Final report on noise and rolling resistance

Transcription

1 PoroElastic Road SUrface: an innovation to Avoid Damages to the Environment EUROPEAN COMMISSION - DG RESEARCH Directorate 1 - Environment Seventh Framework Programme - Contract No Jerzy Ejsmont, Piotr Mioduszewski, Grzegorz Ronowski, Stanisław Taryma, Beata Świeczko-Żurek Document type and No. Deliverable D.6.3 Work Package WP 6 Author(s) Authors' affiliations (acronyms) Contact data for main author Document status and date Dissemination level File Name Jerzy Ejsmont, Piotr Mioduszewski, Grzegorz Ronowski, Stanisław Taryma, Beata Świeczko-Żurek Technical University of Gdańsk (TUG), jejsmont@pg.gda.pl Final Public PERSUADE-TUG-D63-V02-WP Final report on noise and rolling resistance.docx

2 Foreword, acknowledgements and disclaimer The PERSUADE project aims at developing the experimental concept of poroelastic road surfacing (PERS) into a feasible noise-abatement measure as an alternative to, for example, noise barriers. It is expected that PERS may provide substantially higher noise reductions than the best of conventional paving materials. The specific feature of this new type of road surfacing is that it includes a substantial proportion of rubber granulates from recycled car tyres bound with a synthetic resin, such as polyurethane. PERSUADE is the acronym for PoroElastic Road SUrface: an innovation to Avoid Damage to the Environment. The project programme comprises an extensive investigation in the laboratory to develop a durable mixture, the construction of seven test sections in five partner countries, a monitoring effort for the test sections (noise, rolling resistance, skid resistance, durability, winter behaviour, etc.), and a study of all conceivable environmental and economic aspects. The project has been scheduled for a duration of six years. Twelve partners from seven European countries are cooperating in this project, including research institutes, universities and companies representing the involved industry sectors: Belgian Road Research Centre (BRRC), Project coordinator Swedish National Road and Transport Research Institute (VTI) Danish Road Directorate (DRD) NCC Roads AS (NCC) Slovenian National Building and Civil Engineering Institute (ZAG) Polish Road and Bridge Research Institute (IBDiM) Technical University of Gdansk (TUG) Dura Vermeer Infrastructuur NV (DVI) European Tyre Recyclers Association (ETRA) HET Elastomertechnik GmbH ( HET) Katholieke Universiteit Leuven (KULeuven) The French Institute of Science and Technology for Transport, Development and Networks (IFSTTAR) See also the PERSUADE website at This report has been produced within Work Package WP 6 of the project, which deals with Environmental impacts and is led by the Technical University of Gdańsk (TUG). The report has been produced by researchers from the TUG. Only the authors are responsible for the contents of this document. The deliverable has been reviewed by Luc Goubert (BRRC) This work is financed by the European Commission as well as by the Polish National Centre for Research and Development NCBR. By Jerzy Ejsmont et al. I

3 TABLE OF CONTENTS SUMMARY III 1 INTRODUCTION 1 2 LABORATORY TESTS 2 3 ROAD TEST SECTIONS 5 4 RESULTS OF TYRE/ROAD NOISE MEASUREMENTS Comparison of laboratory and road measurements Temperature influence on tyre/road noise on PERS Tyre/road noise of truck tyres tested on PERS 18 5 RESULTS OF ROLLING RESISTANCE MEASUREMENTS Results of laboratory measurements Results of road measurements Temperature influence 33 6 CONCLUSIONS 36 7 REFERENCES 37 By Jerzy Ejsmont et al. II

4 SUMMARY Work Package 6 deals with environmental impacts of PERS use, including noise and rolling resistance. This report covers preliminary noise tests performed in the laboratory and at the road test sites. The obtained results indicate that PERS material reduce noise considerably - up to 12 db in comparison to SMA16 reference surface. Noise reduction properties are especially visible in the case of factory produced PERS slabs supplied by HET Germany. Noise reduction properties of PERS remained almost unchanged after a 3 months long winter time period. Rolling resistance is an important factor in vehicles fuel consumption as well as emission of CO 2 and toxic gases like (NOx, CO). This report covers also rolling resistance tests of PERS performed in the laboratory and on the road. The results indicate that PERS material may reduce rolling resistance in comparison to road surfaces used as references. By Jerzy Ejsmont et al. III

5 1 INTRODUCTION This report summarizes results measurements of tyre/road noise and rolling resistance performed on roadwheel facilities at the laboratory of the Technical University of Gdansk (TUG) in Poland and on road test sections located in Denmark, Sweden and Slovenia. All reported tests were conducted by researchers from TUG. Concerning tyre/road noise this report covers only selected results as most of them are reported in WP5 of the PERSUADE project ( Only a comparison of drum versus road results of tyre/road noise and influence of temperature is covered here. However, in the case of rolling resistance, all relevant results are presented. By Jerzy Ejsmont et al. Page 1

6 2 LABORATORY TESTS TUG has used two out of three of their roadwheel facilities. At first the drum of diameter 1.7 m was equipped with the pre-casted PERS material manufactured by HET (see Fig. 1 and Fig 2). As the 1.7 m drum may only be used for testing passenger car tyres, it was decided to cover with PERS also the 2.0 m drum that may be used for testing of truck tyres (see Fig 3 and Fig. 4). Fig. 1 Mounting of the PERS slabs on the test drum 1.7 m Fig. 2 PERS-HET (on the left) and replica of rough surface dressing APS-4 (on the right) placed on the drum 1.7 m diameter used for rolling resistance and tyre/road noise measurements at TUG. By Jerzy Ejsmont et al. Page 2

