method, limit values that do not reflect state-ofthe-art and several allowances. These developments are illustrated in the figures below that represent type approval values over the last 25 yrs, related to the development of limit values. M+P.MVM03.2.1; revision 2 32

34 Trucks Cars Distribution [%] 82 Distribution [%] Year of inspection Year of inspection S o u n d le v e l p a s s -b y [d B (A )] Sound level pass-by [db(a)] figure 15 Distribution of results of type approval tests of road vehicle over the period in relation to the then valid limit values. The lower data on trucks before 1984 are caused by the change in measurement procedure. The sudden decrease of test data around 1991 reflects the effect of the Austrian Nachtfahrverbot. The values which exceed the limit for passenger cars are caused by allowances for special types (see 7.1.4) It must be feared that the present development of vehicle regulations will not improve the effectiveness of the regulation on short notice because: 1. The introduction of a modified procedure takes transition time to let manufacturers and testing institutes adopt to the new developments, and only after a certain time technology forcing limit values can be applied. 2. The greater contribution of rolling noise to the test result, although in line with actual urban traffic situations, is not a long lasting part of the vehicle characteristic. The OEM tyres, mounted during type approval and COP, will remain on the vehicle only for about km. After that, replacement tyres will be applied, whose properties are, in the worst case, only after 2011 subjected to the tyre noise directive (see also par. 6.4). A possibility to close this gap in the regulation, is to include the replacement tyres in the vehicle regulation. Either directly in the same way as is done in the case of replacement exhaust silencers, or indirectly via a tyre noise index. 7.3 Motorcycles A special case of road vehicles are two wheelers. This category is described in directive EC/97/24 and regulation UNECE R41. Although subjected to type approval regulations that reflect a civilized sound characteristic, the daily presentation is jeopardized by the vast fraction of illegal replacement exhaust silencing systems (IRESS). Estimates are that about 30% is equipped with an IRESS resulting in an increase of the average sound production with about 7 db(a) ref IMMA report the curious silence [figure ]. M+P.MVM03.2.1; revision 2 33

35 7.4 Regulation for tyres for road vehicles and its trailers The regulation for tyre noise is laid down in UNECE R117 and EC/2001/ Test method for tyres The current test method for tyres that is based on coast-by measurements of a vehicle with four test tyres at speeds between km/h for C1 and for C2 and C3 tyres, represents the rolling noise characteristics of tyres under regional road and highway situations. Since the speed exponents of tyres are similar an extrapolation to lower speeds can be done to obtain the characteristics under urban conditions. The test road surface is smooth and with low acoustic absorption (ref ISO [32]). Such a road enhances differences between tyres, especially the acoustic properties of the tyre that are influenced by the tread profile. A point of discussion is the reproducibility and the representativity of the test surface. The reproducibility is not controlled enough by the present ISO standard for test surfaces. A spread up to 5 db has been reported. Secondly the test surface over-emphasizes the effect of the tread profile, but is rather insensible to differences in tyre carcass response. Because of this, a tyre with a low test result will not always be a silent one on the very coarse surfaces applied in certain countries. For the general European road population the effect is about half of what is found on test surfaces (ref [33]) Technical requirements For the noise levels calculated from the measurement at the reference speed of 80 km/h for C1 and 70 km/h for C2 and C3 limit values are defined that for C1 tyres depend on the tyre width and for C2 and C3 tyres depend on tyre type. This is illustrated in the table below. M+P.MVM03.2.1; revision 2 34

36 table VIII Limit values for C1, C2 and C3 tyres as defined in R117. (C1: passenger car tyres, C2 Light Duty vehicle tyres and C3 heavy duty vehilve tyres) Class of tyre Width (in mm) Limit value (db(a)) C1a C1b > 145, C1c >165, C1d >185, C1e > Use C2 Normal 75 C2 Winter (M+S) 77 C2 Special 78 C3 Normal 76 C3 Winter (M+S) 78 C3 Special Administrative rulings and allowances The tyre regulation comes into force over a stretched period, starting at August 2003 for new tyre types, February 2004 for tyres on new vehicle types, February 2005 for tyres on new vehicles and in October 2011 for all after market replacement tyres. For C1 tyres there exists a 1 or 2 db allowance for reinforced and for special purpose tyres. Re-treated and re-profiled tyres are not covered by the tyre directive. There exists, as in the vehicle noise directive, an allowance of 1 db for measurement uncertainty and the final value is rounded off to the lowest integer Discussion The effectiveness of the tyre regulation is jeopardized by several effects. It is generally acknowledged that the present set of limit values for passenger car tyres does not reflect the state-of-the-art, nor will the programmed lowering with max 1 db around 2010 be of any effect (ref [34][35][36]) The majority of tyres on the market is meant for replacement and, when not belonging to a recent introduced new type, is not subjected to the regulation until 2009 (and the wider tyres even until 2011). M+P.MVM03.2.1; revision 2 35

