Question 6 of 7: PGAC grading and the Superpave versus Marshall mix design methods

Question 6: PGAC grading and the Superpave versus Marshall mix design methods (20 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.

(a) What “performance graded” means, and how it differs from penetration grading. A performance grade is written as PG H−L, where H is the highest seven-day average maximum pavement design temperature and L is the lowest one-day minimum pavement design temperature, both in degrees Celsius, at which the binder is still required to meet its specification. The specification criteria themselves — a minimum rutting parameter, a maximum fatigue parameter, a maximum creep stiffness, a minimum m-value — are held constant for every grade; what changes from grade to grade is the temperature at which the criterion must be met. That inversion is the whole idea. A designer picks the grade from the climate at the site (and from the traffic, bumping the high grade one or two steps for slow or standing loads), and the binder supplier is free to reach the criteria by any refining or modification route.

Penetration grading, by contrast, classified a binder by a single empirical measurement — the depth in tenths of a millimetre that a standard needle penetrated the binder in five seconds at 25 °C under a 100 g load — supplemented by ductility, flash point and a thin-film oven residue. The differences that matter in practice are four. Penetration is measured at one temperature only, so it says nothing about behaviour at the summer or winter extremes actually experienced; PGAC measures at the extremes themselves. Penetration is an empirical index with no direct link to a distress; the PGAC parameters are rheological and each is tied to a specific distress. Penetration was blind to modification, so a polymer-modified binder and a straight-run binder of the same penetration were graded identically; the dynamic shear rheometer separates them clearly. And penetration grading involved no ageing conditioning beyond the thin-film oven, whereas PGAC tests the binder in three states — original, short-term aged and long-term aged — matching construction and in-service ageing.

(b) The grade of the tested binder.

Given. A binder that satisfies the high-temperature criteria up to 68.3 °C and the low-temperature criteria down to −33.9 °C (the minus sign is implicit; a low-temperature criterion at +33.9 °C would be meaningless).

Find. The performance grade to be specified.

Standard high-temperature grades are 46, 52, 58, 64, 70, 76 and 82 °C and standard low-temperature grades are −10, −16, −22, −28, −34, −40 and −46 °C. A binder may only be credited with a grade it has actually been shown to satisfy, so the high grade is rounded down to the next standard step and the low grade is rounded up (towards zero). Here 68.3 °C falls between 64 and 70, so the high grade is 64; and −33.9 °C falls between −28 and −34, so the low grade is −28. The binder is therefore graded $$\boxed{\text{PG } 64\text{-}28}$$ The binder misses PG 70-34 by 1.7 °C at the top and 0.1 °C at the bottom, which illustrates why the grading is deliberately conservative: a tenth of a degree of measured performance is well inside the reproducibility of the test.

(c) The principal PGAC tests and the property each targets. Five tests carry the specification, applied to the binder in the ageing state that matches the distress of interest.

(d) Marshall and Superpave mix design. The Marshall method compacts 101.6 mm diameter specimens with a hand or mechanical hammer delivering 35, 50 or 75 blows per face according to the traffic level, at a series of binder contents. Each specimen is tested for bulk relative density, then loaded diametrally in a curved breaking head at 60 °C at 50.8 mm per minute to give a stability (peak load) and a flow (deformation at peak). The designer plots stability, flow, unit weight, air voids and voids filled with asphalt against binder content and selects the binder content that satisfies all the criteria, conventionally at 4 % air voids. The strength of the method is that it is quick, cheap, portable and backed by seventy years of local experience; its weakness is that impact compaction does not reproduce the kneading action of a roller or of traffic, and that stability and flow are empirical indices that correlate only loosely with rutting.

The Superpave method, developed under the Strategic Highway Research Program, replaces the hammer with a gyratory compactor that applies a 600 kPa vertical pressure while the mould gyrates at 1.25° internal angle and 30 revolutions per minute, which reproduces the shearing action of construction and traffic and yields a compaction curve rather than a single point. Specimens are 150 mm in diameter, so a full 25 mm nominal aggregate can be accommodated. The design is purely volumetric: the mix must reach 4 % air voids at the design number of gyrations N_des, must be no denser than 89 % of maximum theoretical density at N_initial (a tender-mix screen) and no denser than 98 % at N_max (a rutting screen), and must satisfy voids in the mineral aggregate, voids filled with asphalt and a dust-to-binder ratio between 0.6 and 1.2. Aggregate quality is controlled by consensus properties — coarse and fine aggregate angularity, flat and elongated particles, sand equivalent — and the gradation must pass through a control-point band while avoiding the restricted zone. The binder is selected as a performance grade for the site climate and traffic, so mix design and binder grading are integrated rather than independent.

Both methods remain legitimate. The practical resolution in Ontario is a matter of scale and risk: Superpave earns its higher cost where traffic is heavy and a rutting failure is expensive, while Marshall remains defensible on low-volume municipal work where the compaction realism matters less and the capital cost of a gyratory compactor cannot be justified.