Attachment J.9 - Historical SATF UTQGS Mix Design.pdf

DESIGN AND CONSTRUCTION

OF

TRACTION TESTING SURFACE

FOR

UNIFORM TIRE QUALITY GRADING SYSTEM

IN

SAN ANGELO, TEXAS

Prepared for

U.S. Department of Transportation

National Highway Traffic Safety Administration

Office of Vehicle Safety Compliance

San Angelo, Texas Test Facility

San Angelo, Texas on

Goodfellow AFB

Prepared by , P.E.

May 31, 2014

“test” means the evaluation of the tire being evaluated or the test tire. Note that the test protocol requires ten individual “good” friction values for both the “control” and “test” tire. Typically 12 to 15 “lock-up” may be required to obtain “good” friction values. Thus, to establish the friction value for a “control” tire, ten friction measurements or “lock-ups” are required and the determination of a friction value for a “test” tire requires ten friction measurements or “lock-ups”. Typical test sequences used on the facility vary with the testing requirements and are illustrated below.

Evaluate one test tire: “control” + “test” + “control” Evaluate two test tires: “control + “test” + “test” + “control” Evaluate eight test tires: “control” + “test” + “test” + “control” + “test” + “test” + “control” + “test” + “test” + “control” + “test” + “test” + “control”

The evaluation of two test tires typically requires about 3 hours of testing and about 60 “lock-ups”. The evaluation of 8 test tires typically requires a long day of testing and about 150 or more “lock-ups”. Some companies would like to evaluate 16 tires over a two day period. This requires between 260 and 300 “lock-ups” typically. The test sequences described above are performed in one “narrow” area of the pad. The “lock-ups” therefore will change the friction value during the test sequence unless the pad is not susceptible to polishing.

Weight on Tire The weight or mass on the friction measuring trailers varies from 600 pounds per tire to 2000 pounds per tire. The weight on the tires can have an effect on the rate of coefficient of friction loss with “lock-ups” or the number of tires being evaluated.

Water Absorption Water is applied by the friction measuring trailer immediately prior to the “lock-up” associated with the friction measurement. The typical sequence of operation is for the trailer to apply water to the surface on the first lap around the oval track followed by measurements of friction on the second and subsequent laps. If the test pad surface absorbs the water at a relatively fast rate, the amount of water on the pad surface can vary with testing. This can present operational problems. Ideally the rate of water absorption on the test pad would be minimal between passes.

Other Facilities/Pads A number of tire test facilities are located in the state of Texas. Many of these facilities have constructed pavement test surfaces to evaluate tire friction, vehicle control and tire durability. A list of the facilities in Texas is shown on Table 1. Information on the year some of the test surfaces were constructed is shown together with the type of facilities located at the site. The Goodrich facility located at Pecos is no longer in operation.

Table 2 contains a listing of some of the friction pads located at these facilities as well at the U.S.

DOT facility at Goodfellow AFB and at Texas A&M University Riverside Campus (1-3). The names of the commercial facilities have been coded as some of this information may be considered proprietary. The information gained from this summary was used in part to select the coarse aggregate type for the U.S. DOT facility at Goodfellow AFB. Additional information in the literature was also reviewed to assist in the development of the friction pad at Goodfellow

AFB (4).

The third test pad at Goodfellow AFB was placed on August 5, 2011. This pad utilized the Streetman light weight aggregate as its coarse aggregate. During the first two years of service, the test pad experienced some performance problems including the following:

1. Friction values in the low 0.40’s

2. Loss of surface aggregate or raveling

3. Soft surface likely due to the use of the relatively soft asphalt cement in the mixture

4. Relatively high water absorption between water application associated with friction testing

The U.S. DOT decided to replace the test pad surface that was placed on August 5, 2011 in the summer of 2012. Three test strips were placed prior to construction of the test pad. Test Strip 7 was placed with Hoban course aggregate. Test Strips 8 and 9 were placed with Streetman lightweight coarse aggregate. The Test Pad was placed with Streetman lightweight coarse aggregate. Dates of construction for the three Test Strips and the Test Pad are provided below

Test Strip 7 November 16, 2012 Test Strip 8 June 26, 2013 Test Strip 9 September 5, 2013 Test Pad September 6, 2013

The mixture design and construction operations associated with the design and construction of the test pad are described below. The Test Pad layout for friction testing is shown on Figure 2.

MATERIALS AND MIXTURE DESIGN CONSIDERATIONS

The design and construction of a traction surface for this facility is largely dependent upon the selection of the asphalt binder and aggregates as well as the mixture design of these materials.

General design criteria as well as available historic asphalt binder, aggregate and mixture designs are reviewed below together with the mixture design information developed for Test Strips 7, 8 and 9 as well as the Test Pad.

General Design Requirements Desirable properties for the asphalt mixture to be used as the surface course on this facility are shown below by general property and materials selection and mixture design considerations. As noted below, a compromise of materials selection and mixture design considerations are necessary to produce the desired mixtures.

Resistance to Permanent Deformation Resistance to permanent deformation or rutting of the pad is not a major design consideration since the magnitude of the weight per tire or load on the facility and the number of repetitions of load at a specific location will be relatively low. To provide resistance to permanent deformation the asphalt binder selected for this facility was a PG 70-22 which will provide adequate resistance to permanent deformation as well as adequate low temperature properties. The 70 designation to this binder implies that an asphalt mixture containing this binder should be capable of providing resistance to permanent deformation for pavements with near surface (20 mm depth) temperatures as high as 70 C.

To insure adequate resistance to permanent deformation in the mixture, the aggregates were selected with mechanically crushed faces and were gap graded. Hveem stability tests and the Hamburg rutting tests were also utilized on these mixtures to insure resistance to permanent deformation. Typically low asphalt binder contents are utilized to control permanent deformation. On this facility durability is more of a potential issue and higher than normal asphalt binder contents were considered.

