DELIVERABLE D3.8. PROJECT START DATE February 1, 2005

Transcription

1 TIP4-CT Page 1 of 95 DELIVERABLE D3.8 CONTRACT N TIP4-CT PROJECT N FP ACRONYM QCITY TITLE Quiet City Transport Subproject 3 Vehicle/infrastructure interface related noise Work Package 3.2 Quantification of effect of mitigation measures for wheel/rail noise Mitigation measures for wheels: conceptual designs Passive wheel noise absorbers for tram car resilient wheel in service acoustic tests Written by Steven Cervello LUC Date of issue of this report February 20, 2006 PROJECT CO-ORDINATOR Acoustic Control ACL SE PARTNERS Accon ACC DE Akron AKR BE Amec Spie Rail AMEC FR Alfa Products & Technologies APT BE Banverket BAN SE Composite Damping Material CDM BE Havenbedrijf Oostende HOOS BE Frateur de Pourcq FDP BE Goodyear GOOD LU Head Acoustics HAC SE Heijmans Infra HEIJ BE Royal Institute of Technology KTH SE Vlaamse Vervoersmaatschappij DE LIJN LIJN BE Lucchini Sidermeccanica LUC IT NCC Roads NCC SE Stockholm Environmental & Health Administration SEA SE Société des Transports Intercommunaux de Bruxelles STIB BE Netherlands Organisation for Applied Scientific Research TNO NL Trafikkontoret Göteborg TRAF SE Tram SA TRAM GR TT&E Consultants TTE GR University of Cambridge UCAM UK University of Thessaly UTH GR Voestalpine Schienen VAS AU Zbloc Norden ZBN SE Union of European Railway Industries UNIFE BE PROJECT START DATE February 1, 2005 DURATION 48 months Project funded by the European Community under the SIXTH FRAMEWORK PROGRAMME PRIORITY 6 Sustainable development, global change & ecosystems

2 TIP4-CT Page 2 of 95 TABLE OF CONTENTS 0 Executive Summary Objective of the deliverable Strategy used and/or a description of the methods (techniques) sed with the justification thereof Background info available and the innovative elements which were eveloped Problems encountered Partners involved and their contribution Conclusions Relation with the other deliverables (input/output/timing) Scope References In service acoustic tests Tram configuration Test site Type of tests Measuring equipment Results of on board tests Results of pass-by tests Conclusions...14 Annex I...18 Annex II...41 Annex III...57 Annex IV...75 Annex V...89

3 TIP4-CT Page 3 of 95 0 EXECUTIVE SUMMARY 0.1 OBJECTIVE OF THE DELIVERABLE Objective of the deliverable is to provide an example of application of possible solutions to reduce the noise emission from tramway resilient wheels. 0.2 STRATEGY USED AND/OR A DESCRIPTION OF THE METHODS (TECHNIQUES) USED WITH THE JUSTIFICATION THEREOF Squealing noise typically takes place in curves and is due to dynamic stick-slip effect between tyre and rail. The main noise source is the tyre that vibrates at specific frequencies. The strategy is then to damp directly the tyre vibration by designing passive dampers with an optimal shape tuned on the frequencies of the tyre. 0.3 BACKGROUND INFO AVAILABLE AND THE INNOVATIVE ELEMENTS WHICH WERE DEVELOPED In the case of the work here presented, the wheels where already in service. And the possibility of mounting dampers on the tyre was already foreseen. During the fist year of service noise measurements were performed and one main squealing frequency was identified. The design of the absorbers were then based on this information. 0.4 PROBLEMS ENCOUNTERED The design of a wheel absorber is not straight forward many experimental trials are necessary before finding the best solution. But experience will improve this point. From a constructive point of view the available space for mounting dampers is normally very limited. 0.5 PARTNERS INVOLVED AND THEIR CONTRIBUTION Lucchini 0.6 CONCLUSIONS The conclusion is that a successful solution was developed able to considerably reduce squealing noise from tram way resilient wheels. The work showed that in the case of squealing noise, is normally difficult to design a best solution before having any information on the actual noise that the wheel produce in service.

