Performance-Based Track Geometry, Phase 1 (2012)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

56 APPENDIX RIDE QUALITY LITERATURE REVIEW TCRP D-7 Track Geometry and Ride Quality Research By C.D. Ketchum, Transportation Technology Center, Inc. EXECUTIVE SUMMARY There are many ride quality standards available to assess passenger comfort on trains. Transportation Technology Center, Inc. (TTCI) has conducted a literature survey to identify how transit authorities around the world measure and assess passenger ride quality and passenger ride comfort. While this project is primarily concerned with rail passenger ride quality, the survey also includes a review of other transport system passenger ride quality analysis techniques. The following four standards are reviewed: 1. ISO 2631 Mechanical vibration and shock — Evaluation of human exposure to whole-body vibration 2. ENV 12299:1999 Railway applications — Ride comfort for passengers — Measurement and evaluation 3. UIC 513 — Guidelines for evaluating passenger comfort in relation to vibration in railway vehicles 4. Sperling Index One of the tasks of the literature review was to determine what measurements and analysis method should be used to accurately correlate ride quality to track geometry. The data collected will eventually be used to help develop a performance-based track geometry (PBTG) method to predict the effects of track geometry on ride quality. There are many ride quality standards available to assess passenger comfort on trains. Not all issues that can affect passenger ride quality are addressed by the standards reviewed. Discrete events will be important in correlating track geometry to ride quality. All the ride quality standards reviewed in this study require similar measurements. TTCI suggests the following measurements be made in order to quantify the relationship between track geometry and ride quality. • Tri-axial accelerometers located a. Over bogie centers (both ends of vehicle) b. Center of vehicle c. Floor in operator’s cabin • Lateral accelerometers located a. Each axle of bogie (so yaw can be calculated and location of curve accurately pinpointed) • Roll rate gyrometer Ride quality will be calculated using all of the standards reviewed in this document. The data will be collected at a filter rate high enough to accommodate all calculation methods. The data will be filtered post-process for each method. This will help determine the best way to relate ride quality to different segments of track such as long tangents, curve entry/exit, embedded track, or separate right of way.

57 A1.0 INTRODUCTION Typical track geometry design and maintenance standards address acceptable geometric restraints based on past safety acceptance levels, but disregard overall performance of various vehicle types and acceptable passenger ride quality standards. In support of the Transit Cooperative Research Program D-7 research program, TTCI is investigating the effect of current track design, geometry, and maintenance standards that will account for vehicle performance and passenger ride quality using a combination of PBTG and NUCARS®9 In Phase I of this work, TTCI has conducted a literature survey to identify how transit authorities around the world measure and assess passenger ride quality and passenger ride comfort. Although this project is primarily concerned with rail passenger ride quality, the survey includes a review of automobile passenger ride quality analysis techniques. Part of the challenge of this research is to address passenger ride quality and comfort for transit authorities from tram systems on street level operations to typical intercity rail transportation. Therefore, this research is intended to encompass a wide range of possible conditions related to passenger ride quality. modeling techniques. This report discusses four common ride quality standards. It highlights similarities and differences among the four. A2.0 BACKGROUND A2.1 Components of Ride Quality Ride quality is a dynamic characteristic of a rail vehicle. It is the effect of the ride environment on the passenger. The following factors can affect passenger perception of ride quality: • Vibration — Human feelings vary with frequency of vibration even if the amplitudes are equal. A frequency weighting factor is used to evaluate ride comfort. • Transient motions transmitted from the vehicle to the passenger and crew • Lighting • Temperature • Humidity • Noise Level — Noise generated by the vehicle, wheel/rail interface, discrete events There are many standards available to analyze ride quality. Most standards address vibration measurements and effects. This study focuses on vibration and transient motions that are related to track inputs. According Forstberg,1 • Average ride comfort level — based on translational accelerations in longitudinal, lateral, and vertical directions with a frequency interval from 0.5 to 80 Hz. The comfort level can be evaluated by different scales, which are provided by ride quality standards such as Sperling, ISO 2631, and ENV 12299. human reaction to the dynamic characteristics can be divided into different categories: • Estimated ride comfort — based on how human subjects rate the ride comfort. *NUCARS is a registered trademark of Transportation Technology Center, Inc.

58 • Comfort disturbances due to motions such as high horizontal acceleration, jerks, and jolts. These disturbances can be due to discrete events that may have both high- and low-frequency content. Discrete events may occur due to transition through a turnout, alignment irregularity, or the combined effect of high lateral forces in circular curves and track irregularities. Comfort disturbances may also result from high lateral acceleration or lateral jerks while negotiating transition curves. • Motion sickness or kinetosis is due to prolonged exposure to low-frequency translational and angular motion. The frequency content related to motion sickness is usually less than 0.5 Hz. A2.2 Factors Affecting Ride Quality The following parameters can affect the motion-caused components of ride quality: • Vehicle suspension — The properties of the vehicle suspension affect the frequency and magnitude of vibration the passenger may feel • Vibration properties of the vehicle/passenger interface ─ Seat, tables, floor • Wheel/rail contact properties — These contact forces are nonlinear functions of displacement and velocity can produce vibrations in the vehicle • Wheel condition — wheel flats (generate inputs) • Vertical and lateral track misalignments • Degree of curvature and cant deficiency • Rail corrugations • Superelevation or cross level (Track Cant) irregularities • Gauge irregularities • Transition curves and superelevation ramps (spirals) • Rail joints, welds • Turnouts • Stiffness transitions (bridges) Vibration is transmitted to the passenger through interfaces such as floor, seat, and tables depending on the position of the passenger. Whole-body vibrations are quantified using the basicentric axes of the human body. Figure A1 shows the axes.2 Table A1 summarizes the passenger interfaces.

59 Table A1. Passenger/Vehicle Interfaces Position Interface Standing • Floor/feet Seated • Seat-supporting surface • Seat back • Floor/feet Recumbent • Surface supporting the pelvis, back, and head Figure A1. Basicentric Axes of the Human Body A3.0 STANDARDS There are many standards available for quantifying ride quality. This report is not all inclusive. In this literature review several of the predominately used standards are presented. A3.1 ISO 2631 Mechanical Vibration and Shock — Evaluation of Human Exposure to Whole-body Vibration 2,3,4 ISO 2631 is well recognized and widely used to quantify ride quality. The standard defines methods for quantifying whole-body vibration and effects on human health and comfort, probability of vibration perception, and incidence of motion sickness. The following types of vibrations are covered in this standard: (Cross-reference ANSI S2.72) • Periodic vibration is oscillatory motion whose amplitude pattern repeats after fixed increments of time. • Random vibration is instantaneous and not specified at any instant of time. • Transient vibration is short duration and caused by mechanical shock. Frequency content, direction, and amplitude of the vibration determine the effect on the passenger. Frequencies of the same amplitude will have different effects on passenger comfort and health. Frequency weightings are required to correctly correlate vibration content to ride

