Transcription

Report No. CDOT·DTD·94·7Influence of Refining Processes and CrudeOil Sources Used in Colorado On Resultsfrom the Hamburg Wheel-Tracking DeviceTim AschenbrenerColorado Department of Transportation4201 East Arkansas AvenueDenver, Colorado 80222Final ReportApril 1994Prepared in cooperation with theU.S. Department of TransportationFederal Highway Administration

The contentstheauthorthe accuracycontents doofwhothisisof thereport reflect the views ofresponsible for the' facts anddata presentednot necessarilyherein.reflect theviews of the Colorado Department ofofficialTransportationor the Federal Highway Administration.does not constituteThis reporta standard, specification,regulation.iTheor

AcknowledgementsThe refineries that provide asphalt cement to Colorado were very cooperative in their assistancewith this research study. Their cooperation was greatly appreciated. The laboratory mixture testswere performed by Kim Gilbert and Nava Far (CDOT). Benja Bemelen (CDOT) performed theasphalt cement tests.The CDOT research panel provided many excellent comments and suggestions for the study.It included: Byron Lord and Kevin Stuart (FHWA - Turner Fairbank Highway Research Center).Doyt Bolling (FHWA - Region 8). Jerry Cloud (FHWA - Colorado Division). Steve Horton and BobLaForce (CDOT - Staff Materials). Ken Wood (CDOT - Region 4 Materials). and Donna Harmelink(CDOT- Research).Special thanks to the panel of Colorado asphalt paving experts who provided numerous ideas andsuggestions which made this study more informational: Bud Brakey (Brakey ConsultingEngineers). Jim Fife (Western Colorado Testing) . Joe Proctor (Morton International). Scott Shuler(Colorado Asphalt Pavement Association) and Eric West (Western Mobile).11

Technical Report Documentation Page1. Repert No.2. Govcnlmellt Accession No.3. Recipient's Catnlog No.CDOT-DTD-R-94-75. Report nate4. ntIt Dnd SubtitleApril 1994Influ ence of Refining Processes and Crude Oil SourcesUsed in ColoradoOn Results from the Hambur!! Wheel-Trackin!! Device6. Perfonnlng OrganIzation Code7. Author(s)8. Performing OrganizAtion Rpt.No.Tim AschenbrenerCDOT-DTD-R-94-710. Work Unit No. (TRAlS)9. Performing Organization Name and AddressColorado Department of Transportation4201 E. Arkansas Ave.Denver, CO. 8022211. Contract or Grant No.13. Type of Rpt. and Period Covered12. SpmlsOI'ing Agency Nome and AddressColorado Department of Transportation4201 E. Arkansas Ave.Denver, CO. 80222Final Reoort14. Sponsoring Agency Code15. Supplementary NotesPrepared in Cooperation with the U.S. Department of Transportation FederalHi ohway Administrationt 6. Ahslrnctlhe H.,mburg wheel.tracking device is known to be very sensitive to aggregate qual ity. It was considered important[(1 invc tigatehow the asphalt-aggregate interaction can influence the results.The purpose of this report is to identify the influence of the aspha1t-aggregate interaction by using results fromtil.:. Hamburg wheel-tracking device. Ten grading AC-lO asphalt cements were used from four refineries who suppJy asphaltin Colorado that used various refiningproce. esand crude oil sources. TIle refining process and crude oil sources doinOuence the test results from the Hamburg wheel-tracking device. Each refinery was able Lo produce an asphalt cementthat pa.,sed in the Hamburg whed-tracking device with the better aggregates.17. Kel Words18. Distribution StatementCrude Oil So urceHamburg Wheel-TrackingDeviceNo Restrictions: This report isavailable to the public throughthe National Technical Info.Service. Springfield, VA 2216120.Securlly ClassiL (page)Z1. No. of PngesRefining Processt9.s c\lrjt ·Clnsslf.(I· port)UnclassifiedUnclassified.III6122. Pl;ce