7 Fig. 3 Mounting of the PERS slabs on the test drum 2.0 m Fig. 4 Testing or rolling resistance on PERS-HET mounted on drum 2.0 m. By Jerzy Ejsmont et al. Page 3

8 Test conditions for drum measurements are summarized in Tab. 1. It must be observed that during tyre/road noise measurements inflation pressure was adjusted on cold tyre (so called caped inflation), while during rolling resistance measurements inflation was regulated on warm tyre for TUG's conditions or caped for ISO conditions. Drum method of tyre/road noise measurements is based on CPX methodology described in ISO-DIS with exception that they are performed on drum covered with replica road surfaces, while torque drum method of RR measurements is described in [1]. Information about replica road surfaces mounted on the drums is presented in Tab. 2. Tab. 1 Load and inflation pressure during laboratory measurements of tyre/road noise and rolling resistance Tyres Rolling resistance (TUG conditions) Rolling resistance (ISO conditions) Tyre/road noise 80% of the Load Load [N] passenger Index (LI) Inflation pressure [kpa] 210 (reg.) 210 (caped) 200 (caped) Load [N] truck Inflation pressure [kpa] (caped) 700 (caped) Tab. 2 Description of surfaces used on the roadwheel facilities Symbol Surface Type Location Description PERSr17 Poroelastic road surface. Roadwheel Facility 1.7m Factory produced slabs. Porous surface made on the basis of mineral and rubber aggregate and polyurethane resin. Pavement suitable for road and drum use, very smooth and flexible. Still in developing stage. For more details refer to [2]. Poroelastic road surface Roadwheel Facility 2.07m Factory produced slabs. Porous surface made on the basis of mineral and rubber aggregate and polyurethane resin. Pavement suitable for road and drum use, very smooth and flexible. Still in developing stage. For more details refer to [2]. Replica of dense asphalt concrete with 16 mm aggregate Roadwheel Facility 2.0 m Polyester laminate replica made on the basis of a typical DAC 16 mm (rather high texture) Replica of ISO reference surface Roadwheel Facility 2.0 m Polyester laminate replica made on the basis of the reference road surface ISO (average texture) APS4r17 Replica of surface dressing 8/10 mm aggregate Roadwheel Facility 1.7m Polyurethane /mineral replica of a single layer surface dressing 11 mm (very high texture) By Jerzy Ejsmont et al. Page 4

9 3 ROAD TEST SECTIONS Road measurements were carried out on two PERS test sections in Denmark (designated here as PERS-DK1 and PERS-DK2), three test sections/subsections in Sweden (designated as PERS-SE1, PERS-SE2-HET and PERS-SE2-VTI) and one test section in Slovenia (designated as PERS-SL). The Danish test sections (Fig. 5) were located on road No 59 between two towns in Southern Denmark: Kalvehave (east) and Stensved (west). PERS-Dk was constructed with, so a called, PERS-ARNAAKE mixture laid on-site. It was laid to occupy the entire width (about 3.5 m) of the Nothern driving lane with a total length of 75 m. The construction of PERS-DK1 was finished on August 30 th, 2013 and on the same day it was opened to road traffic. Fig. 5 Danish test section designated PERS-DK1 During the winter heavy raveling occurred and the PERS pavement was also damaged by a snow plough. It was therefore decided to remove the first PERS test pavement and construct a new improved PERS pavement at the same location instead. The new PERS section was constructed from June 23rd to June 27th 2014 and it was designated PERS-DK2. By Jerzy Ejsmont et al. Page 5

10 The first Swedish test section reported here and designated PERS-SE1 (Fig. 6) was located on the Landbogatan street in Linkoping city in central Sweden. It was built with prefabricated slabs of PERS material manufactured and supplied by HET, Germany. This section was placed as a 1 m wide strip covering the right wheel-track of the south-bound driving lane. The total length of this test section was 25 m. The construction finished on 29 November The section was opened to road traffic on 3 December. Fig. 6 Swedish test section designated PERS-SE1 (photo taken in December 2013 Southern view on left, and in March 2014 Northern view on right) "The second Swedish test section (Fig. 7) was located on Road E 636 at Sjogestad near Linkoping in central Sweden. It was built in two different test subsections, adjacent to each other. The first test subsection was made of prefabricated slabs of PERS material manufactured and supplied by HET, Germany designated here as PERS-SE2-HET. The total length of this test section is 30 m. The second test section was constructed manually on-site by VTI; this is 24 m long and designated as PERS-SE2-VTI. By Jerzy Ejsmont et al. Page 6

11 Both sections were placed as two 1 m wide strips, each covering the right and left wheel tracks of the west-bound driving lane. Surrounding them, before and after the test sections as well as between the wheel tracks and on the shoulder, is a porous asphalt pavement with 11 mm max. chippings. The regular pavement on this road is an SMA 16, of which the eastbound lane and a few hundred meters of the lane west of the PERS test sections are used as a reference surface. The main construction work was made in the beginning of September It was opened to road traffic on 8 September VTI HET Fig. 7 Swedish test section designated PERS-SE2 The test section in Nova Gorica, Slovenia designated here as PERS-SL is composed of two sub-sections referred to as the pre-test track (PERS-SL-a) and the test track (PERS-SL-b) - see Fig. 8. In contrast to other test sections built within PERSUADE project, this section is paved with concrete blocks that are covered by PERS-HET material that was glued on them indoors. The pre-test track was built before the main test track and is worn a little bit more than the test track. Reference surface before and after the test section is 30 years old asphalt concrete AC11. By Jerzy Ejsmont et al. Page 7

