LINKS GAPS

ACPA Links Gaps in Pavement Design, Materials, and Construction

ACPA Links Gaps in Pavement Design, Materials, and Construction

PART I: AASHTO Design Guide

The American Association of State Highway & Transportation Officials pavement design method is based on the results of the AASHTO road test that was conducted in Ottawa, Illinois in the late 1950's. This full-scale test facility evaluated many different pavement sections including both concrete and asphalt. The ultimate goal was to establish a relationship between design features (primarily thickness) and traffic loading.

The construction quality at the road test was the best possible at the time. Fixed form paving was used as were relatively small batch mixers (by today's standards). Close control of the batching, mixing, placement, and other processes was critical in minimizing variability. In other words, having very stringent QA/QC procedures in place removed potential sources of error when developing the design equations.

The road test included only one subgrade soil type (poor) and climatic conditions found in northern Illinois. Traffic was limited in the number of load repetitions and axle weights/configurations. Many of the conditions at the road test were not comparable to those found elsewhere in the U.S. Revisions to the initial design equations developed in the early 1960's addressed these, and other issues.

PART II: Pavement Smoothness

Does pavement smoothness affect anything other than driver comfort? The answer is a resounding YES!

Smoothness has a direct bearing on the long-term performance of concrete pavements, as shown in numerous research studies conducted in the past 10 years. Simply stated, "Smooth pavements stay smoother, longer."

The importance of smoothness has long been recognized as affecting the longevity of concrete pavements. The present serviceability index (PSI) concept was developed to quantify the level of deterioration of the pavements in the American Association of State Highway Officials (AASHO) Road Test, which was conducted between 1958 and 1960. One of the primary elements of the PSI determination was related to pavement smoothness (not necessarily ride quality). The PSI ranges from a 5.0 (perfect pavement) to 0.0 (impassible), although rehabilitation or restoration work would typically begin at a terminal serviceability of 2.0 to 2.5, depending on the functional classification of the roadway.

The current American Association of State Highway & Transportation Officials (AASHTO) design method uses the difference between the as-constructed smoothness (initial serviceability) and terminal serviceability as a direct input in the thickness design determination. To illustrate the importance of this value, a 10-inch thick concrete pavement would carry approximately 20.9 million ESAL's (Equivalent 18,000-lb Single Axle Loads) during its expected life, for a given set of design assumptions. However, if the pavement built at an initial serviceability of 4.3 rather than 4.5, the load carrying capacity would be reduced to 18.8 million ESALs. If it were constructed at 4.7, which is easily attainable using current construction practices, the number of allowable loads increases to 22.9 million. This type of comparison, although not entirely accurate, gives a good indication of the relationship between smoothness and performance. The upcoming 2002 AASHTO Pavement Design Guide will show an even more pronounced effect of smoothness.

Constructing smooth pavements requires attention to detail. Uniformity in concrete production and delivery, support conditions (both under the pavement and in the trackline), placement and consolidation, and surface texturing are the primary construction processes influencing smoothness. However, perhaps the most important details are high quality stringlines (2 lines, adequately tensioned with closely spaced pins) and an accurately set-up paver (including sensor system).

In the past several years, the design thickness of heavily trafficked roads has increased dramatically. This leads to the questions, "How thick is thick enough?" and "What are the implications of short cores?"

Thickness is an important design feature and was a major component in the initial American Association of State Highway & Transportation Officials (AASHTO) design equation that related traffic, thickness, and performance. It often has been stated that the AASHTO design procedure over-estimates the required thickness of concrete pavements, particularly at high levels of traffic and reliability. This assumption seems well founded when comparing the required thickness of pavement using the current AASHTO procedure and a mechanistic-empirical procedure, which is a much more scientific approach to design. Mechanistic means that it is based on engineering principles such as stresses and strains, and empirical means that it has been calibrated using actual real-life situations and projects.

The American Association of State Highway Officials (AASHO -- the forerunner of AASHTO) Road Test, the basis for the current AASHTO procedure, included a number of concrete pavement test sections. The maximum traffic placed onthese sections was not much in excess of one-million equivalent 18,000-lb single axle loads (18-kip ESAL's). The thin sections failed to varying degrees, but the thicker sections did not. The factors leading to the design of excessively thick pavements are based primarily on traffic projections that far exceed the levels at the road test (two to three hundred times, inIn the past several years, the design thickness of heavily trafficked roads has increased dramatically. This leads to the questions, "How thick is thick enough?" and "What are the implications of short cores?"

The effect of thickness can be illustrated by the following example using the same assumptions as last weeks' article. A 10-inch thick concrete pavement would carry approximately 21 million 18-kip ESAL's during its expected life, for a given set of design assumptions. If the thickness is reduced to 9.6 inches, the load carrying capacity is reduced to 16 million. If the thickness is further reduced to 9 inches (requiring removal and replacement in most specifications), the number of loads to failure are reduced to 10 million. On the other hand, if the thickness is increased to 11 inches, the allowable traffic is increased to 37 million 18-kip ESAL's.

In last week's Tech Corner article, the effect of thickness on the load-carrying capacity of concrete pavements was discussed. For a given set of conditions, a 10-inch-thick concrete pavement would carry approximately 21 million 18-kip ESAL's during its expected life. By increasing the thickness to 11 inches, the AASHTO design method estimates approximately 37 million 18-kip ESAL's to failure. If this difference in load carrying capacity were true, designers would always opt for additional thickness providing the maximum "bang for the buck." Unfortunately, pavements don't generally fail in the manner predicted by the AASHTO design equation. This design procedure is heavily weighted towards fatigue damage (e.g., corner breaks) and does not account for the typical modes of failure seen today. Concrete pavements are much more likely to fail due to environmental conditions (freeze/thaw damage) and materials problems (insufficient entrained air) than to fatigue damage, particularly for pavements exceeding 11 to 12 inches.

The foregoing discussion is not intended to downplay the importance of accurate and uniform thickness during construction. A reduction in the as-constructed versus as-designed thickness can have a significant effect on pavement performance, especially for relatively thin pavements. Most agencies have tight specifications regarding thickness and most have a trigger value of 1 inch for requiring removal and replacement. Performance-related specifications, if implemented, will correlate early distress (or a reduction in load carrying capacity to failure) to inadequate or non-uniform thickness in a more rational manner than is currently employed.

Maintaining the design thickness throughout a project requires stringent quality control measures. Paving operations obviously have a direct bearing on thickness and overall pavement quality. Some of the methods that have been used to control thickness (and minimize yield loss) include: using a common, well maintained stringline for all grade preparation and paving operations; accurate grade trimming; delivery and placement of uniform concrete (slump); proper equipment set-up; and a general attention to detail.

Should a contractor working with a grading subcontractor encounter a variable grade that results in a short core (inadequate thickness) in a section of pavement, the loss of structure may be offset by the concrete strength. Under most QA/QC concrete specifications today, contractors provide more strength in the field than needed to meet the design strength. This increase in the as-built concrete strength can often make up for thin pavement, but it must be analyzed carefully. In general, it is always best to grade the base with care and set up the stringline and pavement to meet the minimum nominal thickness.

Next week's Tech Corner will conclude this series with a discussion of the relationship between design and construction. The properties of concrete and how they are accounted for in the design process will be featured.

For more information on thickness design, load carrying capacity, and the AASHTO pavement design method, consult ACPA's WinPAS software and accompanying design manual, code numbers MC016.01P and MC016P.