Unit 4 - Rail Transport

Question 1

three main noise sources on a train for noise

Answer 1

• Rolling Noise is the noise associated with the train wheels rolling along the track. For mainline trains this would be steel wheels rolling on steel rails.
• Traction Noise is the noise generated by the power unit used to propel the train along the track. The category includes the noise from the engine, exhaust and transmission.
• Aerodynamic Noise is the noise that arises as the train moves through the air. At typical train speeds aerodynamic noise is minimal but for modern high-speed trains it may approach significant levels.

Question 2

time history of the noise received at trackside during the pass-by of a train

Answer 2

Two distinct phases are apparent.

• There is an initial peak as the locomotive passes followed by a lower steady level as the remainder of the train passes.
• The peak will consist of Rolling Noise and Traction Noise whereas the lower steady level will be almost entirely rolling noise.
• The relative contribution of Rolling Noise and Traction Noise alter with train speed. Traction Noise is relatively steady.

o Rolling Noise tends to increase by around 9 dB(A) for each doubling of speed.

o At speeds in excess of 160 km/h the Rolling Noise will be in excess of the Traction Noise.

Question 3

Rolling Noise Generation

Answer 3

Rolling Noise is caused when the wheel, the track and the track support structure are forced to vibrate and thereby to radiate noise.

Total Rolling Noise is the sum of wheel radiated noise and track radiated noise.

The balance between these sources, given as contributing equally in the equation below, varies with the detailed design of the wheel and track and operating conditions.

If 𝐿𝑤ℎ𝑒𝑒𝑙 - 𝐿𝑡𝑟𝑎𝑐𝑘 ≥ 10 dB(A), track treatments in isolation will be ineffective.

If 𝐿𝑡𝑟𝑎𝑐𝑘 -𝐿𝑤ℎ𝑒𝑒𝑙 ≥ 10 dB(A), wheel treatments in isolation will be ineffective.

Question 4

tracks with soft pads sources

Answer 4

For tracks with softer rail pads and at lower speeds, the track noise will dominate over wheel noise. In these situations, quieter railways will only result when measures are first applied to the track

Question 5

frequency range of rolling noise

Answer 5

Majority of the noise energy is in the frequency range 250Hz to 4000 Hz

• The driving force for the vibration arises from the interaction of the surface roughness of the contacting wheel rolling across the surface roughness of the rail.
  
  • Wheel roughness is a function of the braking system used on the train.

Question 6

Wheel Vibration

Answer 6

The vibration of the wheel is dominated by some of the natural frequencies that arise from the wheel behaving as a lightly damped resonant body.

The exact frequencies depend on wheel diameter and other design features

• -It is the presence of these natural frequencies which gives the characteristic frequency spectrum of rolling noise which is independent of train speed.

Question 7

Rail Vibration

Answer 7

The rail can be treated as a body of an effectively infinite extent. It is not resonant but acts as a waveguide to structural waves which propagate away from the excitation point.

-Below 1500 Hz the rail behaves as a beam with a rigid cross-section whereas above that frequency crosssectional deformation occurs.

Question 8

Rail Support

in regards to rail vibration

Answer 8

Damping of the propagating waves also take place due to the rail supports and this leads to a decay of vibration with distance along the rail.

Question 9

Why is it important to know the noise levels from each Rolling Noise source (wheel, rail and support)

Answer 9

As with all noise control the loudest source must be the priority for treatment, tackling the quieter noise sources may be ineffective it the loudest source dominates.

Question 10

Traction Noise

Answer 10

The most important source of Traction Noise is the diesel locomotive. Diesel locomotives in current use in the UK range in power from about 1 to 3 MW.

A diesel engine drives a dynamo or alternator which feeds traction motors (usually one per powered axle) for the final drive to the wheels. There is the need to provide cooling to all moving parts

Question 11

Traction Noise

components

Answer 11

Each component will be a potential noise source; these can be summarised as;

• diesel engine
• compressors
• fans
• traction motors

Question 12

diesel engine Exhaust noise

Answer 12

The exhaust noise of the diesel engine will dominate. The low rotational speed of the engines produces a noise spectrum that peaks between 100 Hz and 200 Hz

Question 13

diesel multiple units locomotive

Answer 13

For diesel multiple units, where one or all the passenger vehicles within a fixed formation provide the power for the train, the noise sources are similar to a diesel

Question 14

Electric locomotive

Answer 14

Electric locomotive tends to be much quieter since it derives its electrical power directly from the third rail or overhead catenary

Question 15

Aerodynamic Noise

Answer 15

Generated from fluctuations in the turbulent boundary layer, from flow over sharp edges and protuberances, cavities and the wake. At normal train speeds Aerodynamic noise is insignificant.

