1

In This Issue

NCAT Completes Evaluation ofRelationships of HMA In-PlaceAir Voids, Lift Thickness, andPermeability

— Page 1

Putting Researchinto Practice — Page 4

Asphalt Forum — Page 8

Asphalt ForumResponses — Page 9

Hot Mix Asphalt Technology -50 Years Ago — Page 11

Specification Corner— Page 13

(continued on page 2)

A Publication of the National Center for Asphalt Technology, Auburn University Volume 16, Number 1 SPR ING 2004

AUBURN UNIVERSITY

Placement of hot mix asphalt with different lift thicknesses for NCHRP Project 9-27

NCAT COMPLETES EVALUATION OFRELATIONSHIPS OF HMA IN-PLACEAIR VOIDS, LIFT THICKNESS, ANDPERMEABILITY

The National Center for AsphaltTechnology (NCAT) has completedNat ional Cooperat ive HighwayResearch Program (NCHRP) Project9-27, “Relationships of HMA In-PlaceAir Voids, L i f t Thickness, andPermeability.”

It is generally accepted that propercompaction of hot mix asphalt (HMA)pavements is vital for a stable anddurable pavement. High air voidsresulting from low in-place densityallow water and air to penetrate into

the HMA pavement leading to anincreased potential for water damage(stripping), oxidation, raveling, andcracking. Past research has showndense-graded HMA pavementsbecome excessively permeable at in-place air voids above 8 percent.However, recent studies have indicatedthat coarse-graded Superpave mixturescan be excessively permeable towater at in-place air voids less than8 percent. Therefore, there is a need

2

National Center for Asphalt Technology, 277Technology Parkway, Auburn, Alabama 36830Phone: (334) 844-NCAT (6228) Fax: (334) 844-6248

ASPHALT TECHNOLOGY NEWS (Library of CongressCatalog No. ISSN 1083-687X) is published by the NationalCenter for Asphalt Technology (NCAT) of AuburnUniversity. Its purpose is to facilitate the exchange anddissemination of information in hot mix asphalt technology,trends, developments and concerns. Opinions expressedin this publication by contributors and editors, the mentionof brand names, the inclusion of research results, and theinterpretation of those results do not imply endorsementor reflect the official positions or policies of NCAT or AuburnUniversity.

E. Ray Brown, DirectorPrithvi (Ken) Kandhal, Associate DirectorEmeritus/ EditorJames S. Killian III , Associate Editor, College of Engineering

—NCAT Completes Evaluation(continued from page 1)

(continued on page 3)

to determine the minimum in-place density (maximumallowable air voids) needed for Superpave mixtures toachieve a relatively impermeable and durable HMApavement.

Recent research conducted by the Florida Departmentof Transportation (FDOT) has indicated that HMA liftthicknesses can have an influence on density, and hencepermeability. FDOT constructed several pavement testsections on Interstate 75 that included Superpavemixtures of different nominal maximum aggregate sizes(NMAS) and lift thicknesses. Results of this experimentsuggest that increased lift thicknesses can lead toincreased pavement density and, hence, lowerpermeability. The FDOT study suggested that a liftthickness (t) to NMAS ratio (t/NMAS) of 4.0 ispreferred. In the past, a minimum t/NMAS of 3.0 hasgenerally been used in the industry. Obviously, therewas a need to evaluate this ratio based on NMAS,gradation, and mix type (Superpave and SMA) due tothe potential problems of achieving the desired densityin the field.

Another major concern that has come to the forefrontsince the adoption of Superpave is the proper method

Cores obtained from test sections which used different liftthicknesses for 19 mm coarse-graded Superpave mix

to measure the bulk specific gravity (Gmb

) of compactedHMA samples. Most Superpave mixtures that have beendesigned and placed on the roadway have been coarsegraded. The currently accepted method for determiningthe G

mb is AASHTO T166. This method consists of first

weighing a dry compacted sample in air and thenobtaining the submerged and saturated surface drymasses. This procedure was usually adequate for therelatively fine-graded mixes used in the past. However,with the coarser gradations typically being used today,errors in G

mb measurements have been observed in some

cases at typical in-place air void ranges.The importance of the G

mb measurements lies in the

fact that it is the basis for volumetric calculationsutilized during both HMA mix design and construction.During Superpave mix designs, volumetric propertiessuch as air voids, voids in mineral aggregate (VMA),and voids filled with asphalt (VFA) at the design numberof gyrations are used to evaluate the acceptability ofmixes. All of these volumetric properties are based uponG

mb.

In most states, acceptance of constructed pavementis based upon percent compaction. Pay factors, whetherreductions or bonuses, are generally applied to percentcompaction. Thus, the errors observed when utilizingAASHTO T 166 can significantly affect both the ownerand the contractor.

The objectives of this major research project were(1) to determine the minimum ratio of lift thickness tand nominal maximum aggregate size (t/NMAS) neededfor desirable pavement density levels to be achievable,and thus impermeable pavements; (2) to evaluate thepermeability characteristics of compacted samples at

3

different thicknesses; (3) to evaluate factors affectingthe relationship between in-place air voids, permeability,and lift thickness; and (4) to recommend a propermethod for determining the bulk specific gravity (G

mb)

of compacted HMA.All four objectives of this research project are

interrelated. Permeability has been shown to be relatedto pavement density. Increased lift thickness has beenshown to allow desirable density levels to be more easilyachieved. Permeable mixes tend to increase the potentialfor errors in G

mb measurements when using AASHTO

T166.The evaluation of the relationship between thickness,

density, and permeability was conducted both in thelaboratory and the field. Various HMA lift thicknesseswere placed in a Superpave gyratory compactor anddensity determined. The variables included: 3aggregates, 4 gradations, 3 nominal maximum aggregatesizes (NMAS) for Superpave mixes, and 3 NMAS forSMA mixes. After the mix designs were performed forthese mixes they were compacted in the Superpavegyratory (100 gyrations) to t/NMAS ratios of 2.0, 3.0,and 4.0. The effect of t/NMAS on density was thendetermined. The plan was to select the t/NMAS thatgave optimum density. However, the results did notprovide a clear answer.

