Evaluating the Field Performance of Asphalt Mixtures In the Lab, January/February 2005 Public Roads

FHWA is assessing the Simple Performance Tester to determine its effectiveness for use in the field.

The National Cooperative Highway Research Program (NCHRP) is developing advanced pavement technology that supports the Federal Highway Administration's (FHWA) mission to achieve long-lived asphalt pavements. One of these advanced technologies is the Simple Performance Tester (SPT), which evaluates asphalt mixtures to determine their response to permanent deformation (rutting) and fatigue cracking. FHWA is supporting NCHRP's research by evaluating how well the SPT works in a field laboratory environment.

Highway engineers have been looking for a simple, inexpensive, and accurate performance test for hot-mix asphalt (HMA) since HMA first came into use in the late 19th century. One of the first tests was the Hubbard-Field method, developed in the 1920s, which measured the strength of the asphalt by using a punching shear load to test for failure.

In the 1930s and 1940s, the industry developed new tools, including the Marshall Stability test to determine the maximum load resistance that an asphalt test specimen will develop at 60 degrees Celsius (140 degrees Fahrenheit) and the Hveem Stabilometer test, which applies vertical loads to compacted specimens to measure the resulting lateral pressures and thus the internal friction within a mixture. During the 1970s and 1980s, a new approach used a loaded wheel tester to apply a load over a compacted slab of specimen mix.

"All of these tests had shortcomings," says John Bukowski, senior engineer, FHWA. "The tests provided only relative rankings of the performance of the mixtures, and they were unable to indicate how each of the variables in the pavement mixture affected performance."

SPT and Superpave™ Mix Design

From 1987 through 1993, the Strategic Highway Research Program (SHRP) conducted a $50 million research effort to develop new ways to specify, test, and design asphalt materials. The final product of the SHRP asphalt research program was a new system referred to as Superpave, which stands for Superior Performing Asphalt Pavements. The Superpave method specified how asphalt could be mixed, designed, and analyzed.

"Superpave included a suite of tests and equipment to characterize the performance of mix designs," says Dr. Ed Harrigan, senior program officer for NCHRP, "and the Superpave mix design method has been adopted almost exclusively around the country by State departments of transportation. But the equipment, which is very good, has been found to be too sophisticated and too complex for routine day-to-day operations."

In 2000, the Transportation Research Board's Superpave Mixture and Aggregate Expert Task Group surveyed materials engineers with State agencies and found that they need a simple performance test during mix design, and possibly for field quality control, to evaluate how well the HMA resists permanent deformation and fatigue cracking.

In response to this need, NCHRP developed and evaluated a Simple Performance Tester to complement the Superpave mix design method by providing a test to evaluate the asphalt mix design quickly. "The goal was to come up with equipment that was reliable but inexpensive enough to be placed in all State and contractors' labs," says Harrigan. "It was designed to be small enough to be used either in a fixed or mobile lab and also to be available at a practical cost."

Why Conduct Performance Testing?

Materials engineers use Superpave mix designs that are appropriate for their region. In Florida, for example, the asphalt binder must be stiffer at higher temperatures than would be required in a State such as Maine with its cooler climate. The SPT can be used to test HMA mixtures in any region to determine how the mixture will perform in the field.

The tester is used to help design both the pavement structure and the HMA material used in the pavement. "The Simple Performance Tester helps us get engineering strength values for hot-mix asphalt and mechanistic input values for pavement design modeling," says Dr. Murari Pradhan, bituminous engineer with the Utah Department of Transportation (UDOT), "which is really helpful because it is not only good for materials design, but also it is going to be an integral part of pavement design."

The tester will be useful in designing the pavement structure (for example, determining the appropriate thickness of each layer of pavement), because it outputs a stiffness, or E*, value. As described in "Three Tests Included in the Superpave Performance Tester" on this page, this value is used as an input in the overall design of the pavement structure in NCHRP's Mechanistic-Empirical Pavement Design Guide. (For more information about the mechanistic-empirical guide, see "Designing Tomorrow's Pavements" in the September/October 2004 issue of PUBLIC ROADS.)

The SPT also is used to design the pavement material based on the desired pavement performance. For example, a materials engineer who wants to achieve less than 12.7 millimeters (0.5 inch) of rutting can use the SPT to test the asphalt pavement mixture and determine how stiff the mixture should be to meet that criterion.