Under normal circumstances, even up to speeds of 300 km/h, there is little evidence of aerodynamic noise

the maximum speed of trains in the UK will be held below 250 km/h for the foreseeable future so aerodynamic noise is not likely to be a major source

Aerodynamic noise is a major contributor to the internal noise within a train

Question 16

Railway nose control - Noise at Source.

Answer 16

Noise level is dependent on train speed, train length and distance from the track.

Useful reductions in noise level can be achieved by large reductions in train speed

• not compatible with the operation of a commercially competitive railway
• diesel locomotive Traction Noise is a major contributor, noise energy can increase with a reduction in train speed.

studies have indicated that it is possible to achieve worthwhile reductions from low noise wheel and track designs.

Question 17

Railway Noise Control

Answer 17

The logical method is to reduce the noise at the source first followed by attenuating along the transmission path and, finally, providing some protection at the receiving position[

Question 18

Railway nose control -Wheel Noise

Answer 18

Wheel noise can be reduced by:

• Smooth wheels and rails
  
  • optimisation of the cross-section to minimise axial response due to radial forces
• reduction of wheel diameter
• additional damping
• screening of the wheel web

Question 19

Railway nose control -improvements of Wheel Design

Answer 19

Reductions in wheel radiated noise of about 7 dB(A) have been achieved with optimised wheel design including tuned absorbers.

A further 2 dB(A) reduction can be achieved using shields on the wheel web.

Question 20

Railway nose control -Track Noise

Answer 20

Track noise can be reduced by:

• smooth wheels and rails
  
  • Use of stiff rail pads
  • Rail tuned absorbers
  • Reduction of rail foot width
  • Optimised sleeper

To reduce noise, the rail pads need to be stiff but softer rail pads reduce dynamic fatigue loads in sleepers and track support structures.

• Aim is to reduce the length of rail which radiates noise by increasing damping

Question 21

Railway nose control - Bogie Shrouds and Low Trackside Barriers

Answer 21

noise shields to the vehicle body are not practical as rotation of the wheel assemblies (boogies) on tight radius curves would foul the shields.

boogie shrouds are possible but the need to accommodate the suspension movements and track gaugingspecifications limit closeness and effectiveness

Used with a low barrier a 3 dB(A) reduction in Rolling Noise was achieved with boogie shrouds.

Question 22

Railway nose control - Cost Benefit Analysis

Answer 22

With the introduction of vehicle and track noise mitigation at source it is possible to achieve target noise levels at lower costs than the use of noise barriers alone

• Has a significant positive effect on the visual impact of a railway which included noise control at source.
• Cost benefit analysis therefore favours reduction of noise at source

Question 23

Public Response to Train Noise - main conclusions

Answer 23

Their main conclusions were:

• the most appropriate noise unit relating pass-by noise to annoyance is 𝐿Aeq,24h measured at the external facade
• annoyance increases steadily as the noise level increases
• there is no threshold for the onset of annoyance
• there was no evidence that the number of trains at night had a special effect on annoyance beyond that already included in 𝐿Aeq,24h

Question 24

Public Response to Train Noise - further conclusions

Answer 24

• overhead electric routes are less annoying than diesel routes at the same noise level
• an increase in the proportion of freight increases annoyance slightly for the same noise level
• when joint noise is audible there is more annoyance

Railway noise causes less general annoyance than either aircraft or road traffic at the same noise level

Question 26

Railway noise - Noise Insulation Regulations

Answer 26

The Noise Insulation Regulations affect new and additional systems, for which the regulations are mandatory, and altered systems, for which the regulations are discretionary, but are covered by the same noise conditions.

Condition 1

Sound insulation should be provided where the external noise, 1m in front of the most affected façade, is predicted to exceed 68 dB(A) 𝐿eq during the day (0600 - midnight) or 63 dB(A) 𝐿𝑒𝑞 during the night(midnight - 0600).

Condition 2

The highest noise within 15 years of opening must exceed the pre-construction noise by at least 1 dB(A) 𝐿𝑒𝑞 in the time intervals.

Condition 3

The noise from vehicles using the new, additional or altered system must make an effective contribution of at least 1 dB(A) 𝐿𝑒𝑞 to the total railway noise within the time intervals.

Future conditions more harsh than for current track

Question 27

Railway noise - Environmental Appraisal

Answer 27

requirement to prepare an environmental assessment for a new railway scheme although there are no defined appraisal rules.