Table of Contents. . . . . . . . . . . . . . . . . .12.0 Experimental Grid . . . . . . . . . . . . . . . . . . . . . .2.1 Materials. . . . . . . . . . . . . . . . . . .2.1.1 Asphalt Cements . . . . . . . . . . . . . . . . . . . . . . .2.1.2 Aggregates . . . . . . . . . . . . . . . . . .2.1.3 Hot Mix Asphalt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.2 Tests . . . . . . . . . . . . . .2.2.1 Hamburg Wheel-Tracking Device . . . . . . . . . . .2.2.2 Asphalt Cement Tests . . . . . . . . . . . . . . .2.3 Experimental Grid . . . . . . . . . . . . . . . . . . . . . .2222333663.0 Results and Discussion . . . . . . . . . . . . . . . . . . . .3.1 Asphalt Cement Test Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.2 Hamburg Wheel-Tracking Device Test Results . . . . . . . . . . . . .3.2.1 Refinery A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.2.2 Refinery B . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.2.3 Refinery C . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.2.4 Refinery D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.2.5 Venezuelan . . . . . . . . . . . . . . . . . "3.2.6 Summary . . . . . . . . . . . . . . . . . . .3.3 Correlation of Asphalt Cement and Hamburg Wheel-Tracking Results . . . . .3.3.1 Refinery A . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.3.2 Refinery B . . . . . . . . . . . . . . . . . . . . . . .3.3.3 Refinery C . . . . . . . . . . . . . . . . . . . .3.3.4 Refinery D . . . . . . . . . . . . . . . . . . .3.3.5 Venezuelan . . . . . . . . . . . . . . . . . . .3.3.6 Summary . . . . . . . . . . . . . . . . . . . . .3.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .88910101010111111111111121212161.0 Introduction4.0 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185.0 Implementation . . . . . . . . . . . . 196.0 References . . . . . . . . . . . . . . . . . . . . 19IV

List of TablesTableTableTableTable1. Experimental Grid for the Hamburg Wheel-Tracking Device Tests . . . . . . "72. Experimental Grid for the Asphalt Cement Tests. . . . . . . . . . . . . . . . . . . . . . . . ·73. Results from the Asphalt Cement Testing . . . . . . . . . . . 84. Results from the Hamburg Wheel-Tracking Device - Deformation (mm) After20,000 Passes. . . . . . . . . . . . . . . . . . . . 9Table 5. Comparison of Results By Mixture. . . . . . . . . . . . . . . . . 16Table 6. Comparison of Results By Refinery . . . , . . . . . . . . . . . . . . . , . . 17List of FiguresFig. 1. The Hamburg Wheel-Tracking Device. . . . . . . . . . . . . . . . . . 4Fig. 2. Close-Up of the Hamburg Wheel-Tracking Device. . . ,. . . . . . . . 4Fig. 3. Summary of Results from the Hamburg Wheel-Tracking Device. . . . . . . . 5Fig. 4. Results from the Hamburg Wheel-Tracking Device as a Function of AsphaltCement Stiffness as Measured by Absolute Viscosity. . . . . . . 13Fig. 5. Results from the Hamburg Wheel-Tracking Device as a Function of AsphaltCement Stiffness as Measured by Penetration. . . . . . . . . 14Fig. 6. Results from the Hamburg Wheel-Tracking Device as a Function of AsphaltCement Stiffness as Measured by the Dynamic Shear Rheometer. . . . . . . . 15AppendicesAppendix A: Summary of Asphalt Cement Test ResultsAppendix B: Summary of Hamburg Wheel-Tracking Resultsv

Influence of Refining Processes and Crude 011 SourcesUsed in ColoradoOn Results from the Hamburg Wheel-Tracking DeviceTim Aschenbrener1.0 IntroductionIn September 1990, a group of individuals representing AASHTO, FHWA, NAPA, SHRP, AI, andTRB participated in a 2-week tour of six European countries. Information on this tour has beenpublished in a "Report on the 1990 European Asphalt Study Tour" (1). Several areas for potentialimprovement of hot mix asphalt (HMA) pavements were identified, including the use ofperformance-related testing equipment used in several European countries.The ColoradoDepartment of Transportation (COOT) and the FHWA Tumer-Fairbank Highway Research Center(TFHRC) were selected to demonstrate this equipment.The first priority of the demonstration was to verify the predictive capabilities of this equipmentby performing tests on mixtures of known field performance. The Hamburg wheel-tracking devicewas compared to several mixtures of known stripping performance (2), both good and bad.Results from the test correlated very well with field performance.The next step was to attempt to improve the stripping resistance of the HMA placed in Coloradoby using the Hamburg wheel-tracking device. The test is known to be very sensitive to aggregatequality (2), and methods to determine potential areas of improvement with the aggregate wereinvestigated (3). Besides aggregate quality, asphalt-aggregate interaction is also important. Onone project the results were very sensitive to the asphalt-aggregate interaction (4). Since theasphalt-aggregate interaction could influence the results, it was considered important toinvestigate the potential frequency and magnitude of this occurrence in a systematic manner.The purpose of this report is to identify the influence of the various refining processes and crudeoil sources used in Colorado on the test results from the Hamburg wheel-tracking device. Theresults of this study will provide importa:nt guidance to asphalt paving contractors that bid onfuture projects that use the Hamburg wheel-tracking device as a specification.1