Resistance to Raveling Loss of fine and coarse aggregate from the surface with use and aging will be controlled by specifying an asphalt with good high and low temperature performance characteristics that is polymer modified (PG 70-22) as well as using hydrated lime to help control the water sensitivity of the mixture. Additionally, the asphalt binder was selected from a refinery which has improved aging properties as compared to an asphalt binder produced at a relatively close proximity refinery. The asphalt binder content was also higher than normal to help control the raveling.

Resistance to Bleeding Bleeding is defined as the accumulation of asphalt binder on the surface of the pavement. Bleeding will likely not be problem on this surface due to the relatively low number of relatively low weight loads. The relatively high asphalt binder content used on this facility would typically be a concern with regard to bleeding. Higher binder contents were used to help control raveling, provide better resistance to the presence of water and to provide durability.

Resistance to Fatigue Cracking Fatigue or alligator cracking is associated with a large number of heavy loads on a pavement that does not have adequate load carry capability. The test pad at this location has adequate load carrying capability and fatigue cracking should not be an issue.

High asphalt binder content improves the resistance to fatigue cracking. Low in-place air void contents will also improve resistance to fatigue cracking.

Thermal Cracking (Block Cracking) Transverse cracking describes cracking that is transverse relative to the direction of travel. This is a fairly common type of cracking in west Texas. Block cracking is a type of cracking that is both transverse and parallel relative to the direction of travel. This type of cracking is also present in west Texas and can be associated with parking lots and other large areas of paving. Transverse and longitudinal cracking is not present on the existing facility.

Mitigation of transverse and block cracking is associated with the selection of binder with good low temperature properties (PG 70-22). The -22 indicates that an asphalt mixture produced with this asphalt cement will be polymer modified and will have resistance to these forms of cracking to an air surface pavement temperature of -22 C. In addition, relatively high binder contents were selected to insure that transverse cracking would be mitigated.

Durability Durability is a term used to describe the ability of an asphalt binder and asphalt mixture to withstand long term use. This includes resistance to aging and water susceptibility.

Asphalt mixtures with high binder contents typically provide durable mixtures. Mixtures with low in-place air voids also provide improved durability. The inclusion of hydrated lime improves water susceptibility. Since the friction value is measured in the presence of water, water susceptibility is of importance.

Absorption of Surface It is desirable to produce a surface that will not quickly absorb water placed on the surface by the friction measuring trailer. The absorption of water on the surface is controlled by the aggregate gradation, asphalt binder content and in-place air void content. Dense graded aggregate gradations are desirable together with high asphalt binder contents. Low in-place air voids are also of importance. Figure 3 shows Test Strip 7 after water application and indicates non-uniform water absorption into the surface.

Since it is desirable to have relatively large volumes of coarse aggregate in the mixture, it is difficult to obtain both the desired friction and low water absorption.

Uniform Friction It is desirable to place a test pad with uniform friction that does not vary in the transverse and longitudinal direction, does not change with use and is in the range of 0.40 to

0.60. Coarse aggregate selection, gradation of the aggregate and uniformity of construction are important to achieve this goal. The aggregate gradation was selected on the original test pads and subsequent pads to provide relatively high quantities of coarse aggregate at the surface of the mixture. This gradation is typically defined as a “gap gradation”.

Constructability The designed mixture must be able to be produced, placed and compacted with conventional construction equipment and processes. Since the mixture design is to be somewhat open or gapped graded, draindown of the mixture is of some concern. A draindown laboratory test will be performed to asses this property.

Mixtures constructed with polymer modified asphalt binders, relatively high asphalt binder contents and mixtures containing lightweight aggregates with high water absorption are prone to being “tender” during construction. “Tender” asphalt mixtures can shove under the weight of the rollers, the mat can be “cut” at the edges of the steel wheel roller and the formation of small scale surface cracking can occur. These potential problems will be identified during construction of the Test Strips.

Summary Based on the information presented above on mixture design and the desired criteria for the test pad, compromises for asphalt binder selection, asphalt binder content, aggregate type, aggregate gradation and in-place air voids are necessary. The major considerations for mixture design to insure test pad performance are summarized below.

1. Relatively high asphalt binder contents to insure durability and resistance to raveling, water susceptibility, aging and thermal/block cracking while not creating problems with rutting and bleeding.

2. Asphalt binder with good high and low temperature properties which contains a polymer to provide resistance to thermal/block cracking, provide a surface that can be used without damage on high air temperature days, provide a surface that will not experience rutting and provide a surface that will not bleed.

3. Asphalt binder with relatively low aging properties in-service to reduce the occurrence of raveling and thermal/block cracking.

4. Aggregate with a history of providing friction values within the desired range.

5. Relatively high coarse aggregate contents (greater than 50 percent by volume) in mixture to provide the desired friction values as the friction will largely be determined by the micro texture of the coarse aggregate as well as the macro texture of the test pad asphalt mixture surface.

6. Relatively open mixtures to provide exposure of the coarse aggregate (gap rather than dense graded aggregates) and to provide for good macro pavement surface texture.

7. Hydrated lime to assist in the control of water susceptibility and aging of the asphalt binder as the test pad will be “wetted” continually with testing.

8. Low laboratory compacted air voids to provide high binder contents in the mixture for the desired properties described above.

9. Low in-place air voids to provide for durability, resistance to raveling, fatigue cracking, thermal/block cracking and to help control applied water absorption.

10. Draindown of the asphalt binder as well as shoving may be present during construction.

Historic Information Table 4 contains information on the asphalt binder source, grade and percent as well as the aggregated blended for mixture design purposes for facilities placed at Goodfellow AFB. Note that details were not available for all mixtures placed at the facility. Table 5 contains the gradation of the aggregates used for various facilities at Goodfellow AFB on both a weight and volume gradation basis. Asphalt binder contents are also provided. Gradations for these same mixtures on a passing/retained basis are shown on Table 6 on both a weight and volume basis.