It was then decided to use a vibratory compactor. Itwas believed that the vibratory compactor would bettersimulate field compaction and would provide morereasonable answers. However, the test results from thevibratory compactor were also unable to provide a clearanswer.

Therefore, the effect of t/NMAS on compaction wasevaluated in a field study where the lift thicknesses werevaried and compacted under similar conditions. TheNCAT test track was being rebuil t during thisevaluation. All of the mixes to be placed at the trackhad to evaluated in a test section prior to actualplacement on the track. Seven mixes from the track wereconstructed on a paved surface adjacent to the track tolook at the effect of lift thickness on density. A generaldescription of these seven mixtures is provided in thetable. The seven mixes were compacted at l i f tthicknesses varying from 2 to 5 t/NMAS. For some ofthese seven mixes one side was compacted withvibratory roller and the other side was compacted withvibratory and rubber tire roller. The test data wasevaluated and provided reasonable results forestablishing minimum t/NMAS.

Another part of this field study examined the effectof lift thickness on permeability. Permeabilitytests were conducted in-place with a fieldpermeameter and in the laboratory with alaboratory permeameter using core specimensfrom the seven test sections.

Based upon the results of this study, thefollowing conclusions have been drawn.

• The ratio of lift thickness t and the nominalmaximum aggregate size (t/NMAS) is clearlyrelated to the density that can be obtained undernormal rolling condit ions. For improvedcompactibility, it is recommended that the t/NMAS be at least 3 for fine-graded mixes andat least 4 for coarse-graded mixes. Ratios lessthan these suggested numbers can be used butit will likely require more compactive effort toobtain the desired density. Care must beexercised when the thickness gets too large toensure that adequate density is obtained.

(continued on page 7)

Mix Information for Field Density Study

Section NMAS Gradation Asphalt Type Aggregate Type

1 9.5 mm Fine-Graded Superpave Unmodified Granite and Limestone

2 9.5 mm Coarse-Graded Superpave Unmodified Limestone

3 9.5 mm SMA Modified Granite

4 12.5 mm SMA Modified Limestone

5 19.0 mm Fine-Graded Superpave Unmodified Granite and Limestone

6 19.0 mm Coarse-Graded Superpave Unmodified Granite

Effect of t/NMAS on permeability of compacted HMA at 7% air voids

4

PUTTING RESEARCH INTO PRACTICE

The following papers were presented at the annualmeeting of the Transportation Research Board(TRB) held in Washington, DC in January. We arereporting observations and conclusions from themwhich may be of value to field engineers. Thesecomments are obtained mostly from researchprojects with a limited scope; before application topractice we recommend that you read the entirepaper to determine its limitations. Titles of thepapers are given, with names of authors inparentheses, followed by a brief summary.

1. WORKABILITY OF HOT MIX ASPHALT( G u d d i m e t t l a , C o o l e y , a n d B r o w n)

One of the important issues in the HMA industry isthe compactibility of mixes. This concern has arisenfrom the increased use of coarse-graded mixes and/orthe increased use of polymer-modified binders, whichare difficult to compact because of low workability. Ifthe workability of a mix can be quantified, it will alsoindicate the compactibility of the mix.

The primary objective of this study was to develop adevice to measure the workability of hot mix asphaltmixes. Workability was used in this study to describethe ease with which a HMA could be constructed. Basedon the test results and analyses conducted in this study,the following conclusions were drawn:

• The prototype workability device developed in thisstudy was successful in differentiating various mixcharacteristics. The workability of mixes was measuredby pushing a paddle through hot HMA placed in a bucketand measuring the torque required to maintain a specificrate of revolution. Measurements can be made atdifferent temperatures as the mix in the bucket coolsdown.

• The workability of HMA was affected by aggregateparticle shape and angularity. Mixes prepared with acubical, angular granite were less workable (requiredmore torque at a given temperature) than mixes preparedwith a semi-angular, crushed gravel.

• The workability of HMA was affected by thenominal maximum aggregate size (NMAS) of the mix.As the NMAS increased for a given aggregate type,gradation shape, and binder type, the workabilitydecreased.

• Binder type significantly affected the workabilityof mixes. As expected, mixes containing a PG 76-22binder were significantly less workable than mixescontaining a less stiff, unmodified PG 64-22 binder for

a given temperature.• As expected, the temperature of the mix

significantly affected the workability of HMA. Therewas a relationship between workability and temperaturethat showed increased workabi l ity at h ighertemperatures.

2. EVALUATION OF NEW TEST PROCEDURESFOR DETERMINING THE BULK SPECIFICGRAVITY OF FINE AGGREGATE USINGAUTOMATED METHODS (Prowell and Baker)

Determining the bulk specific gravity of fineaggregate is very important when designing HMAmixtures and for other uses. The bulk specific gravityis used in calculating the voids in the mineral aggregate(VMA) of an HMA mixture. The current method ofdetermining the dry bulk specific gravity (Gsb) of fineaggregates using AASHTO T 84 or ASTM C 128, usesa cone and tamp to determine the saturated surface-dry(SSD) condition of a fine aggregate. This method doesnot work well when determining the SSD condition ofangular or rough fine aggregates because they do notreadily slump. Therefore, a more accurate and morerepeatable method of determining Gsb is needed toprovide lower variability between operators and toaddress problems associated with angular materials. Inorder to solve this problem, a method that is moreautomated and less user dependent is needed todetermine both Gsb and absorption of fine aggregates.

The objective of this study was to evaluate twoautomated methods for determining the Gsb of fineaggregates. Both methods were evaluated against thestandard method described in AASHTO T 84. A roundrobin was conducted according to ASTM C 802 for eachmethodology. The round robin data were used tocompare the Gsb, Gsa (apparent specific gravity), andabsorption expected from a cross section of laboratoriesand materials. The round robin data were also used todevelop a precision statement for each automatedmethod.

The fol lowing two automated devices wereevaluated:

• SSDetect. This device uses an infrared light sourceand detector to determine when the fine aggregate hasreached the SSD condition.