4 TIP4-CT Page 4 of 95 For this reason is always better to design the final damper after the new wheel has entered into service. On the other hand it s always important to foresee before the mechanical fixing of the dampers to the tyre by designing the tyre with the necessary screw holes. 0.7 RELATION WITH THE OTHER DELIVERABLES (INPUT/OUTPUT/TIMING) This deliverable is a solution to be implemented in the validation sites that will have identified the problem of squealing noise. It is then in relation with deliverables of SP5.

5 TIP4-CT Page 5 of 95 1 SCOPE This report describes the results of the in service experimental tests brought over on the Amsterdam Combino Tram resilient wheels equipped with absorbers developed and manufactured by Lucchini Sidermeccanica (figure 1). The absorbers where developed for this particular wheel to solve the squealing noise problem. Previous in service acoustic tests had found a high noise resonance at 2 khz occurring when approaching curves. The problems found in the development of the absorbers was the very small space left for the application due to the fact that for the powered wheels, the gear was directly linked to the major part of the centrewheel; whereas for the trailer wheels, much space was already occupied by the braking system. Never the less after various optimisation iterations of the absorber s geometry, it was possible to define a solution which proved to be able to considerably reduce squealing noise.

6 TIP4-CT Page 6 of 95 2 REFERENCES ISO/DIS (2001) Railway applications - Acoustics - Measurement of noise emitted by railbound vehicles RS_RV_FWG08_02 Passive wheel noise absorbers for Tram car resilient wheel

7 TIP4-CT Page 7 of 95 3 IN SERVICE ACOUSTIC TESTS The absorbers performance has been tested in service conditions in Amsterdam on the Combino tram N TRAM CONFIGURATION The tyre profiles were newly machined and no lubrication was active on the wheel during the tests. In figure 2 there is the layout of the train set with the position of the powered (PA and PB) and trailer (T) bogies, the wheel position where the 4 microphones where placed (M1, M2, M3 and M4) (see figure 3) and the front driving direction. During the tests, the wheels were equipped with absorbers in different configurations, which are listed in table 1. Configuration A Configuration B Configuration C Table 1 Bogies PA, T and PB equipped with wheel absorbers Bogies T and PB equipped with absorbers and PA without Bogies PA, T and PB equipped without absorbers 3.2 TEST SITE The tests sites name and description are listed in table 2. Curve A Curve B Curve C In the city Table 2 Narrow curve on the ring track inside the GVB maintenance depot (see fig. 4) Medium curve on the ring track inside the GVB maintenance depot (see fig. 4) Large curve on the ring track inside the GVB maintenance depot (see fig. 4) Different curve track conditions around the city In figure 4 there is a sketch of the ring track with three main curves: a narrow one called curve A, a medium radius called B and a larger one called C (on this last one pass by tests where performed.

8 TIP4-CT Page 8 of TYPE OF TESTS Two type of tests have been performed: On board acoustic tests The tram was in configuration B and four microphones were mounted near to four different wheels (see figure 2 and 3): M1 front right powered wheel of bogie PA; M2 front left powered wheel of bogie PA; M3 front right powered wheel of bogie PB; M4 front left powered wheel of bogie PB. Microphone signals where recorded at a sampling frequency of 10 khz after an antialiasing filer set at 5 khz during different running conditions: at curves A and B at normal speeds; at curves C at different speeds; at different curves around the city. Pass by acoustic tests The tram was in two different configuration: A (all wheels with absorbers) and C (all wheels without absorbers). Two fixed microphones were placed in the inner side of curve C with the tram passing at different speeds, direction. The microphones were: M1 placed at 7.5 m from the central track, at 1.5 m height from the ground; M2 placed at 1 m on the ground from the nearer rail. Microphone signals where recorded at a sampling frequency of khz after an antialiasing filer set at 8 khz. The curve track site was part of the ring track inside the GVB maintenance depot and was the one called C according to figure 4.