12 PERS-SL-HET1 PERS-SL-HET2 Fig. 8 Slovenian test section. On the first plan, the test-track PERS-SL-b, on the second plane the pre-test track PERS-SL-a Road measurements of tyre/road noise were performed by CPX method using Tiresonic Mk.4 trailer presented in Fig. 9. In 2013 rolling resistance was measured by the R 2 trailer (see Fig. 10) while in 2014 and 2015 new trailer R 2 Mk.2 was used (see Fig. 11). It was noticed, that Trailer R 2 Mk.2 gives more reliable results with less run-to-run variation. Statistics show, that results obtained with new trailer are lower than in case of the old one, and that may be attributed to better aerodynamic screening and more precise mechanisms. Load and inflation pressure during road measurements are summarized in Tab. 3. Fig. 9 Noise tests with Tiresonic Mk.4 trailer on the PERS-DK1 section By Jerzy Ejsmont et al. Page 8

13 Fig. 10 Test trailer R 2 during rolling resistance measurements on the PERS-SE1 section in Linkoping, Sweden. Fig. 11 Test trailer R 2 Mk.2 during rolling resistance measurements in By Jerzy Ejsmont et al. Page 9

14 Tab. 3 Load and inflation pressure during road measurements of tyre/road noise and rolling resistance Tyres Rolling resistance Tyre/road noise Load [N] passenger Inflation pressure [kpa] 210 (reg.) 200 (caped) By Jerzy Ejsmont et al. Page 10

15 4 RESULTS OF TYRE/ROAD NOISE MEASUREMENTS 4.1 Comparison of laboratory and road measurements The same pre-casted PERS material manufactured by HET was used on the roadwheel facility at TUG as was used on test sections in Sweden (located in Linkoping on the Landbogatan street). Noise measurements utilizing the two reference tyres according to the latest version of ISO/TS specification (designated P1 and H1 in the ISO/DIS ) were performed with the speeds of 50 and 80 km/h on all these surfaces. Temperature correction of 0.07 db/ºc (actual air temperature regarding the reference temperature of 20ºC) was applied to all the obtained values. Road tests were conducted twice: when the PERS pavement was new laid and after the winter period. The results of measurements are summarized in Tab. 4. Noise levels obtained for P1 (SRTT) and H1 (AAV4) tyres as well as calculated CPX index (average of noise levels of the two tyres) are also presented in Fig. 12. Tab. 4 Comparison of SPLs obtained during laboratory and road measurements on the test sections applied in Sweden. Noise levels in db(a): Tyre P1 (SRTT) Tyre H1 (AAV4) CPX Index Surface name 50 km/h 80 km/h 50 km/h 80 km/h 50 km/h 80 km/h PERSr17 ( ) 84,8 89,8 85,4 90,0 85,1 89,9 PERS SE1 ( ) 82,2 88,3 85,6 92,1 83,9 90,2 PERS SE1 ( ) 82,7 88,0 83,5 90,4 83,1 89,2 PERS SE2R-HET ( ) 83,5 89,2 84,9 90,8 84,2 90,0 PERS SE2L-HET ( ) 82,5 88,6 84,1 90,1 83,3 89,3 PERS SE2R-HET ( ) 82,4 89,3 82,6 89,9 82,5 89,6 PERS SE2L-HET ( ) 83,8 91,3 82,5 89,8 83,1 90,5 By Jerzy Ejsmont et al. Page 11

16 A-weighted Sound Pressule Level [db(a)] PERSr17 ( ) PERS SE1 ( ) PERS SE1 ( ) PERS SE2R-HET ( ) PERS SE2R-HET ( ) PERS SE2L-HET ( ) PERS SE2L-HET ( ) 76 Tyre P1 Tyre H1 CPX Index Tyre P1 Tyre H1 CPX Index 50 km/h 50 km/h 50 km/h 80 km/h 80 km/h 80 km/h Fig. 12 Comparison of SPLs obtained during laboratory and road measurements Depending on test tyre, speed and particular road test section the differences in noise levels between laboratory and road measurements are up to 2.9 db(a) but the average difference for both tyres and two speeds is only 0.6 db(a). Fig show noise frequency spectra for tyre P1 (SRTT) and H1 (AAV4) tested at the speed of 50 and 80 km/h respectively. The figures do not indicate any significant differences between all PERS sections with one exception. Only the PERS surface placed on the laboratory drum is characterized by significantly lower levels in the frequency range of Hz. By Jerzy Ejsmont et al. Page 12

17 Sound Pressure Level [db(a)] Tyre P1 (SRTT) km/h Sound Pressure Level [db(a)] PERSr17 ( ) 60 PERS SE1 ( ) PERS SE1 ( ) PERS SE2R-HET ( ) 55 PERS SE2R-HET ( ) PERS SE2L-HET ( ) PERS SE2L-HET ( ) Frequency [Hz] Fig. 13 Noise frequency spectra for tyre P1 (SRTT) at 50 km/h Tyre P1 (SRTT) km/h PERSr17 ( ) PERS SE1 ( ) PERS SE1 ( ) PERS SE2R-HET ( ) PERS SE2R-HET ( ) PERS SE2L-HET ( ) PERS SE2L-HET ( ) Frequency [Hz] Fig. 14 Noise frequency spectra for tyre P1 (SRTT) at 80 km/h By Jerzy Ejsmont et al. Page 13