2.0 Experimental Grid2.1 Materials2. 1.1 Asphalt CementsThe fou r refineries that provide most of the asphalt cement used for CDOT projects provided theasphalt cement in this study. The asphalt cement from each refinery was labeled A, B, C, or D.The refineries produced different "types" of asphalt cement by changing crude oil sources orrefining processes. Asphalt cements were all grading AC-1 0 (AASHTO M 226, Table 2). For thisstudy, the refineries produced 1 to 3 different types of AC-1 o.Additionally, an asphalt cement refined from a Venezuelan crude oil was used. This was usedprimarily for information arid labeled ·Vn". There was only one "type".2.1 .2 AggregatesAggregates used for lIlis study came froni a variety of sources willl a variety of performancehisl-Jries. The aggregate combinations were selected to provide a variety of results, good to poor,in the Hamburg wheel-tracking device. All mixtures used quarried aggregate. Two different typesof natural sands were added to help vary the performance of each mixture. All aggregates weretreated with 1% hydrated lime prior to mixing with the asphalt cement.The aggregates for Mix 1 were entirely from a quarried source that has had a history of goodperformance. The aggregates for Mix 2 were primarily from a different quarry with a good historyof performance. However, a poor quality natural sand was added. Although the natural sand isnor.-plastic, it does have clay present.The aggregate for Mix 3 was from a quarry with a mixed history of good and marginalperformance. A problematiC natural sand lIlat has been associated with many HMA pavementstha: have stripped was added. The natural sand does not adhere to asphalt cement very well andcontains clay. The aggregate for Mix 4 was from a quarry with a history of poor performance.The poor quality natural sand with clay used in Mix 2 was also added to Mix 4.2

2. 1.3 Hot Mix AsphaltThe optimum asphalt content for each of the mixtures was determined with the Texas gyratoryin general accordance with ASTM 0 4013.The pre-gyration stress, end point stress andconsolidation stress used were 210. 690, and 17,240 kPa (30, 100, 2500 psi), respectively.These stresses simulate the loads applied to the HMA pavements by high levels of traffic inColorado. All mixes contained 1% hydrated lime.2.2 Tests2.2. 1 Hamburg Wheel- Tracking DeviceThe Hamburg wheel-tracking device is manufactured by Helmut-Wind Inc. of Hamburg, Germanyas shown in Figs. 1 and 2. A pair of samples are tested simultaneously. A sample is typically260 mm (10.2 in.) wide, 320 mm (12.6 in.) long, and 40 mm (1 .6 in.) deep. A sample 's mass isapproximately 7.5 kg (16.5 Ibs.), and it is compacted to 6% 1% air voids. For this study,samples were compacted with the linear kneading compactor. The samples are submergedunder water at 50 C (122 F), although the temperature can vary from 25 C to 70 C (7rF to158'F). For this study, all of the samples were an AC-l0 grading so the test was performed at45 C (5). A steel wheel, 47 mm (1 .85 in.) wide, loads the samples with 705 N (158 Ibs.) Thewheel makes 50 passes per minute over each sample. The maximum velocity of the wheel is34 cmlsec (1 .1 ftlsec) in the center of the sample. Each sample is loaded for 20,000 passes oruntil 20 mm of deformation occurs. Approximately 6-1/2 hours are required for a test.The results from the Hamburg wheel-tracking device include the creep slope; stripping slope andstripping inflection point as shown in Fig. 3. These results have been defined by Hines (6). Thecreep slope relates to rutting from plastic flow. It is the inverse of the rate of deformation in thelinear region of the deformation curve, after post compaction effects have ended and before theonset of stripping. The stripping slope is the inverse of the rate of deformation in the linear regionof the deformation curve, after stripping begins and until the end of the test. It is the number ofpasses required to create a 1 mm impression from stripping. The stripping slope is related to theseverity of moisture damage. The stripping inflection point is the number of passes at theintersection of the creep slope and the stripping slope. It is related to the resistance of the HMAto moisture damage.3