37 The tyre regulation does not apply to re-treaded or re-profiled tyres. It is estimated that the majority of the tyres used for HDV are of this type, and therefore do not have to comply with the limit values. Possible development and marketing of low noise tyres will come from marketing objectives of tyre manufacturers and customer requirements in the OEM market. 7.5 Regulation for road surfaces Although one of the major factors in the noise production of road traffic, the surface quality of the road is not subjected to regulations. This can be explained by the objective of the regulations presented above, that is removal of trade barriers and since road surfaces are not (yet) a commodity, regulating it is not foreseen. In the 5th framework R&D project SILVIA a European acoustic qualification system is under development that at least will harmonize acoustic qualification of surface properties in different countries. M+P.MVM03.2.1; revision 2 36

38 8 ANNEX 2: NOISE MAPS OF 3 EUROPEAN CITIES 8.1 Introduction The effect of source measures on the noise production of road traffic in relation to the number of annoyed and highly annoyed people is calculated for three major European cities: Amsterdam, Madrid and Munich. These calculations are carried out on base of: a noise calculation model based on a general accepted propagation calculation method and the Netherlands emission model [1], but modified with the recent information from the EU 5th and 6th framework projects Harmonoise and Imagine in which and rolling noise and engine noise are separated (see 8.2); the general dose effect relations for road traffic noise as developed by the European Commission, and a special one addressing interrupted traffic flows (see8.3); GIS-based noise maps of Amsterdam, Madrid and Munich made by M+P in a project for the European Environment Agency (EEA) containing figures on inhabitants, vehicle fleet and infrastructure for the year 2010 (see 8.4). The calculations are carried out for three scenarios: - Scenario I: 2010: no extra effort - Scenario II: 2010: extra effort - Scenario III: 2010: much extra effort In this study, the effort parameter indicates the amount of effort that is put into the noise source policy for vehicles, tyres and road surfaces. The quantitative definition of the scenarios is given in 8.5 The results were then calculated in terms of the exposure of the population in 1 db classes. With these data and the indicated dose-effect relations the fraction of annoyed and highly annoyed persons in the population was found. These results are presented in Noise emission model The calculation models for the three cities are built up from the following parts: a model describing the noise production of road vehicles, distributed in a rolling noise part and a propulsion noise part, for different speeds, vehicle classes and road surfaces. In this model we distinguish between free flowing traffic and interrupted traffic; a model defining the properties of the propagation from source to receiver based on a straight forward propagation in which the effect of screening by housing rows is taken into account in a general way; M+P.MVM03.2.1; revision 2 37

39 a GIS model of the three mentioned cities, giving for each road section in the city the vehicle fleet composition, vehicle intensity, road surface and speed. Next the distance from the housing facades to the centre of the road and the density of the population, build up from the number of dwellings at each layer per meter, the number of layers and the average occupation per dwelling Noise emission constant flow For constant driving conditions the emission formulae of method I (SRMI) of the Dutch calculation scheme are regarded. The noise emission factors for light (lv) and heavy (hv) vehicles are based on measurements of traffic on several one to three year old dense asphalt road surfaces. The emission factors are given as a function of the average vehicle speed v and the number of vehicles per hour Q: v lv E lv = 69,4 + 17,6.lg + v ref with v ref,lv = 80 km/h 10.lg ( Q ) lv (1) v with vref,hv = 70 km/h hv E hv = 69,4 + 7,9.lg + v ref 10.lg ( Q ) Now, for both light and heavy vehicles formulae are deduced for the rolling noise and engine noise separately: lv, rolling hv ( v ) 10.lg( Q ) lv lv (2) E = 7,3 + 21,5.lg + (3) lv, engine ( v ) 10.lg( Q ) E = 20, + 13,6.lg + (4) hv, rolling lv lv ( vhv ) 10.lg( Qhv ) ( v ) 10.lg( Q ) E = 15,0 + 21,5.lg + (5) E = 51,0 + 2,0.lg + (6) hv, engine hv The Harmonoise emission formulae assume a linear relation between the level of the propulsion noise and the vehicle speed but comparison between the two approaches demonstrated that this caused no major difference [2]. The relations between vehicle speed and noise emission are depicted in figure 1 for light and heavy vehicles respectively. The energetical sum of the noise emission of rolling and engine noise in both cases matches the SRMI-values for the overall noise emission for all relevant speeds. Equations (3)-(6) are used in this report to describe the noise emission of light and heavy vehicles. hv M+P.MVM03.2.1; revision 2 38