Gradations and asphalt binder contents on a weight basis are shown on Table 7 for these same mixtures. The mixture design for the Ranger lightweight aggregate test pad placed in 1991 is contained in reference 5.

Mixture Designs for Test Strip 7 The mixture design for Test Strip 7 was performed by in their San Angelo laboratory. The laboratory mixture design for the Test Strip is shown on Table 8. The percent of the various aggregates is shown together with the target asphalt binder content, combined aggregate gradations and the laboratory compacted volumetrics. The original mixture design was altered somewhat and laboratory testing performed on Sept 6, 2012.

Hot plant, field mixtures were produced on Sept 12, 2012, October 18, 2012 and November 12, 2012 to refine the mixture design and to determine uniformity of the produced mixtures. Asphalt binder contents, aggregate blend quantities, aggregate gradations and laboratory volumetrics are shown on Table 8. Results from the Hamburg rut tester and tensile strength are shown for one of the field produced trial batches.

Mixture Designs for Test Strips 8 and 9 The laboratory mixture designs used as a starting point for Test Strips 8 and 9 is contained in references 6 and 7. The mixture design for the Test Pad constructed in 2008 (Streetman lightweight aggregate) is contained in reference 6 and reference 7 contains the new mixture design for the Streetman lightweight aggregate.

Properties of the PG 70-22 asphalt binder available at Heartlands Lubbock facility as reported by TxDOT in August of 2012 are provided below.

Original Asphalt Binder G*/sin delta at 70 C 1.32 kPa Elastic Recovery 70 % RTFO Aged Asphalt G*/sin delta at 70 C 2,86 kPa PAV Aged Asphalt Binder G*sin delta at 25 C 1.24 MPa BBR-Stiffness at -12 C 70 MPa BBR-m value at -12 C 0.349

Properties of the Streetman lightweight aggregate as published in TxDOT’s Monthly Quality Control Report (March 2012) are provided below.

L.A. Abrasion 33 % for Grading B and 27 % for Grading C Sulfate Soundness 0.8 % (No. 4 aggregate) Absorption 11.54 % Dry Bulk Sp. Gr. 1.369 Absolute Sp Gr (Rice) 1.489 Freeze-Thaw Loss No. 3-2.7 and No. 4-1.3 Pressure Slaking No. 3-4.5 and No. 4-3.0

Table 9 shows a comparison of the two mixture designs in terms of the aggregate blends, asphalt binder contents and aggregate specific gravity. Individual aggregate gradations on a wash and sieve analysis basis are shown on Table 10. These aggregates were used are representative of those used for the construction of Test Strip 9 and the Test Pad. The same aggregate production streams were used for construction of Test Strip 8.

The laboratory mixture design for Test Strip 8 and contained in Reference 7 is shown on Table

11. This mixture design was used to produce trial batches produced in the hot mix plant on June 22, 2013 and June 24, 2013. Table 11 shows the asphalt binder contents, the percent of the various aggregates, the aggregate gradations and the laboratory compacted volumetrics for these trial batches. Hamburg rut test values are also shown for two of the field trial batches.

Field trial batch test results for Test Strip 9 placed the day before the construction of the test pad are shown on Table 12. Three trial batches were produced on August 30, 2013, September 4, 2013 and September 5, 2013 to insure that the aggregate available and the mixture produced from these aggregates were nearly identical to those used for Test Strip 8. Note that small changes were made in aggregate blends to better meet the desired gradations and volumetric properties. These trial batches were used to verify the hot mixture prior to placement of the test pad. Information for aggregate blends, asphalt binder content and laboratory compacted volumetrics are shown on Table 12.

Mixture Design for Test Pad

The mixture design used for the test pad is that shown on Table 13 and resulted from the trial batches and the construction of Test Strips 8 and 9. Aggregate blends, aggregate gradations and asphalt binder contents are shown on the table under Job Mix Formula (revised).

CONSTRUCTION CONSIDERATIONS

Information is contained below which describes the plant production of the hot mix material, the construction operation, the quality control approach for the construction and the burnishing approach for the Test Strips as well as the Test Pad.

Hot Mix Plant The hot mix asphalt was produced by materials company. A Barber Greene counter flow dryer drum was utilized with an external, twin shaft mixer. Six cold feed bins were available for aggregates. The cold feeds were calibrated for the materials utilized on the test strips and test pad. Bag house fines were returned to the hot mix plant. Dry lime was introduced with a vane feeder control into the return system for the primary collector and bag house fines. The combined materials were introduced into the front of the twin shaft continuous mixers near the point of entry for the asphalt binder. Burner fuel and some waste oil were used as a fuel source for the plant.

The hot mix plant typically produces at 100 to 125 tons per hour. Production in the range of 60 tons per hour was used to produce the mixture for the lightweight aggregate Test Strips 8 and 9 as well as the Test Pad. Recall that Test Strip 8 and 9 as well as the Test Pad were produced with a lightweight coarse aggregate.

A storage silo is used with the continuous production facility. The plant is located at near San Angelo.

Construction Operation Conventional construction operations were used for laydown and compaction. A materials transfer device was used to place Test Strip 9 and the Test Pad. A steel wheel, vibratory roller was used for breakdown, a pneumatic roller was used for intermediate rolling and a steel wheel static roller was used for finish rolling. The breakdown rolling operation varied in number of passes and use of the vibrator mode of operation depending on the mixture characteristics.

A two inch overlay was used for the Test Pad construction as well as the Test Strips. Tack coats were applied prior to the placement of the asphalt mixture overlay.

Quality Control Program Quality guidelines and a quality control testing program were used to provide quality control for the project. All sampling and testing was performed by personnel with the exception of some of the indirect tensile testing. Guidelines supplied to the construction contractor for the quality control operations are shown below. The construction contractor will provide the following personnel, facility and test equipment:

1. Certified technicians

2. Accredited laboratory

3. TxDOT and Superpave laboratory compaction devices

4. Ignition oven

5. Sieves (gradation)

6. Rice specific gravity of mixture

7. Bulk specific gravity of mixture

8. Hveem Stabilometer

9. Hamburg rut tester

10. Tensile strength tester

11. Coring equipment

12. Other equipment for mixture design purposes (ovens, scales, etc.)

Plant production and construction operation were nearly identical for the construction of the Test Strips as well as the Test Pad.