𝐿𝐴𝑒𝑞 is the most relevant acoustic indicator for predicting annoyance from rail traffic.

Criteria suggested by assessments of railway noise include:

• an acceptable daytime level of 55 dB 𝐿𝐴𝑒𝑞,1ℎ at local residences resulting from a new development.
• If 𝐿𝐴𝑒𝑞 is below 50 dB, the population quickly become used to the noise.
• For all situations the noise levels should not exceed 85 dB 𝐿𝐴𝑚𝑎𝑥,𝐹.

Question 28

Railway noise - Construction and Use

Answer 28

European Commission introduced Interoperability Directives44 and a high-speed train Technical Specification for Interoperability (TSI)

A TSI for conventional vehicles has been prepared and includes limits for starting noise, pass-by noise and stationary noise.

Special derogation has been given to UK due to reduced loading gauge making noise control more difficult.

The noise limits will be applied to all newly designed trains in the UK.

Question 29

CRN process

Answer 29

The prediction model sums the noise from all trains, on all tracks to determine the total noise level. The order in which the calculations are carried out will depend on the situation being investigated.

Question 30

CRN - SEL abbreviation differences to CRTN

Answer 30

Note that CRN uses the abbreviation SEL to denote LAE.

• The method is similar to that in CRTN but differs in the source term.
  
  • For road traffic, the flow density is sufficiently high to allow the noise to be modelled as a continuous linear ource,
  • trains each individual train on each track is treated a discrete source which can be expressed as a reference SEL

SEL is made up of rolling noise and power train

Question 31

Railway noise - Track nature reference level

Answer 31

Track nature will change ref level

• Ref level assumes welded track on concrete sleeper
  
  • Corrections required for wooden sleepers etc

Question 32

CRN propagation

Answer 32

The propagation terms in CRTN and CRN differ because the road traffic is a line source and a train can be a combination of a finite line source (rolling noise) and a moving point source (diesel traction unit ).

Question 33

CRN attenuation terms

Answer 33

CRN uses air attenuation term, ground attenuation term, a correction for a ballast support layer and attenuation due to barriers.

• For barrier calculations

o rolling noise source location is taken to be at the rail head of the nearest rail

o diesel power the noise source is located on top of the locomotive

Question 34

Railway Noise - Barrier effectiveness

Answer 34

Barrier design is important in determining its effectiveness.

It is unlikely that barriers will be built closer than 3m to the rail

• the effect of a reflected wave between the side of the rail vehicle and the barrier can be an influencing factor

Question 35

CRN - Splitting Segments

Answer 35

For most situations, the noise is determined by the train when it is closest to the receiver

-assumption of uniform source characteristics or propagation factors over the whole segment may not be valid.

• necessary to split the railway into segments where each segment defines a straight-line railway with uniform source characteristics and constant propagation conditions.
  
  • the calculation for each of the segments and the sum each segment
  • railway should be divided into a sufficient number of segments to ensure that variation is less than 2 dB(A)

Question 36

Causes of train vibration

Answer 36

Vibration caused by a moving train at a point beside the track, results from a combination of

• the vibration generated at the track at all points of wheel/rail contact the
• propagation of this energy to the point in question

Question 37

two forms of excitation for railway vibration

Answer 37

• The unsprung mass of the wheel being excited vertically as it moves over the irregular profile of the track and due to its own unevenness referred to as ‘dynamic excitation’
• The displacement caused by the full axle load as the wheel runs along the track, the combination of successive axles giving rise to a coherent pattern of moving axles, referred to as ‘quasi-static excitation

Question 38

Vibration from surface railway

Answer 38

For a surface railway

-frequency range of the vibrations is between 5 and 25 Hz for heavy freight trains on soft grounds and from 10 to 50Hz for fast passengert rains.

railway in a tunnel

-frequencies of up to 160Hz can been encountered.

Question 39

Railway vibration propagation

Answer 39

Vibration is transmitted to the surrounding ground via

• the rail and track support structure including ballast
• when track passes through tunnels,
• tunnel structure.

Question 40

railway vibration wave types in solid materials

Answer 40

Solid materials (including ground) support two body wave types.

• One is a compressional wave which is the same as a sound wave in air
• second type is a shear wave

At a free surface, these waves combine and produce a Rayleigh wave.

Question 41

Two aspects for design aspects for railway noise

Answer 41

Two aspects need to be considered when attempting to determine acceptable levels

• low frequency audible rumble
• perceptible vibration

Question 42

What is UK, BS 6472 used for

Answer 42

UK, BS 647249 is used to determine the likelihood of adverse comments from vibration.