• Corelok Device. This device uses a combinationof a calibrated pycnometer and a vacuum-sealing deviceto determine Gsb and absorption.

(continued on page 5)

5

Six different aggregates were selected for the roundrobin testing. A wide variety of aggregates were chosenin an attempt to cover a wide range of materialproperties. A limestone, a medium and high dustdiabase, a slag, a rounded sand, and a crushed gravelwere selected for the study.

The following conclusions were drawn from thisstudy: • The CoreLok and SSDetect methods of determiningfine aggregate specific gravity offer significant timesavings over AASHTO T 84. Neither method requiresthe 16-hour soak period included in AASHTO T 84.This means that belt sweep samples of aggregates couldbe taken in a HMA plant to verify bulk specific gravitiesduring construction for volumetric calculations in atimely manner.

• Both the Corelok and SSDetect methods generallyproduce Gsb results that are similar to AASHTO T 84.Where statistical differences occurred for Gsb betweenthe new methods and AASHTO T 84, the differenceswere smaller for the SSDetect method. It is believedthat AASHTO T 84 may not produce accurate resultsfor angular materials with high dust contents. Morefrequent statistical differences exist between AASHTOT 84 and both the Corelok and SSDetect Gsa (apparentspecific gravity) and water absorption results. However,Gsa and water absorption are not used in volumetriccalculations for hot mix asphalt.

• The SSDetect offers improved precision as comparedto AASHTO T 84. The precision of the Corelok methodis not as good as AASHTO T 84. It is possible that theprecision of the Corelok method will improve astechnicians become more familiar with the procedure.

NCAT is planning to conduct a similar evaluationfor the Gilson automated SSD device.

3. LABORATORY PERFORMANCE TESTINGOF OGFC MIXTURES (Watson, Cooley, Moore,and Williams)

A tentative mix design system was recentlyrecommended by the National Center for AsphaltTechnology (NCAT) for new generation open-gradedfriction course (OGFC) mixes. The new-generationOGFC mix has a coarser gradation than typically usedin the past and includes fiber stabilizer and/or polymermodified asphalt binder to resist draindown. Polymermodified asphalt binder is also used to improve mixturedurability and resist raveling.

Mix is compacted using 50 gyrations of a Superpavegyratory compactor and air void contents are determinedby dimensional analysis. The Cantabro abrasion test isthen conducted on unaged and aged (7 days at 60°C ina forced draft oven) samples. The Cantabro abrasiontest is used in several European countries. A singleMarshall compacted specimen is placed in a LosAngeles abrasion machine without the steel balls andis allowed to tumble in the drum for 300 revolutions.The resulting loss in mass from abrasion is thenmeasured. An asphalt content that meets the followingcriteria is selected as optimum asphalt content.

1. Air Voids. A minimum of 18 percent based ondimensional volume is acceptable, although highervalues are more desirable. The higher the air voids are,the more permeable the OGFC.

2. Abrasion Loss on Unaged Specimens. The

NCAT Applications Steering Committee met at NCAT last December

(continued on page 6)

6

(continued on page 7)

abrasion loss from theCantabro test should notexceed 20 percent.

3. Abrasion Loss on AgedSpecimens. The abrasion lossfrom the Cantabro test shouldnot exceed 30 percent.

4. Draindown. Draindownshould not exceed 0.3 percentby total mixture mass.

The objective of this studywas to evaluate the laboratoryperformance of OGFCmixtures based on previousrecommendations and, i fnecessary, refine the tentativeOGFC mix design proceduredeveloped by NCAT based onresults of the laboratory performance tests. Thelaboratory tests include a comparison of Cantabro testresults for 50-blow Marshall and SGC compactedspecimens; evaluation of a modified version ofAASHTO T283 for moisture conditioning, whichinvolved up to five freeze-thaw cycles and a comparisonof the effect of fibers and polymer modified asphalt inreducing draindown.

Three aggregate types: granite, siliceous crushedgravel, and traprock were used for this study. ThreeSuperpave performance graded asphalt binders wereutilized: PG 67-22, PG 76-22 (SBS modified) and PG76-34 (rubber modified).

The following conclusions were drawn from thisstudy:

• The addition of fiber stabilizers significantlyreduced the potential for draindown.

• Sample size did not significantly affect theCantabro test results when comparisons were madebetween the 100 mm (4 in) diameter Marshall specimensand the 150 mm (6 in) diameter SGC samples.

• Unconditioned SGC samples should have Cantabrostone loss of no more than 20 percent. Since there is nosignificant difference between unaged and aged sampleresults, the aging procedure is not necessary.

• Results from this study show no significantdifference in tensile strength when 1, 3, or 5 freeze-thaw cycles are used in the moisture conditioningprocedure. Therefore, only one freeze-thaw cycle isneeded.

4. DETERMINING AIR VOIDS CONTENT OFCOMPACTED SMA MIXTURES (Xie andWatson)

Stone Matr ix Asphalt (SMA) has been usedsuccessfully in the United States since its firstintroduction in 1991. SMA uses a larger proportion ofcoarse aggregate compared to conventional dense gradedmixtures. The resulting coarse texture of an SMAmixture may result in more internal air voids connectedto surface voids (texture) than conventional mixturesalthough the volume of air voids is the same.Consequently, it becomes difficult to define thevolumetric properties of compacted SMA mixtures. Itis believed that volumetric properties of compactedSMA, especially the air voids content, are the mostimportant parameters for determining the long-termperformance of SMA pavement. Air void content incompacted SMA mixture is calculated from the bulkspecific gravity and the maximum theoretical specificgravity of the mixture.

The standard method currently used to measure theSMA bulk specific gravity is AASHTO T 166, BulkSpecific Gravity of Compacted Bituminous MixturesUsing Saturated Surface-Dry Specimens, commonlyknown as the saturated surface dry (SSD) method.Another alternative method recently used is the CoreLokmethod.

The objective of this study was to compare theCoreLok and SSD methods for determining the bulkspecific gravity of compacted SMA mixtures, and con-sequently the mix design volumetr ic propert iesespecially air voids content.