Guidelines for Asphalt Mixture Production The following guidelines were provided to the construction contractor.

1. Calibrate asphalt binder pump/delivery system

2. Calibrate cold feed bins for each aggregate used in the mixture

3. Calibrate the metering system for the bag house return system

4. Calibrate the metering system for the lime introduction

5. Record production rate

6. Record aggregate and mixture temperature

7. Record the time the asphalt paving mixture was produced and placed

8. Record temperature of mixture as placed in the truck at the plant, at the paving site and immediately after the laydown operation

9. Record air temperatures and wind velocity at paving site

Guidelines for Laydown and Compaction The following guidelines were provided to the construction contractor.

1. Use materials transfer device for Test Strips if available

2. Materials transfer device must be used on Test Pad

3. Stage trucks with sufficient material to finish one “pull/pass” without stopping the laydown machine by using the material transfer device

4. Adjust the speed of the paver to finish one “pull/pass” without stopping the paver and waiting for the asphalt mixture to be transferred from the truck to the paver

5. Consider using two breakdown rollers if air temperature are low at the time of paving

6. Start breakdown rolling as soon as possible after placement of the asphalt mixture

7. Use pneumatic roller for intermediate or secondary rolling

8. Provide release agent for rubber tire roller since the area paved will be relatively small and the rubber tired wheels will have little time to warm to prevent adhesion of the mixture

9. Use steel wheel roller for finish rolling to remove roller and paver marks

10. Unsupported and supported longitudinal joints should be rolled with the first pass of the steel wheel roller about 4 to 6 inches from the edge of the freshly laid asphalt mixture.

The second pass of the steel wheel roller should overlap the previously un-compacted strip and as well as the adjacent pavement (for a supported longitudinal joint). A judgment will need to be made in the field if aggregate crushing is a problem with this technique.

11. Use calibrated nuclear gage or other suitable device to help control the sequence of rolling to achieve the desired in-place air voids. Gage calibration should be made with cores if possible.

12. In-place air void content should be in the range of 5 to 7 percent

Guidelines for Burnishing Operations Burnishing refers to the operations conducted after the placement and compaction of the paving mixture to remove the asphalt binder film from the aggregate surface. Burnishing techniques are not standardized and a variety of techniques have been used over the years on tire test facilities in the United States. Presently this is a trial and error type of operation. Examples of burnishing techniques used include the following:

1. Pressure washing

2. Dragging tires without abrasive material

3. Dragging tires with abrasive materials (sand, fly ash, etc.)

4. Skid tires on area with friction measuring equipment

All of these techniques are very time consuming and their effectiveness varies. Friction measurements need to be made in concert with the burnishing approaches. The friction values will change as the number of passes of the various techniques are more or less effective.

Guidelines for Sampling and Testing Guidelines for sampling and testing associated with the trial batches, Test Strip and Test Pad were provided.

Trial Batches The guidelines for the trial batches produced as part of the mixture design are provided below.

1. Sample loose mixture at three random locations (truck samples at plant)

2. Perform the following tests:

a. Asphalt binder content with calibrated ignition oven

b. Aggregate gradation after ignition testing

c. Laboratory compacted sample volumetrics including air voids, VMA, VFA

d. Hveem stability

e. Hamburg rutting

f. Tensile strength

3. Retain 5 gallons of loose mixture for future testing

Test Strips Guidelines for sampling and testing associated with the Test Strips are provided below.

1. Sample loose mixture at five random locations (truck samples at plant)

2. Perform the following tests:

a. Asphalt binder content with calibrated ignition oven

b. Aggregate gradation after ignition testing

c. Laboratory compacted sample volumetrics including air voids, VMA, VFA

d. Hveem stability

e. Hamburg rutting

f. Tensile strength

3. Retain 10 gallons of loose mixture for future testing

4. Obtain cores at 3 to 5 locations to determine in-place air void content

5. Retain core samples for future testing as needed

Test Pad Guidelines for sampling and testing associated with the Test Pad were the same as the Test Strips.

TEST STRIP 7

Test Strip 7 was placed with the Hoban aggregate as the coarse aggregate on November 16, 2012. The Hoban aggregate had been used successfully on the original friction pad placed at Goodfellow AFB in 1972. In addition, other facilities in Texas were constructed with the same aggregate and had good performance histories.

Summary of Mixture Design As discussed above the mixture design for the Hoban mix was developed by on September 6, 2012. Trial mixtures were produced at the hot mix plant on September 12, October 18 and November 12, 2012. Table 8 contains a summary of the information obtained as part of this effort. A mixture was produced with the desired gradation which included a relatively high percentage of coarse Hoban coarse aggregate, a relatively high asphalt binder content and with laboratory air voids in the range of 3 to 4 percent.

Construction and Quality Control The Hoban test strip was placed on November 12, 2012. A Lincoln 660 AXL pick-up machine was used rather than a materials transfer device. A Barger Greene BG 240B paver with a Caterpillar 10-20B Extend-A-Mat screed was used for laydown. A Dynapac CC 522 steel wheel vibratory roller was used in combination with a Dynapac CP 271 pneumatic or rubber tired roller. Finish rolling was provided by the Dynapac CC 522 and a smaller Wacker 5215.

The asphalt mixture was produced at temperatures in the range of 330 to 350 F. The mixture was placed in a windrow by trucks at a temperature in the range of 315 to 330 F. The temperature of the asphalt mixture behind the laydown machine was 300 to 320 F. The matt temperature during breakdown rolling was in the temperature range of 270 to 300 F.