• These levels are equivalent to 0.2 – 0.4 mm/s for continuous vibration in the daytime. Lower levels are felt necessary at night

Question 43

two damage levels from BS6472

Answer 43

ground vibration levels and identified two damage levels

1. Architectural damage. This could be in the form of plaster cracking. The damage is easily repaired and has no bearing on the safety of the structure.
2. Damage to load bearing components.

Question 44

vibration control possibilities for a surface railway

Answer 44

The vibration control possibilities for a surface railway may be summarised as follows:

• ballast depth - Ineffective in the mitigation of vibration
• rail pad stiffness - Ineffective in the mitigation of vibration from ballasted track
• sleeper spacing, continuous rail support - Results inconclusive in identifying potential benefit
• booted sleepers - Up to 20 dB insertion loss at frequencies above 63 Hz
• ballast mats - Between 8 and 18 dB insertion loss for frequencies greater than 63 Hz
• slab track - Theoretical studies indicate reduction at lower frequencies due to higher precision of railfixation but an increase in vibration at ground borne noise frequencies
• floating slabs and other mass spring systems - Mass spring systems with resonance frequency of 5 to 6 Hz have been installed in Metro systems. 10 dB insertion loss can be achieved for frequencies above 16 Hz, rising to 25 dB at 125 Hz.
• soil stiffening including wave impedance blocks - Varying treatments have been developed includinglime modification, lime injection and jet grouting. These are predicted to give benefits of up to 12 dB forfrequencies between 4 and 31.5 Hz
• trenches - A limited effect is observed only in close proximity to the trenchay

Question 45

A Comparison of CRTN with CRN

Answer 45

CRTN : 1988

CRN : 1995

Purpose

Calculation of noise from road traffic as defined by the Traffic Noise InsulationRegulations, 1975.

Also widely used in planning.

Calculation of noise from railway systems as defined by the Rail NoiseInsulation Regulations, 1995.

Also used in planning.

Contents

Section I : Prediction Method, calculation

Section II: Prediction Method (additionalprocedures)

Section III: The Measurement Method.

Section I: Prediction Method calculation

Section II: Prediction Method (additionalprocedures)

Section III: The Measurement Method

Section I

The PredictionMethod;PreferredOption

Stage 1: Divide Road into homogeneoussegments

Stage 2 : Calculate the basic noise levelat a distance of 10m from near-sidecarrageway for each segment.

Basic Noise level a function of;Flow,

Speed,

Percentage Heavies,Gradient,

Road SurfaceBasic Noise Level either

hourly L10 or

18 hour L10

Stage 3 : Corrections to L10 fordistance and eitherground absorption orscreening.

Stage 4 : Corrections for site layout;reflections and

angle of view.

Stage 5 : Combine contributions from allsegments to give predicted noise levelas

18 hour L10 ( 06.00 - 00.00)

Stage 1: Divide Railway intohomogeneous segments

Stage 2 : Calculate the reference SEL ata distance of 25m from near-siderailhead for each segment.

Reference SEL a function of;

Speed,Power,

Length of train,Type of track,

Track support systemReference SEL calculated for each

train and

track

Stage 3 : Corrections to SELref fordistance

ground absorptionair absorptionscreening

angle of view

Stage 4 : Corrections for site layoutreflections

Stage 5 : Convert corrected SELrefvalues into LAeq taking into account

the time period and

the number of trains.

Stage 6 : Combine contributions from allsegments and all tracks to givepredicted total noise level as

daytime 18 hour L Aeq (06.00 - 00.00)and

night time 6 hour Leq ( 00.00 - 06.00)

Section IIAdditionalProcedures

Covers specific cases were the generalmethod does not work. These include;Low traffic flow,

End of scheme,Curved roads,

Multiple roads and junctions,Houses fronting onto a main road,Multiple screening

Combined screening and reflectioneffects.

Covers specific cases were the generalmethod does not work. These include;Sidings

StationsStopping places

Guided systems other than railways.

Section IIIMeasurementMethod

Measurement should be used when;

i) traffic conditions fall outside the validity of the Charts,

1. traffic conditions are complex rendering traffic dataunreliable,
2. measurement provides a more economic method.

Measurement should be used when;

i) traffic conditions fall outside thevalidity of the Charts,

1. traffic conditions are complex rendering traffic dataunreliable,
2. guided rail system is not arailway where the vehicles run onsteel rails
3. use of new vehicles and where it is impractical to obtain reference SELs.