Fig. 1. The Hamburg Wheel-Tracking Device.Fig. 2. Close-Up of the Hamburg Wheel-Tracking Device.4

0 IPost Compaction I2'-SI .I1\EE . . . .J . . . . L.:.·6'·4rCreep Slope I-. I.-·1 6!. . . . - . . -----. --·1 8.-. .----. .-----.- "' '0.--11iIoI-- --.,-- -.:.·-·--f·- .-.-.-. -L.·20I· . . .:- P\ --.-f-.--.I!.IStripping Inflection POint 1'--.-.----L -- '" .---.l .J lJ Pin.g- ope 1 --:r-)- . . . . . .1 - --.-·h - I24681012 .1416No. of Pa.sses(Thousands)Fig.3. Summary of Results from the Hamburg Wheel-Tracking Device.5s.,II1.-'t.-.18I20

The City of Hamburg requires less than 4 mm rut depth after 20,000 passes. The COOT hasindicated this specification is very severe (2). A more reasonable specification for the traffic andenvironment in Colorado is a deformation less than either 4 mm after 10,000 passes, or 10 mmafter 20,000 passes. A project in Colorado will be using the results from the Hamburg wheeltracking device as a specification for incentive payment only. Since that specification requiresless than 10 mm after 20,000 passes, the deformation after 20,000 passes is used in this report.2.2.2 Asphalt Cement TestsA series of tests were performed to characterize the asphalt cements used in this study. If resultsfrom the Hamburg wheel-tracking device indicated differences between the asphalt cements, theasphalt cement tests might potentially explain differences.The traditional asphalt cement tests specified by COOT were performed: penetration at 25 C(AASHTO T 49) and absolute viscosity at 60 C (AASHTO T 202).Additionally, the asphaltce ment tests developed by SHRP were also performed. A full series of tests were performed todetermine the high temperature SHRP Performance Grade of each asphalt cement. The testingdevice was the Dynamic Shear Rheometer (OSR) and tests were performed on asphalt cementsthat were unaged (tank) and Thin Film Oven Test (TFOT) aged (AASHTO T 179).The DSR is used to measure the ability of the asphalt cement to resist permanent deformationat high temperatures.Since the Hamburg wheel-tracking device is performed at hightemperatures, results from the OSR may be appropriate to understand differences in test results.2.3 Experimental GridEach of the types of asphalt cement from each refinery were mixed with the four differentaggregates and tested in the Hamburg wheel-tracking device. The experimental grid is shownin Table 1. All samples were tested at 45"C in the Hamburg wheel-tracking device since eachasphalt cement was an AC-10 grading (5).6

Table 1. Experimental Grid for the Hamburg Wheel-Tracking Device Tests.Type. 123Mix 1XXXXXXXXXXMix 2XXXXXXXXXXMix 3XXXXXXXXXXMix 4XXXXXXXXXXRefmei)'E[@E:tjEBAX - Replicate samples were mixed with AC-10 grading asphalt cement and tested at 45 C.For characterization purposes, all of the asphalt cements were tested with the current asphaltcement tests specified by the CDOT and the high temperature asphalt cement tests developedby SHRP. If any differences between the asphalt cements were identified by the results from theHamburg wheel-tracking device, the tests on the asphalt cement might potentially explain thedifferences. The experimental grid of the asphalt cement tests performed is shown in Table 2 .Table 2. Experimental Grid for the Asphalt Cement Tests.E[@E[@Et3EBRefin Vn.Type1Penetrationand XXXXXXX1231231217

3.0 Results and Discussion3.1 Asphalt Cement Test ResultsEach of the asphalt cements used in this study was tested with the standard asphalt cement testsspecified by the CDOT and the new SHRP binder tests, The results are summarized in Table3 and all of the tests are reported in Appendix A. The absolute viscosity of all of the asphaltcements meets the requirement for an AC-10 grading (AASHTO M 226, Table 2). Additionally,the high temperature properties of the asphalt cements meet the SHRP high temperatureperformance grade of 58. In general, a penetration range of 85 to 100 dmm is expected to beapproximately an AC-10. Although most of the penetration results fell within this range, therewere two that were quite a bit softer: particularly Refinery B Type 2 and Vn.Table 3. Results from the Asphalt Cement sDSR(tank)·[email protected] kPaDSR(TFOT)·[email protected] 3106062.562.7581100101061.862.658290100062.4. 61.258D187110062.360.758Vn.1129104061.762.158 .ABC. 8208