60 quality. Different frequency weightings are used for different axes of vibration. Frequency weighting curves have developed over years with experiments using human subjects. Figure A2 shows the ISO 2631 frequency weightings. Figure A2. ISO 2631 Frequency Weightings For example, the frequency weighting for motion sickness (Wf ISO 2631 provides several analysis methods for measured accelerations. The following is a summary of the methods and details can be found in Appendix AA. ) shows that the lower frequency vibrations have greater potential to cause motion sickness. • Basic method is used for a general evaluation of ride quality. If a more in-depth analysis is required, one of the other methods should be used. A weighted root-mean-square (RMS) acceleration method is used when the crest factor is less than 9. ─ Crest factor is the modulus of the ratio of the maximum instantaneous peak value of the frequency-weighted acceleration signal to its RMS value • Running RMS method is used to evaluate vibration with occasional shocks and transient vibration. • Fourth power vibration dose method is more sensitive to peaks. Once the weighted acceleration has been calculated, the ride quality effects on health, comfort perception, and motion sickness can be determined. Figure A3 shows vertical vibration exposure criteria curves defining fatigue decreased proficiency boundaries. ISO 2631 provides contours, as Figure A3 shows, for ride quality measures including reduced comfort, fatigue decreased proficiency, and exposure limit. The weighted acceleration can be plotted on the graph to determine if ride quality exceeds the health and comfort boundaries.

61 Figure A3. Lateral Exposure Criteria Curves showing Reduced Comfort Boundaries Table A2 shows how the weighted accelerations are related back to passenger comfort perception. There are two ways provided by the standard to quantify passenger comfort level. 1. A single number can be calculated for the vibration magnitude and related directly to a passenger comfort level described in Table 2. 2. The weighted RMS value of acceleration can also be plotted against the boundaries shown in Figure A3. See Appendix AA for details on the calculation of aw Table A2. Vibration Magnitude and Corresponding Comfort Level . This allows the analyst to determine which frequency ranges of vibration are contributing most to the ride quality. Vibration Magnitude (meters/second Comfort Level 2) aw Not uncomfortable <0.315 0.315 <a w A little uncomfortable <0.63 0.5<a w Fairly uncomfortable <1 0.8<a w Uncomfortable <1.6 1.25<a w Very uncomfortable <2.5 a w Extremely uncomfortable >2 A3.1.1 Required Measurements It is important to take measurements at the passenger interfaces. Measurements should be carried out at both ends and at the middle of the test vehicle. For a double-decker (multilevel) vehicle, measurements should be taken on both upper and lower decks (all levels).4 ISO 10056 Mechanical vibration – Measurement and analysis of whole-body vibration to which passengers 1.E-03 1.E-02 1.E-01 1.E+00 1.E+01 1 10 100 A cc el er at io n rm s 3rd Octive Frequency (Hz) Lateral Vibration exposure criteria curves 1 Minute 16 Minute 25 Minute 1 Hour 2.5 Hour 4 Hour 8 Hour 16 Hour 24 Hour A cc el er at io ns R M S 1/3 Octave Frequency (Hz)

62 and crew are exposed in railway vehicles5 Vibration accelerations can be characterized by translational and rotational components. However, in this standard it was assumed that for rotational vibration the center of rotation is large enough to consider the vibration as translational. Therefore, the physical parameters usually measured are the translational accelerations at the passenger interfaces. is a supplement to ISO 2631 that describes ride quality measurements on a railway vehicle in detail. Table 3 shows the equipment, measurement locations, and measurement directions needed to accurately measure accelerations. Table A3. Summary of Measurement Requirements Equipment Measurement Locations Measurement Directions Notes • Accelerometers • Filters (band limitation and frequency weighting) • Data Recorders • Floor ─ Over bogie centers ─ Center of vehicle ─ Vestibule floor (optional) • Seat ─ On and under seat at center of vehicle ─ Both ends of vehicle (optional) • Operator’s Cabin ─ Near where seat is mounted NOTE: Accelerometers should be mounted on the floor as close as possible (less than 100 millimeters if possible) to the vertical projection of the center of the seat pan, and on the vestibule floor when studying the standing position for local transport. X-axis – longitudinal, along the direction of travel Y-axis – Lateral at right angle to direction of travel Z-axis – vertical, upwards perpendicular to floor Measurements shall be collected for a duration of no less than 20 minutes, divided into representative sequences of 5 minutes each to assure statistical significance A3.1.2 Report Format ISO 100565 • Basic Aim of the test specifies that the following information shall be reported: • Evaluation Methods • Test Conditions ─ Description of vehicle  Vehicle (railcar, coach, locomotive, etc.)  Type  Load Conditions

63  Structural arrangements (steel, aluminium, type of suspension, type of bogie, wheel profile) ─ Description of seat  Type (single, multiple, etc.)  Covering (synthetic, fabric)  Special features (arm-rest, foot-rest, foldaway table, reclined, etc.)  Position (in a row, face to face, location, and orientation in vehicle) ─ Description of occupant  In cases where the measurements are carried out at the man-seat interface, the height and mass of the occupant shall be indicated  Age and gender ─ Description of Track  Route section and geographical location  Type of track (gauge, sleeper type, rail-support system, rail profile)  Description of track quality  Detail of track (radius of curvature, turnouts, etc.) ─ Running Speed  Vehicle speeds during test • Measurement Setup • Measurement Results ─ Spectral Analysis ─ Statistical Results  RMS values of accelerations • Histogram and cumulative histogram • Width of class and number of classes  Evaluated statistical parameters • Mean value • Standard deviation • 95th percentile, etc. A3.2 ENV 12299:1999 Railway Applications — Ride Comfort for Passengers — Measurement and Evaluation ENV 12299:1999 is a European Standard with the status of a Swedish Standard. This standard is applied to passenger comfort influenced by the dynamic behavior of the vehicle. Discomfort associated with relatively low levels of acceleration is considered in this standard; however, health risk effects are not considered. Table 4 summarizes the applications for this standard. 6

64 Table A4. ENV 12299:1999 Railway Applications Description Included Excluded Effect of vibration • Comfort • Health • Activities Transmission • Whole body through passenger interface • Single body part • Whole Surface • Vehicle Motion Type of vehicle • Railway vehicles designed for carrying passengers • Other types of railway vehicle (e.g. locomotive) Test Procedure • Definitions • Reference System • Requirements • Measure and evaluation rules • Report Rules • Limiting Values Position of passenger • Standing • Seated • Lying • Performing specific actions, (e.g. writing) Analysis Methods • Indirect Measurements ─ Simplified Mean Comfort ─ Complete Mean Comfort ─ Comfort on Discrete Events ─ Comfort in Curve • Direct Measurement • Combined Measurement Figure A4 shows the frequency weightings applied in this standard.

65 Figure A4. Frequency Weighting Curve for Weighting Factor Wab and Wad Several analysis methods are provided for measured accelerations. The following is a summary of the methods, and details can be found in Appendix AB. • Simplified Mean Comfort Method is based on measurements at the floor. This method is adequate for a general assessment of the ride quality.