Five loose samples of the mix were obtained from the truck at the hot mix plant. Gradations, asphalt binder contents, laboratory compacted air void contents, Hamburg test and tensile test results are shown on Table 14. Table 15 shows the statistical summary of the quality control tests for gradation asphalt binder content and laboratory compacted air void contents. Table 16 contains the statistical summary for the Hamburg and tensile strength test results for Test Strip 7.

The data contained on Tables 14, 15 and 16 indicate that good control was obtained on aggregate gradation, asphalt binder content and laboratory compacted air voids. The produced mixture was near the target values and had low variability associated with gradation, asphalt binder content and laboratory compacted air voids as measured by standard deviation.

A single core was taken and the in-place air voids content measured at 6.9 percent.

Burnishing and Friction Measurements A summary of the burnishing and friction measurements for this Test Strip are shown on Table

17. Figure 4 shows the Test Strip 7 prior to burnishing. Figure 5 represents Test Strip 2 after burnishing while Figure 6 shows Test Strip 7 after burnishing. A drag apparatus to pull tires over the pad with and without sand was used on this pad. Sand was used as a “grinding” aid as well for some of the burnishing operation. A power wash burnishing operation was also tried on this pad using the equipment in Figures 7 and 8. Some removal of the asphalt film from the aggregate was evident after this process (Figures 9 and 10).

Friction values reported on Table 17 indicated inadequate performance as the friction values were mostly in the high to mid forty range at the start of friction testing and in several cases decreased to below the criteria of 0.40. These values were not expected. Visual examination of the Test Strip 7 surface (Figures 9 and 10) indicated the presence of a relatively large number of uncrushed surfaces. The Hoban aggregate supplier was contacted and indicated that they had changed their crushing and sizing operation in the last few years. These changes likely resulted in a reduced number of crushed faces and hence lower friction values.

As stated previously Hoban has had a good experience record as a suitable aggregate to produce friction values in the low 0.50’s to mid 0.40’s range. The first Test Pad placed at Goodfellow as well as several other surfaces placed at tire testing facilities utilized the Hoban aggregate. Hoban was used as the coarse aggregate for the “standard” friction pad from 1972 to 1991.

Based on these test results and after trying several burnishing techniques, a decision was made to place a Test Strip containing the Streetman lightweight aggregate. This resulted in the construction of Test Strip 8 which is discussed below.

TEST STRIP 8

Test Strip 8 was placed with the Streetman lightweight aggregate as the coarse aggregate on June 26, 2013. The Streetman lightweight aggregate was used on test strips 5 and 6 placed in February and August of 2009. The Test Pad placed on August 5, 2011 also utilized the Streetman lightweight aggregate. These test strips and the test pad had friction values mostly in the mid

0.40. These values were considered acceptable from a friction point of view. The Streetman lightweight Skid Pad placed on August 5, 2011 experienced raveling and some non-uniformity problems as well as being soft in hot weather and not retaining water placed by the friction trailer during testing.

The only lightweight aggregate available near the test site was the Streetman aggregate.

Streetman lightweight aggregate placed on highways in Texas have good friction performance histories.

Summary of Mixture Design As discussed above the mixture design for the Streetman lightweight aggregate mix was developed by (7) on May 21, 2013. Trial mixtures were produced at the hot mix plant on June 22 and 24, 2013. Table 11 contains a summary of the information obtained as part of this effort. The mixture design target gradation and asphalt binder content was changed somewhat for production of the Test Strip. The revised gradation and asphalt binder asphalt content is shown on Table 18. The trial batch mixtures produced at the plant (Table 11) have the desired laboratory compacted air voids and Hamburg rutting test values. Figure 11 shows Hamburg rut test results.

The resulting mixture had a relatively high percentage of coarse Streetman lightweight coarse aggregate, a relatively high asphalt binder content and with laboratory air voids in the range of

2.5 to 3.5 percent.

Construction and Quality Control The Streetman test strip was placed on June 26, 2013. A pick-up machine was used rather than a materials transfer device. A conventional paver and conventional rollers were used for construction. The mixture was tender during the compaction operation. The amount of vibratory rolling during breakdown rolling was reduced and the mat was allowed to cool below about 180 F once the tenderness or shoving under the rolling started. Pneumatic rolling started at a temperature in the range of 160 to 180 F. Breakdown rolling at temperatures starting in the range of 270 F caused shoving, cutting and checking of the mat.

Gradations of the aggregates sampled from the collector belt were obtained and gradations determined and are reported on Table 18.

Three loose samples of the mix were obtained from the truck at the hot mix plant. Gradations, asphalt binder contents and laboratory compacted air void contents are shown on Table 18.

Tables 19 to 21 show a summary of gradation and in-place air void determinations from core samples taken from the constructed Test Strip. Table 22 shows the statistical summary of the quality control tests for gradation asphalt binder content and laboratory compacted air void contents. Table 23 contains the statistical summary for the Hamburg and tensile strength test results for Test Strip 8.

The data contained on Tables 18 to 23 indicate that good control was obtained on aggregate gradation, asphalt binder content, laboratory compacted air voids and field compacted in-place air voids. The produced mixture was near the target values and had low variability associated with gradation, asphalt binder content and laboratory compacted air voids as measured by standard deviation.

Nine cores were taken and the in-place air voids content averaged 7.7 percent. The variability of air void content was higher than typically expected on highway construction projects.

Burnishing and Friction Measurements A summary of the burnishing and friction measurements for this Test Strip are shown on Table

24. Figure 12 shows Test Strip 8 after construction and before burnishing. A burnishing trailer with multiple tires was used for some of the burnishing operation (Figure 13 and 16). Surface texture of Test Strip 8 is shown before and after 20 passes of the burnishing trailer in Figures 14 and 15.

Friction values were initially in the mid 0.50’s range and after approximately 450 friction measurements were reduced to the high 0.40’s range. The initial values and the retained values of friction after a large number of friction measurements were acceptable and construction of the Test Pad with the Streetman aggregate proceeded.