18 Sound Pressure Level [db(a)] Tyre H1 (AAV4) km/h Sound Pressure Level [db(a)] PERSr17 ( ) PERS SE1 ( ) 60 PERS SE1 ( ) PERS SE2R-HET ( ) 55 PERS SE2R-HET ( ) PERS SE2L-HET ( ) PERS SE2L-HET ( ) Frequency [Hz] Fig. 15 Noise frequency spectra for tyre H1 (AAV4) at 50 km/h Tyre H1 (AAV4) km/h PERSr17 ( ) PERS SE1 ( ) PERS SE1 ( ) PERS SE2R-HET ( ) PERS SE2R-HET ( ) PERS SE2L-HET ( ) PERS SE2L-HET ( ) Frequency [Hz] Fig. 16 Noise frequency spectra for tyre H1 (AAV4) at 80 km/h By Jerzy Ejsmont et al. Page 14

19 4.2 Temperature influence on tyre/road noise on PERS An experiment to determine the temperature influence on noise generated by car tyres when rolling on the PERS surface was conducted on the roadwheel facility at the Technical University of Gdańsk in Poland. The experiment was described in detail in the PERSUADE Technical Report R6.1 see [3]. The pre-casted PERS material manufactured and delivered by HET, Germany was placed on the test drum of 1.7m diameter. The air temperature in the laboratory room was controlled within a range from 5º to 37ºC using a climate control unit (chiller/heater). Together with air temperature, the PERS surface temperature was acquired during measurements. It was changing within a range from 8º to 40ºC. Twelve car tyres were selected for this purpose. Among the selected tyres there were two sets (differing in rubber hardness) of proposed reference tyres for CPX method (designated P1 and H1 in the ISO standard) and eight other tyres including tyres specially designed for electric vehicles. For detailed description of the tested tyres please refer to [3]. The derived temperature correction factors, separate for three test speeds and each test tyre, are summarized in Tab. 5 (for air temperature) and Tab. 6 (for PERS surface temperature). The emoticons indicate correlation between temperature and noise levels according to the values given in Tab. 7. Tab. 5 Air temperature correction factors CT t derived for tested tyres Tyre Air temperature 50 km/h 80 km/h 100 km/h CT t [db/ºc] R 2 CT t [db/ºc] R 2 CT t [db/ºc] R (SRTT) -0,053 0,95-0,031 0,93-0,026 0, (AAV4) -0,089 0,91-0,065 0,95-0,083 0, (SRTT) -0,046 1,00-0,017 0,33-0,045 0, (AAV4) -0,099 0,97-0,084 0,93-0,073 0, ,057 0,89-0,010 0,84-0,027 0, ,050 0,99-0,038 0,90-0,009 0, ,032 0,97-0,025 0,95-0,015 0, ,057 0,86-0,026 0,56-0,026 0, ,028 0,80-0,012 0,55-0,021 0, ,018 0,87-0,034 0,80-0,034 0, ,025 0,79-0,019 0,46 0,011 0, ,024 0,64-0,015 0,85-0,034 0,95 By Jerzy Ejsmont et al. Page 15

20 Tab. 6 PERS surface temperature correction factors CT t derived for tested tyres Tyre Surface temperature 50 km/h 80 km/h 100 km/h CT t [db/ºc] R 2 CT t [db/ºc] R 2 CT t [db/ºc] R (SRTT) -0,057 0,96-0,033 0,94-0,028 0, (AAV4) -0,091 0,95-0,065 0,96-0,092 0, (SRTT) -0,046 1,00-0,016 0,30-0,045 0, (AAV4) -0,106 0,99-0,088 0,90-0,075 0, ,062 0,88-0,010 0,85-0,030 0, ,055 0,99-0,043 0,92-0,010 0, ,035 0,95-0,027 0,95-0,016 0, ,057 0,87-0,027 0,59-0,027 0, ,030 0,86-0,012 0,49-0,022 0, ,019 0,85-0,038 0,78-0,038 0, ,028 0,80-0,021 0,47 0,012 0, ,026 0,64-0,016 0,83-0,037 0,97 Tab. 7 Evaluation of coefficient of determination R 2 Coefficient of determination R 2 Unsatisfactory R 2 < 0.5 Poor 0.5 < R 2 < 0.6 Satisfactory 0.6 < R 2 < 0.8 Good 0.8 < R 2 < 0.9 Very good R 2 > 0.9 One can observe that for higher speeds (80 and 100 km/h) for two tyres, designated 1071 and 1112, the correlation between temperature and noise emission is unsatisfactory or poor the coefficient of determination R 2 is very low. A similar behaviour can be observed also for tyre 1077 (SRTT) and 1076, but only when tested at 80 km/h. In such a case one cannot talk about any correlation, so the calculated factors have been excluded from further analysis. All other tested tyres present very good, good or at least satisfactory correlation. The average value of coefficient of determination R 2 for all tyres with acceptable correlation is 0.88 (from 0.64 to 1.00). The values of calculated temperature correction factors CT t slightly differ depending on measured temperature (air or PERS surface), test speed and, of course differ much more depending on the selected tyre. The lowest absolute value is db/ºc, the By Jerzy Ejsmont et al. Page 16