66 • Complete Mean Comfort Method is based on measurements at the floor and seat interface. This method correlates better to passenger perception and is recommended to be used where practical. • Comfort on Discrete Events Method is a measure of passenger comfort for individual discrete events such as local track irregularities without evaluation of cumulative effects. • Comfort in Curves Method gives a measure of passenger comfort for an individual curve transition. It is a measure of a single event without evaluation of cumulative effects. Once the weighted accelerations have been measured and analyzed by one of the methods above, it can be related to passenger comfort according to the scale shown in Table 5. The scale is only for the passenger comfort index calculated using the simplified or complex method. There is no scale for discrete events or curves. As a rule, the larger the discrete event values are, the poorer the ride quality. Appendix AC contains details of calculating comfort index N. Table A5. Comfort Index Correlated to Passenger Perception Comfort Index Passenger Perception N<1 Very Comfortable 1<N<2 Comfortable 2<N<4 Medium 4<N<5 Uncomfortable N>5 Very Uncomfortable A3.2.1 Required Measurements Translational accelerations are measured at the floor and interface between the passenger and the seat to quantify mean passenger comfort. Measurements should be taken in the center and at both ends of the vehicle. For a double-decker (multilevel) vehicle, measurements should be taken on both upper and lower decks (all levels) in the center of the vehicle. Comfort in curve transitions is quantified by measuring lateral accelerations, carbody roll, speed, and tilting angle if applicable. Comfort on discrete events is quantified by measuring lateral accelerations and speed. The measurements are taken at the following locations for both curve transitions and discrete events: • Center of carbody floor • Above leading and trailing bogies • Axle box lateral accelerations A3.2.2 Report Format The test report should include the test specification, the characteristics of the tested vehicle, the track characteristics, and a precise description of the actual test conditions; include necessary measurements, statistical results, and evaluation of comfort. A3.3 UIC 513 Guidelines for Evaluating Passenger Comfort in Relation to Vibration in Railway Vehicles UIC 513 can be applied to vibrations normally encountered in the railway environment. This standard provides recommendations on measurements, analysis, and evaluations of vibrations to quantify passenger comfort. The evaluation of vibration comfort is based on the relationship, 7

67 obtained over 5-minute periods, between the accelerations measured in the vehicle and the average of the vibration comfort ratings given by a representative group of passengers. The evaluation methods in this standard are based on the following: • Low level vibration • Large part of energy contained below 3 Hz • Physiological weightings have been made in particular in the frequency range of 0.5 to 5 Hz. • Translational measurements are made at standard points of the vehicle and seat. Rotational accelerations are not measured because a minimal contribution to passenger comfort is assumed. • Statistical evaluation method is based on the correlations between objective measurements and subjective impressions of passengers. • Statistical evaluation is made with weighted RMS values calculated over 5-second periods. Two methods of evaluation are presented in this standard. 1. Simplified method is based solely on the accelerations measured at the floor level. 2. Fuller method is based on accelerations measured at the floor level, seat pan, and seat back. The measurements are conducted with two test persons weighing 114.64 pounds (52 kilograms) and 198.42 pounds (90 kilograms) to represent 5th percentile of women and the 95th percentile of men respectively. The test may also be conducted with two persons weighing 154.32 pounds (70 kilograms) each representative of the 50th Once the weighted accelerations have been measured and analyzed according to one of the methods above, a comfort level can be determined according to the scale shown in Table 6. percentile. Table A6. Comfort Index and Passenger Perception Comfort Index Passenger Perception N<1 Very good comfort 1<N<2 Good comfort 2<N<4 Moderate comfort 4<N<5 Poor comfort N>5 Very poor comfort The weighting curves for UIC 513 are the same as those used in ENV 12299 (shown in Figure A4). A3.3.1 Required Measurements Translational accelerations are measured at the floor and seat/person interfaces. Measurements should be taken at the center and at both ends of the vehicle. For a double-decker (multilevel) vehicle, measurements should be taken on both upper and lower decks (all levels) in the center of the vehicle, and at each end of the lower deck. More measurement points may be selected depending on the objective of the test. Accelerometers, conditioning amplifier and filters, and data recorders will be needed to record specified measurements.

68 A3.3.2 Report Format The following information should be reported according to UIC 513: • Subject of the test • Method of evaluation ─ Simplified ─ Full • Test conditions • Description of vehicle ─ Type of vehicle (motor car, passenger coach, locomotive, etc.) ─ Vehicle loading conditions ─ Structural details (steel, aluminium, type of suspension, etc.) ─ Wheel profiles and actual conicity • Description of seat ─ Type ─ Covering ─ Special features ─ Position • Description of seat occupant ─ Height ─ Weight ─ Age and sex • Description of track ─ Geographical location and kilometer points of measurement ─ Track type (gauge, type of sleeper, type of rail, etc.) ─ Description of track quality ─ Special track features (curvature, turnouts, level crossings) • Running speed • Measuring chain ─ Example: Accelerations, filters, recorder • Vibration characterization • Spectral analysis • Statistical results • Comfort rating A3.4 Sperling Index Sperling Index is one of the first methods developed for quantifying ride quality and comfort in railway vehicles. The Sperling method is based on a series of studies performed by the Rolling Stock Test Department of the Reischbahn at Berlin-Brunewald in 1941. In this standard, the estimate of ride quality is an evaluation of the vehicle itself, while ride comfort is the correlation of vehicle performance to perceived passenger comfort. The equations for expressing ride quality and ride comfort are the following: 8 • Ride quality 10/1 3 )(896.0 f aWZ =

69 • Ride comfort 10/1 3 )]()[(896.0 fF f aWZ = a is the peak acceleration f is the oscillation frequency )( fF is the weighting factor The Sperling Index is calculated from measured data and can be evaluated using the following scales. Table 7 shows the ride index corresponding to vehicle ride quality, and Table 8 shows the ride index corresponding to ride comfort. Table A7. Ride Index Corresponding to Vehicle Ride Quality Ride Index W Ride Quality Z 1.00 Very good 2.00 Good 3.00 Satisfactory 4.00 Acceptable for running 4.50 Not acceptable for running 5.00 Dangerous Table A8. Ride Index Corresponding to Passenger Ride Comfort Ride index W Comfort (vibration sensitivity) Z 1.00 Just noticeable 2.00 Clearly noticeable 2.50 More pronounced but not unpleasant 3.00 Strong, irregular, but still tolerable 3.25 Very irregular 3.50 Extremely irregular, unpleasant, annoying, prolonged exposure intolerable 4.00 Extremely unpleasant, prolonged exposure harmful Both lateral and vertical accelerations are evaluated at the carbody floor in the center and at both ends of the vehicle. Figure A5 shows the frequency weighting curves for Sperling Ride comfort index. 9

70 Figure A5. Sperling Frequency Weighting Curves -30 -25 -20 -15 -10 -5 0 5 0.01 0.1 1 10 100 1000 G ai n (d B ) Frequency (Hz) Sperling Weighting Curves Lateral Vertical