TEST STRIP 9

Test Strip 9 was placed with the Streetman lightweight aggregate immediately prior to the construction of the Test Pad to insure that aggregate stockpiles had not changed significantly and to further optimize the mixture design. Test Strip 9 was placed on September 5, 2013.

Summary of Mixture Design As discussed above the mixture design for the Streetman lightweight mix was developed by

(7) on May 21, 2013 for . Trial mixtures were produced at the hot mix plant on August 30, September 4 and Sept 5, 2013. Table 12 contains a summary of the information obtained as part of this effort. A mixture was produced with the desired gradation which included a relatively high percentage of coarse Streetman lightweight coarse aggregate, a relatively high asphalt binder content and with laboratory air voids in the range of 2.5 to 3.5 percent range. Based on these test results slight changes were made in the mixture design.

Construction and Quality Control This second Streetman lightweight Test Strip was placed on September 5, 2013. A conventional paver and rollers were used for construction. Bottom dump trucks were used to transport the asphalt paving mixture from the plant to the laydown side. An Astec (Roadtec) materials transfer device was used to pick-up the materials from the roadway and deposit the paving materials into the paving machine hopper. The paving machine, a Vogele (Wirtgen Group) Super 2100-2 was used for laydown. Breakdown rolling was performed with a Hamm HD+ 120 oscillating, 24 ton vibratory steel wheel roller. A Dynapac CP 271 10 wheeled pneumatic roller (80 psi tire pressure) was used for intermediate rolling. 15 passes of intermediate rolling were utilized. The finish rolling was performed with a steel wheel Dynapac CC 622 HF, 23 ton roller. One to two passes of finish rolling was utilized.

The mixture was not tender during the compaction operation. This Test Strip was a “mill and fill” operation that constrained the sides of the mat.

Breakdown rolling was initiated when the matt temperature was between 240 and 270 F.

Three loose samples of the mix were obtained from the truck at the hot mix plant. Gradations, asphalt binder contents and laboratory compacted air void contents were determined. Table 25 shows the individual test results. The gradations were in the desired range with the asphalt binder content high. Laboratory compacted air voids were at 2.5 percent, an acceptable level.

Since the Test Pad was to be placed the next day, the in-place air void content was not measured.

Burnishing and Friction Measurements This Test Strip was not burnished and friction measurements were not made as the Test Pad was placed the next day.

TEST PAD

The Test Pad was placed with the Streetman lightweight aggregate as the coarse aggregate on September 6, 2013. As stated previously the Streetman aggregate had a history of providing satisfactory friction values on Test Strips as well as TxDOT highways. Test Strips 8 and 9 allow for the refinement of the original laboratory mixture design.

Summary of Mixture Design As discussed above the mixture design for the Streetman lightweight mix was developed by

(7) on May 21, 2013 for . Test Strips were placed on June 26, 2013 and September 5, 2013 to refine the mixture design, construction operations and to evaluate the friction characteristics. Table 13 shows the mixture design for the test pad.

Construction and Quality Control The Streetman lightweight aggregate Test Pad was placed on September 6, 2013. Conventional construction equipment was utilized. The construction equipment used for Test Strip 9 was used to construct the Test Pad. The paver speed was 18 to 22 feet per minute.

The asphalt paving mixture was produced at 320 to 340 F at the plant. The paving operation started at 8:20 am on Lane 1. Breakdown rolling was started immediately behind the paver.

Intermediate rolling was started at 10:00 am. Paving was started on Lane 2 at 10:23 and paving started on Lane 3 at 12:00 noon.

The mixture was tender during the compaction operation. The amount of vibratory rolling during breakdown rolling was reduced and the mat was allowed to cool below about 180 F once the tenderness or shoving under the rolling started. Pneumatic rolling started at a temperature in the range of 160 to 180 F. Breakdown rolling at temperatures below about 260 F caused shoving and cutting of the mat.

The likely causes of the tender mixture during construction was the use of the polymer modified asphalt binder, the rounded coarse aggregate, the high water content in the lightweight coarse aggregate (caused a foaming of the mixture), the design of the gap graded mixture and the relatively high asphalt binder contents. It is interesting to note that the same materials and mixture design was placed the previous day with the same construction equipment without shoving and cutting of the matt.

Five loose mix samples were obtained from the truck at the hot mix plant. Gradations, asphalt binder contents, laboratory compacted air void contents, Hamburg test and tensile test results are shown on Table 13 and Tables 26 to 29. Core samples were obtained from the Test Pad and results are shown on Tables 30 to 34. Table 35 shows the statistical summary of the quality control tests for gradation asphalt binder content and laboratory compacted air void contents.

Table 36 contains the statistical summary for the Hamburg and tensile strength test results for the Test Pad.

The data contained on these tables indicate that good control was obtained on aggregate gradation and laboratory compacted air voids. The standard deviation on asphalt binder content was somewhat high for typical highway construction. The produced mixture was near the target values and had low variability associated with gradation and laboratory compacted air voids as measured by standard deviation.

Five sets of cores were obtained. The average in-place air voids content were 9.4 percent which was higher than desired.

Burnishing and Friction Measurements A summary of the burnishing and friction measurements for this Test Strip are shown on Tables 37A and 37B. The Test Pad surface is shown on Figure 17. Friction values collected during the burnishing operation are shown on Table 37A. Values prior to burnishing were in the high 0.50’s to the low 0.60’s. Friction values after burnishing and a number of friction measurements are also shown on Table 37A. Values of friction were reduced to the mid 0.40’s after 150 measurements.

Table 37B presents a summary of friction measurement made by tire manufacturer’s representatives on the Test Pad. Starting friction values in the high 0.40’s are indicated after burnishing. Friction values decreased to the low 0.40’s with multiple friction measurements.

Monitoring and management of the lanes assigned for tire evaluation allow this test pad to be utilized. With time additional asphalt binder will wear and weather from the surface of the aggregate and improved friction is anticipated (Table 37B).