40 90 90 Noise emission level [db(a)] Speed [km/h] ROLLING NOISE ENGINE NOISE ROLLING + ENGINE NOISE REFERENCE: SRMI Noise emission level [db(a)] ROLLING NOISE ENGINE NOISE ROLLING + ENGINE NOISE REFERENCE: SRMI Speed [km/h] figure 16 Noise emission of vehicles, left light vehicles, right heavy vehicles Noise emission interrupted flow For a description of the noise emission of an interrupted traffic flow the formulae describing the engine noise are rewritten by adding 5 db(a) to the engine noise emission formulae (7) and (9) providing the formulae (10) and (11). This factor of 5 db(a) represents the observed propulsion noise increase due to an acceleration of 0,8 m/s2 for cars and about 0,4 m/s2 for HDV s. The rolling noise component remains the same. E = 25,3 + 13,6.lg (v ) 10.lg(Q ) (7) E lv, engine lv + = 56,0 + 2,0.lg (v ) 10.lg(Q ) (8) hv, engine hv + In figure 3 the resulting noise emission of accelerating light and heavy vehicles at different speeds is shown graphically. lv hv Noise emission level [db(a)] Speed [km/h] ROLLING NOISE ENGINE NOISE ROLLING + ENGINE NOISE SRMI Noise emission level [db(a)] Speed [km/h] ROLLING NOISE ENGINE NOISE ROLLING + ENGINE NOISE SRMI figure 17 Noise emission of accelerating light vehicles (left) and heavy vehicles (right). 8.3 Noise exposure The noise nuisance, in terms of fraction of the population that is annoyed (%A) and fraction of the population that is highly annoyed (%HA), is calculated using the dose-effect relations for road traffic noise exposure [3]: %A den den den = 1, (L - 37) + 2, (L - 37) (L - 37) (9) %HA = 9, (L - 42) + 1, (L - 42) + 0,512(L - 42) (10) den den den M+P.MVM03.2.1; revision 2 39

41 Based on the conclusion in [4] in urban situations interrupted vehicle flows cause higher annoyance than free flow road traffic at comparable LDN levels. At levels of around 65 db(a) the % highly annoyed (%HA) is stated to be 6% higher than for constant speeds. This finding is included in this model by adapting the formulae (12) and (13) in such a way that in both cases at levels of around 65 db(a) the % (highly) annoyed is 6% higher than for the condition at constant speed, resulting in: %A den den den = 1,80.10 (L - 37) + 2,11.10 (L - 37) (L - 37) (11) %HA = 9,87.10 (L - 42) + 1,44.10 (L - 42) + 0,512(L - 42) (12) den The four dose-effect relations which are used in this study are presented graphically in figure 5. den den % of exposed people 100% 80% 60% 40% %A - constant speed %HA - constant speed %A - interrupted flows %HA - interrupted flows 20% 0% Road traffic noise exposure level in Lden [db(a)] figure 18 Road traffic noise dose-effect relations for flows with a constant speed and interrupted flows 8.4 Noise maps The M+P-noise maps of Amsterdam, Munich and Madrid are described in previous reports [5][6]. Examples are shown in figure 6, 7 and 8 respectively. The noise maps are made in a Geographical Information System (GIS). The noise maps are street-orientated, which means that all relevant data on traffic, population and infrastructure are designated to street sections from crossing to crossing. All data on traffic, population and infrastructure are based on predictions for the year The input data in the noise maps include: - number of light/heavy vehicles per day/evening/night; - speed of light/heavy vehicles; - average distance to first line dwellings; - noise barrier between road and first line dwellings; - road surface type (dense or porous); - number of inhabitants on first/second line dwellings (the effect of a noise barrier is estimated to be on average 10 db(a)). M+P.MVM03.2.1; revision 2 40

42 figure 19 M+P noise maps of Amsterdam (top), Munich (middle) and Madrid (bottom) M+P.MVM03.2.1; revision 2 41