71 A4.0 STANDARD COMPARISIONS Table 9 summarizes the standards reviewed in this literature review. Table A9. Summary of Ride Quality Standards Reviewed Standard ISO 2631 ENV 122992,3,4 UIC 5136 Sperling7 Effect of movement 8 • Health (0.5 to 80 Hz) • Comfort/Perception (0.5 to 80 Hz) • Motion Sickness (0.1 to 0.5 Hz) • Comfort (0.4 to 20 Hz) • Comfort (0.5 to 40 Hz) • Comfort (3-8 Hz**) Transmission Whole Body Whole Body Whole Body Whole Body Vehicle Motion Position of Passenger • Standing • Seated • Recumbent • Standing • Seated • Standing • Seated Type of Vehicle • ISO 10056 – Railway vehicles • Railway vehicle designed for carrying passengers • Railway vehicle designed for carrying passengers • Railway vehicle designed for carrying passengers Measurement Type • Translational • Translation • Rotational • Translational • Translational Analysis Methods • Basic Method • Running RMS method • Fourth Power Vibration Dose Method • Simplified Mean Comfort • Complete Mean Comfort • Simplified Method • Full Method Discrete Events Analyzed Separately N/A • Comfort on Discrete Events • Comfort in Curves N/A N/A Persons N/A N/A 2 persons • 114.64 lb (52 kg) • 198.42 lb (90 kg) N/A Minimum Instrumentation Requirement 3 tri-axial accelerometers • Over bogie center • Center of vehicle • Floor in operator’s cabin 3 tri-axial accelerometers • Both ends of vehicle • Center of vehicle Lateral accelerometer • Axle box accelerations Car body roll speed** Tilting angle if applicable Speed 3 tri-axial accelerometers • Over bogie center • Center of vehicle • Floor in operator’s cabin Biaxial accelerometer on vehicle floor

72 A5.0 STANDARDS AND PASSENGER PERCEPTION Many passengers who commute on trains use the time read, write, or work on laptop computers. The rail vehicle, in many cases, becomes an extension of a person’s office. The ability to perform some of these tasks may affect a person’s perception of ride quality. Several studies have been done to assess the accuracy of standards in correlation with passenger perception. A study was conducted on passenger trains in Sweden.10 Vibration measurements were taken at five locations: seat pan, backrest, floor, laptop, and table. The measurements on the laptop and table are not included in either standard, but were included to assess the vibrations at these locations. Accelerations were measured in x, y, z directions. Vibration measurements and passenger surveys were conducted simultaneously on trains to determine the correlation between vibration and passenger activities. ISO 2631 and Sperling Ride Index were the standards used in this study to evaluate ride comfort. The passenger survey was conducted simultaneously on all trains at the same time vibration measurements were taken. The survey consisted of 30 questions divided into 6 parts: 1. General background of participants 2. Information about journey 3. Types of sedentary activities and time spent on each activity 4. Postural positions related to reported activity 5. Short typing test 6. Feeling from disturbances from noise, vibration, jerks, etc. Evaluation of the questions did not show any significant difference between gender, age, or sitting positions in judgment of ride comfort. An evaluation of the measured vibration using both standards showed reasonably good ride comfort. However, the passenger surveys indicated that a significant number of passengers had difficulties performing activities such as reading or working on a laptop computer. This indicates that the standards may not evaluate the effect on sedentary activities accurately. It also indicates that low levels of vibration may have an effect on passenger activities. A similar study was conducted in India.11 The ride quality measurements and assessment indicated a ride comfort in the medium to comfortable range. However, the surveys again indicated passengers were moderately affected by train vibrations while performing sedentary activities such as writing and working on a laptop computer. The motions that were reported to have the greatest effect were lateral vibrations and occasional jerk and vertical vibrations. The study also indicated that low levels of vibrations can affect passenger activities. Vibration measurements and surveys were conducted simultaneously. The survey questions were categorized similarly to the study done in Sweden. However, the ENV 12299:1999 was used to analyze the accelerations and quantify the ride quality.

73 It may be necessary to look at discrete events individually to determine effects on ride quality. Some of the discrete events that may cause lateral vibration, jerk, and vertical vibrations are transition curves, turnouts, and corrugations to name a few. It will be important to quantify the vibrations caused by these events and to look at the contribution to passenger discomfort in detail. A6.0 RIDE QUALITY AND TRACK GEOMETRY Track geometry can affect ride quality. Typical track geometry design and maintenance standards address acceptable geometric restraints based on past safety acceptance levels, but usually disregard overall performance of various vehicle types and acceptable passenger ride quality standards. The following are parameters to consider when quantifying passenger ride quality: • Vertical and lateral track misalignments • Corrugations • Cant irregularities • Gauge irregularities • Transition curves and superelevation ramps (spirals) • Rail joints, welds • Turnouts • Stiffness transitions (bridges) There are systems available to automatically measure track geometry and report exceptions to the safety standards. These systems give a report as to the size of defect and location, so it can be maintained as necessary. However, these systems do not usually take into account how the track geometry defects will affect passenger ride quality. Many of the current standards do not specifically quantify discrete events and relate them to ride quality. A study on ride comfort of high-speed trains traveling over railway bridges was conducted.12 • To investigate how rail roughness level influences ride comfort The ride comfort was studied using the Sperling Comfort Index, and the maximum level of accelerations measured. Some of the objectives of this study are listed below: • To investigate the influence of ballast stiffness on ride comfort • To investigate the effects of train speed on ride comfort A parametric study was done using a time domain model. Timoshenko beam theory was used to model the rail and bridge. Parallel damped springs and masses were used to model rail pads, sleepers, and ballast. A random, irregular vertical track profile was modeled. Figure A7 shows the vertical track profile and roughness index for each of the three different classes of track modeled. Nonlinear Hertz theory was used to model the wheel-rail contact. A 300-meter carriage type of Japan’s Shinkansen (SKS) system was modeled. Three different types of suspension for this vehicle were modeled: 1. Linear primary and secondary suspension (base model)

74 2. Primary suspension is unchanged from base model and secondary suspension is modeled by a nonlinear rubber element 3. Primary suspension system is modeled, and nonlinear rubber springs and the secondary suspension are unchanged from the base model. Results were calculated for train speeds from 0 to 250 miles per hour (400 kilometers per hour). Figure A6 shows the calculation procedure. Figure A7 shows track geometry deviations and roughness levels. Figure A6. Calculation Procedure Input Data Modal Analysis Galerkin’s Method of Solution Response in time Domain FFT Analysis Filtering Procedure Output 1 (max. Acceleration) Output 2 (Comfort Index)

75 Figure A7. Track Geometry Deviations and Roughness Levels Figure A8 shows the calculated ride comfort for different track roughness levels shown in Figure A7. For track with no irregularities (smooth), comfort index is independent of speed. However, as track roughness increases, the comfort index changes with speed. Track roughness also has a significant effect on ride quality.