21 highest is db/ºc. All derived values are summarized in Tab. 8. The results indicate that the temperature correlation varies from tyre type to tyre type. Tab. 8 Summary of temperature correction factors CT t derived for tested tyres with acceptable correlation Air temperature PERS surface temperature Tyre 50 km/h 80 km/h 100 km/h Average 50 km/h 80 km/h 100 km/h Average CT t [db/ºc] CT t [db/ºc] 1097 (SRTT) -0,053-0,031-0,026-0,037-0,057-0,033-0,028-0, (AAV4) -0,089-0,065-0,083-0,079-0,091-0,065-0,092-0, (SRTT) -0,046-0,045-0,046-0,046-0,045-0, (AAV4) -0,099-0,084-0,073-0,085-0,106-0,088-0,075-0, ,057-0,010-0,027-0,031-0,062-0,010-0,030-0, ,050-0,038-0,009-0,032-0,055-0,043-0,010-0, ,032-0,025-0,015-0,024-0,035-0,027-0,016-0, ,057-0,057-0,057-0, ,028-0,021-0,025-0,030-0,022-0, ,018-0,034-0,034-0,029-0,019-0,038-0,038-0, ,025-0,025-0,028-0, ,024-0,015-0,034-0,025-0,026-0,016-0,037-0,026 Average: -0,048-0,038-0,037-0,041-0,051-0,040-0,039-0,043 The temperature effect on noise emission of P1 (SRTT) and H1 (AAV4) tyres was analyzed separately. Due to extremely low correlation of tyre 1077 at 80 km/h this value has been excluded from the analysis. The summary of temperature correction factors derived for ISO reference tyres is presented in Tab. 9. Tab. 9 Summary of temperature correction factors CT t derived for ISO reference tyres Air temperature PERS surface temperature Tyre 50 km/h 80 km/h 100 km/h Average 50 km/h 80 km/h 100 km/h Average CT t [db/ºc] CT t [db/ºc] 1097 (SRTT) -0,053-0,031-0,026-0,037-0,057-0,033-0,028-0, (AAV4) -0,089-0,065-0,083-0,079-0,091-0,065-0,092-0, (SRTT) -0,046-0,045-0,046-0,046-0,045-0, (AAV4) -0,099-0,084-0,073-0,085-0,106-0,088-0,075-0,090 Average: -0,072-0,060-0,057-0,062-0,075-0,062-0,060-0,064 One can observe that the temperature correction factors for tyres P1 (SRTT) and H1 (AAV4) differ very much the values for tyre P1 are about a half of the values for H1 tyre. Thus, By Jerzy Ejsmont et al. Page 17

22 according to the authors, separate temperature correction factors CT t should be used for PERS surface in the temperature correction procedure. The values, averaged for two tyres, are given in Tab. 10 and Tab. 11. Tab. 10 Temperature correction factors CT t derived for P1 (SRTT reference tyre) Air temperature PERS surface temperature Tyre 50 km/h 80 km/h 100 km/h Average 50 km/h 80 km/h 100 km/h Average CT t [db/ºc] CT t [db/ºc] 1097 (SRTT) -0,053-0,031-0,026-0,037-0,057-0,033-0,028-0, (SRTT) -0,046-0,045-0,046-0,046-0,045-0,046 Average: -0,050-0,031-0,035-0,041-0,052-0,033-0,036-0,042 Tab. 11 Temperature correction factors CT t derived for H1 (AAV4 reference tyre) Air temperature PERS surface temperature Tyre 50 km/h 80 km/h 100 km/h Average 50 km/h 80 km/h 100 km/h Average CT t [db/ºc] CT t [db/ºc] 1087 (AAV4) -0,089-0,065-0,083-0,079-0,091-0,065-0,092-0, (AAV4) -0,099-0,084-0,073-0,085-0,106-0,088-0,075-0,090 Average: -0,094-0,074-0,078-0,082-0,098-0,076-0,084-0,086 The following conclusions can be drawn based on the performed experiment on air temperature influence on CPX measurements on PERS pavements: 1. The temperature correction factor derived for all tyres with acceptable correlation is changing within a range from to with the average value of There is a big difference in temperature correction factor for tyres P1 (SRTT) and H1 (AAV4) the values for tyre P1 are about a half of the values for H1 tyre. 3. The average temperature correction factor for tyre P1 (SRTT) is The average temperature correction factor for tyre H1 (AAV4) is Tyre/road noise of truck tyres tested on PERS Seven truck tyres were tested on the roadwheel facility with 2.0 m drum. The tyres used in the experiment are presented in Tab. 12. By Jerzy Ejsmont et al. Page 18

23 Tab. 12 Parameters of truck tyres tested on the roadwheel facility Designation Manufacturer Tread Size Date code Remarks 1084 DUNLOP SP /65R K BRIDGESTONE R /65R K SAVA CIT U3 275/70R /145J 4208/0314 retreaded 1116 DUNLOP SP372 CITY 275/70R /145J MICHELIN XDU 275/70R /145J 1210 retreaded 1118 BRIDGESTONE M /70R /145J 3807/3114 retreaded 1119 DUNLOP SP372 CITY 275/70R /145J 4309 retreaded Noise measurements of truck tyres were performed on the PERS surface () and two other reference surfaces ( and ) at the speeds of 50 and 80 km/h. Results of the measurements are presented in Tab. 13 and in Fig Tab. 13 Results of noise measurements A-weighted Sound Pressure Levels of truck tyres Noise levels [db(a)]: Speed [km/h] Speed [km/h] Speed [km/h] Truck tyre DUNLOP SP242 83,3 90,2 91,6 99,1 93,7 101, BRIDGESTONE R168 81,9 88,6 91,2 98,3 93,2 101, SAVA CIT U3 83,4 90,7 90,0 97,1 91,6 99, DUNLOP SP372 CITY 82,6 89,6 92,0 98,9 93,7 101, MICHELIN XDU 84,1 91,6 90,4 97,7 91,9 99, BRIDGESTONE M758 89,8 96,9 93,6 100,3 94,7 102, DUNLOP SP372 CITY 83,9 91,5 90,8 98,3 92,2 100,0 The was the noisiest surface for the tested truck tyres. Depending on the tyre, the difference between and surfaces was in a range from 4.9 db up to 12.8 db independently on test speed. For the surface the difference was from 3.3 db up to 9.7 db. Frequency spectra of the truck tyres measured at 80 km/h are shown in Fig. 19. Noise reduction properties of PERS surface can be observed in the whole frequency range. By Jerzy Ejsmont et al. Page 19