9 TIP4-CT Page 9 of MEASURING EQUIPMENT The measuring equipment listed in table 3 was used for the tests. M1 Free field microphone B&K 4189 sn M2 Free field microphone B&K 4189 sn M3 Free field microphone B&K 4189 sn M4 Free field microphone B&K 4189 sn Microphone calibrator B&K 4231 sn Calibrated on channel DAC board NI PCI 4452 Portable PC Dolch Flex Pac Acquisition Software Lab View Custom Program Table 3 Before each type test all microphones were calibrated. 3.5 RESULTS OF ON BOARD TESTS In order to analyse the data, all the microphone signals have been first A weighted in the time domain and then processed in the time and frequency domain using a Hanning window of samples and an overlap of 30%, obtaining a coloured map of the sound pressure level with a time resolution of s and a frequency resolution of 9.77 Hz from 0 to 5000 Hz. Following, the Hz frequency component time history is considered as this is where the most important squealing noise takes place. Finally a time history of the over all or equivalent sound pressure level is calculated with a time resolution of s. Results of these analysis are in annex I, II and III which are respectively about measurements along curve A and B, curve C and city track. For each measurement the following graphs are given concerning the microphones that are considered to be more important (M1 or M2 and M3 or M4): Hz frequency component time history comparing microphones sound pressure level; equivalent sound pressure level time history comparing microphones sound pressure level; time frequency sound pressure level coloured map per microphone. Following, table 4 summarises and compares the maximum sound pressure levels of the analysed microphones and average and standard deviation values are calculated. The following graph is an average 1/3 octave band spectra of the noise measured by the four microphones averaged over runs 11, 1, 7, 8.

10 TIP4-CT Page 10 of 95 On board test - Curve A at the depot Lp max at 1934Hz OA Lp max Standard Absorbers Delta Standard Absorbers Delta Run N dba dba dba dba dba dba On board test - Curve B at the depot Lp max at 1934Hz OA Lp max Standard Absorbers Delta Standard Absorbers Delta Run N dba dba dba dba dba dba On board test - Curve C at the depot Lp max at 1934Hz OA Lp max Speed Standard Absorbers Delta Standard Absorbers Delta Run N km/h dba dba dba dba dba dba On board test - around the city Lp max at 1934Hz OA Lp max Standard Absorbers Delta Standard Absorbers Delta Run N dba dba dba dba dba dba Average Standard deviation Table 4

11 TIP4-CT Page 11 of 95 Analysis information Standard: Frequency range: Band width Mode: Time constant: Averaging type: Weight filter: ANSI Hz 1/3 octave Fast 125 ms linear A On board tests - Average levels M1 without absorbers M2 without absorbers M3 with absorbers M4 with absorbers , OA dba (re 2e-5 Pa) 1/3 octave bands (Hz)

12 TIP4-CT Page 12 of RESULTS OF PASS-BY TESTS The same analysis applied for the on board tests is used for the pass-by tests. All the microphone signals have been first A weighted in the time domain and processed in the time and frequency domain using a Hanning window of samples and an overlap of 30%, obtaining a coloured map of the sound pressure level with a time resolution of s and a frequency resolution of 16 Hz from 0 to 8000 Hz. Following the Hz frequency component time history is considered as this is where the most important squealing noise take place. Finally a time history of the over all or equivalent sound pressure level is calculated with a time resolution of s. Results of this analysis are in annex IV that is about pass by measurements along curve C. For each measurement the following graphs concerning the microphones M1 at 7.5 m from the track and M2 at 1 m from the track are given: Hz frequency component time history comparing microphones sound pressure level; equivalent sound pressure level time history comparing microphones sound pressure level; time frequency sound pressure level coloured map per microphone for some of the measurements. Following, table 5 summarises and compares the maximum sound pressure levels of the analysed microphones and average and standard deviation values are calculated. Pass-by test - Curve C at the depot Lp max at 1934Hz OA Lp max Microphone M1 at 7,5 m Microphone M2 at 1 m Microphone M1 at 7,5 m Microphone M2 at 1 m Speed Standard Absorbers Delta Standard Absorbers Delta Standard Absorbers Delta Standard Absorbers Delta Run N km/h dba dba dba dba dba dba dba dba dba dba dba dba < > < > <-- 6/ > 5/ Average Standard deviation Table 5 The following graphs are narrow band spectra of the noise measured by the near microphone over comparing different runs with and without absorbers at different speeds.