76 Figure A9 shows comfort index related to ballast stiffness. Bξ is the ratio between the actual ballast stiffness and the change in stiffness. The ballast stiffness has little effect on the comfort index. FigureA8. Comfort Index Related to Track Roughness Figure A9. Comfort Index Related to Ballast Stiffness

77 Results of the study showed that ride comfort can be affected significantly by track roughness. Speed also has an effect on ride comfort. Ballast stiffness did not have a significant effect on track roughness. The referenced study shows that track parameters can have a significant effect on ride quality. Most ride quality standards do not address discrete events separately. In order to understand specifically what the contributions of track roughness, corrugation, and track alignment are, it may be important to relate these specific events to ride comfort. A7.0 OTHER TRANSPORT SYSTEMS Other transportation modes also address ride quality and passenger comfort issues. The automotive industry also uses ISO 263113,14,15,16 Marine transport systems also use ISO 2631 to quantify passenger comfort. to assess the effect of vibration on ride comfort. Studies have been done to determine the correlation between the subjective measure of passenger perception and the quantitative measure of vibration. One study indicated that overall, the correlation between fourth power method and passenger perception was good. The fourth power method is more sensitive to discrete events than the simplified or full methods in ISO 2631. This indicates the need to individually correlate discrete events to passenger comfort. 17 Aircraft transportation systems have recognized that airport pavement roughness does affect passenger comfort and safety during landing and takeoff. This transportation system is particularly interested in vibration effects on motion sickness. ISO 2631 is the only standard reviewed in this literature study that addresses motion sickness. In the study by Prince, the correlation between ISO 2631 and passenger perception was good in addressing whether or not motion sickness would occur, but not to the degree of motion sickness. 18 Railways also have corrugations that develop in the running surface. The ride quality standards used to address railway ride comfort do not directly address the wavelength content of corrugations and the effect on passenger safety and comfort. The study looked at the relationship between pavement profile wavelengths and vertical vibrations. This study indicates there is a correlation between pavement profile wavelength and passenger safety and comfort. Uncomfortable and unsafe frequencies were identified. The evaluation index for highway pavement roughness, i.e., International Roughness Index (IRI), was determined to be unsuitable for this application. A8.0 CONCLUSIONS There are many ride quality standards available to assess passenger comfort on trains. Not all issues that can affect passenger ride quality are addressed by the standards reviewed. Discrete events will be important in correlating track geometry to ride quality. Passenger comfort is a subjective measure and the standards do not always correlate to passenger perception. This is especially true when passengers are performing sedentary activities. One reason for the discrepancy may be due to the vibrations from discrete

78 events being averaged with the rest of the route. It may be necessary to analyze discrete events independently to get a more accurate picture of ride quality. One of the objectives of the literature review was to determine what measurements and analysis method should be used to accurately correlate ride quality to track geometry. It will be important to quantify the overall ride quality and vibrations induced from discrete events. The data collected will eventually be used to help develop a PBTG method to predict the effects of track geometry on ride quality. All the ride quality standards reviewed in this study require similar measurements. TTCI suggests the following measurements be made in order to quantify the relationship between track geometry and ride quality: • Tri-axial accelerometers located ─ Over bogie centers (both ends of vehicle) ─ Center of vehicle ─ Floor in operator’s cabin • Lateral accelerometers located ─ Each axle of bogie so yaw can be calculated and location of curve accurately pinpointed • Roll rate gyrometer Ride quality will be calculated using all of the standards reviewed in this document. The data will be collected at a filter rate high enough to accommodate all calculations. Then, the data will be filtered according to the requirements for each method. This will help determine the best way to relate ride quality to different segments of track such as long tangents, curve entry/exit, embedded track, or separate right of way. Bibliography 1. Förstberg, Johan. 2000. “Ride comfort and motion sickness in tilting trains.” Doctoral thesis, TRITA-FKT Report 2000:28, ISSN 1103-470X, ISRN KTH/FKT/D—00/28—SE, Department of Vehicle Engineering, Royal Institute of Technology, Stockholm 2. International Organization for Standardization (ISO).1997. Mechanical vibration and shock — Evaluation of human exposure to whole-body vibration Part 1: General requirements. ISO 2631-1:1997 (E), Second edition corrected and reprinted 1997-07-15, Switzerland 3. International Organization for Standardization (ISO).1989. Evaluation of human exposure to whole-body vibration Part 2: Continuous and shock-induced vibration in buildings (1 to 80 Hz). ISO 2631-2:1989 (E), First edition 1989-02-15, Switzerland 4. International Organization for Standardization (ISO).2001. Mechanical vibration and shock — Evaluation of human exposure to whole-body vibration Part 4: Guidelines for the evaluation of the effects of vibration and rotational motion on passenger and crew comfort in fixed-guideway transport systems. ISO 2631-4:2001 (E), First edition 2001-02-01, Switzerland

79 5. International Organization for Standardization (ISO). 2001. Mechanical vibration — Measurement and analysis of whole-body vibration to which passengers and crew are exposed in railway vehicles. ISO 10056:2001(E), First edition 2001-08-15, Switzerland 6. CEN. European Committee for Standardization. February 1999. Railway applications — Ride comfort for passengers — Measurement and evaluation. European Standard ENV 12299:1999 E, ICS 45.060.20, Brussels 7. International Union of Railways. 1995. Guidelines for evaluating passenger comfort in relation to vibration in railway vehicles. UIC Code 513 R, 1st 8. Sperling, E. and Ch. Betzhold. August 1957. “Contribution to the evaluation of comfortable running of railway vehicles.” Bulletin-International Railway Congress edition, 1.7.94, Paris, France 9. Kumar, Vivek and Vikas Rastogi. 2009. “Investigation of vertical dynamic behavior and modeling of a typical Indian Rail Road Vehicle through Bond Graph.” World Journal of Modeling and Simulation Vol. 5 No. 2, pp 130-138 10. Narayanamoorthy, Ramasamy, Shafiquzzaman Khan, Mats Berg, Virendra Kumar Goel, V Huzur Saran, and S.P. Harsha. May 2008. “Determination of Activity Comfort in Swedish Passenger Trains,” Paper 24112008 Proceedings World Congress on Railway Research, Seoul, Korea 11. Narayanamoorthy, R., V.K. Goel, and V.H. Saran. 2008. “Vibration Measurement — A Tool for Evaluating Activity Comfort in Trains.” Research Paper 17112008, 12. Kargarnovin, M.H., D. Younesian, D. Thompson, and C. Jones. March 2005. “Ride comfort of high-speed trains traveling over railway bridges.” Vehicle System Dynamics, Vol. 43, No. 3, pp 173-199 http://www.iitr.ac.in/outreach/web/CIRCIS/publication/Research%20Paper-moorthi- 17112008.pdf 13. Cann, Adam P., E. Kent Gillin, Alan W. Salmoni, Peter Vi, and Tammy R. Eger. 2005. “Transmissibility of Whole-body Vibration from Floor to Seat Experienced by Scraper Operators in the Construction Industry.” http://www.nexgenergo.com/ergocenter/trends/20051305.pdf 14. Giacomin, J. and T.M. Hacaambwa. May 23-25, 2001. “Performance of ISO2631 and BS6841 Comfort Criteria for Evaluating Automobile Road Vibrations.” Paper 01A1083. ATA 7th International Conference on the Role of Experimentation in the Modern Automotive Product Development Process, Florence, Italy. 15. Valasek, Michael, Jan Pelikan, Jaroslav Ulehla, Ondrej Vaculin, and Pavel Steinbauer. 2008. “Experimental Verification of Correlation between Objective and Subjective Evaluation of Passenger Car Vibration Comfort. Bulletin of Applied Mechanics, Vol. 4, Issue 15, pp 99-101. 16. Eger, Tammy, Martin Smets, Sylvain Grenier, and Vibration Research Group. 2005. “Whole-body Vibration Exposure Experienced during the Operation of Small and Large Load-Haul-Dump Vehicles.” http://www.nexgenergo.com/ergocenter/trends/20051343.pdf