Table 1: Tire Test Facilities in Texas Location Company(s) Year Constructed Facilities Comments

Ft. Stockton Bridgestone/Firestone 1955

• Oval

• Meandering

• Pads

• Shared facility between Bridgestone/Firestone

Laredo Goodrich, Toyo, Michelin, Yokohama, MB Tech • Shared facility

Pearsall Cooper 1988

• Oval

• Meandering

• Pads

• Noise

Pecos Goodrich 1986

• Oval

• Meandering

• Pads

• Not active

San Angelo

Goodfellow 1972 • Pads • U.S. DOT facility

Goodyear 1957

• Oval

• Meandering

• Pads

• Other

• Two major wet friction pads plus other facilities

Uvalde General Tire/Continental

• Oval

• Meandering

• Pads

• Shared facility between General Tire and Continental

Table 2: Other Facility Pads-Coarse Aggregate Type and Friction Values

Facility Designation Date

Placed Coarse

Aggregate

Typical Friction Values

Comments Mean Representative

Range

U.S. DOT-

San Angelo Pad 1972 Hoban • Goodyear

• Several pads placed

U.S. DOT-

San Angelo Pad Aug

2010 Hoban • Goodyear

A Pad 1972 Hoban • Pecos

Texas A&M

Pad No. 2 Jennite Flush Seal 0.17-0.20

• Low friction pad

• Scheduled for replacement

Pad No. 3 Limestone

HMA

• Terrazzo Finish

• Low friction

• Low macrotexture

Pad No. 4 Crushed Gravel

HMA

0.43-0.46

Pad No. 5 Rounded Gravel

HMA

0.41-0.43

Pad No. 6 Lt. Wt.

Chip Seal 0.52-0.54

Pad No. 7 Rounded Gravel Chip Seal

0.410-0.42

Pad No. 8 Lt. Wt.

HMA 0.47

Table 2: Other Facility Pads-Coarse Aggregate Type and Friction Values (Continued)

Facility Designation Date

Placed Coarse

Aggregate

Typical Friction Values

Comments Mean Representative

Range

B Pad Lt. Wt.

HMA

• Firestone/Bridgestone

• Same design as

Goodfellow

C Knippa

• General/Continental

• High tire wear

• Also used imported

German aggregate

Pecos

Hoban 0.32-0.38

• TTI surfaces at old

Goodrich facility

• Representative values for PFC, fine SMA and CAM mixes

• Lab samples polished and tested in the laboratory

Eastland- Limestone 0.22-0.27

Delta- Sandstone 0.33-0.42

Jones Mill -Igneous 0.3.4-0.45

TCS-

Limestone 0.26-0.37

TTI Study 5627-2

A-crushed siliceous gravel

• TxDOT project

• SN-trailer values

• Aggregates used in seal coats, PFC and Type C HMA

• 2 to 5 years of data

• Pavement age from 2 to 11 years

D-crushed Granite

H- Sandstone 0.42 0.23-0.44

L-Lt Wt Aggregate 0.62 0.38-0.68

O-Trap Rock 0.33 0.19-0.40

P-Trap Rock 0.28 0.27-0.32

S-Rhyolite T-Granite

Table 3: Goodfellow Test Strips and Pads-Coarse Aggregate Type and Typical Friction Values

Designation Date Placed Coarse

Aggregate

Typical Friction Values

Comments Mean Representative

Range

Pad 1972 Hoban 0.40’s

• First Pad placed at Goodfellow,

• other pads placed at Goodyear, Goodrich and other tire test facilities and Texas A&M

Test Strip 0 Sept 8, 2006 Limestone-Del Rio Kelly Pit

• Not tested

• Polished under service

Test Strip 1 Sept 11, 2006 Hoban 0.48-0.55

• Friction drop with use

(0.54 to 0.48, 0.50 to 0.47, 0.42 to 0.39)

Test Strip 2 April 10, 2007 Hoban 0.48-0.55

• Lowered value to

0.42 with burnishing

• Values recovered without use

Test Strip 3 Aug 11, 2007 Limestone 0.42-0.48

• Values drop to 0.42 with use

• Values recovered without use

• Removed to place Test Strip 5

Test Strip 4 April 18, 2008 Sandstone 0.40-0.50 • Values drop to 0.38 with use

Test Strip 5 Feb 4, 2009 Lt. Wt.- Streetman 0.45-0.58

• Highly variable with burnishing

• 0.49 to 0.51 under test

Test Strip 6, Lane 1 Aug 22, 2009 Lt. Wt.-

Streetman • Not burnished

Test Strip 6, Lane 2 Aug 22, 2009 Lt. Wt.-

Streetman 0.45-0.49 • Power washed only

Test Strip 6, Lane 3 Aug 22, 2009 Lt. Wt.-

Streetman 0.43-0.50 • Power washed

• Acid washed

Test Strip 6, Lane 4 Aug 22, 2009 Lt. Wt.-

Streetman 0.45-0.53 • Sand

Pad Aug 5, 2011 Lt. Wt.- Streetman 0.42-0.51

• Drop close to 0.40 with use

• Rock loss on surface

• Variable surface

• Porous surface

Table 4: Mixture Designs-Goodfellow

Designation Date

Placed Coarse Aggr

Aggregate A Aggregate B Aggregate C Aggregate D Aggregate E Asphalt Binders

Source

Source

Source

Source

Source

G rade

Source

Pad 1972

H oban

A C

-10

Big Spring

Pad 1991

Lt. W t.-

R anger

A C

-10

Big Spring

Test Strip 0 Sept 8, Lim estone

Test Strip 1 Sept 11, H oban

Test Strip 2 April 10, H oban

H oban-

Type D

Turner

Screenings

32 5.5

64-16

Alon

Test Strip 3 Aug 11, Lim estone

Test Strip 4 April 18, Sandstone

D elta-G

r. 5

Turner W ashed Screenings

Turner B low Sand

10 6.7

64-22

SEM

Test Strip 5 Feb 4, Lt. W t.

Streetm an

Lt. W t.

