In britain, frost heave (and also thaw weakening) is prevented by ensuring that all materials within 450 mm of the road surface are non-frost susceptible as defined by the transport and road research laboratory test, the latest version of which is known as the sr829 test. Materials are not frost susceptible if their mean heave in valid sr829 tests is not greater than 12 mm. For materials close to the borderline, tests are required at three different laboratories to define the mean heave. The test, which now has improved reproducibility, is to be included in british standard specification 812. In this paper the evolution of the test is summarized and the implications of the revisions and of other factors in design, such as depth of frost penetration and the availability of water, are examined critically. Results of a series of tests are presented that show that the british frost-susceptible materials correlate with the "highly frost-susceptible" classification in the united states and french systems. Particle size distribution and saturation moisture content are shown to be poor indicators of the frost susceptibility of aggregates, because neither can reflect the contribution of between-particle and within-particle pore size distributions. Suction characteristics, determined by an osmotic technique, overcome this shortcoming and correlate well with the frost susceptibility of the materials tested. Thispaper appeared in transportation research record no. 1146, Frost protection and insulation for transportation facilities. For coveringabstract see IRRD no 817763.
Abstract