80 17. Price, Brian L., Mary Claire Froelich, Eric Pierce, Jeffrey Blankenship, Joseph LaBrecque, and Michael McCauley. 2008. “Motion Sickness Incidence (MSI) Results using Power Spectral Density (PSD) Technique.” Proceedings International Maritime Conference, Sydney, Australia. http://www.highspeedcraft.org/abcd.php 18. Chen, Yi-hsien and Chia-pei Chou. 2004. “Effects of Airport Pavement-Profile Wavelength on Aircraft Vertical Responses.” Transportation Research Record: Journal of the Transportation Research Board, No. 1889, TRB, National Research Council, Washington, D.C., pp 83-93. Other References 19. AEA Technology Plc. June 7-8, 2004. “Derailment Prevention and Ride Quality.” Competitive and Sustainable Growth Programme, Light Rail Thematic Network, Prague, Czech Republic. <www.libertin.info> 20. American Public Transportation Association. 2007. “Standard for Definition and Measurement of Wheel Tread Taper.” 17. APTA SS-M-017-06, Volume V- Mechanical. 21. Andersson, Evert, Anneli Orvnäs, and Rikard Persson. August 2009. “On the Optimization of a Track-Friendly Bogie for High Speed.” Proceedings of the 21st International Symposium on Dynamics of Vehicles on Roads and Tracks, IAVDS'09, Stockholm. 22. Australian Rail Track Corporation Ltd. April 2007. ARTC Track and Civil Code of Practice. “Section 5: Track Geometry” Adelaide, South Australia. 23. Bamberg, Wolfgang and Hans Ludwig. 1979. “New Measuring Method for Ride Comfort in Transit Vehicles.” Vehicular Technology Conference, 1979. 29th IEEE. 24. Berggren, Eric G., Martin X. D. Li, and Jan Spannar. September 2006. “A New Approach to the Analysis and Presentation of Vertical Track Geometry Quality and Rail Roughness with Focus on Train-Track Interaction and Wavelength Content.” Proceedings 7th 25. Clark, Jr., Alfred. 1989. “A Primer on Roller Coaster Dynamics, Part II — You Can Bank on Them.” RollerCoaster! Volume X, Issue 1, pp 32-37 International Conference on Contact Mechanics and Wear of Rail/Wheel Systems, Brisbane, Australia. 26. Cleon, Louis-Marie and Guido Lauriks. May 1996. “Evaluation of Passenger Comfort in Railway Vehicles.” Journal of Low Frequency Noise and Vibration, Vol. 15 No. 2, pp 53-69 27. Donohew, B. and M.J. Griffin. July 2003. “Simulator for Optimizing Train Passenger and Crew Environment.” Rail Research UK Report: RRUK/C1/1, Railway Safety and Standards Board, UK 28. Dukkipati, Rao V. and Joseph R. Amyot. 1988. Computer-Aided Simulation in Railway Dynamics. New York: Marcel Dekker, Inc. 29. El-Sibaie, Magdy, et al. March 1997. “Engineering Studies in Support of the Development of High-Speed Track Geometry Specifications.” IEEE/ASME Joint Railroad Conference, Boston, MA

81 30. Esveld, Coenraad. 2001. Modern Railway Track. Second Edition. Zaltbommel: MRT-Productions 31. Fazio, A.E. and J.J. Cunningham. 1991. “An Analytic Approach to the Maintenance of High-Speed Track.” Proceedings of the 1991 IEEE/ASME Joint Railroad Conference, St. Louis, MO 32. Förstberg, J., M. Griffin, and B. Donohew I. 15 March 2004. “Appearance of onset of nausea: Transfer of knowledge; suggestion of laboratory studies.” ERRI/FACT/MM 16/17/02/04, Brussels 33. Förstberg, Johan. 2003. “Fast and comfortable trains (FACT).” Presentation for Nordic Rail 2003 by VTI, Sweden 34. Garg, Vijay K. and Rao V. Dukkipati. 1984. Dynamics of Railway Vehicle Systems. Academic Press Canada: Ontario 35. Hylén, Bertil. October 7-9, 2003. “EU Railway Research Presentation.” VTI Conference 20 — VTI’s fack- och monterseminarier vid Nordic Rail 2003, Sweden 36. ITTC — Recommended Procedures and Guidelines. 1999. “Testing and Extrapolation Methods High-Speed Marine Vehicles Excerpt of ISO 2631, Seasickness and Fatigue.” International Towing Tank Conference 20th 37. Iwnicki, Simon, ed. 2006. Handbook of Railway Vehicle Dynamics. CRC Press, Taylor & Francis Group: Boca Raton ITTC 1993, pp 384-392 38. Jacobson, D., R.W. Barker, B.D. Pepler, and L.L. Vallerie. 1977. “Comparison of Passenger Comfort Models in Buses, Trains, and Airplanes.” Transportation Research Record 646. Transportation Research Board: National Academy of Sciences, Washington, D.C. 39. Karakasis, Kleomenis, Dimitris Skarlator, and Tilemachos Zakinthinos. 2005. “A factorial analysis for the determination of an optimal train speed with a desired ride comfort.”Applied Acoustics 66 (2005) 1121-1134, 40. Kazato, A. and S. Kamoshita. 2008. “Measures against high-frequency vibration for Next-Generation Tilt Control System.” Proceedings 8 www.sciencedirect.com th 41. Kim, Y-G., H-B. Kwon, S-W. Kim, C-K. Park, and T-W. Park. 2003. “Correlation of ride comfort evaluation methods for railway vehicles.” Proceedings IMechE Conference Vol. 217 Part F World Congress on Railway Research, Seoul, Korea 42. Klauder, Jr., Louis T. June 20, 2007. “Roll-Guided Spirals for Railroad Tracks.” Proc. Railway Engineering – 2007 Conference, London, UK 43. Klauder, Jr., Louis T. and Sherman A. Clevenson. 1975. “Evaluation of Ride Quality Measurement Procedures by Subjective Experiments using Simulators.” Proceedings of the 1975 Ride Quality Symposium, NASA TM X-3295, Paper No. 11 44. Klauder, Jr., Louis T., Steven M. Chrismer, and John Elkins. 2002. “Improved Spiral Geometry for High-Speed Rail and Predicted Vehicle Response.” Transportation Research Record: Journal of the Transportation Research Board, Volume 1785/2002, Washington, D.C.

