|Research Home | Pavements Home|
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-01-167
Date: April 2005
Structural Factors of Jointed Plain Concrete Pavements: SPS-2—Initial Evaluation and Analysis
Appendix B. Summary of SPS-2 Project Construction and Deviation Reports
After the construction of each SPS-2 site, a comprehensive construction report was prepared to provide a general description of the project, summarize the construction activities, describe the paving materials, and note any key observations about the project. In many cases, a deviation report was also prepared to indicate significant events and deviations from the planned factorial designs. These construction and deviation reports provide valuable information about the SPS sites and are essential to a good understanding of these projects and their performance.
These reports are not readily available to the general public. Therefore, a summary of each of these reports is provided in this appendix. The report summary is presented in the following alphabetical order:
The Arizona SPS-2 project site is located in the eastbound lanes of I-10 in southwestern Arizona, approximately 56 km west of Phoenix. I-10 is a rural interstate that carried average annual daily traffic (AADT) of 15,900 (1992 statistics). The SPS-2 project was constructed as part of the rehabilitation of I-10. The typical pavement design consists of two 3.66-m-wide lanes, an outside shoulder width of 3.05 m, and an inside shoulder width of 1.22 m. The SPS-2 test sections were constructed on a portion of I-10 that is relatively straight and flat.
This test site is classified by LTPP to be in the dry no-freeze zone. The automated weather station has been functional since July 1994.
International Road Dynamics bending plate WIM equipment was installed in the fall of 1993. Calibration was completed on January 24, 1994.
Construction of the SPS-2 site started in June 1993 with the removal of the existing pavement. Construction of individual test sections was completed and opened to traffic on October 1, 1993.
All required core sections were constructed. Supplemental State test sections constructed included the following:
Materials and Construction Methods of Individual Pavement Layers
Overexcavation of the subgrade for test section construction was accomplished with a front-end loader, a sheepsfoot roller, scrapers, and a water truck. Compaction of the subgrade ranged from 95 to 97 percent of maximum density. Only 8 percent of the surveyed elevations ranged from the SHRP tolerance of 13 mm in 15.2 m.
The DGAB consisted of a crushed river gravel with a 25.4-mm top size. Six percent of the fraction passed the No. 200 sieve. The DGAB was placed (belly dump trucks unloaded on grade) in 102- to 152-mm lifts and was compacted with a rubber tired roller. A CMI TM trimming machine was used to achieve proper layer thicknesses; however, layer thicknesses were generally in excess of design requirements.
The PATB contained an AC-20 binder at 2.5 percent of the total mix volume. This layer was placed (124 °C laydown temperature) at a 114-mm lift thickness with a Blaw-KnoxTM PF500 track- mounted paver. Profile grade control was maintained with a wire guide line and/or paving skis. The PATB was compacted by two steel-wheeled vibratory steel-wheeled rollers. They achieved a rolldown of approximately 19 mm at a mat temperature under 77 °C.
The mix design for this material included 105 kg of Type II cement, 23 kg (50 lb) Class F fly ash, 1,505 kg coarse aggregate, water reducers, and air entraining agents. The mix had a water-cement ratio of 0.88.
The LCB was paved with a GomacoTM GP3000 slip-form paver and a Gomaco 9500 spreader, which had a laydown width of 6.7 m. Initial laydown of the LCB mix produced a dry mix, which resulted in considerable surface checking and dragging. These areas were hand patched with grout. Batch plant modifications remedied the mix problems (water demand), and construction of a smooth LCB mat was maintained for the remainder of the project. Finishing was initially provided with an automatic float and a burlap drag, but was later changed to a hand float and no texturing. A membrane-curing compound was applied to the LCB. Partial monolithic construction into the outside shoulder was achieved (single pass 0.9 m to 6.7 m right of the centerline). Three transverse cracks developed in section 040217 while 8, 4, and 17 cracks developed in sections 040218, 040219, and 040220, respectively.
Portland Cement Concrete
The 3.8 MPa and 6.2 MPa SHRP mixes had a maximum aggregate size of 25.4 mm, while supplemental State sections (undoweled) used a 3.8 MPa flexural strength AZDOT mix with a maximum aggregate size of 38.1 mm.
All mix design utilized on this project utilized Type II cement.