Five aggregates with a range of LA abrasion loss from17 to 36 percent were selected for the study. Threenominal maximum aggregate size (NMAS) mixtureswere selected: 19 mm, 12.5 mm, and 9.5 mm. All the

Attendees of the meeting of the Task Group on Perpetual Pavement. Sitting (L to R): DaveNewcomb, Jim Musselman, Dave Timm, and Dean Weitzel. Standing (L to R): Brian Prowell,Prithvi Kandhal, Ray Brown, Joe Mahoney, and Harold Von Quintus.

7

samples were compacted by a Superpave gyratorycompactor (SGC) at 100 gyrations.

The CoreLok method uses a vacuum-sealing device.Under vacuum a plastic bag tightly conforms to the sidesof the sample and prevents water from infiltrating intothe sample. The volume of the specimen encapsulatedby the bag is considered as the bulk volume of thesample.

The following conclusions were drawn from thisstudy:

• There is a significant difference between air voidscontent measured by the SSD method and theuncorrected CoreLok method. The difference in airvoids between these two methods had a range from 0.4to 4.6 percent with a maximum air voids content of 7.1percent by the SSD method. The uncorrected CoreLokmethod provided an average of 1.3 percent higher airvoid content than the SSD method.

• The difference between these two methods dependson air voids level and the NMAS. As the air voids levelincreases, the difference between the two methodsincreases as well. With an increase in the NMAS, thedifference between the two methods increases also.

• The CoreLok method was found to have a systemerror that overestimated the air voids content at zero tovery low air void levels. The correction factor for themachine used in this study was determined to be 0.5percent. The correction factor for each machine used inthe Corelok method should be determined by testing asolid metal sample that is approximately the same sizeand shape as the asphalt specimen to be tested.

• The difference between corrected CoreLok andSSD results was found to be within one percent air voidsfor all 9.5 mm NMAS mix samples. The differencebetween the two methods was more than one percent

JOIN USWe at NCAT hope you enjoy this issue ofAsphalt Technology News. It is provided freeof charge. If you wish to be added to our mailinglist, please send your business card or your name,company, mailing address, and email address(ATN is sometimes distributed by email only)to us and also indicate whether you prefer toreceive ATN in the mail or by email, send yourinformation to:

Prithvi (Ken) KandhalAssociate Director Emeritus, NCAT277 Technology ParkwayAuburn, AL 36830pkandhal@eng.auburn.edu

• The relationship between permeability and t/NMAS is illustrated in the figure on page 3. The resultswere mixed concerning the permeability of coarse-graded (below the restricted zone or BRZ) and fine-graded (above the restricted zone or ARZ) mixtures. Insome cases there appeared to be no difference and inothers the coarse-graded mixtures appeared to havehigher permeability. Most work that has been doneindicates that coarse-graded mixes become permeableat a lower void level than fine-graded mixes.

Permeability testing needs to be conducted on localmaterials to develop an under-standing of the effect ofair voids, NMAS, and grading on permeability. Ifpermeability appears to be a problem considerationshould be given to making permeability a part of thespecification requirements.

• Four test methods were evaluated to determinethe bulk specific gravity of HMA: the dimensionalmethod, the gamma ray method, the wa t e rdisplacement method (AASHTO T166), and thevacuum-sealing method.

When laboratory prepared samples having low levelsof water absorption were evaluated, the dimensionalmethod resulted in the highest air void contentsfollowed by the gamma ray method. The vacuum-sealing and water displacement methods resulted insimilar air void contents when the water absorption levelwas low.

At low levels of water absorption, the waterdisplacement method is an accurate measure for bulkspecific gravity. The error develops when removing thesample from water to determine the saturated, surface-dry weight. The allowable absorption level to use thedisplacement test method is now 2 percent in AASHTOT166 but this can create accuracy problems accordingto this study. This number should probably be reducedto 1 percent or even lower for better accuracy. However,it is recommended that the absorption limit for the waterdisplacement test method be set at 1 percent due topractical considerations.

For mix design samples and other laboratory samplesthat are compacted to relatively low voids, thedisplacement method will provide reasonably accurateanswers. However, for field samples where the voidlevels will typically be 6 percent or higher, the vacuum-sealing method may often need to be used.

—NCAT Completes Evaluation(continued from page 3)

air voids for 19 mm NMAS samples with air voidsgreater than 3.7 percent, and for 12.5 mm NMASsamples with air voids greater than 5.6 percent.

8

Colorado (Bill Schiebel, Colorado DOT)Colorado is interested in any agency experience or

any field performance data comparing projects builtwith polymer modified binders and projects built withacid or chemically modified binders. Are any otheragencies taking action to limit the use of chemicallymodified binders in Superpave mixtures?

South Carolina (Milton Fletcher, South CarolinaDOT)

How are other states dealing with the issue of acidmodified binders? Are restrictions being placed on allbinder testing to minimize their use, such as an elasticrecovery specification, or are the restrictions such thatno acid modification is allowed at all?

NCAT (Prithvi “Ken” Kandhal, Associate DirectorEmeritus)

Since 1992, the National Center for AsphaltTechnology (NCAT) has conducted extensive researchon different techniques of constructing longitudinaljoints. The objective was to recommend techniqueswhich give durable longitudinal joints. NCAT Research

Reports 97-4 and 02-03 pertaining to this research canbe downloaded free from our website, <http://www.ncat.us>. The main conclusions from this researchproject are as follows:• The longitudinal joints with high densities right

at the joint generally show better performancethan those with relatively low densities.Highway agencies should specify minimumcompaction levels to be achieved at thelongitudinal joint to ensure good performance.It is recommended that the density right on thejoint be not more than two percent lower thanthe density specified in the lanes away from thejoint.

• Application of rubberized asphalt tack coat onthe unconfined edge of the first paved laneresulted in a durable longitudinal joint.

• The Michigan notched wedge joint (12.5 mmvertical offset and 12:1 taper) has a goodpotential of obtaining a satisfactory longitudinaljoint.