24 50 km/h Sound Pressure Level [db(a)] Tyre 1084 Tyre 1085 Tyre 1114 Tyre 1116 Tyre 1117 Tyre 1118 Tyre 1119 Fig. 17 A-weighted Sound Pressure Levels of truck tyres at 50 km/h 80 km/h Sound Pressure Level [db(a)] Tyre 1084 Tyre 1085 Tyre 1114 Tyre 1116 Tyre 1117 Tyre 1118 Tyre 1119 Fig. 18 A-weighted Sound Pressure Levels of truck tyres at 80 km/h By Jerzy Ejsmont et al. Page 20

25 Sound Pressure Level [db(a)] Sound Pressure Level [db(a)] Tyre 1084 Tyre km/h km/h Frequency [Hz] Frequency [Hz] Tyre 1114 Tyre km/h km/h Sound Pressure Level [db(a)] Sound Pressure Level [db(a)] Frequency [Hz] Frequency [Hz] Tyre 1117 Tyre km/h km/h Sound Pressure Level [db(a)] Sound Pressure Level [db(a)] Frequency [Hz] Frequency [Hz] Tyre km/h Sound Pressure Level [db(a)] Frequency [Hz] Fig. 19 Noise frequency spectra for truck tyres at 80 km/h By Jerzy Ejsmont et al. Page 21

26 5 RESULTS OF ROLLING RESISTANCE MEASUREMENTS 5.1 Results of laboratory measurements During drum measurements 16 different car tyres were tested on PERS-HET and three replica road surfaces described in Tab. 2. Basic tests were performed according to the ISO methodology (ISO 18164), that is with tyres loaded to 80% of max. load (LI) and inflated to 210 kpa, but instead of steel drum surface, appropriate replica road surfaces were used. Tests were performed at 30, 50, 80 and 100 km/h. Set of tested tyres included three tyres considered to be "informal reference tyres", that is tyre Michelin PRIMACY HP 225/60R16, Uniroyal TIGER PAW (SRTT) P225/60R16 and Avon SUPERVAN AV4 195R14C. Results of measurements are presented in Tab. 14 and Tab. 15. In Figs the data obtained on different surfaces for passenger car tyres are compared for all test speeds. It is clearly visible, that rolling resistance of passenger car tyres rolling on PERS-HET is higher than for smooth and medium textured road surfaces ( and ) but at the same time it is clearly lower than for very rough replica of surface dressing. Tab. 14 Coefficients of Rolling Resistance of passenger car tyres tested on the 1.7 m drum facility TYRE PERSr17 APS4r17 DESIGNATION MANUFACTURER MODEL SIZE T-1068 MICHELIN PRIMACY HP 225/60R16 98V T-1077 UNIROYAL TIGER PAW P225/60R16 97S M+S T-1063 AVON SUPERVAN AV4 195R14C 106/104N T-1067 CONTINENTAL CONTIECOCONTACT 5 195/60R15 88H T-1071 VREDESTEIN QUATRAC 3 195/60R15 88V M+S T-1066 WANLI S /60R15 88H M+S T-1072 YOKOHAMA Wdrive 195/50R15 88T M+S T-1070 WREDENSTEIN WINTRAC NEXTREME 205/55R16 94V M+S T-1064 MICHELIN PRIMACY HP 225/60R16 98V T-1069 VREDESTEIN SLICK 235/45R17 94W T-1073 AVON ZV5 195/65R15 91V T-1078 GOODYEAR EFFICIENT GRIP PERFORMANCE 195/65R15 91H T-1079 BRIDGESTONE ECOPIA EP001S 195/65R15 91H T-1080 MICHELIN ENERGY SAVER X GREEN 215/55R17 94H T-1081 DUNLOP SPORT BLURESPONSE 195/65R15 91H T-1082 MICHELIN ENERGY SAVER + Extra Load X GREEN 195/65R15 95T T-1073 AVON ZV5 195/65R15 91V Averaged By Jerzy Ejsmont et al. Page 22

27 Tab. 15 Coefficients of Rolling Resistance of passenger car tyres tested on the 2.0 m drum facility TYRE DESIGNATION MANUFACTURER MODEL SIZE T-1068 MICHELIN PRIMACY HP 225/60R16 98V T-1077 UNIROYAL TIGER PAW P225/60R16 97S M+S T-1063 AVON SUPERVAN AV4 195R14C 106/104N T-1067 CONTINENTAL CONTIECOCONTACT 5 195/60R15 88H T-1071 VREDESTEIN QUATRAC 3 195/60R15 88V M+S T-1066 WANLI S /60R15 88H M+S T-1072 YOKOHAMA Wdrive 195/50R15 88T M+S T-1070 WREDENSTEIN WINTRAC NEXTREME 205/55R16 94V M+S T-1064 MICHELIN PRIMACY HP 225/60R16 98V T-1069 VREDESTEIN SLICK 235/45R17 94W T-1073 AVON ZV5 195/65R15 91V T-1078 GOODYEAR EFFICIENT GRIP PERFORMANCE 195/65R15 91H T-1079 BRIDGESTONE ECOPIA EP001S 195/65R15 91H T-1080 MICHELIN ENERGY SAVER X GREEN 215/55R17 94H T-1081 DUNLOP SPORT BLURESPONSE 195/65R15 91H T-1082 MICHELIN ENERGY SAVER + Extra Load X GREEN 195/65R15 95T T-1073 AVON ZV5 195/65R15 91V Averaged Fig. 20 Relation between rolling resistance on PERS-HET and different replica road surfaces at 30 km/h. By Jerzy Ejsmont et al. Page 23