SPT Facilitates Flexibility In Designs

The SPT offers unprecedented flexibility when designing asphalt pavement materials. "Right now we have specified parameters for controlling volumetrics and mix design, which is kind of like a recipe," says Judie Ryan, asphalt mixtures specialist at the Wisconsin DOT (WisDOT). "And the recipe calls for 1 cup of sugar and 2 teaspoons of salt, and so forth, and it makes something called an 'apple pie.' As you start to develop your own taste, you may change what's in the recipe, say by adding less sugar, but it's still called an apple pie. But with our current parameters, we can only keep asking for apple pies. If I want something else, say a pie with blueberries, I'm out of luck; our current requirements never allow us to get to any new type of pie. We have a finite set of ingredients and combinations, and while other performance testers can measure symptoms that define poor product, the SPT may be that additional tool that helps prevent it."

Ryan continues: "The SPT probably isn't as simple as we all had hoped, but the new technology provides more information than we've ever had before. It's very specific to application and use, where that's been a weakness in our specifications up to this point. So with the SPT, the opportunity is there to find that next best way to design mixtures since we can relate it to actual performance parameters."

Performance testing is also of interest to contractors. "Wisconsin was one of the first States to develop a warranty specification for HMA," says Ryan. The current warranty period is 5 to 7 years and requires the contractor to perform remedial or corrective work whenever a distress threshold is exceeded. "The intention of the warranty is to give the contractors as much freedom as possible while assuring a quality product," says Ryan, "so the warranty specification allows contractors to select their own materials, mix design, construction techniques, and so forth. What this means for the SPT is that if contractors have that next great idea, they can use the SPT as a tool to test and evaluate [the mix] prior to producing it. The risk has shifted to them as they provide that product. In the event that there are some failures, then they have the liability to maintain the roadway for at least the length of the warranty period."

Mobile Asphalt Lab Brings The SPT to the States

At the invitation of State highway agencies, FHWA's experienced technicians and engineers have been traveling with the agency's Mobile Asphalt Lab to project sites across the country. To date, the Mobile Asphalt Lab has visited about 15 States to help them implement new pavement technologies, including the SPT.

Utah is one of the States visited. "The FHWA Mobile Asphalt Lab was here for 2 weeks," says UDOT's Pradhan. "During those 2 weeks, FHWA conducted a presentation to about 30 to 40 people, including all of our pavement and materials engineers from both regional and central offices, as well as contractors and consultants. It was a good opportunity to educate everyone on pavement design and volumetrics design. Now all our engineers, including consulting and contracting engineers, are familiar with the SPT and how it can improve our material mixes, and also how it is going to be part of the new mechanistic-empirical pavement design procedures."

Fabrication of Test Specimens

A rigorous evaluation of the SPT in the field, using data from several of the States visited by the Mobile Asphalt Lab, provided promising results that demonstrate SPT's applicability in a field laboratory environment. FHWA analyzed both laboratory-blended and plant-produced materials. For laboratory-blended samples, the researchers brought the samples to testing temperatures with approximately 1 hour of oven heating. After measuring the mixture properties, they compacted specimens in the Superpave gyratory compactor according to the target air void content in the field, as indicated in construction specifications. Specimens typically were compacted to 180 to 185 millimeters (mm) in height, and the average air void content was typically 1.5 to 2 percent higher than the target for the final specimen.

The technicians measured the height and average diameter (at the mid and third points of the specimens along two perpendicular axes) using a digital caliper ruler. They calculated the standard deviation of the resulting six measurements, recording to the nearest 1.0 mm. Specimens were required to have a standard deviation less than 1.0 mm. For the ends of the specimens, the technicians recorded parallelism, using a machinist's square and feeler gauges, and flatness, using the feeler gauges and a straight edge. The technicians then cored specimens using a conventional core drill to obtain a sample 100 mm in diameter. Finally, the specimen ends were trimmed with a masonry tile saw to obtain a height of 150 mm.

Tests Conducted in Field Lab

The FHWA engineers analyzed the asphalt mixtures in the SPT using two performance tests, dynamic modulus and repeated load test, as described in "Three Tests Included in the Superpave Performance Tester" on page 39. The FHWA technicians conducted the tests in accordance with protocols currently being established through NCHRP research projects 9–19 and 9–29. In the dynamic modulus test, the technicians applied a continuous haversine axial compressive load and used the resulting stresses and strains to calculate the dynamic modulus and phase angle. The testing protocols included running the dynamic modulus test at six different frequencies to represent traffic loads traveling at low to high speeds.