NCAT would like to ask the highway agencies whatsteps they have taken in recent years for obtaining adurable longitudinal joint. Please report the techniquesand related specification criteria implemented. This willfacilitate exchange of ideas on this important subject,which will be published in the next issue of the AsphaltTechnology News.

ASPHALT FORUMNCAT invites your comments and questions. Questions and responses are published in eachissue of Asphalt Technology News . Some are edited for consistency and space limitations.

PROFESSOR TRAININGCOURSE IN ASPHALT

TECHNOLOGY

NCAT has written and published an up-to-date college textbook on asphalt technology.NCAT has also developed a training programfor college and university civil engineeringfaculty that will allow them to offer state-of-the-art undergraduate and elective courses inasphalt technology. This 8-day intensivecourse is conducted at NCAT in June everyyear. It will be held on June 15-24 this year.The course has been updated to includeSuperpave binder and mix technology, andstone matrix asphalt (SMA). Some financialassistance in attending this course is possible.Please call NCAT at (334) 844-NCAT forbrochure or information or visit our web siteat <http://www.ncat.us>.

DOWNLOAD NCAT RESEARCHREPORTS AT NO COST

Over 75 NCAT research reports are now availableas PDF (portable document format) files whichcan be easily downloaded at no cost from our website. You will need the Adobe Acrobat Reader,which can also be downloaded free from ourhomepage, to open these files. Visit our web siteat http://www.ncat.us and click on NCATPublications. Previous editions of AsphaltTechnology News are also available from ourhomepage.

•

•

•

9

ASPHALT FORUM RESPONSES

The following responses have been received to questions raised in the Fall 2003 Asphalt Forum.

What are the experiences of other states with the fineaggregate angularity (FAA) requirement in Superpavemixture design? We are interested in finding out if ASTMC1252 Method A has been preferred or if alternatemethods/technologies have been used for evaluating thefine aggregates. Have some agencies discontinued theFAA evaluation? (Ed Johnson, Minnesota DOT)

Colorado (Bill Schiebel, Colorado DOT)Colorado continues to require FAA minimum of 45.

We use AASHTO T304 for determining the FAA value.

Indiana (Tommy Nantung, Indiana DOT)We require that the fine aggregate angularity value

of the total blended aggregate material and recycledmaterials should meet or exceed the minimum valuesfor the appropriate ESAL category and position withinthe pavement structure.

Kansas (Dick McReynolds, Kansas DOT)The Kansas DOT conducted research prior to

Superpave to adopt the original angularity test method(ASTM C1252) for field use with Marshall mixes. TheKansas Test Method KT-50 “Uncompacted Void Contentof Fine Aggregate” proved extremely successful becauseit allowed aggregate suppliers, contractors, and theKansas DOT field staff to determine angularity in about30 minutes. The test method uses a volumetric flask todetermine the equivalent specific gravity required bythe standard test method. When Superpave wasimplemented, further research was done to confirm thatthe test method would also work with mixed fineaggregates (including limestone). The test method is stillin routine use.

Kentucky (Allen Myers, Kentucky TransportationCabinet)

Kentucky utilizes AASHTO T304 (Method A) fordetermining FAA values. We test our asphalt mixturesfor FAA at the laboratory design stage only, and ourFAA requirements are very similar to AASHTO MP2with one minor exception: for projects with less than0.3 million 20-year ESALs, we require a minimum FAAof 40 for mixtures located both < 100 mm and > 100mm from the surface.

New Mexico (John Tenison, New Mexico DOT)AASHTO T304 (Method A) is used to determine

FAA for hot-mix asphalt aggregates. We have notdiscontinued its required use and have no plans in thefuture to do so.

Maine (Dale Peabody, Maine DOT)Maine uses AASHTO T304 (Method A) for testing

the “uncompacted void content of fine aggregate” forSuperpave mixtures. We do not use the ASTM C1252.To date, we have not experienced any problems withour aggregates meeting the minimum FAA valuesrequired for Superpave designed mixes.

Missouri (Mark Shelton, Missouri DOT)Missouri uses AASHTO T304 (Method A) for

determining FAA and all of our crushed materialgenerally have FAA values equal to or above 45 percent.A large percentage of our Superpave mixtures have the45 percent minimum requirement, and therefore, fewcontractors have looked at blending rounded naturalsands in order to develop a mixture that compacts moreeasily.

Nebraska (Laird Weishahn, Nebraska Departmentof Roads)

Nebraska uses AASHTO T304 (Method A) and wehave actually increased our FAA requirements for lowertraffic level mixtures due to the abundance of naturalsands and rounded river gravels in Nebraska. The rivergravels mostly have to be crushed to meet the FAArequirement.

Nevada (Michael Dunn, Nevada DOT)We have very limited experience with evaluating fine

aggregates using the Superpave system. Over the lasttwo years we have only constructed four Superpave testsections and the fine aggregates were evaluated usingAASHTO T304. In each case, the FAA barely met theminimum requirement.

New Hampshire (Alan Perkins, New HampshireDOT)

We use AASHTO T304.(continued on page 10)

10

West Virginia (Richard Genthner, West VirginiaDivision of Highways)

As noted in the Specification Corner, we arelowering the requirement for FAA to 43 percent forblends comprised of 100 percent crushed aggregate. Wehave experienced some problems with a couple oflimestone sources that would not always meet the 45percent requirement. These were sources that hadprovided quality, rut-resistant mix designs for several

RESEARCH IN PROGRESSWe have discontinued the publication of this

column in this newsletter because it can now beaccessed on NCAT’s homepage (http://www.ncat.us). Click on “Information” at the topof the page, then “Research in Progress.” It isupdated frequently based on the informationreceived f rom the Depar tments ofTransportation and other sources.

New Jersey (Frank Stia, New Jersey DOT)The New Jersey DOT uses AASHTO T304 (Method

A) which is similar to the ASTM method C1252.

South Carolina (Milton Fletcher, South CarolinaDOT)

South Carolina has only used FAA as a tool to assistin the approval of Superpave mix designs. ASTM C1252(Method C) has been utilized with good success, butwe have noted that this procedure could eliminate knownangular material.