13 TIP4-CT Page 13 of 95 Pass by acoustic test without abs run n 1 10 km/h without abs run n 2 10 km/h with absorbers run n 1 10 km/h with absorbers run n 2 10 km/h 5 dba (re 2e-5 Pa) Frequency (Hz) - (16Hz band) Pass by acoustic test without abs run n 5 20 km/h without abs run n 6 20 km/h with absorbers run n 1 20 km/h with absorbers run n 2 20 km/h 5 dba (re 2e-5 Pa) Frequency (Hz) - (16Hz band) Pass by acoustic test without abs run n 9 30 km/h without abs run n km/h 5 dba (re 2e-5 Pa) Frequency (Hz) - (16Hz band)

14 TIP4-CT Page 14 of 95 4 CONCLUSIONS The processing of all the signals recorded during the tests show that the absorbers developed for the Combino tram resilient wheel are able to reduce considerably squealing noise. At the frequency of Hz, where the highest squealing resonance take place, there is: an average reduction of 24 db (Lpmax) when measuring with microphones mounted on the bogie near to the wheel along different type of curves; an average reduction of 34 db (Lpmax) when measuring the pass by noise of the tram first completely equipped with absorbers then without, at a speed of 20 km/h, along curve C. For the overall sound pressure level (Lpeq) there is: an average reduction of 13 db (Lpeq) when measuring with microphones mounted on the bogie near to the wheel along different type of curves; an average reduction of 17 db (Lpeq) when measuring the pass by noise of the tram first completely equipped with absorbers then without, at a speed of 20 km/h, along curve C. The standard deviation in some cases is quite high (see table 4); this is due to the fact that different conditions are considered and also because the squealing noise phenomena is quite complex and depends on many parameters. If we consider other resonance different from the Hz one we could state that on the damped wheel (with absorbers) there are two important frequencies: around 800 Hz and Hz. A further improvement of the absorbers should take into account also these frequencies.

15 TIP4-CT Page 15 of 95 Figure 1

16 TIP4-CT Page 16 of 95 Figure 2

17 TIP4-CT Page 17 of 95 Figure 3 Figure 4

18 TIP4-CT Page 18 of 95 ANNEX I

19 TIP4-CT Page 19 of 95 Combino on board Curve A at depot (run N 1) right curve Time history signal for 1934Hz component M2: Lpmax=128dB M3: Lpmax=83dB Delta =45dB Abstract of STFT map Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , M2: Lpmax=130dB M3: Lpmax=107dB Delta =23dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 3 Plot 4 Plot 5 Plot 6 Cursor ,41 0 0

20 Combino on board Curve A at depot (run N 1) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 20 of Cursor 0 27, ,36 64,

21 Combino on board Curve A at depot (run N 1) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 21 of Cursor 0 27, ,36 58,

22 Combino on board Curve A at depot (run N 1) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 22 of Cursor 0 27, ,36 73,

23 Combino on board Curve A at depot (run N 1) M4 db ref 2,0E-5 Pa 5000 TIP4-CT Page 23 of Cursor 0 27, ,36 63,

24 TIP4-CT Page 24 of 95 Combino on board Curve A at depot (run N 11) right curve Time history signal for 1934Hz component M2: Lpmax=125dB M3: Lpmax=87dB Delta =38dB Abstract of STFT map Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , M2: Lpmax=130dB M3: Lpmax=108dB Delta =22dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 2 Plot 4 Plot 5 Plot 6 Cursor ,70 0 0

25 Combino on board Curve A at depot (run N 11) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 25 of Cursor 0 19, ,59 14,

26 Combino on board Curve A at depot (run N 11) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 26 of Cursor 0 19, ,59 33,

27 Combino on board Curve A at depot (run N 11) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 27 of Cursor 0 19, ,59 35,

28 Combino on board Curve A at depot (run N 11) M4 db ref 2,0E-5 Pa 5000 TIP4-CT Page 28 of Cursor 0 19, ,59 29,

29 TIP4-CT Page 29 of 95 Combino on board Curve A at depot (run N 11) Standard: Frequency range: Band width Mode: Time constant: Averaging type: Weight filter: ANSI Hz 1/3 octave Fast 125 ms linear A right curve On board tests - Annex I - Run 11 Curve A - front direction M1 without absorbers M2 without absorbers M3 with absorbers M4 with absorbers , OA dba (re 2e-5 Pa) 1/3 octave bands (Hz)

30 TIP4-CT Page 30 of 95 Combino on board Curve B at depot (run N 2) right curve Time history signal for 1934Hz component M2: Lpmax=125dB M3: Lpmax=108dB Delta =17dB Abstract of STFT map ,0 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , M2: Lpmax=125dB M3: Lpmax=110dB Delta =15dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 15,0 Plot 4 Plot 5 Plot 6 Cursor ,84 0 0

