Report

Accelerated load testing of pavements. HVS-NORDIC tests in Sweden 1999

Reseach area: Traffic analysis, Infrastructure maintenance, Road and rail engineering
Year: 2001
VTI-code: R477A
Price: 160 kr

 
VTI Publication R477A (91 pages, 2 131 kB. Written in English)

In 1997 Finland and Sweden jointly invested in a Heavy Vehicle Simulator, HVS, from South Africa. After the first period in Finland the HVS-NORDIC was moved to Sweden and VTI in Linkoping in September 1998. A program for research co-operation in the area of accelerated pavement testing has been agreed between Finland and Sweden covering the years 1997–2003.

The general objective of the research co-operation is to learn more about pavement response and pavement performance. A second but most important objective is to learn about the HVS-machine itself and the way it simulate traffic and deteriorate the pavements and how this is related to real pavement performance.

During the first year in Sweden, 1999, tests were performed as planned in the research program on typical Swedish low volume road structures.

During the second year, 2000, two typical Icelandic pavement structures were tested on request from the Icelandic Road Administration and two steel fabric reinforced structures which were tested as a part of an European research project, REFLEX (Reinforcement of Flexible Road Structures with Steel Fabrics to Prolong Service Life). These tests are and will be reported elsewhere.

The tests during the first year on typical Swedish low volume road structures are presented in this report.

The main objective of these first tests was to get pavement response and pavement performance results that could act as reference in coming tests.

The results presented in this report are data from the documentation of the test structures and data from the measurements of pavement performance during the tests. The aim of the report is rather informative than scientific. The considerable amount of data from the response measurements will be evaluated and presented later in other reports.

The HVS-NORDIC is a mobile linear full-scale accelerated pavement-testing machine (HVS Mark IV), figure 1. The machine can be run over a short distance by itself at walking speed and can be moved as a semi-trailer over longer distances.

The HVS-NORDIC has a heating/cooling system and thus temperature can be held constant. The air inside the insulating box is heated or cooled and controlled in order to keep the pavement temperature constant. The standard pavement temperature is selected at +10°C. The HVS can be run by diesel or by electric power. The diesel engine also provides power for the heating/cooling system and thus it is independent of external power.

The main technical characteristics are as follows:

- Loading wheels, dual or single
- Loading can be in one or both directions
- Number of loading are 22 000 in 24 hours (including daily maintenance)
- Loading wheel lateral movement is 1.5 m.

In total the length is 23 m, width 3.5 m, height 4.2 m and weight 46 ton. The wheel load can be varied from 30 kN to 110 kN (corresponding axle loads 60...220 kN) at speeds up to 12 km/h. The machine can be run 24 hours a day, during the nights and weekends without any staff present.

At VTI there is an indoors full-scale pavement test facility where pavements can be constructed by ordinary road construction machines. This facility comprises three test pits and two of these are used for the accelerated pavement testing, figure 4. The size of the test pits is 3 m in depth, 5 m in width and 15 m in length. The use of two test pits means that one test section can be constructed while the test is running on the other, figure 5.

During the first year in Sweden four structures were tested (SE01–SE04). The test pits were filled with fine sand to a thickness of 2.5 m to act as a subgrade to the test structures. The first two test (SE01 and SE02) were performed on the newly built pavements while the other two were performed after rehabilitation of the first two after completed tests. The rehabilitated pavements were called SE03 and SE04 and the rehabilitation measure was milling and levelling (overlay in the milled area) of the surface layer to get a new even surface without rutting.

During the construction of the test structures a lot of sensors were installed in the structures to be used in the response measuring program. Most of the sensors are located in the centre line of the loaded area (6x1 m). The following sensors was used in test structure SE01:

- Strain gauges on top and at the bottom of the asphalt concrete layer in longitudinal and transverse direction. The sensors were retrofit strain gauges of the same kind as was used previous in the tests in Finland.
- Soil pressure cells, from University of Nottingham, on three different levels.
- Surface deflection was measured by LVDT on top of steel rods anchored at the concrete bottom of the test pit.
- Deflection of base and subgrade surface were measured by plates connected to steel rods through the concrete bottom of the test pit.
- Temperature sensors in the asphalt concrete layer.

Before each accelerated loading test pre-loading and a comprehensive response measurement program were performed.

The pre-loading was done in order to relax possible residual stresses and cause some post-compaction. This was done during one day with a lower wheel load (30 kN single wheel load) and with a specified lateral distribution. The size of the single wheel was 425/65R22.5.

The response measurements embrace a considerable amount of measurement of stresses, strains and deflections at different positions in the test structures and at different test-loads, test-wheels, tyre pressures, lateral positions, speeds and temperatures.

After the response measurements the accelerated loading test was started. Normal running was day and night seven days a week with interruptions only for daily service of the machine, which means about 22 000 loading per day including both directions. The following standard set of test parameters have been used: dual wheel load 60 kN, tyre pressure 800 kPa, wheel speed 12 km/h and pavement temperature +10°C. The size of the dual wheel was 295/80R22.5.

Pavement performance has been studied by daily visual inspections and measurements of cross profiles three times a week in five locations for rut depth calculations.

All collected data will be stored in a common Finnish-Swedish database including information on test sites, pavement structures, sensors, materials, response and performance measurement results.

During the first year in Sweden four accelerated pavement tests was carried out with the new HVS-machine. The tested structures were typical Swedish low volume road structures as they were presented in the former Swedish national specifications, BYA 84. The first two tests were performed on newly built structures and the second two tests were performed on the same structures after rehabilitation. Both structures were of gravel-bitumen type with thin asphalt surface layers and a total pavement thickness of 150 mm and 300 mm respectively.