Mississippi DOT (MDOT) and HMA industry personnel visited NCAT Test Track and laboratory facilities in April. The DOTpersonnel included: Wayne Brown (Chairman of the Mississippi Transportation Commission); Butch Brown (MDOT ExecutiveDirector); Harry James (MDOT Chief Engineer); Joy Portera (MDOT Assistant Chief Engineer, Operations); David Foster(MDOT Assistant Chief Engineer, Preconstruction); Melinda McGrath (MDOT Assistant Chief Engineer, Maintenance); RandyBattey (MDOT State Research Engineer); and Richard Sheffield (MDOT Assistant State Materials Engineer). The group isseen here examining the two Mississippi test sections on the NCAT Test Track.

years, so we did not feel comfortable in limiting theiruse based solely on the FAA criteria.

Ontario (Kai Tam, Ontario Ministry of T rans-portation)

A minimum FAA value of 43 is acceptable.

11

Asphalt technology papers published many years agomake an interesting and informative reading.Sometimes, it seems we are trying to reinvent the wheel.

The following are excerpts from the Proceedings ofthe Association of Asphalt Paving Technologists(AAPT), Volume 23, 1954. We hope you will find theminteresting. The comments in parentheses are those ofthe editor (Prithvi “Ken” Kandhal).

F.B. Odasz and D.E Nafus, “Statistical QualityControl Applied to an Asphalt Mixing Plant”

“At the present time, a great deal of sampling andtesting is being done in asphalt paving work. Besidesthe Marshall tests, this work takes the form of tests forvoids content, analysis of the mix for asphalt content,screen analyses, the Hubbard-Field test and others. Howmuch more wisely these results could be used! Howmuch better basis for action the same, or even a lesser,amount of work would provide if the data were used inthe form of control charts than just a running record!”

“Statistical quality control chart is a simple anduseful tool that will signal the user when something iswrong and provide him with a basis for action to rectifythe situation and control the variability of the process.Prompt and directed action will reward him with aneconomic advantage by reducing waste motion andmaterial, and by assuring consistent operation at highquality.”

(This paper illustrates the use of control charts forquality control of HMA. Individual test results are

plotted on the charts which show the job-mix formulatarget value, lower control limit, and upper control limit.Whereas some states have used these charts for anumber of years, others have just started to use them intheir quality control/quality assurance program.)

At this AAPT meeting held in Louisville, Kentucky,a symposium titled “Control of Plant and PavingOperations in Asphalt Concrete Surface Construction”was also held. Some interesting observations were madeduring this symposium.

Roger Martin , who chaired the symposium, madethe following introductory remarks:

“We are assembled today to discuss these problemsfrom the “how-to-do-it” standpoint, more effectivecontrol in plant manufacture, and more efficientlaydown methods. After we have exchanged informationalong this line, possibly, some of the more brilliantmembers will turn their thoughts toward rationalevaluat ion of plant manufacture and laydowntechniques. Questions pop into your mind in rapid-fireorder. What is the exact mechanism of internal moisturein aggregates, which seemingly defies all laws of natureby running out the back end of a truck containing hotmix at a temperature 75°F above the boiling point ofwater? What is the boiling point of water in the capillaryspaces? Does its heat of vaporization differ appreciablyfrom that of surface water? What is the relation between

HOT MIX ASPHALT TECHNOLOGY - 50 YEARS AGO

(continued on page 12)

Pam Turner

PAMELA TURNER JOINS NCATPamela Turner joined the NCAT staff, as a laboratory engineer, in March. Pamela

received her Bachelor’s degree in Civil Engineering from the University of Kentucky in1996. While attending UK, she worked as a co-op student at the Asphalt Institute laboratory.After completing her degree, she accepted a full-time position with the Asphalt Instituteas a Materials Engineer, focusing primarily on asphalt binder testing. Pamela is experiencedin all aspects of asphalt binder testing, including conventional test methods and theSuperpave system. She has participated in many research projects, both at the nationallevel and for private clients, and has supervised the everyday activities of one of thepremier asphalt binder testing facilities in the country. Although her primary field of expertise is asphaltbinder testing, Pamela is also familiar with the Superpave mix design system, Superpave performance tests,and many other asphalt related tests.

12

the time required for this water to dissipate itself andthe size of the rock and the nature of the rock? A rollerwill shear and “squash” a pavement at too high atemperature, but over a certain range of temperaturesthe aggregate will “lock” and satisfactory compactiontakes place. If it is rolled after it is too cool, permanentcracks are created. What method can be found toestablish, in advance, the optimum rolling temperaturelimits? These are only a few of the questions which needto be answered.”

W.H. Campen: “It is my opinion that the ChiefInspector at the plant ought to really know asphalticmixes. He ought to know bituminous mixes, from thebottom up. He ought to know how to analyze them. Heought to know if they are right when he looks at them.As far as the contractor is concerned, he ought to havegood enough men at the plant so that they can keep themix the way it has been set up for them. The contractorshouldn’t expect the inspector to guarantee his mix. Inother words, the inspector can see that the mix is setright and then the contractor has to more or less keep itthat way.”

(Those were the days when the highway departmentinspector more or less controlled the asphalt plants.)

Roger Martin: “Every time I hear this argumentabout batch plants and continuous plants I feel that it isa “tempest in a teapot.” Everybody gets good resultswith both of them. I think the main factor is the menrunning the plant. That is more important than what kindof plant you have.”

(This still applies to batch plants and drum plants.)

George Birney: “I think one tool that should behighly recommended is a pick handle on the back of ascreed of a paving machine for those individuals whoare anxious to jiggle that screed up and down. Anotherthing I have noticed on the spreading machine is thatthe operator will keep his feeding pan that leads intothe hopper wide open, causing an uneven flow of mixto the screed.”