31 Combino on board Curve B at depot (run N 2) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 31 of Cursor 0 14, ,59 61,

32 Combino on board Curve B at depot (run N 2) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 32 of Cursor 0 14, ,59 68,

33 Combino on board Curve B at depot (run N 2) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 33 of Cursor 0 14, ,59 69,

34 Combino on board Curve B at depot (run N 2) M4 db ref 2,0E-5 Pa 5000 TIP4-CT Page 34 of Cursor 0 14, ,59 69,

35 TIP4-CT Page 35 of 95 Combino on board Curve B at depot (run N 12) right curve Time history signal for 1934Hz component M2: Lpmax=126dB M3: Lpmax=109dB Delta =17dB Abstract of STFT map ,0 25,0 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 15,00 79, M2: Lpmax=126dB M3: Lpmax=112dB Delta =14dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 15,0 25,0 Plot 4 Plot 5 Plot 6 Cursor 0 15,00 95,35 0 0

36 Combino on board Curve B at depot (run N 12) M1 TIP4-CT Page 36 of 95 db ref 2,0E-5 Pa Cursor 0 24, ,59 60,

37 Combino on board Curve B at depot (run N 12) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 37 of Cursor 0 24, ,59 60,

38 Combino on board Curve B at depot (run N 12) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 38 of Cursor 0 24, ,59 54,

39 Combino on board Curve B at depot (run N 12) M4 db ref 2,0E-5 Pa 5000 TIP4-CT Page 39 of Cursor 0 24, ,59 72,

40

41 TIP4-CT Page 41 of 95 ANNEX II

42 TIP4-CT Page 42 of 95 Combino on board 30 km/h (run N 9) Curve CLeft curve back direction Time history signal for 1934Hz component M2: Lpmax=124dB M3: Lpmax=98dB Delta =26dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 11,34 113, M2: Lpmax=130dB M3: Lpmax=112dB Delta =18dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , Combino on board 30 km/h (run N 9)

43 TIP4-CT Page 43 of 95 M1 db ref 2,0E-5 Pa Cursor 0 27, ,59 54,

44 Combino on board 30 km/h (run N 9) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 44 of Cursor 0 27, ,59 68,

45 TIP4-CT Page 45 of 95 Combino on board 25 km/h (run N 8) Curve CRight curve front direction Time history signal for 1934Hz component M2: Lpmax=110dB M3: Lpmax=110dB Delta =0dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 11,34 60, M2: Lpmax=115dB M3: Lpmax=110dB Delta =5dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor ,49 0 0

46 Combino on board 25 km/h (run N 8) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 46 of Cursor 0 27, ,59 73,

47 Combino on board 25 km/h (run N 8) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 47 of Cursor 0 27, ,59 61,

48 TIP4-CT Page 48 of 95 Combino on board 25 km/h (run N 7) Curve CRight curve front direction Time history signal for 1934Hz component M2: Lpmax=120dB M3: Lpmax=91dB Delta =29dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , M2: Lpmax=125dB M3: Lpmax=110dB Delta =15dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor ,93 0 0

49 Combino on board 25 km/h (run N 7) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 49 of Cursor 0 27, ,59 65,

50 Combino on board 25 km/h (run N 7) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 50 of Cursor 0 27, ,59 70,

51 TIP4-CT Page 51 of 95 Combino on board 20 km/h (run N 6) Curve CLeft curve back direction Time history signal for 1934Hz component M2: Lpmax=110dB M3: Lpmax=102dB Delta =8dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , M2: Lpmax=112dB M3: Lpmax=109dB Delta =3dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor ,19 0 0

52 Combino on board 20 km/h (run N 6) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 52 of Cursor 0 27, ,59 80,

53 Combino on board 20 km/h (run N 6) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 53 of Cursor 0 27, ,59 67,

54 TIP4-CT Page 54 of 95 Combino on board 20 km/h (run N 5) Curve CRight curve front direction Time history signal for 1934Hz component M2: Lpmax=126dB M3: Lpmax=100dB Delta =26dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor , M2: Lpmax=126dB M3: Lpmax=100dB Delta =26dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor ,57 0 0

