Question 4 of 7: Mix design methods, and the three principal HMA mixture types

Question 4: Mix design methods, and the three principal HMA mixture types (10 + 10 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

Part (a) — The three mix design methods

(i) The Marshall method. Developed by Bruce Marshall of the Mississippi Highway Department and adopted by the US Army Corps of Engineers during the Second World War, the Marshall method compacts a 100 mm diameter specimen with a 4.54 kg hammer falling 457 mm, 35, 50 or 75 blows per face according to the design traffic level, and then loads the specimen diametrally in a curved testing head at 60 degrees Celsius to measure stability, the peak load, and flow, the deformation at that peak. Its merits are practical rather than theoretical. The equipment is inexpensive, portable and robust enough for a field laboratory; the procedure is quick, so a full series of asphalt contents can be run in a day; the technicians to run it are everywhere; and the density and voids analysis it produces is genuinely useful and survives into modern practice. Its limitations are equally clear. Impact compaction bears no relationship to the kneading action of a roller and produces an aggregate structure that is not representative of the field, particularly for large-stone and crushed mixes; stability is an empirical index with no direct engineering meaning and cannot be used in any structural calculation; the test is performed at a single temperature and takes no account of the climate the pavement will actually see; and it gives no measure of the mixture’s resistance to permanent deformation under repeated loading, which is the distress that most often ends the life of a modern heavily trafficked surface.

(ii) The Hveem method. Francis Hveem of the California Division of Highways built his procedure around the mechanics of the mixture rather than an index. Specimens are compacted by a kneading compactor, which genuinely simulates the rolling action of field compaction, and are then tested in the stabilometer, a triaxial-type device that measures the lateral pressure transmitted when a vertical load is applied. This yields a stabilometer value that is a real measure of internal friction and particle interlock. The companion cohesiometer measures the tensile cohesion of the binder, and the swell test assesses moisture susceptibility. The merits follow: the compaction is realistic, the stabilometer value has a defensible mechanical interpretation, and the method contains an explicit moisture-damage check that both of the other methods lack. Its limitations are why it never spread far beyond California. The kneading compactor and stabilometer are expensive, heavy and unsuited to a field laboratory; the procedure is more time-consuming and requires greater operator skill; the design binder content is selected largely on the pyramid of a minimum stabilometer value and a minimum air void content, which tends to produce lean mixtures that are durable in a dry climate but can be under-asphalted and prone to ravelling elsewhere; and, like Marshall, it does not address long-term rutting or fatigue directly.

(iii) The Superpave volumetric method. Superpave emerged from the Strategic Highway Research Program of 1987 to 1993 and is a system rather than a single test. Compaction is by gyratory compactor, which combines vertical pressure with a 1.25 degree gyration and reproduces field densification and aggregate orientation well; the number of gyrations is tied to the design traffic and the climate. Aggregate quality is controlled explicitly through consensus properties — coarse and fine aggregate angularity, flat and elongated particles, and clay content — and the gradation is constrained by control points and a restricted zone on the 0.45 power chart. The binder is specified by performance grade, selected from the actual pavement temperature extremes at the site, which for a Canadian project might be a PG 58-34 or PG 52-40. The merits are that the method is founded on climate and traffic rather than on convention, that the volumetric criteria are consistent across the country, and that binder selection at last matches the environment the pavement will experience. The limitations are the capital cost of the gyratory compactor and the binder-grading equipment, which puts a full Superpave laboratory out of reach of a small agency; the reliance on volumetric criteria alone in the routine level of the method, with no strength or performance test at all in the standard procedure, so a mixture can satisfy every volumetric requirement and still rut; and the additional complexity and training burden. Canadian agencies have therefore tended to adopt Superpave binder grading and aggregate requirements while retaining familiar mixture designations, and increasingly supplement the volumetric design with a performance test such as the Hamburg wheel-tracking test.

Part (b) — Three types of hot-mix asphalt

(i) Dense-graded mixes. A dense-graded mix has a continuous, well-distributed gradation from the nominal maximum size down to the filler, producing a low-void, largely impermeable mixture. Its advantages are that it is the general-purpose workhorse of asphalt paving: it is strong and stiff, it is essentially impermeable when compacted to 92 to 96 percent of maximum theoretical density and therefore protects the layers beneath it, it is durable, it can be used in the surface, binder and base courses alike, and it is the cheapest and most widely understood mixture available. Its disadvantages are all consequences of that impermeability and fine texture: surface water cannot drain into it, so it ponds and sprays, reducing wet-weather visibility and skid resistance; the surface texture is comparatively smooth, which raises tyre noise; and because the load is carried partly through the fine aggregate and mastic rather than purely by stone-on-stone contact, a dense-graded mix is more susceptible to rutting under very heavy, slow or channelised traffic than a stone-skeleton mixture is.

(ii) Open-graded friction course. An open-graded, or uniformly graded, friction course is a thin surfacing, typically 20 to 40 mm, made almost entirely of a single coarse size with very little fine aggregate, giving an in-place air void content of 15 to 25 percent. Its advantages are drainage-related. Water entering the mat drains laterally within it to the shoulder instead of standing on the surface, so splash and spray are dramatically reduced, hydroplaning risk falls, wet-night visibility of pavement markings improves, and tyre noise drops by several decibels because the porous structure absorbs it. Wet skid resistance is high. Its disadvantages are that it contributes essentially nothing structurally and must be placed on a sound impermeable dense-graded layer; that the pores clog with detritus over time and the drainage benefit decays; that the very open structure exposes a large surface area of binder to air and water, so ravelling and oxidative ageing are the usual failure modes and a polymer- or fibre-modified binder is normally necessary; that it is vulnerable to damage from snowplough blades and to stripping in freeze-thaw climates, which limits its use in much of Canada; and that it cannot be used where studded tyres or heavy winter maintenance are expected.

(iii) Stone mastic asphalt. Stone mastic asphalt, sometimes written stone matrix asphalt and abbreviated SMA, is a gap-graded mixture developed in Germany in the 1960s to resist studded-tyre wear. It contains a high proportion of coarse aggregate, 70 to 80 percent, so that the stones bear directly on one another in a stone-on-stone skeleton, with the voids between them filled by a rich mastic of filler, fine aggregate, a high binder content of 6 to 7 percent and a stabilising additive, usually cellulose fibre. Its advantages are excellent resistance to permanent deformation, because the load path is through the stone skeleton rather than the mastic; long service life, because the high binder content and thick films resist oxidation and fatigue; good macro-texture and hence good wet friction and reduced spray; and strong resistance to abrasion and studded-tyre wear. Its disadvantages are cost, since the mixture needs high-quality cubical crushed aggregate, more binder and a fibre additive; the narrow construction window, because the rich mastic will drain from the aggregate if the mixing or laydown temperature is too high and the mixture is unforgiving of poor temperature control; the tendency of the fresh surface to be slippery until the initial binder film has worn off, which sometimes requires an aggregate dressing on opening; and the need for a well-controlled plant, since fibre dosing and the tight gradation window demand more of the supplier than a conventional dense mix.