W.H. Campen: “Last year I was called to a projectwhere an asphaltic concrete mixture was being preparedin a continuous mixing machine. After the mixture wasgiven a preliminary rolling, it was allowed to cool forabout 20 minutes. When rolling was resumed, themixture was so unstable as to shove and flow out fromunder the roller. An investigation revealed a veryabsorptive aggregate. As the mixture came out of themixer, water vapor issued from it. We finally attributed

the lack of stability to condensed water vapor whichbecame entangled in the mixture and prevented adhesionand cohesion.”

(This aggregate had a water absorption of 5 percent.)

D.D. Dagler: “Years ago a plant would put out threeor four hundred tons in a ten-hour day and the contractorhad two rollers on the job. Now he puts out a thousandtons a day and he adds another roller and he thinks thatis sufficient. You just don’t get it and you just don’t getthe workmanship.”

J.J. Forrer: “I have had the opportunity of lookingon both sides of the fence, on the highway department’spoint of view and now on the contractor’s point of view.I’ve had to change my views somewhat since I havebeen working on the contractor’s side. I have found thatif the highway engineer could more clearly specify theend results that he wants, the contractor can get that forhim and probably at less cost to the taxpayers thanputting a whole lot of these gadgets and things like thaton asphalt plants. I’ve seen these gadgets used and mustsay that the plants today are vastly improved over thefirst one I saw operated in 1912 in Memphis.”

“The point that I am trying to get at is I am wonderingif we shouldn’t begin to think about a change ofprocedure; that is leave it up to the contractor. Say tohim, “I want this. It’s up to you to get it. If you want tomake your mix in a concrete mixer, go ahead. I amlooking to you for a result in the end.” That proceduremay be a little bit extreme and I purposely painted itextreme but I think that a good contractor can get theexcellent results the highway engineer

wants without too much additional refinements onthe equipment. It makes no difference whether he getsthat from continuous mix or batch mix, or whether hedoes it with a wheelbarrow or the newest equipment.”

(Now, that is called “end result specification” thesedays, which is being adopted or considered by manystate DOTs.)

NCAT has a new domain name, to make iteasier to f ind our web page. <http://www.ncat.us> is the new address, but youcan also reach our web page by going to ourold address, <http://www.eng.auburn.edu/center/ncat/>.

NCAT HAS NEWDOMAIN NAME

13

SPECIFICATION CORNER

(continued on page 14)

Alabama - Effective in April this year, the AlabamaDOT is adopting the locking point mix design concept– similar to that used by the Georgia DOT. The lockingpoint is where the sample being gyrated loses less than0.1 mm in height between successive gyrations. Thespecifications require the locking point to be between60 gyrations and the design number of gyrations. If themix locks up outside this range, either 60 or the designnumber of gyration are used instead of the locking point.

After the locking point for the aggregate structure isdetermined, the optimum asphalt content is refined usingthe locking point as the new design number of gyrations.This approach is being taken to increase the overallasphalt contents in Superpave mixtures.

Colorado - Last year, the Colorado DOT made changesto the method of establishing production targets on allHMA projects with the purpose of increasing the asphaltcontent in the mixes. The historic Superpave mix designremains the same, but production targets for asphaltcontent, VMA, and other parameters are establishedbased on the mix design data after a shift down in voidsof up to 1.0 percent. This method helps ensure that thevoids in produced HMA are filled with asphalt binderrather that other material.

Full incentive/disincentive specifications that impactpayment for all HMA placed on the project were adoptedlast year. These specifications include longitudinal jointdensity requirement, which was necessitated due toinadequate performance of joints in the state. A targetdensity of 92 percent is required at the longitudinal jointwith a tolerance of “ 4 percent. The new joint densityspecification increased the overall density and decreasedthe variability during last year. The performance of tenprojects before and ten projects after the newspecification was adopted, is being tracked.

The Colorado DOT has begun to develop a pavementsegregation specification. Segregation has also longbeen a major concern both to the DOT and the industryin the state. A pilot specification will be used this yearto gather density profile data using methods similar tothose used in Texas and Washington.

The Colorado DOT has banned the use of acid andalkali modification of binders for all 2004 projects.Detection methods are currently being finalized, andinclude pH and X-ray methods. Comparison testing ofbinders containing acids and polymers is beingconducted using the Hamburg wheel device.

A pilot project of testing aggregates using the MicroDeval test has also begun.

Kansas - Since January 2001, only major Superpaveprojects had provisions for bonus and paymentdeduction for air voids and density based on percentwithin limits (PWL). Starting in October 2003, PWLspecifications for air voids were added to all Superpaveprojects. Based on a review of data from the completedprojects, the acceptable range for air voids was changedfrom 2.75-5.25 percent to 3.0-5.0 percent.

Kentucky - It is planned to publish a new specificationsmanual in March this year that contains several changesto HMA requirements. The major revisions involve theHMA acceptance program. The present paymentschedule will be divided into three schedules dependingon the method of density acceptance. The paymentschedule for VMA will be modified to permit a threepercent bonus for values above the AASHTO MP2specified minimums and full pay for values down to0.5 percent below the MP2 minimums. The paymentschedules for air voids and lane density will correspondto the applicable ESAL level. A linear payment schedulewill be introduced for air voids such that each 0.1-percent change in air voids results in a one-percentchange in pay. A payment schedule for joint density forsurface mixtures has also been implemented. Finally, aprocedure has been standardized for determining whenHMA with substandard properties should be removedand replaced versus accepting that pavement in placewith a reduced pay factor.

Missouri - Missouri’s Standard Specifications havebeen reviewed and revised. The new specifications willbe effective in July this year. The specifications werereviewed to incorporate performance-related criteria andquality control and quality assurance (QC/QA), wherepossible. Some of the revisions are: (a) addition oftensile-strength ratio (TSR) for plant-produced HMA;(b) elastic recovery requirement for asphalt bindersbased on the spread between the upper and lowertemperature gradings; (c) an acid insoluble residuerequirement for aggregates used in HMA surfacemixtures for high traffic pavements; and (d) requirementto conduct dry back procedure when the Rice test isconducted on a mixture containing aggregate with more

14

than 2 percent water absorption.The new “Performance Specifications” can be

viewed at <http://www.modot.mo.gov/design/specbook/perffiles.htm>.