82 45. Kufver, Björn. 2001. “Research on Track Geometry for High-speed Trains.” Swedish National Road and Transport Research Institute (VTI), No. 2, pp 16-18 46. Kufver, Björn and Ola Rydell. June 24-25, 2009. “Certain Aspects of the CEN Standard for Alignments through Switches and Crossings.” Proceedings 10th 47. Kufver, Björn. October 2003. “Track/vehicle interaction and alignment design.” Paper presented at seminar arranged by the VTI at Nordic Rail in Jönköping International Railway Engineering Conference, London, UK 48. Kufver, Björn and Ola Rydell. October 2003. “Dynamic vehicle response on horizontal curves without transition curves, including turnout curves through Swedish UIC60 turnouts.” Paper presented at seminar arranged by the VTI at Nordic Rail in Jönköping 49. Luber, Bernd, Andreas Haigermoser, and Gerald Grabner. 2010. “Track Geometry Evaluation Method Based on Vehicle Response Prediction.” Vehicle System Dynamics. Vol. 48, Supplement, 2010, 157-173 50. MacMillan, Renata and Jingnan Guo. January 2009. “Reading — Assessment and Control of Whole-Body Vibration.” Safetyline Institute, 51. Madejski, J. 2004. “Autonomous track geometry diagnostics system.” Journal of Materials Processing Technology 157-158 (2004) 194-202 www.worksafe.wa.gov.au/institute 52. Mancini, Giampaolo, Sergio Pelacchi, Monica Balli, and Alessandro Corbizi Fattori. 2008.“New developments with the Italian solution for tilting trains: optimization of tilting system on new generation of Pendolino trains.” Proceedings 8th 53. Mansfield, Neil J. 2006. “Literature Review on Low Frequency Vibration Comfort.” EU Asia-Link ASIE/2005/111000 CIRCIS World Congress of Railway Research, Seoul, Korea http://www.uic.org/cdrom/2008/11_wcrr2008/pdf/R.2.4.3.5.pdf 54. Mee, Brian, Brian Whitte, and Boris Neijikovsky. 1995. “Railway testing using a portable ride quality and vibration measurement system with GPS.” Proceedings SPIE Vol. 2458, pp 138-147 55. Mian, Zack. March 2009. “Wheel Inspection System Environment Qualification and Validation.” Report No. C-06-31 for the New York State Energy Research and Development Authority and New York State Department of Transportation, Albany, NY 56. Nakagawa, M. and M. Miwa. 2008. “Next-Generation Management of Ride Quality in the Tokaido Shinkansen.” Proceedings 8th 57. Parkinson, H. and S. D. Iwnicki. 2004. “An Intelligent Track Monitoring System.” Manchester Metropolitan University, Manchester, UK World Congress of Railway Research, Seoul, Korea. http://www.uic.org/cdrom/2008/11_wcrr2008/pdf/I.3.4.1.4.pdf 58. Picu, Ana. 2009. “An Investigation of the Human Sensation Induced by Wheel Hand Arm Vibration and Seat Whole-body Vibration using Single Axis and Tri-axial Methods — A Study using the Steven’s Power Law.” Proceedings of the http://www.railtechnologyunit.com

83 ACOUSTICS High Tatras 2009 34th 59. Popprath, S., C. Benatzky, C. Bilik, M. Kozek, A. Stribersky, and J. Wassermann. July 2006. “Experimental Modal Analysis of a Scaled Car Body for Metro Vehicles.” Thirteenth International Congress on Sound and Vibration, Vienna, Austria. International Acoustical Conference — EAA Symposium. 60. Prem, Hans and Geoff Ayton. 2005. “Improved Techniques for Assessing Ride Quality on Concrete Pavements.” Eighth International Conference on Concrete Pavements: Innovations for Concrete Pavement: Technology Transfer for the Next Generation, pp 733-754. 61. Presle, Gerard, Werner Hanreich, and Paul Mittermayr. 2000. “Austrian Track Testing and Recording Car EM 250, Source for Wheel-Rail Interaction Analysis.” Transportation Research Record: Journal of the Transportation Research Board, 1713, Paper No. 00-0743, Transportation Research Board of the National Academies, Washington, D.C., pp 22-28. 62. Raymond, Gerald P. and Richard J. Bathurst. March 1994. “Repeated-load response of aggregates in relation to track quality index.” Canada Geotech. J. 31, pp 547-554 63. Scharr, Richard L. and Raymond P. Owings. 1975. “Intercity Rail Passenger Car Ride Quality Test Program.” Proceedings of the 1975 Ride Quality Symposium, NASA. pp 141-158 64. Schmidt, Tom. September 2005. “Why we need dynamic car-measurement tools.” Railway Age 65. Sundström, Jerker. 2008. “Difficulties to Read and Write under Lateral Vibration Exposure: Contextual studies of train passengers’ ride comfort.” Doctoral thesis. Royal Institute of Technology, Stockholm, Sweden 66. Track Access. May 1998. “Track Geometry — Inspection and Assessment.” Track Access Issue 2.3 67. Vermeij, David J. 2000. “Design of a high-speed track.” HERON, Vol. 45, No. 1 (2000) ISSN 0046-7316 68. Winter, Theo, Peter A.J. Meijvis, Wijnand J.M. Paans, Michael J.M.M. Steenbergen, and Coenraad Esveld. 2007. “Track Quality Achieved on HSL-South.” Project Organization HSL-Zuid. European Railway Review, Issue 3, pp 48-53 69. Yoo, Wan-Suk, Chang-Hwan Lee, Wui-Bong Jeong, and Sang-Hyun Kim. 2005. “Development and Application of New Evaluation System for Ride Comfort and Vibration on Railway Vehicles.” Journal of Mechanical Science and Technology, Vol. 19, No. 7, pp 1469-1477 70. Zhou, Hinsong, Gang Shen, Hong Zhang, and Lihui Ren. 2008. “Application of modal parameters on ride quality improvement of railway vehicles.” Vehicle Systems Dynamics, Vol. 46, Supplement, pp 629-641

84 Appendix AA – International Standard for Organization ISO 2631 The detail for the analysis methods included in ISO 2631 are given in this appendix. AA1. The Basic Method Basic evaluation method is weighted root-mean-square (RMS) acceleration for both translational and rotational vibration. The weighted RMS acceleration is calculated by: 2 1 0 2 )(1       = ∫ T ww dttaT a )(taw is the weighted acceleration as a function of time (m/s2 or rad/s2 T is the duration of the measurement (s) ) AA2. The Running RMS Method The running RMS method takes into account occasional and transient vibration by use of a short integration time constant. ∫ − = 0 0 2 1 2 0 })]([ 1{)( t t ww dttata ττ )( 0taw is the instantaneous frequency-weighted acceleration τ is the integration time for running averaging t is the time (integration variable) 0t is the time of observance (instantaneous time) The maximum transient vibration value is defined as: )](max[ 0taMTVV w= This is the highest magnitude of )( 0taw read during the measurement period. It is recommended that τ is 1s in measuring MTVV. AA3. The Fourth Power Vibration Dose Method The fourth power vibration dose method is more sensitive to peaks than the basic evaluation method by using the fourth power instead of the second power of the acceleration time history as the basis for averaging. 4 1 0 4 })]([{ dttaVDV T w∫= VDV is the fourth power dose value in m/s 1/4 or rad/s )(taw 1/4 is the instantaneous frequency-weighted acceleration