3.2 Hamburg Wheel-Tracking Device Test ResultsThe test results from each mix tested in the Hamburg wheel-tracking device for each refinery andtyPE: of asphalt produced from each refinery are shown in Table 4. The results are plotted inAppendix B. The results shown are the mm of deformation at 20,000 passes. Each test had areplicate sample, so the left (L), right (R), and average (Avg) are reported in Table 4. In someinstances, the samples in the Hamburg wheel-tracking device had greater thandefcrmation before20 mm of20,000 passes. These samples failed dramatically and are reported as " 20"in Table 4. The test results were considered acceptable when the average of the two sampleshad a deformation of less than10 mm after 20,000 passes. The cells in Table 4 were shadedto indicate when the average of the two samples were greater thanTable 4.10 mm after 20,000 passes.Results from the Hamburg Wheel-Tracking Device - Deformation (mm) After20',0'0'0' Passes.c::J12Refinery ARefinery BRef. CDVn123123121LlOA5.17.99.412.312.29.96.9 2012.1R8.111 .87.85.017.216.36.81.9 8Hi28.4 0018.56.0H'l. l5.111.3R9.08.914.310.0 '4.0t 2.77.3'ttl.5'L14.4 2011.69.4 2"017.59.54.9 20 20R2"1.31Mt2li20.8,,2015.7l1H5.6 20)0'20Avg17;9' 21ftaf t 5.fl o'2Q't 61l' 14."5.2. 00' 20'L3.214.6HtO9.9 2022.S 207.0 00 20R5.211.l17.321.4,,20lU 204.8 20 20"Avg4.215.9't7.ff1;5.6 20"17.4' 20"5.9 .2(f all""'Fc.iling test result (shaded)L Left sample9R - Right sampleAvg - Average

3.2. 1 Refinery AThe Types 1 and 2 asphalt cements performed as expected by passing with the better mixes(Mixes 1 and 2). Th e Type 2 asphalt cement also performed as expected by failing with thepoorer mixes (Mixes 3 and 4). It was pleasantly surprising that Type 1 passed with one of thepoorer mixes (Mix 4). Only one of the mixes (Mix 1) passed with the Type 3 asphalt cement.Refinery A's Type 1 asphalt cement would be very desirable to use on projects with theseaggregates.3.2.2 Refinery BThe Type 1 asphalt cement performed as expected. The better mixes (Mixes 1 and 2) passed,and the poorer mixes (Mixes 3 and 4) failed. No mixes passed with the Type 2 asphalt cement,and three failed dramatically. It would be desirable to avoid Refinery B's Type 2 asphalt cement.Although Types 1 and 2 were supplied by the refinery, the Type 3 asphalt cement was sampledfrom an asphalt paving contractor's plant. Based on the asphalt cement tests, Types 1 and 3appeared identical. Therefore, it might be expected that the results from the Hamburg wheeltracking device would also be similar. For the poorer mixes (Mixes 3 and 4), the results werevery similar as expected. However, for the better mixes (Mixes 1 and 2) the Type 3 asphaltcement had twice the deformation as the Type 1. This difference is perplexing.3.2.3 Refinery CThe Type 1 asphalt cement performed as expected. Th e better mixes (Mixes 1 and 2) hadpassing results and the poorer mixes (Mixes 3 and 4) had failing results. Only Mix 4 faileddramatically. The Type 2 asphalt cement is the only asphalt cement that had passing results withboth of the poorer mixes (Mixes 3 and 4). Mix 2 barely failed with the Type 2 asphalt cement.Refinery C's Type 2 asphalt cement would be very desirable to use on projects with theseaggregates.3.2.4 Refinery DOnly one type of asphalt cement was provided by Refinery D and it performed miserably. Threeof the mixes (Mixes 1, 3, and 4) failed, and they all failed dramatically. This asphalt cement didnot perform very well with the aggregates used in this study.10