Nebraska - The HMA specification now has tolerancelimits for VMA of the plant-produced mix. Theminimum VMA is based on Superpave specifications.If two consecutive individual VMA values are belowthe minimum by more than 0.75 percent, the HMApayment will be reduced by 20 percent.

Nevada - The PG76-22NV asphalt binder has been usedin the southern part of Nevada for the last two years.Preliminary results indicate that this binder is out-performing the conventional asphalt cements (AC-30and AC-40) used in the past. Starting in 2004, the AC-20P binder will be replaced with PG64-28NV binder inthe northern part of Nevada to have a uniform binderspecification throughout the state.

New Hampshire - QA specifications will now be

appl icable to th in asphal t over lays as wel l .Characteristics to be measured are: gradation, asphaltcontent, and in-place voids.

New Mexico - The Micro-Deval Apparatus (AASHTOTP 58) is being evaluated to establish maximum percentloss that will be allowed for HMA, concrete, andaggregate base course materials. This evaluation shouldresult in a specification requirement for the 2005construction season.

An IRI based smoothness specification has beenadopted that will be used during the 2004 constructionseason. Initially, the 100 percent pay band will be set to62.0 inches/mile but will be reduced by 2.0 inches/mileper year to 56.0 inches/mile by the 2007 constructionseason for all new and reconstruction projects. The 100percent pay band for all pavement preservation andrehabilitation construction projects is 6.0 inches/milehigher than that for new and reconstruction projects.

Oregon - After the Superpave control points wereintroduced for HMA gradation, a trend toward finer

NCAT's Asphalt Technology CourseMarch 1-5, 2004

Back, L-R: Myles Tatum, Robert James (Instructor), Ron Elrod, Don Chappel, Clint Andrews, Doug Hanson (Instructor) Front, L-R: Don Watson (Instructor), Aaron Warren, James Dowland, Wen Hsiao, Ronald Martinez

15

gradations with more sand and excessive minus 200sized material has been observed. This has resulted inthe production of mixes, which are believed to have poorlong-term durability and stripping resistance. To addressthis concern, dust to effective asphalt content ratio (0.8to 1.6) has been added as a specification requirementduring production. More restrictive dust to effectiveasphalt content ratio requirements are being consideredwhen the percentage passing the No. 8 sieve exceeds38 percent.

Rhode Island - An open-graded asphalt friction course(OGFC) mix is being developed based on the currentNCAT pooled fund study. Work is continuing to developa chemically modified, low viscosity crack sealer withfibers, that can be applied at a temperature of about300EF.

South Carolina - South Carolina is in the process ofrevising and updating the Standard Specifications,which are scheduled to be published in January 2005.It is proposed to use the Superpave gyratory compactorfor al l mix designs in the state. This involvesmaintaining the present gradation limits on all mixesand switching to the gyratory compactor instead of theMarshall hammer.

Other key issues in the specification will include theallowance of reclaimed asphalt pavement (RAP) in highvolume surface mixes that utilize the PG 64-22 binderand incorporating an asphalt plant certif icationprocedure that will replace the AASHTO M 156requirements of the current specification.

West Virginia - The fine aggregate angularity (FAA)requirement is being lowered from 45 to 43 percent forblends comprised of 100 percent crushed aggregate.Some problems were experienced with a couple oflimestone sources that would not always meet the 45percent requirement. These sources have providedquality, rut-resistant mix designs for several years.

Ontario Ministry of Transportation1. It is planned to use Ontario’s new special

provision for SMA on all new contracts with more than30 million ESALS (during 20 years’ design life).

2. It is proposed to develop a new IRI-basedsmoothness specification utilizing lightweight profilersfor trial contracts starting in the 2005 constructionseason.

3. An end-result specification for acceptance ofsurface course thickness using 50 mm cores and anotherusing a nondestructive test method is currently being

used on trial contracts. Two test methods have also beenfinalized in conjunction with these specifications tomeasure layer thickness using nondestructive methods.Two trials have been completed successfully and up toeight additional ones are planned for this year.

4. An end-result specification for longitudinal jointcompaction is being developed for use in 2004 HMApaving contracts.

5. A material (and a modified construction)specification for a Superpave 25 mm “rich bottom” mixhas been developed. This will be used for pavements tobe constructed this year, which will be based on theperpetual pavement design concept.

An international symposium on Design andConstruction of Long-Lasting Asphalt Pavementswill be held in Auburn, Alabama June 7-9, 2004.The International Society for Asphalt Pavements(ISAP) is the primary sponsor of this specialtyconference. The National Center for AsphaltTechnology (NCAT) will host the conference inAuburn, Alabama (100 miles southwest of Atlanta,Georgia). The conference is also co-sponsored bythe Federal Highway Administration, the NationalAsphalt Pavement Association, and the AlabamaDOT. The objective of this conference will be toaddress materials, mix design, and constructionprocedures to ensure long-lasting (perpetual)asphalt pavements. Topics will include: perpetualpavement design, materials and mix design,quality control/quality assurance, constructionissues, contracting methods, and other relatedsubjects. On-line registration is available throughthe NCAT website, <http://www.ncat.us>. Foradditional information concerning the conference,contact ISAP at <isap@eng.auburn.edu> or call(334) 844-6228.

INTERNATIONALSYMPOSIUM ONLONG-LASTING

ASPHALT PAVEMENTS

16

Back, L-R: Jason Hale, Doug Hanson (Instructor), Brian Prowell (Instructor), Josh Harvill, Bill Fogarty, Paul Gluszak, Bill Pinkerton, Ralph KinderFront, L-R: Chris NeSmith (Instructor), Don Watson (Instructor), Bob Pelland, Bill Taylor, Robert Smith, John Bourda, Nick Andrade

PresortedStandard

Non-Profit OrganizationAuburn University, AL 36849

PERMIT NO. 9

277 Technology ParkwayAuburn, AL 36830

NCAT's Asphalt Technology CourseFebruary 2-6, 2004