28 Fig. 21 Relation between rolling resistance on PERS-HET and different replica road surfaces at 50 km/h. Fig. 22 Relation between rolling resistance on PERS-HET and different replica road surfaces at 80 km/h. By Jerzy Ejsmont et al. Page 24

29 Fig. 23 Relation between rolling resistance on PERS-HET and different replica road surfaces at 100 km/h. Also 6 truck tyres were tested on the roadwheel facility with 2.0 m drum. Truck tyres were tested only at 50 and 80 km/h. Results of measurements are presented in Tab. 16 as well as in Fig. 24 and Fig. 25. In the case of truck tyres rolling resistance on PERS-HET is considerably higher (30-40%) than on other medium or low textured replicas. In order to investigate reason for a substantial increase of rolling resistance for truck tyres rolling on PERS-HET thermal images of the drum and tyre were taken as they help to find where the energy loses are located. The thermograms for truck tyre rolling on and PERS-HET are presented in Fig. 26. Tab. 16 Coefficients of Rolling Resistance of truck tyres tested on the 2.0 m drum facility TYRE Designation Manufacturer Model Size T-1114 SAVA CIT U3 275/70R22,5 148/145J T-1115 MADE IN CHINA GT /70R22,5 148/145J T-1116 DUNLOP SP372 CITY 275/70R22,5 148/145J T-1117 MICHELIN XDU 275/70R22,5 148/145J T-1118 BRIDGESTONE M /70R22,5 148/145J T-1119 DUNLOP SP372 CITY 275/70R22,5 148/145J Averaged By Jerzy Ejsmont et al. Page 25

30 Fig. 24 Relation between rolling resistance of truck tyres on PERS-HET and different replica road surfaces at 50 km/h. Fig. 25 Relation between rolling resistance of truck tyres on PERS-HET and different replica road surfaces at 80 km/h. By Jerzy Ejsmont et al. Page 26

31 Drum Tyre Drum Tyre Fig. 26 Thermograms of truck tyre rolling on replica road surface (upper picture) and on PERS-HET (lower picture), speed 80 km/h. By Jerzy Ejsmont et al. Page 27

32 Analyses of Fig. 26 indicate that in the case of truck tyres rolling on PERS-HET a lot of energy dissipates in the PERS material. After warming up on the temperature of tyre shoulder was ºC and max. temperature of replica was 30.6 ºC. For the same conditions when tyre was rolling on PERS-HET its shoulder temperature increased to ºC while max. temperature of the PERS-HET surface was as high as ºC. This means that temperature of the tyre shoulder increased by about 3ºC while temperature of the pavement increased by 15 ºC. Ambient temperature during all tests with thermography was 20 ºC. Truck tyres were tested according to ISO conditions specified in Tab. 1. However, for passenger car tyres the situation is very different - see Fig. 27. After warming up on the temperature of tyre shoulder was 37-38ºC and max. temperature of replica was 27 ºC. For the same conditions when tyre was rolling on PERS-HET its shoulder temperature increased to ºC while max. temperature of the PERS-HET surface increased only to 29 ºC. This means that temperature of the tyre shoulder increased by about 1ºC while temperature of the pavement increased by 2 ºC. Passenger car tyres were tested according to TUG conditions specified in Tab. 1. Results presented above indicate that hardness of PERS-HET surface is well adjusted to the contact pressure provided by passenger car tyres, but is too low for truck tyres. Truck tyres deflect the pavement too much from the point of view of rolling resistance optimisation. 5.2 Results of road measurements TUG tested rolling resistance on the test sections in Denmark (PERS-DK1 and PERS-DK2), Sweden (PERS-SE1, PERS-SE2-VTI, PERS-SE2-HET) and in Slovenia (PERS-SL). Original asphalt concrete surfaces used on the road nearby the test sections were used as reference surfaces. Tests were performed on dry road surfaces with load and inflation pressure set according to Tab. 1. The results with temperature correction are summarized in Tables 17. By Jerzy Ejsmont et al. Page 28

33 Drum Tyre Drum Tyre Fig. 27 Thermograms of passenger car tyre rolling on replica road surface (upper picture) and on PERS-HET (lower picture), speed 80 km/h. By Jerzy Ejsmont et al. Page 29

34 Tab. 17 Coefficients of Rolling Resistance tested by the trailer on the road tests sections (corrected for temperature). Blue colour indicates results obtained with "old' trailer R 2. Pavement Air temp. [ºC] RWT indicates Right Wheel Track, LWT - Left Wheel Track, BWT - Between Wheel Track Graphic representation of the results is presented in Figs It is obvious that data obtained for reference surface adjacent to PERS-SE1 do not fit to other data. For this surface rolling resistance coefficients are much higher than for PERS-SE1. It is very difficult to find the reason for this deviation as the measurements were done in a series and all runs show the same relations - see Fig. 33. Tyre / speed AAV4 MCPR SRTT T1066 T km/h 50 km/h 80 km/h 50 km/h 80 km/h 50 km/h 80 km/h 50 km/h 80 km/h 50 km/h PERS DK PERS Dk2 (2015) Ref. DAC 11 11/ PERS-SE Ref. DAC PERS-SE2-HET, RWT PERS-SE2-HET, LWT PERS-SE2-VTI, RWT PERS-SE2-VTI, LWT Drainage BWT Ref. SMA PERS-SL Ref. AB Fig. 28 Rolling resistance coefficients obtained on PERS-DK1 and reference road surface DAC11 at 80 km/h. By Jerzy Ejsmont et al. Page 30