3.2.5 VenezuelanThe crude oil source of the Venezuelan asphalt tested was not known. All four of the mixes failedthe requirement. The two better mixes (Mixes 1 and 2) barely failed, and the two poorer mixes(Mixes 3 and 4) failed dramatically.3.2.6 SummaryAll of the refineries were able to produce an asphalt cement that performed well in the Hamburgwheel-tracking device when using the better aggregates, except Refinery D. It appears there maybe a unique asphalt-aggregate interaction that is dependent on each asphalt cement and eachaggregate used in the mixture.3.3 Correlation of Asphalt Cement and Hamburg Wheel-Tracking Results3.3. 1 Refinery AThe Type 1 asphalt cement was the stiffest of Refinery A's grading AC-1 0, and it also performedthe best. The Type 1 not only passed with the better aggregates, but it also did exceptionally wellwith one of the poorer aggregates (Mix 4). Although the Type 2 asphalt cement was softer thanType 3, the Type 2 performed better. Type 2 passed with all of the better aggregate, while Type3 did not.3.3.2 Refinery BThe Type 2 asphalt cement was an exceptionally soft grading of AC-10, and it performed verypoorly. Types 1 and 3 had virtually identical stiffnesses to each other and were closer to themiddle of the expected ranges for AC-1 0 than Type 2. Despite Types 1 and 3'5 similarity, theyperformed dramatically differently. Type 1 passed with the better aggregates; while Type 3 failedwith all aggregates.3.3.3 Refinery CThe Types 1 and 2 asphalt cements both performed remarkably well in the Hamburg wheeltracking device. The stiffnesses were fairly similar to each other and in the middle of the rangenormally expected for a typical grading AC-10 asphalt cement.11

3.3.4 Refinery 0With the aggregates used in this study, this was one of the poorest performing asphalt cements.Although the viscosity was in the middle of the expected range of stiffness; the penetrationindicated a very stiff AC-10; and the DSR indicated one of the softest grading AC-10 asphaltcements. Although the asphalt cement tests provided conflicting information about the stiffness,the results in the Hamburg wheel-tracking device clearly indicated this was one of the poorerasphalt cements with the aggregates used in this study.3.3.5 Venezuelan ·The viscosity and DSR of this asphalt cement was in the range of expected values.Thepenetration indicated this was a very soft grading AC-10. Although the asphalt cement testsprovided conflicting information about the stiffness, the results in the Hamburg wheel-trackingdevice clearly indicated this was one of the poorer asphalt cements with the aggregates used inthis study.3.3.6 SummaryThe deformation recorded in the Hamburg wheel-tracking device is plotted versus the asphaltcement stiffness as measured by the absolute viscosity (Fig. 4), penetration (Fig. 5), and dynamicshear rheometer (Fig. 6). In each of the plots, it is not possible to identify a direct correlationbetween the asphalt cement stiffness and results in the Hamburg wheel-tracking device.However, each of the points are labeled with their aggregate mixture. In general, the distinctionbetween the mixtures with the better aggregates (Mixes 1 and 2) can be made from the mixtureswith the poorer aggregates (Mixes 3 and 4), regardless of asphalt cement stiffness.Some of the Types of asphalt cement tested indicated the softer grading AC-1 0 asphalt cementsdid poorer than the stiffer grading AC-10 asphalt cements.However, there were numeroussamples with very similar stiffnesses that performed dramatically differently in the Hamburg wheeltracking device. The variable that influence the results from the Hamburg wheel-tracking deviceare a function of: 1) aggregate quality, 2) refining process and crude oil source, and 3) asphaltcement stiffness.12

,20.0.------ - r-----l.,.-- , }---, - J!II--- ,-- , --, - ,,j····· ··· ·· ·····;····· ····· ·r···········r····· ····18.0!o c1;*1it16.0 ··· ····· · ·r···· · ··· ···r····· ····n l············ · ·r ···· ······: ·· ······ ···I··· · · · · ··· · r ·· ········I··· ······ · · · · · · ······ · 14.0 ···r·········:·········· ·I ······ ··· ·· · · · ·· · ·t ·······"1"8··,r --, Mix 1' 12.0-I --tt --r--rr-·Mix2: !.:Mix3DMix4···· ···· · · l ··· ········r···· ··········I ······· ···· ·· · · ···· r*······· · ······· ····l· · · · · · · ·"!" · · · ···· . . . . . '---------'l1 O.0 li1i··· - ····t-· ····· ··:-·· ·· ·· · :· · · · -t- ···· ··1···········t-······· ··· · · ·· ·· ···:······ ····j . . .! !! iI.! .o- -i - - ·I I --r:-:X--t-l --!-- r --rt-t-r--lt- i i-r!:i::604. 1 1 11 11 1 1 1 Viscosity, poisesFlg.4. Results from the Hamburg Wheel-Tracking Device as a Function of Asphalt CementStiffness as Measured by Absolute Viscosity.13