Streetm an

Type FF

Turner Screenings

C

SA

Sand

12 7.9

A C

-10

Alon

Test Strip 6 Aug 22, Lt. W t.

Streetm

Pad Aug. 5, Lt. W t.

Streetm

Table 5: Mixture Design Summary-Accumulative Percent Passing

Sieve Size

Ranger 1991 Streetman

Hoban

Sandstone

Hoban

Streetman

Vol Wt Vol Wt Wt/Vol Wt/Vol Wt/Vol Vol Wt 1 in ¾ in ½ in 100 100 100 100 100 99.9 100 100 100 3/8 in 99.1 99.3 99.8 99.8 98.1 97.9 98.6 99.9 99.9 No. 4 60.6 68.1 56.2 65.4 44.5 61.3 59.6 54.4 63.2 No. 8 40.3 No. 10 28.3 36.8 26.3 35.8 28.9 26.9 25.6 33.5 No. 16 No. 30 22.5 No. 40 16.2 21.2 17.9 24.4 14.0 15.3 14.8 19.2 No. 50 16.1 No. 80 7.5 9.7 11.7 15.9 10.0 9.9 11.0 14.1 No. 100 No. 200 4.8 6.1 4.5 6.0 4.6 4.4 5.4 4.9 6.0 Asphalt Binder, Percent

8.7? 7.9 5.5 6.7 5.8 6.3

Ranger 1991-Pad that performed until 2011 Streetman 2008-Test Strip 5 and 6 Hoban 2007-Test Strip 2 Sandstone 2008-Test Strip 4 Hoban 2012-Test Strip 7 Streetman 2013-Test Strip 8

Table 6: Mixture Design Summary-Percent Passing/Retained

Passing/Retained Sieves

Ranger

Streetman

Hoban

Sandstone

Hoban

Streetman

Vol Wt Vol Wt Wt/Vol Wt/Vol Wt/Vol Vol Wt ¾ x ½ in 100 100 100 100 100 100 100 100 100 ½ x 3/8 in 0.9 0.7 0.2 0.2 1.9 2.1 1.4 0.1 0.1 3/8 x No. 4 39.1 31.2 43.6 34.4 53.6 36.6 39.0 45.5 36.7 No. 4 x No. 10 32.3 31.3 29.9 29.6 15.6 21.0 32.7 28.8 29.7 No. 10 x No. 40 12.1 15.6 8.4 11.4 14.9 17.8 11.6 10.8 14.3 No. 40 x No. 80 8.7 11.5 6.2 8.5 4.0 6.4 5.4 3.8 5.1 No. 80 x No 200 2.7 3.6 7.2 9.9 5.4 11.7 5.5 6.1 8.1 Minus No. 200 4.8 6.1. 4.5 6.0 4.6 4.4 5.4 4.9 6.0 Asphalt Binder, Percent

8.7? 7.9 5.5 6.7 5.8 6.3

Ranger 1991-Pad that performed until 2011 Streetman 2008-Test Strip 5 and 6 Hoban 2007-Test Strip 2

Table 6: Mixture Design Summary-Percent Passing/Retained (Continued) Sandstone 2008-Test Strip 4 Hoban 2012-Test Strip 7 Streetman 2013-Test Strip 8

Table 7: Gradation of Aggregates after Extraction (Weight Basis)

Sieve Size

Ranger 1991 Streetman 2008

Hoban

Sandstone

Hoban

Streetman

Streetman

Vol Wt

TB

Wt

TB

Wt

TS

Wt

TS

Wt Wt/Vol Wt/Vol Wt/Vol Vol Wt Vol Wt 1 in ¾ in ½ in 100 100 100 100 100 100 100 100 100 100 100 100

3/8 in 99.2 99.5 99.1 99.3 98.4 97.3 99.0 98.5 99.7 99.7 99.7 99.8 No. 4 68.8 66.8 66.3 68.4 53.1 43.8 60.3 57.7 54.1 63.3 59.4 67.7

No. 8 35.7

No. 10 36.9 35.4 34.9 35.3 28.6 25.3 28.2 28.9 37.6 31.1 38.1 No. 16 No. 30 20.3 No. 40 21.9 21.4 20.6 22.0 18.9 13.4 16.4 16.5 21.3 16.9 21.1 No. 50 14.6 No. 80 8.5 12.5 11.1 12.6 9.9 9.5 9.1 11.6 11.6 14.5

No. 100 10.3 No. 200 4.2 4.1 3.5 4.4 3.1 5.4 4.2 5.7 4.2 5.1 7.4 9.0 Asphalt Binder, Percent 8.7 6.5 7.4 6.0 6.5 7.3 Ranger 1991-Pad that performed until 2011 (average of three test results after extraction) Streetman 2008-Trial Batch 1 (TB 1) and Trial Batch 2 (TB 2) after extraction Streetman 2008-Test Strip 5 (TS 5) and Test Strip 6 (TS 6) after extraction Hoban 2007-Test Strip 2 after extraction Sandstone 2008-Test Strip 4 after extraction Hoban 2012-Test Strip 7 after extraction Streetman 2013-Test Strip 8 after extraction Streetman 2014-Test Pad after extraction

Table 8: Lab Mix Design and Field Trial Batches-Hoban-Test Strip 7 (Weight Basis)

Property Design Target (9-06-12) (1) Lab Mix Design Field Trial Batch

Sept 6, 2012

No. 1 Sept 6, 2012

No. 2

Sept 12, No. 1

Oct 18, No. 2

Nov 12, No. 3

Aggregate Hoban-Ty D 53 58 58 58 58 Hoban-Gr 6 19 14 14 14 14 Turner-Sc 22 22 22 22 22

Table 8: Lab Mix Design and Field Trial Batches-Hoban-Test Strip 7 (Weight Basis) (Continued)

Property Design Target (9-06-12) (1)

Lab Mix Design

Field Trial Batch

Turner-Sand 5 5 5 5 5 Other…