The 3.8 MPa mix had a water-cement ratio of 0.47 and a 14-day flexural strength of 3.9 MPa. The 6.2 MPa mix had a water-cement ratio of 0.36 and a 14-day flexural strength of 5.8 MPa.
The PCC was paved with a Gomaco GP3000 slip-form paver and a Gomaco 9500 spreader/distributor. Cold joints were installed at the end of the day in several of the test sections. Test sections were paved in September. Corresponding concrete temperatures in some test sections were at least 27 °C.
Construction Report and Data Evaluation
AASHTO No. 57 coarse aggregate was utilized as the backfill material in the pavement base drain.
A taper transition of the PATB into the DGAB could not be achieved. This resulted in the PATB being placed against the DGAB at the end of section 040263.
The width of the Class B geotextile supplied was too short to be wrapped around the PATB edge according to SHRP specifications. This could facilitate soil intrusion from the adjacent DGAB.
Transverse drains were installed perpendicular to the roadway centerline instead of in a herringbone fashion.
A 0.9-m-wide roll of filter fabric was placed with a 0.305 m width under the median edge of the PATB. The remaining width was wrapped around the median edge of the PATB to prevent soil infiltration.
Transverse cracking occurred in the LCB in sections 040217 through 040220 before placement of the PCC.
Longitudinal tie bars were uncoated and were only 508 mm in length. SHRP specifications require epoxy-coated rebar, 762 mm in length.
Paving was intermittently stopped in several of the test sections due to concrete availability, mix adjustments, and equipment failure.
PCC segregation and/or slump variations occurred in several of the sections.
The concrete temperature throughout construction generally ranged from 28 °C to 31 °C.
The Arkansas SPS-2 project site is located in the westbound lanes of I-30 in west central Arkansas. The project is located just to the west of the I-70/I-30 interchange. I-30 is classified as a rural interstate with a 1993 AADT of 18,000 and 45 percent heavy trucks. The SPS-2 project was included in the reconstruction of I-30. Of the 12 test sections required for the SPS-2 project, 3 were located in original construction fill areas, 6 were located in original construction cut areas, and 3 sections were at-grade. The roadway typical for this project consists of two 3.66-m-wide lanes, an outside asphalt shoulder width of 3.05 m, and an inside shoulder width of 1.22 m.
This test site is in the wet no-freeze zone. The subgrade is classified as fine-grained. A weather monitoring station and WIM equipment were installed onsite.
Construction of the SPS-2 site began in November 1993 with the removal of the existing pavement. Construction of individual test sections was completed on October 1, 1995, and the project was opened to traffic on November 1, 1995.
Test Sections Constructed
All required core sections were constructed.
The existing subgrade is classified as a fine-grained material.
The PCC was batched in 6.9 m3 loads and delivered to the project, which was located approximately 5 km from the batch plant.
For the 6.2 MPa mix (sections 050204, 050208, and 050212), the air content ranged from 4.9 to 6.0 percent while the slump ranged from 25.4 to 38.1 mm.
Construction Report and Data Evaluation
Existing edge drains from the original I-30 roadway construction were removed during construction of the SPS-2 test sections.
On section 050208 (279 mm 6.2 MPa JPC on 152 mm LCB), the vibrators of the slip-form paver became entangled with the dowel basket assembly at station 2+50. This caused the augers of the paver to stop operating. The contractor removed and replaced the affected concrete and dowel basket assembly.
Longitudinal joints were not sealed until early 1997. By this time, pumping was evident through these joints.
Data Collection and Materials Sampling and Testing Deviations
The Colorado SPS-2 project is located in the eastbound lanes of I-76 in central Colorado, approximately 32 km northeast of Denver. I-76 is a rural interstate with a 1988 AADT of 8,400 and 16 percent heavy trucks. The design KESALs for this project is 15,600 for a 20-year design life. Seven SPS-2 test sections were included in the phase 1 section of I-76, which was constructed on a new alignment (sections 080217, 080220, 080221, 080222, 080223, and 080224; station 155+90 to station 227+90). The remaining six test sections (sections 080213, 080214, 080215, 080216; 080218, and 080219) were located within the phase 2 section of I-76, which was being reconstructed (station 101+40 to station 155+60).
The 136th Street interchange bisects this SPS-2 site. Sections 080213 through 080216 and 080218 through 080219 are located south of the 136th Street exit. Sections 080221, 080222, and 080223 are located north of the 136th Street entrance ramp. Sections 080217, 080220, and 080224 are located between the 136th Street exit and entrance ramps. No appreciable difference in traffic loadings is expected due to the presence of this interchange.
