Evaluation Of The Current Test Methods Of Water Sensitivity And Permanent Deformation

Abstract

The adaption of the functional characterization of asphalt concrete mixtures by the Dutch authorities, led to a big necessity in acquiring a fundamental understanding on its properties. In order to achieve this, the long-term project NL-Lab was initiated by InfraQuest in 2012. With the advantage of its longevity and the multiple parties contributing with material, testing and analysing, the knowledge aimed for, is gradually accumulating with the time passing. After 5 years of running, a considerable number of data have been collected and analysed, leading to many observations, that in turn raised questions and doubts. These doubts are the subject of this research, focusing on the functional characteristics of water sensitivity and permanent deformation.
A common problem for both properties regards the production and compaction of type-testing specimens. The practice so far suggests both stages to take place in laboratory conditions. However, there were indications that the performance of the same asphalt mixtures taken from the field instead of the lab, differed. Testing and comparing in terms of water sensitivity and permanent deformation, these indications were indeed repeated. Lab and field specimens did not match in performance in the majority of the tests, suggesting that differences between the two conditions are projected in differences in performance.
The second common problem also evolves around the type-testing, this time regarding the test procedure as a whole. It is a highly time and money consuming process that has to be repeated in a trial-and-error manner, until a satisfying mixture in terms of performance is found. The aim of this research was to limit this process as much as possible by replacing it with regression relations. For the water sensitivity test, the relations obtained were of a fairly good predicting quality, and could be eventually used in the preliminary mixture design. The permanent deformation relations however, were at low levels of predictability, due to the test’s high sensitivity in being reproduced and the inconsistent test setups used by the various contractors taking part in the project.
Focusing only on water sensitivity, the conditioning protocol described in the current standard, is considered as not very capable of capturing the property it aims for, and that it is conservative in the damage it causes to the mixture. Utilizing two conditioning protocols, MIST and Freeze-Thaw protocol, the standard method indeed was found to be the least damaging of all. With the MIST being the most destructive, at extreme levels in most cases, and the Freeze-Thaw following it, each method’s pros and cons were found.
The last aspect tested referred to the friction reduction method used during the permanent deformation test (triaxial cyclic compression). The old and new standard versions both prescribe two different materials, while the requirements for rutting remained unchanged. Testing using these two materials, Teflon and Latex, plus one new, it was found that the differences in the triaxial test output were major. Consequently, this mismatch means that inconsistent comparisons take place, leading to conclusions on rutting performance not representative of the actual situation.