Measurements on the small scale Sterrebeek and full scale Herzele test sections in Belgium

Two small scale 'pilot' test sections have been constructed in Sterrebeek on a private road without traffic to gain experience with the construction method. These construction works demonstrated that a 'regular' asphalt paver can be used for onsite PERS constructions. Then a full scale test track has been constructed on a road with regular traffic in Herzele. The result of the first small scale construction was an inhomogeneous test section as the mix hardened too quickly and the construction had to be done in five parts. Skid resistance was not fulfilling the minimum requirement. The amount of accelerator in the mix influences skid resistance significantly. The more accelerator, the more foaming, the less skid resistance. The result of the second small scale construction was a more homogeneous test section with a slightly improved skid resistance, using a mix with less accelerator. Unfortunately, local unevenness's of the surface were caused by infiltration of water during the construction. Both small scale test sections were monitored yielding the following results: " Initial skid resistance was found to be insufficient based on dynamic measurements. It has been improved by sanding the surface with porphyry 0/2. An amount of 200 g/m² yields sufficient skid resistance. The sandblasting technique may be used as a solution in case skid resistance problems rise on a trafficked road. Attempts to measure sound absorption on drill cores of laboratory samples indicate that possibly absorption is not affected by sanding but more tests are needed to confirm this. " Drill cores taken from the first test section demonstrate a good adhesion to the sub layer. The adhesion is apparently weakened by some repeated action. Settling of the underlayer by the forces exerted by heavy vehicles is the main suspect. " Test slabs made during the second site construction were tested in laboratory: o Measurements on thin and plane sections reveal an average air void content of 21.4 % and a good coating of aggregates and rubber by polyurethane. o A permeability of 3.2 seconds was measured, which is comparable to new two-layer porous asphalt. o Mechanical impedance measurements show an estimated dynamic Young modulus of 526 MPa, which is very high compared to the other tests on PERS that were performed (64 to 208 MPa). o Acoustical absorption was measured on three drill cores. The peak frequencies of were found to be 995, 970 and 980 Hz. The maximum absorption coefficients were found to be 0.93, 0.94 and 0.95. These results are similar to Arnakke. " CPB measurements were performed on the second test sections with a Citroen C1. A noise reductions of 7.6 dB(A) at 50 km/h was measured with respect to an old dense asphalt concrete. This is surprisingly high if one keeps in mind the numerous unevenness's which caused tyre vibrations and extra noise. A small speed dependency was observed (more noise reduction at higher speeds). As CPX measurements could not be performed according to the standard and as the test section was too short, no reliable conclusions can be drawn from these measurements on both test sections. The full scale test section showed among others that: " An ordinary asphalt finisher can be used to get a surface with a nice texture and an acceptable evenness; although for the latter there is scope for improvement which is relatively easy to achieve by more practicing with the laying and the compacting " Porphyry 0/2 sand can be used to enhance the initial skid resistance, but it wears off after some time before the stone aggregates are exposed by wearing off of the polyurethane film. The skid resistance dropped during the first months without becoming dramatically low and then stabilized " The noise reduction is remarkable: and lies between 6 and 9 dB(A). " A CPB measurement with the same vehicle (Citroen C1) as in Sterrebeek showed a similar noise reduction: 7.9 dB(A). " CPX and SPB measurements showed no significant deterioration of the acoustic properties during the nine months the test track existed " Absorption appeared to be relatively low according to field test and quite high according to lab testing (considered the most reliable results). Mechanical impedance relatively high for a poroelastic pavement " Although about 2000 vehicles passed per day over the test track with about 5% heavy vehicles, this test track showed a remarkable good raveling resistance during the 10 months of its existance. No glimpse of local or global raveling was observed as texture or colour change of the pavement. Neither were loose aggregate ever found in the gutter. " In the 8th month of its lifetime some loosening was observed at the edge and in the 9th also some in the first 10 m, apparently caused by settlings in the underlayer which did not resist to the direct exposure of the forces exerted by heavy vehicles. " The situation deteriorated quickly and the experiment had to be aborted in the 10th month after the construction, i.e. June 2015.