This SPS-2 site is located in a dry-freeze climate. The subgrade for this project is coarse grained and predominantly consists of sand to clayey sand. The replicate SPS-2 project for this site is located in northern Nevada.
All test sections are on a tangent. The vertical grade averages +1.4 percent in the direction of traffic. Six test sections were located in a cut (sections 080217, 080218, 080219, 080220, 080223, and 080224), while all other sections were located in on fills. The roadway typical for this project consists of two 3.66-m-wide lanes, an outside shoulder width of 3.05 m, and an inside shoulder width of 1.22 m.
A weather monitoring station was installed on-site. Construction began on July 1, 1993, and was completed on November 1, 1993. Phase 1 work (new alignment) was completed first and opened to traffic on October 7, 1993. Phase 2 work (I-76 reconstruction) was started after phase 1 was open to traffic. Phase 2 was opened to traffic on January 5, 1994.
All required core sections were constructed. Supplemental State test sections constructed included the following:
Materials and Construction Methods of Individual Pavement Layers
The existing subgrade varies from clayey sand to sandy clay, but is classified as coarse grained for this experiment. Profile grade construction in phase 1 was accomplished by constructing embankments in fill from subgrade removed from the mainline in cut areas. Subgrade compaction was accomplished through the weight of construction traffic, which consisted of scrapers and dozers. Compaction was monitored with nuclear density gauges. No moisture was added to the prepared subgrade since the soil exhibited an acceptable level of moisture due to a locally high water table.
The subgrade in section 080222 appeared to be poorly compacted but was not recompacted. The subgrade in section 080220 appeared to have high water content, and pumping was observed in the transition area between this section and section 080224. Two areas in the travel lane in section 080217 required undercutting. Subgrade was removed to a depth of 1.22 to 1.83 m and a width of 1.83 m and replaced with fine sand. Several areas within this test section pumped due to the weight of the end-dump trucks during paving of the LCB. Subgrade compaction with steel-wheeled rollers was performed immediately in front of the trucks, but this action did not solve the pumping problem. LCB paving proceeded with no additional subgrade stabilization being performed. All other test sections in the phase 1 area appeared to be well compacted.
During construction, a temporary access road crossed section 080221. Local traffic and construction traffic used this access road. This resulted in a varying compactive effort for the subgrade in this test section. Additionally, the subgrade was constructed at different times in this test section due to construction sequencing of the access road.
Embankment construction in the phase2 area consisted of removing the existing bituminous and concrete layers of the old roadway, crushing this material to 152-mm-diameter pieces, and recompacting this material along the mainline fill sections in the phase2 area. Approximately 0.61 m of fine sand cover material (obtained onsite) was placed on top of the crushed roadway material. The depth of crushed roadway material ranged from 0.91 to 1.22 m in section 080213 and 080216, to 1.83 m in section 080214, and to 3.05 m in section 080215. Sections 080218 and 080219 were cut with individual cuts ranging from 1.83 to 2.44 m in depth.
Dense-Graded Aggregate Base
The DGAB was a pit run material that conformed to SPS-2 and Colorado Class 5 specifications. Class 5 material has a 38.1-mm top size, 30 to 70 percent of the material passing the No. 4 sieve, and 3 to 15 percent of the material passing the No. 200 sieve. The DGAB was delivered in belly-dump trucks, which were permitted to drive on grade, bladed with a Caterpillar (CAT) 140G, compacted with a CAT CS563 steel-wheeled roller, and trimmed to the desired thickness.
Weather conditions during placement of the DGAB in the phase1 sections were cool and overcast, with light to heavy rain on several occasions. Phase 2 was constructed under better weather conditions, with variations from clear and sunny to cool and overcast with light rain.
The PATB contained an AC-10 binder at 2.5 percent with no antistrip agent. This layer was delivered in end-dump trucks from an offsite drum plant and placed with a Blaw-Knox track-mounted paver. Grade control was maintained with a stringline. The PATB was placed in 122- to 127-mm lifts at 82 to 93 °C in 4.11- m widths (except for the last pass, if the required paved width was less than 4.11 m). The PATB was compacted by two passes from a 5-ton steel-wheeled roller. Excessive fines were noticed in sections 080221, 080222, and 080223. The PATB was removed and replaced in section 080221 due to excessive fines. An additional 50.8 mm of PATB were added (day after original paving) in the right shoulder area of section 080221 for 38.1 m when construction inspection personnel discovered that this section was too low.