55 Combino on board 20 km/h (run N 5) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 55 of Cursor 0 27, ,59 56,

56 Combino on board 20 km/h (run N 5) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 56 of Cursor 0 27, ,59 63,

57 TIP4-CT Page 57 of 95 ANNEX III

58 TIP4-CT Page 58 of 95 Combino on board in the city (run N 1) (inverted direction) Time history signal for 1934Hz component M2: Lpmax=126dB M3: Lpmax=102dB Delta =24dB Abstract of STFT map Roundabout - left 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,76 58, M2: Lpmax=128dB M3: Lpmax=123dB Delta =5dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,83 88,09 0 0

59 Combino on board in the city (run N 1) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 59 of Cursor ,77 73,

60 Combino on board in the city (run N 1) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 60 of Cursor ,77 86,

61 Combino on board in the city (run N 1) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 61 of Cursor ,77 44,

62 Combino on board in the city (run N 1) M4 db ref 2,0E-5 Pa 5000 TIP4-CT Page 62 of Cursor ,77 64,

63 TIP4-CT Page 63 of 95 Combino on board in the city (run N 1) (inverted direction) Standard: Frequency range: Band width Mode: Time constant: Averaging type: Weight filter: ANSI Hz 1/3 octave Fast 125 ms linear A On board tests - Annex III - Run 1 Roundabout - inverted direction Roundabout - left M1 without absorbers M2 without absorbers M3 with absorbers M4 with absorbers , OA dba (re 2e-5 Pa) 1/3 octave bands (Hz)

64 TIP4-CT Page 64 of 95 Combino on board in the city (run N 3) Right curve Straight Left curve (very noisy) Time history signal for 1934Hz component M2: Lpmax=122dB M3: Lpmax=80dB Delta =42dB Abstrac t of ST FT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,76 55, M2: Lpmax=130dB M3: Lpmax=105dB Delta =25dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,83 88,53 0 0

65 Combino on board in the city (run N 3) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 65 of Cursor 0 32, ,59 62,

66 Combino on board in the city (run N 3) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 66 of Cursor 0 32, ,59 60,

67 TIP4-CT Page 67 of 95 Combino on board in the city (run N 7) Left curve Straight Right curve Time history signal for 1934Hz component M2: Lpmax=120dB M3: Lpmax=96dB Delta =24dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,76 70, M2: Lpmax=124dB M3: Lpmax=116dB Delta =8dB Leq time history 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,83 95,54 0 0

68 Combino on board in the city (run N 7) M2 db ref 2,0E-5 Pa 5000 TIP4-CT Page 68 of Cursor 0 32, ,59 90,

69 Combino on board in the city (run N 7) M4 db ref 2,0E-5 Pa 5000 TIP4-CT Page 69 of Cursor 0 32, ,59 58,

70 TIP4-CT Page 70 of 95 Combino on board in the city (run N 7) Standard: Frequency range: Band width Mode: Time constant: Averaging type: Weight filter: ANSI Hz 1/3 octave Fast 125 ms linear A Left curve Straight Right curve On board tests - Annex III - Run 7 Left curve - Straight - Right curve M1 without absorbers M2 without absorbers M3 with absorbers M4 with absorbers , OA dba (re 2e-5 Pa) 1/3 octave bands (Hz)

71 TIP4-CT Page 71 of 95 Combino on board in the city (run N 8) Roundabout - Left Time history signal for 1934Hz component M2: Lpmax=134dB M3: Lpmax=85dB Delta =49dB Abstract of STFT map 1 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,76 59, M2: Lpmax=133dB M3: Lpmax=122dB Delta =11dB Leq time history 1 13 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 2,83 88,78 0 0

72 Combino on board in the city (run N 8) M1 db ref 2,0E-5 Pa 5000 TIP4-CT Page 72 of Cursor 0 32, ,53 58,

73 Combino on board in the city (run N 8) M3 db ref 2,0E-5 Pa 5000 TIP4-CT Page 73 of Cursor 0 32, ,53 54,

74 TIP4-CT Page 74 of 95 Combino on board in the city (run N 8) Standard: Frequency range: Band width Mode: Time constant: Averaging type: Weight filter: ANSI Hz 1/3 octave Fast 125 ms linear A Roundabout - Left On board tests - Annex III - Run 8 Roundabout - Left M1 without absorbers M2 without absorbers M3 with absorbers M4 with absorbers , OA dba (re 2e-5 Pa) 1/3 octave bands (Hz)