.----------s----------- -- -- ----tlh20.0 ----,,l.-:::-- -:l- -r::r::r T :-r l::l--:14. O* ' 12.0 - . ·.··.·. 1 0.0Ex-···· ··-····· ·:···· ······· -···· ·r· · · · ··· · ··j · · ··· ·· . · · t···· · · ·· ···· ·· · · · ····· · ··· ·· · ··· ·· ·i . . .j. . .· .t··. ···.,.j.··· .· .··.t . ·.· .· · ····· · ·· ··· I · · · · · · ·t·· · · · · · · · I · · · · -··-T-··-··· ·;8.0.;- · ······ · · ···! ·,· ··· · ···J····· · · ' T·········· -·t.·········- -··· · ··· t·-·· ··-·-· · I·-·· · ·····- -/- ···-····- .x ii;;4.01 ----- 4*-- D--- I; ----- ---- ------ -I---- ---- ----4-I---- 859095100105110115120125130PenetrationFlg.5. Results from the Hamburg Wheel-Tracking Device as a Function of Asphalt CementStiffness as Measured by Penetration.14xMix 1 Mix2*Mix3DMix4 -

20.0.- ---. ----,----- ,--.Br-- ---- ---- ---. r----,xMix 1i'] Mix216.0EE14.0i*x I,,,1*; ,,,., . .,. . . '' . ,.,. . . ' .1- ., . .-.- . . . :·,···· · ·· ·t · ··· ·· · ·i ·,· ·· ·,,· · ·,··t ·· · ,· · ·'t······ ,·· ·,1: !!co:;1\'1E. C1 2-0 .i :1.!." . ".j. .". . . :."",.". . .".".- ., . · ·"··"I· ···"··· ·"*·"·· -"l·"·· ··"--· --"·" -·t·-······-.1 0.0-,, · ·jir , ···· ······"· · ·1" ··"· ·,,·· ···,,··,, ·- ·,-··, ·i ·-···-··i,, -··· ·,,· ·i,,-·,,-·"i""······, ··i , ·· ····,,·· !!jX .O-- - --I- -- -- I --! -r·-i- i-t-.,.j"·-···-"·· -"·-,,·· · i, ··,,···"····i"·· ······t· · " · " · i"·· · , · ·, ··· · , · · ····i, · , · · ··'1"···"····1"···"· ' ·', xi''04. 01-----;'---- ------r-----r-I----6rI2----- I----6'r3-----r-I----6 14 -- 6.0-59.56060.56161.562.563.564.5Dynamic Shear Rheometer,CFIg. 6. Results from the Hamburg Wheel-Tracking Device as a Function of Asphalt CementStiffness as Measured by the Dynamic Shear Rheometer.15*Mix3oMix4

3.4 SummaryResults shown in Table 5 indicate that aggregate differences can make a large difference in theresuits. The better mixes, Mix 1 and Mix 2, had passing test results with many of the asphaltcements used.For the poorer aggregates, virtually none of the samples tested passed,regardless of the asphalt cement. However, there were a few asphalt cements that worked wellwith the poor aggregates.When a sample has greater than 20 mm of deformation before the entire 20,000 passes, thefailure is dramatic. This occurred only once each for the better mixes (Mix 1 and 2). For thepoorer aggregate, the dramatic failure occurred for almost half of the asphalt cements tested.TableS. Comparison of Results By Mixture.MixPassFail 10 mm 10 mm 20mm1641.255131944284Results from the Hamburg wheel-tracking device are significantly influenced by the aggregate.When results from the device are una

in Colorado that used various refining proce. es and crude oil sources. TIle refining process and crude oil sources do inOuence the test results from the Hamburg wheel-tracking device. Each refinery was able Lo produce an asphalt cement that pa.,sed in the Hamburg whed-trac