Range of Stiffness Data at Effective Pavement Temperatures for Various Mixtures

Project ID	Mix ID	Eff Temperature for Fatigue(°C)	Dynamic Modulus, E* (MPa)		Eff Temperature for Rutting (°C)	Dynamic Modulus, E* (MPa)
Project ID	Mix ID	Eff Temperature for Fatigue(°C)	Min	Max	Eff Temperature for Rutting (°C)	Min	Max
Coarse 19 mm PG 76-16	Laboratory-Blended	22.0	1594	7298	44.0	169	2061
Coarse 19 mm PG 76-16	Plant-Produced	22.0	1316	8499	44.0	203	2694
Fine 19 mm PG 64-22	Laboratory-Blended	15.6	1744	13050	31.2	183	6501
	Plant-Produced	15.6	1126	11885
	Laboratory-Blended	19.6	1679	11920
	Plant-Produced	19.6	823	10863		224	6644
	Laboratory-Blended	23.6	874	9784
	Plant-Produced	23.6	437	8555
12.5 mm PG 76-22 & PG 67-22
	Coarse 76-22	20.1	2663	10313	53.4	272	2530
	Fine 76-22		2782	10063		190	2466
	Coarse 67-22 Top Lift		2675	11100		118	2742
	Coarse 67-22 Bottom Lift		2080	11696		110	2523
Coarse 19 mm PG 58-28
	Laboratory-Blended	17.0	1017	9576	40.0	92	2315
	Plant-Produced	17.0	1424	10106	40.0	92	2315
	Laboratory-Blended	23.0	882	7226	40.0	114.3	2453
	Plant-Produced	23.0	582	6697	40.0	114.3	2453

For the repeated load test, the technicians applied a pulsating load for 0.1 second, followed by a 0.9-second rest period and contact stress of 30 kilopascals (kPa). They applied an axial stress of 600 kPa to simulate an average tire contact stress of a mixed traffic loading.

They ran the dynamic modulus tests to obtain stiffness values for a pavement at the two temperatures—the effective pavement temperatures for fatigue damage and for rutting, respectively. Effective pavement temperatures were calculated using equations developed in NCHRP and SHRP research programs, and were based on climatic data from the NOAA and LTPP weather databases.

Dynamic Modulus Database

The FHWA engineers developed a database to present the range of stiffness values for the different mixtures investigated in the Mobile Asphalt Lab. The findings, shown in the table above, provide the minimum and maximum stiffness values for the laboratory-blended and plant-produced specimens at the various temperatures for fatigue cracking and rutting.

The engineers assessed the volumetric data to help explain differences in stiffness values that were observed. Note that for all data from one State, the 12.5-mm PG 76–22 and PG 67–22 mixes are based on plant-produced specimens only. The differences in stiffness between lab-blended and plant-produced mixes are consistent, although slightly less pronounced for rutting. This result implies that using stiffness values from lab-blended specimens in a mechanistic pavement design or asphalt mix design procedure could possibly under- or overestimate the material stiffness.

Evaluation of Data from Repeated Load Test

The repeated load test conducted in the SPT is also used for predicting the rutting performance of the HMA pavement. Technicians ran the test at the effective pavement temperature for rutting for each mix. A mixture with a low value of microstrains, which represents strain (deformation) expressed as parts per million, can be expected to exhibit better rutting performance than a mixture with a high amount of strain. That is, a low number of microstrains means that the asphalt mix does not flow as easily under a heavy pulsing load and could be expected to perform well in the field under the axle loads applied by traffic. Typical flow numbers result in 4,000 to 25,000 microstrains. The flow number is the number of load repetitions when shear deformation occurs under constant volume. The total accumulated microstrains is the amount of deformation in the specimen at the end of the repeated load test.