- The first experience of the test was that the first pavement lasted longer than expected. This was also the case in the first test in Finland.
- In the first test, with the thin pavement, the rutting was relative large during the first part of the test (pre-loading and response measurements).
- In dry condition even the thin structure could carry rather high wheel load. An increase of wheel load from 60 kN to 80 kN and corresponding increase in tire pressure didn’t increased the rutting very much.
- Most of the surface rutting before rehabilitation could be found as deformation of the sand subgrade while after rehabilitation up to half of the surface rutting was due to deformation of the asphalt layers.
- The effect of moisture was clear. The pavement life (number of loads) at 10 mm rut depth was 2.5 to 5 times more in dry condition compare with wet condition.
- The thicker structure (SE02) showed largest difference between dry and wet condition. This was due to good performance at dry condition.
- The pavement life (number of loads) of the thicker structure (SE02) was 4 times more in dry condition compared with the thinner structure (SE01). The total thickness of pavement SE02 was twice that of pavement SE01 but also the actual asphalt surface layer was thicker, 62 mm compared with 49 mm for SE01 pavement.
- This difference in layer thickness leads to the conclusion that it’s not an easy task to construct pavements with high accuracy in layer thickness. This was also one of the experiences from the first tests in Finland.
- The difference in pavement performance at wet conditions was less than expected. One reason for this could be that the comparison wasn’t done at the same condition. The thinner structure was exposed to more loading before water was added and thus had got more deformations in the dry situation.
- The effects of the overlays differ in a way that could be expected. The thicker overlay on the thinner structure reduced the rutting more than the thinner overlay on the thicker structure.
- Another effect after overlay, as mentioned above, was that almost half of the rutting after overlay could be found in the overlay and not as before overlay, most of the rutting in the subgrade. This points out the importance to chose a rut resistant mix to the overlay.

The results evaluated in this report are the results from the pavement performance measurements, rutting and deformation. A large amount of data has been collected during the pavement response measurements. These data, stresses, strains and deflections during a variety of test conditions will be used later for comparison of the test structures and for analysis of relations between pavement response and pavement performance.

As described in the introduction a program for research co-operation in the area of accelerated pavement testing has been agreed between Finland and Sweden covering the years 1997–2003.

During the first year in Sweden, 1999, tests were performed as planned in the research program on typical Swedish low volume road structures and these are presented in this report.

During the second year, 2000, two typical Icelandic pavement structures were tested on request from the Icelandic Road Administration and two steel fabric reinforced structures which were tested as a part of an European research project, REFLEX (Reinforcement of Flexible Road Structures with Steel Fabrics to Prolong Service Life). The Icelandic tests are reported in [2] and the REFLEX tests will be reported in the autumn of 2001.

In November 2000 the HVS machine was taken by land- and sea transport back to Finland for the second period of testing there.

In this second period in Finland tests will be concentrated on low volume road structures. The first test will be on a pavement frost protected with a 200 mm layer of EPS (expanded polystyrene) on the subgrade. The granular base layer, (450 mm), will be reinforced by steel fabric on half of the test area and the surface asphalt layer will be 50 mm thick.

The second and third tests will be on thin structures, 50 mm asphalt surface layer on 250 mm unbound base layer, one with and the other without steel fabric reinforcement in the unbound base layer, as a part of the REFLEX project.

The next three pavements to be tested are also for low-volume roads. The aim is to study the importance of the road cross section to the structural properties. Two different slopes are built and as a reference a “horizontal” slope. The pavement response due to moving wheel load with several offsets will be measured and finally the pavement performance is evaluated with accelerated testing.

Around APT, (Accelerated Pavement Testing), there is international co-operation. The first international conference on APT was held in October 1999 at Reno, Nevada. It included two presentations from Finland and one presentation from Sweden concerning the HVS studies. The TRB, (Transportation Research Board) Task Group is well known within countries that have APT facilities. In November 2000 this Task Group was appointed to an ordinary TRB Committee, A2B09 FULL-SCALE/ACCELERATED PAVEMENT TESTING, and Finland and Sweden has members in this committee.

FEHRL, (Forum of European National Highway Research Laboratories), proposed a group on accelerated pavement testing in Transport Research in COST (European Co-operation in the Field of Scientific and Technical Research). COST is an organisation of the European Commission but may have members in its groups also from non-EU European countries. All the countries that have APT facilities, including Finland and Sweden, will participate in this action. This COST 347 action was started in November 2000 and is planned for a period of 3.5 years.

Author(s)
Wiman, Leif G

Related reports

- 2/16/2010
Monitoring of Long Term Pavement Performance (LTPP) – Progress Report, December 2009
- 2/10/2010
Evaluation of the LTL-M – Mobile measurement of road marking
- 2/10/2010
Milled rumble strips in the middle of the lane – a test in Skaulo
- 1/25/2010
Estimation of energy consumption and carbon dioxide emissions during construction, maintenance and operation of roads
- 1/25/2010
Condition assessment of road markings in three regions in Sweden in 2009
- 1/15/2010
Rubber asphalt – laboratory experiments
- 1/15/2010
Traffic safety problems for pedestrians – Injuries according to police reports and health service
- 12/15/2009
Road Marking Management System – a pre-study
- 11/26/2009
Equipment detecting cycles. A literature study
- 11/26/2009
Development of Road Marking Tester. Status of RMT Version 2