75 TIP4-CT Page 75 of 95 ANNEX IV

76 TIP4-CT Page 76 of 95 Combino no absorbers. Pass by at the depot curve C (30km/h) (run N 10) Left curve Back direction Time history signal for 1934Hz component M2: Lpmax=84dB M3: Lpmax=101dB Delta =17dB Abstract of STFT map 2 3 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 28, M2: Lpmax=88dB M3: Lpmax=104dB Delta =16dB Leq time history 5 3 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 58,31 0 0

77 TIP4-CT Page 77 of 95 Combino no absorbers. Pass by at the depot curve C (30km/h) (run N 10) M1 (7.5 m) db ref 2,0E-5 Pa Cursor 0 15, ,00 50,

78 TIP4-CT Page 78 of 95 Combino no absorbers. Pass by at the depot curve C (30km/h) (run N 10) M2 (1 m) db ref 2,0E-5 Pa Cursor 0 15, ,00 55,

79 TIP4-CT Page 79 of 95 Combino no absorbers. Pass by at the depot curve C (30km/h) (run N 9) Time history signal for 1934Hz component M2: Lpmax=74dB M3: Lpmax=92dB Delta =18dB Abstract of STFT map Right curve Front direction 2 3 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 17, M2: L pmax =81dB M3: L pmax =96dB Delta =15dB Leq time history 5 3 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 55,15 0 0

80 TIP4-CT Page 80 of 95 Combino no absorbers. Pass by at the depot curve C (25km/h) (run N 8) Time history signal for 1934Hz component M2: Lpmax=88dB M3: Lpmax=104dB Delta =16dB Abstract of STFT map Left curve Back direction 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 24, M2: Lpmax=94dB M3: Lpmax=108dB Delta =14dB Leq time history 5 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 53,05 0 0

81 TIP4-CT Page 81 of 95 Combino no absorbers. Pass by at the depot curve C (25km/h) (run N 7) Time history signal for 1934Hz component M2: Lpmax=88dB M3: Lpmax=100dB Delta =12dB A bstrac t of ST FT map Right curve Front direction 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 31, M2: Lpmax=89dB M3: Lpmax=103dB Delta =14dB Leq time history 5 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 54,45 0 0

82 TIP4-CT Page 82 of 95 Combino no absorbers. Pass by at the depot curve C (20km/h) (run N 6) Time history signal for 1934Hz component M2: Lpmax=92dB M3: Lpmax=105dB Delta =13dB Abstract of STFT map Left curve Back direction 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 30, M2: Lpmax=94dB M3: Lpmax=107dB Delta =13dB Leq time history 5 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 53,05 0 0

83 TIP4-CT Page 83 of 95 Combino no absorbers. Pass by at the depot curve C (20km/h) (run N 5) Time history signal for 1934Hz component M2: Lpmax=82dB M3: Lpmax=97dB Delta =15dB Abstract of STFT map Right curve Front direction 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 28, M2: Lpmax=87dB M3: Lpmax=100dB Delta =13dB Leq time history 5 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 54,52 0 0

84 TIP4-CT Page 84 of 95 Combino with absorbers. Pass by at the depot curve C (20km/h) (run N 2) Time history signal for 1934Hz component M2: Lpmax=53dB M3: Lpmax=74dB Delta =21dB Abstract of STFT map Left curve Back direction 2 3 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 23, M2: Lpmax=74dB M3: Lpmax=89dB Delta =15dB Leq time history 5 3 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 53,67 0 0

85 TIP4-CT Page 85 of 95 Combino with absorbers. Pass by at the depot curve C (20km/h) (run N 2) M1 (7.5 m) db ref 2,0E-5 Pa Cursor 0 8, ,00 43,

86 TIP4-CT Page 86 of 95 Combino no absorbers. Pass by at the depot curve C (20km/h) (run N 2) M2 (1 m) db ref 2,0E-5 Pa Cursor 0 8, ,00 53,

87 TIP4-CT Page 87 of 95 Combino with absorbers. Pass by at the depot curve C (20km/h) (run N 1) Time history signal for 1934Hz component M2: Lpmax=54dB M3: Lpmax=72dB Delta =18dB Abstract of STFT map Right curve Front direction 2 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,69 33, M2: Lpmax=74dB M3: Lpmax=90dB Delta =16dB Leq time history 5 Plot 0 Plot 1 Plot 2 Plot 3 Plot 4 Plot 5 Plot 6 Cursor 0 1,72 66,39 0 0