35 Fig. 29 Rolling resistance coefficients obtained on PERS-SE1 and reference road surface DAC11 at 50 km/h. Fig. 30 Rolling resistance coefficients obtained on PERS-SE2 and reference road surface SMA16 at 50 km/h. By Jerzy Ejsmont et al. Page 31

36 Fig. 31 Rolling resistance coefficients obtained on PERS-SE2 and reference road surface SMA16 at 50 km/h. Fig. 32 Rolling resistance coefficients obtained on PERS-SL and reference road surface at 80 km/h. By Jerzy Ejsmont et al. Page 32

37 DAC 11 PERS Fig. 33 Distance history of 10 rounds of measurements performed on PERS-SE1 and reference road surface DAC11. Tyre AAV4, speed 50 km/h. In all other cases rolling resistance measured on PERS is at about 10% higher than on reference road surfaces. Very similar differences were recorded for all tested tyres. 5.3 Temperature influence Rolling resistance of tyres is very dependent on their temperature. However, it is not very feasible to relate rolling resistance to temperature of the tyre, as each part of a tyre has its own temperature so moving the measuring point by a few centimetres may change the estimation of tyre temperature considerably. Ambient temperature seems to be a better descriptor of tyre temperature than the tyre itself, provided that the tyre temperature is stabilized. In 2014 TUG performed an extensive measuring campaign of temperature influence on rolling resistance. The campaign was partly sponsored by projects ROSANNE [4] and LEO [5]. As the problem of temperature influence is very important for PERSUADE project, selected results of measurements are presented below. Laboratories of TUG have standard heating and A/C systems designed to ascertain that the air temperature near roadwheel facilities during measurements is kept within the range of ºC. In order to perform tests of temperature influence it was necessary to supplement the system with a 50 kw heat pump unit with four portable heat exchangers that were located near the roadwheel facilities. The system made it possible to test tyres in By Jerzy Ejsmont et al. Page 33

38 temperature range 5-40 ºC. During the tests passenger car tyres were loaded according to TUG methodology while truck tyres were loaded according to Tab. 1. Results obtained for passenger car tyres were very similar on all tested surfaces. The influence observed on all surfaces was nearly linear so it was possible to establish coefficients of temperature influence - K t that are valid for the whole tested temperature range. The coefficients are presented in Fig. 34. However, for the truck tyres the situation was different. On stiff replicas, like and the temperature influence was much higher for low temperatures than for high temperatures. This is illustrated in Fig. 35. On PERS-HET temperature influence of truck tyres is rather linear, but what is more interesting, differences between two tyres nearly diminishes - see Fig. 36. It indicates that on PERS rolling resistance of truck tyres is controlled to a great extent by energy losses in the pavement, thus relative differences caused by differences in tyre construction are less important. Fig. 34 Coefficients K t for different passenger car tyres and road surfaces. By Jerzy Ejsmont et al. Page 34

39 Fig. 35 Influence of ambient temperature for two truck tyres tested on replica at 80 km/h. Fig. 36 Influence of ambient temperature for two truck tyres tested on at 80 km/h. By Jerzy Ejsmont et al. Page 35

40 6 CONCLUSIONS The results of noise measurements performed within WP6 indicate that both types of PERS material (on-site produced PERS and HET) provide significant noise reduction in comparison to reference surfaces. However, the HET material shows better noise reduction properties than the on-site produced Mix. When comparing laboratory drum and road CPX measurements of the PERS-HET material the observed differences in noise levels were up to 2.9 db(a) depending on test tyre, speed and particular road test section. The average difference for both tested tyres and two speeds was 0.6 db(a). The performed experiment on air temperature influence on CPX measurements on PERS pavements can be concluded: the temperature correction factor derived for all tyres with acceptable correlation is changing within a range from to with the average value of , there is a big difference in the average temperature correction factors for tyres P1 (SRTT) and H1 (AAV4): and correspondingly. The laboratory measurements of noise emission of truck tyres can be concluded: the was the noisiest drum replica surface, depending on the tyre, the difference between and surfaces was in a range from 4.9 db up to 12.8 db independently on test speed, for the surface the difference was from 3.3 db up to 9.7 db. On PERS-HET rolling resistance of passenger car tyres is lower than on rough surface dressing but higher than on dense asphalt concrete and an SMA16. For truck tyres rolling on PERS-HET the coefficient of rolling resistance is higher by % than in the case of running on dense asphalt concrete. For passenger car tyres averaged coefficient of temperature influence K t = By Jerzy Ejsmont et al. Page 36

41 7 REFERENCES [1] Ejsmont J., Swieczko-Zurek B.: Methods of Tire Rolling resistance Measurements, COTUME 2014, March 2014, Sousse, Tunisia [2] Świeczko-Żurek B., Ejsmont J., Motrycz G., Stryjek P. (2014). Risks related to car fire on innovative Poroelastic Road. Fire and Materials, March 2015, Vol. 39, Issue 2, pp also published online in Wiley Online Library (wileyonlinelibrary.com). DOI: /fam [3] Mioduszewski P., Ejsmont J., Taryma S., Woźniak R.: Temperature influence on tire/road noise on the poroelastic surface evaluated by the drum method. PERSUADE- TUG-R01-V01-WP , Technical Report R6.1, (2015). [4] ROSANNE ROlling resistance, Skid resistance, ANd Noise Emission measurement standards for road surfaces, Project under the 7th Framework Program (FP7/ ) grant agreement No (2013), website: [5] LEO Low Emission Optimised tyres and road surfaces for electric and hybrid vehicles, Project within Polish-Norwegian Research Programme CORE, grant agreement No (2013), website: By Jerzy Ejsmont et al. Page 37