The importance of capturing the total accumulated microstrains was even more pronounced in data derived exclusively from plant-produced specimens, as seen in the figure above. If only the microstrains at flow number are captured, both fine- and coarse-graded mixtures, regardless of the binder type and applied stress level, appear to behave similarly. The general trend is that if the test is stopped at microstrains at flow number, there is no discernable difference between the four mixes. A review of the total accumulated flow (microstrains) for these specimens, however, tells a very different story. Once the test is carried out to the total accumulated microstrains, the data clearly demonstrate the difference in rutting performance between the modified and unmodified coarse and fine mixtures. In doing so, results make engineering sense in that the modified (PG 67) mixes are more resistant to rutting than the unmodified neat (PG 76) mixes. Therefore, although the current test protocol in NCHRP 9-19 recommends flow number as the determining criterion, the investigation indicated that important trends will be missed; and the behavior of the mix in the field may be miscalculated if flow number is used to define mixture rutting performance. These data suggest that the more appropriate parameter may be the total accumulated microstrains, and this parameter should be used to evaluate mixtures in the repeated load test for rutting.

Key Findings from Investigation

The most important finding of the overall investigation is that the SPT can be used routinely and can be used in a field laboratory. FHWA has used the SPT for more than 2 years on construction sites and has successfully tested field mixes in a field laboratory, facilitating a quick and easy prediction of the asphalt performance.

A second finding is that any of the tests included in the SPT (such as the dynamic modulus test, repeated load test, or static creep test) appear to verify a clear relationship between the volumetric properties (the makeup of the mix) and composition, and the predicted performance in the field.

A third study finding is that, for the repeated load test, analysts should review not only the flow number but also the total accumulated microstrain, which will provide a more accurate indication of the flow characteristics of the asphalt mix.

States Look Forward to Using the SPT

To date, only one State, Maryland, owns the latest model of the SPT. But other States plan to purchase it once testing and refinement of the SPT is complete and it is available commercially, which is expected within a year.

"From a State point of view, there's a real need for the equipment," says Larry Michael, regional engineer and asphalt team leader for the Maryland State Highway Administration. "We need a quick strength test for HMA mixes, and the equipment will work well with the new Mechanistic-Empirical Pavement Design Guide, which is of interest to Maryland. The equipment itself is user friendly and performs well. We've all been surprised at how good the early tests have been."

Utah also was pleased. "FHWA demonstrated that [the] SPT can be used in the field," says UDOT's Pradhan. "We ran two performance testers, and we found that the results were consistent, and they can be reproduced. We're going to use the SPT for our [mixture] design, and as we get comfortable with the design, we may be using it on the field mixtures too."

According to UDOT, the SPT may be particularly helpful in the field when there's a dispute between the agency and the contractor. "When the contractor lays down the pavement, we take a sample and test for the volumetrics," says Pradhan. "Sometimes this does not match up with what is required, which can lead to a dispute. So in this situation the SPT will be helpful since we can quickly conduct the tests to provide some objective data."

Wisconsin also plans to use the SPT. "In Wisconsin, up until the early 1990s, the department would just go out and kind of kick the tires and, if we thought the pavement mixture needed something else, we kind of guessed, turned some dials, and that's what it was," says WisDOT's Ryan. "So we're very excited that, in this last decade, we've learned more about the materials that are going into the mixture and what the end-product is coming out. And the SPT should allow us now to define what it is that we actually want to come out, in terms of an expected performance."

Next Steps

FHWA plans to continue evaluating the SPT throughout the country, visiting as many locations in the Mobile Asphalt Lab as possible. Because only two mixes at each location will be tested, State transportation agencies eventually will need to have their own SPTs.

To further assist States, FHWA will post data on the agency's pavement technology Web site showing the range of stiffness values found at given temperatures for various types of asphalt mixes around the country. So as States begin to purchase SPTs and conduct tests, they can compare their test results to the posted data. FHWA expects that States will find data from a region comparable to their own to confirm that their tests are functioning properly.

Leslie Ann Myers, Ph.D., is program director for the Mobile Asphalt Laboratory, which evaluates new and research-grade asphalt testing technology in the field by visiting States throughout the United States. She is a member of the Asphalt Team in FHWA's Office of Pavement Technology. She is also a member of the Design Guide Implementation Team and has worked on pavement design and analysis for the last 10 years. She holds a doctorate in civil engineering from the University of Florida; her research focused on top-down cracking in asphalt pavements.

John D'Angelo is an asphalt materials engineer in the Office of Pavement Technology at FHWA. He has been with FHWA for 27 years. For the past 12 years, he has been involved with the implementation of Superpave. He has published numerous papers on material testing and quality control.

January/February 2005

Evaluating the Field Performance of Asphalt Mixtures In the Lab by Leslie Ann Myers and John D'Angelo