85 T is the duration of measurement Note: When the vibration exposure consists of two or more periods of different magnitudes, the vibration dose value for the total exposure should be calculated from the fourth root of the sum of the fourth power of individual dose values: ∑= i itotal VDVVDV 4 1 4 )( AA4. Ratios Used for Comparing Basic and Additional Methods of Evaluation Additional methods other than the basic method should be used if the following ratios are exceeded. 5.1= wa MTVV 75.1 4 1 = Ta VDV w AA5. Frequency Weighting Frequency content of the vibration determines the effect on the passenger. Frequencies of the same amplitude will have different effects on passenger comfort and health. Frequency weightings are required to correctly correlate vibration content to ride quality. Different frequency weightings are used for different axes of vibration. Table A1 summarizes the application of the frequency weighting curves. Table AA1. Summary of Frequency Weighting Curves Frequency Weighting Health Comfort Perception Motion Sickness W z-axis, seat surface k z-axis, seat surface, standing, vertical recumbent (except head) x,y,z axes, Feet (sitting) z-axis, seat surface, standing, vertical recumbent (except head) N/A W x-axis, seat surface d y-axis, seat surface x-axis, seat surface y-axis, seat surface x,y-axes, standing, horizontal recumbent y,z axes, seat back N/A W N/A f N/A N/A Vertical W x-axis, seat back c x-axis, seat back x-axis, seat back N/A W N/A e rx,ry,rz r-axes, seat surface x,ry,rz N/A -axes, seat surface W N/A i Vertical recumbent (head) Vertical recumbent (head) N/A AA6. Combining Vibrations in More Than One Direction The following equation is the total value of weighted RMS acceleration for vibrations in more than one direction: 2 1 222222 )( wzzwyywxxv akakaka ++= a wx, a wy a, wz k are the weighted RMS accelerations with respect to the orthogonal axes x k, y k, z are multiplying factors

86 AA7. Multiplying Factors Table AA2. Multiplying Factor for Seated Persons – Seat Surface Effect of Movement x-axis y-axis z-axis rx -axis ry -axis rz -axis Health Wd W, k=1.4 d W, k=1.4 k , k=1 Comfort Wd W, k=1 d W, k=1 k W, k=1 e W, k=0.63 m/rad e W, k=0.4 m/rad e Perception , k=0.2 m/rad Wd W, k=1 d W, k=1 k , k=1 Table AA3. Multiplying Factor for Seated Persons – Back Rest Effect of Movement x-axis y-axis z-axis Health Comfort Wc W,k=0.8 d W, k=0.5 d Perception , k=0.4 Table AA4. Multiplying Factor for Seated Persons –Feet Effect of Movement x-axis y-axis z-axis Health Comfort Wk W,k=0.25 k W,k=0.25 k Perception ,k=0.4 Table AA5. Multiplying Factor for Standing Persons Effect of Movement x-axis y-axis z-axis Health Comfort Wd W, k=1 d W, k=1 k Perception , k=1 Wd W, k=1 d W, k=1 k , k=1 Table AA6. Multiplying Factor for Recumbent Persons: Under Pelvis Effect of Movement Horizontal Axes Vertical Axes Health Comfort Wd W, k=1 k Perception , k=1 Wd W, k=1 k , k=1

87 APPENDIX AB – EUROPEAN STANDARD ENV 12299:1999 The detail for the analysis methods included in European Standard ENV 12299:1999 are given in this appendix. AB1. Mean Comfort – Simplified Method Comfort Index Calculation for Simplified Method: ))()()((*6 295 2 95 2 95 Wab ZP Wad YP Wad XPMV aaaN ++= W is the weighted frequency value Wab: vertical direction = Wa * Wb Wad: lateral direction= Wa*Wd a is the RMS acceleration subscripts x,y,z indicate direction p indicates floor interface 95 indicates 95 percentile RMS value AB2. Mean Comfort – Complete Method Seated comfort index )(*4)()(*2)(*4 95 2 95 2 9595 Wac XD Wab ZA Wad YA Wab ZPVA aaaaN +++= Standing comfort index )(*5))()(*4)(*16(*3 95 2 50 2 50 2 50 Wad YP Wab ZP Wad YP Wad XPVD aaaaN +++= Weightings Wab = Wa*Wb Wac = Wa*Wc Wad = Wa*Wd Acceleration RMS aXD a : Seat back level YA, aZA a : Seat pan level XP ,aYP, aZP : Floor level AB3. Comfort on Curve Transitions The comfort index gives a measure of passenger comfort for an individual curve transition, referred to as single events without an evaluation of cumulative effects. This measure is applicable to conventional and tilting vehicles at any speed and at medium or high levels of uncompensated lateral acceleration.

88 This index is based on the relationship between the relevant magnitudes of lateral jerk, body roll speed, variation of lateral acceleration level, and the average value of the comfort information given. Comfort Index on Curve Transitions E CT DCyByAP ϑ *)**( +−+= A, B, C, D, E are constants Condition A B C D E In rest – standing 28.54 20.69 11.1 0.185 2.283 In rest – seated 8.97 9.68 5.9 0.120 1.626 PCT y – Comfort Index on curve transitions - Maximum value of lateral acceleration in the carbody averaged on 1s base shifting (1/10) s, in the interval between the beginning of the entry or reverse transition and the end +1.6s y - maximum jerk, evaluated as maximum variation of two subsequent values of lateral acceleration scaled of 1s, in the time interval 1s before the beginning of the entry or reverse transition and the end of the same. Eϑ - maximum absolute value of carbody roll speed 1ϕ averaged on 1 s base shifting by (1/10)s from the beginning to the end of the transition Location of Measurements Lateral accelerations – center of carbody floor and above the leading axle (and trailing if possible) Non-compensated lateral accelerations at the axlebox Carbody roll speed in a suitable location on carbody Tilting angle Speed AB4. Comfort on Discrete Events Comfort Index of discrete events is a measure of passenger comfort resulting from the interaction of the rail vehicle and local track irregularities. cybyaP mPDE −+=  ** PDE a,b,c – constants - Comfort index for discrete events within 2 s intervals shifting by (1/10)s Condition a b c In rest – standing 16.62 27.01 37.0 In rest – seated 8.46 13.05 21.7 Py - difference between maximum and minimum value of lateral accelerations measured within an interval of 2 s my - average value of the lateral acceleration in the same 2 s interval

89 APPENDIX AC – UIC 513 The detail for the analysis methods included in UIC 513 are given in this appendix. AC1. Simplified Method for Seating or Standing Position 2 95 2 95 2 95 )()()(6 Wd ZP Wd YP Wd XPMV aaaN ++= • MVN is the simplified method comfort index • a is the effective value of acceleration • Wd is the frequency weighting in the horizontal direction • P indicates floor • 95 indicates 95th percentile AC2. Full Method in Seated Position )(*4)()(*2)(*4 95 2 95 2 9595 Wc XD Wb ZA Wd YA Wb ZPVA aaaaN +++= VAN is the full method comfort index for seated position • a is the effective value of acceleration • Wb is the frequency weighting in the vertical direction • Wc is the frequency weighting for the seat back • Wd is the frequency weighting in the horizontal direction • A indicates seating surface • D indicates seat back • P indicates floor • 95 indicates 95th percentile AC3. Full Method in Standing Position )(*5)()(*4)(*16*3 95 2 50 2 50 2 50 Wd YP Wb ZP Wd YP Wd XPVD aaaaN +++= VDN is the full method comfort index for standing position • a is the effective value of acceleration • Wb is the frequency weighting in the vertical direction • Wd is the frequency weighting in the horizontal direction • P indicates floor • 95 indicates 95th • 50 indicates 50 percentile th percentile