88

89 TIP4-CT Page 89 of 95 ANNEX V Scope The scope of this laboratory experimental analysis is to perform vibration measurements able to compare the vibro-acoustic parameter levels of a tram car resilient wheel in different configurations. The different configurations are to take into account the two levels of wheel stiffness and the effect of applying a set of passive dampers absorbers specially developed to damp a critical resonance at 2000 Hz verified to produce a high level of squeal in servce. There 5 tested configuration are: A. Powered wheel (66L101_1A rev.1) with N 22 original rubber blocks (CRS/DP_002 rev.0). B. Powered wheel (66L101_1A rev.1) with N 22 original rubber blocks (CRS/DP_002 rev.0) and 1 type absorbers (66L101_1a_ABSO). C. Powered wheel (66L101_1A rev.1) with N 22 original rubber blocks (CRS/DP_002 rev.0) and 2 type absorbers. D. Powered wheel (66L101_1A rev.2) with N 26 new rubber blocks (66L101_HTS rev.2). E. Powered wheel (66L101_1A rev.2) with N 26 new rubber blocks (66L101_HTS rev.2) and 2 type absorbers. The result of this report will be the basis to start an investigation on how an active damper could be developed to be more effective than the passive one also at lower frequencies like the 800Hz one. The passive absorbers The absorbers developed for this resilient wheel (fig. 1) are plates made of two stainless steel layers coupled together with a thin visco-elastic and bi-adhesive polymer. The plates are bolted to the lateral side of the tyre which is considered to be the dominant noise source of the wheel. The shape of the absorber is iteratively modified in order to reach modal frequencies near to the tyre ones and so able to damp the tyre fundamental modes. In this sense these absorbers can be considered to be tuned passive absorbers. Test set-up The tests are performed on a complete powered wheelset (two powered wheels and one axle bridge) (fig. 2). The wheel set axle bridge leans through two rubber block on the ground of the laboratory.

90 TIP4-CT Page 90 of 95 The exiting force is an impulse in two fixed positions, first on the rim rolling surface, radial direction, then laterally on the tyre flange. The acceleration measurements are done by one triaxial accelerometer applied on the rim, external side. One of the two wheels if equipped alternatively with different type of rubber blocks and absorbers to have the different 5 configurations previously described. Results Test results are given in the form of Frequency Response Functions (FRFs) between the force excitation and the acceleration response (m/s 2 /N). In the next pages, FRF plots are given first for radial excitation tests and then lateral flange excitation tests. Comparison are done first for the present existing in service configurations (A, B and C) with original rubber blocks, then the new configurations (D and E) are added. Special attention is given to the resonance around 2000 Hz which was the squealing tone found during the in service preliminary tests. The FRF graphs show that both the absorbers and the new configuration of rubber blocks, should improve the squealing noise problem. It can be seen that especially around 2000 Hz, solution C is better than B and A; and when new rubber blocks are used, solution D is better than A and E is better than B. Comments The result of the design of this type of absorber was satisfactory for the 2000 Hz but attempts to damp also the 800 Hz weren t successful. Figure V1 The resilient wheel with 4 passive absorbers mounted on the tyre

91 TIP4-CT Page 91 of 95 Figure V2 Wheelset measuring set-up configuration

92 TIP4-CT Page 92 of 95 Radial excitation comparing: A) standard wheel (green), B) with prototype absorbers (yellow) and C) with serial absorbers (red)

93 TIP4-CT Page 93 of 95 Radial excitation comparing: A) standard wheel with original rubber blocks (green), D) standard wheel with new rubber blocks (yellow), B) wheel with original rubber blocks and prototype absorbers (red), E) wheel with new rubber blocks with serial absorbers (blue)

94 TIP4-CT Page 94 of 95 Lateral flange excitation comparing: A) standard wheel (green), B) with prototype absorbers (yellow) and C) with serial absorbers (red)

95 TIP4-CT Page 95 of 95 Lateral flange excitation comparing: A) standard wheel with original rubber blocks (green), D) standard wheel with new rubber blocks (yellow), B) wheel with original rubber blocks and prototype absorbers (red), E) wheel with new rubber blocks with serial absorbers (blue).