The mix design for this material included 92 kg of Type I/II low-alkali cement, 720 kg crushed No. 57 stone, 698 kg (1550 lb) sand fine aggregate, 27 kg Class F fly ash, and 115 kg water. The water-cement ratio was 0.96. This mix design produced a 7-day compressive strength ranging between 3.8 and 5.2 MPa.
The LCB was hauled in end-dump trucks, which were permitted to back down the subgrade directly in front of the slip-form paver (CAT SF550). The LCB was finished with a wet burlap drag, hand trowels, and steel floats. The LCB was paved at laydown width of 11.58 m in phase 1 sections and 9.75 m in the phase 2 sections. Air temperatures during placement averaged 21 °C during phase1 and 7.2 °C during phase 2.
The subgrade was pumping during paving in sections 080217 and 080220. The finished surface appeared rough throughout the entire length of each of these Phase 1 test sections at 0.76 m (2.5 ft) right of the proposed centerline. The LCB mix changed from a 102-mm (4-in) slump to a 50.8-mm (2-in) slump near the end of section 080217 (the first LCB section to be paved). Several transverse cracks, segregated areas, and depressions were noticed in the LCB mat after paving and curing.
Both LCB sections constructed in phase 2 were placed during cool rainy conditions. Both sections had broken edges (shoulders added at a later date) and a muddy finished surface with water stains.
All mix designs on this project utilized a Type I/Type II low-alkali cement. The 3.8 MPa mix included Class F fly ash. The 3.8 MPa mix had average flexural strengths of 3.6 MPa at 7 days and 3.9 MPa at 14 days. The 6.2 MPa mix had average flexural strengths of 5.8 MPa at 7 days and 6.2 MPa at 14 days. The water-cement ratio averaged 0.47 for the 3.8 MPa mix and 0.29 for the 6.2 MPa mix.
Side-dumping concrete into a track- mounted Gomaco PS60 spreader completed paving. This machine augered the concrete across the lane, which was consolidated with a track- mounted CAT SF550 slip-form paver. A power screed, wet burlap drag, and a mechanical float were used for intermediate finishing, while hand floats and an astroturf drag were used for final finishing. A transverse tining machine completed the texturing.
Joint sawing was performed within 8 hours of concrete placement. Joints were sawn and sealed according to SPS-2 specifications. The shoulder joint was sawn full-depth along all sections.
Paving of phase1 sections was completed on September 3, 1993 for sections 080221 and 080223 (3.8 MPa mix) and section 080222 (6.2 MPa mix). The air temperature ranged from 13 to 20 °C during section 080223 paving, 21 to 26 °C for section 080222 paving, and 24 to 27 °C for section 080221 paving. At station 186+00 in section 080221, the paving train pulled out a dowel basket. This assembly was not replaced.
Sections 080224, 080220 (6.2 MPa mix), 080217 (3.8 MPa mix), and 080259 (CO DOT 4.5 MPa mix) were paved from September 7 to 9, 1993. Paving in section 080224 was discontinued due to heavy rain. The edge drains and geotextile were damaged during paving but were not repaired. A construction joint was formed in this test section, and paving was resumed on September 8. The belt on the feeder to the spreader broke after completion of this test section. Paving operations were then discontinued for the day. The remaining sections were paved on September 9 and 10, 1993. The air temperature ranged from 13 to 20 °C during section 080220 paving, 22 to 27 °C for section 080217 paving, and 13 to 27 °C for section 080259 paving.
Paving of the phase 2 sections was completed from October 11 through October 22, 1993. Sections 080216 (6.2 MPa mix) and 080213 (3.8 MPa mix) were paved on October 11, 1993. The air temperature ranged from 7.2 to 21 °C during the section 080216 paving and 21 to 22 °C for the section 080213 paving. Section 080214 (6.2 MPa mix) was paved at air temperatures of 10 to 18 °C. Sections 080215, 080218, and 080219 were paved at air temperatures of 18 to 22 °C, 4 to 13 °C, and 3.3 to 9 °C, respectively. Sections 080213 and 080215 (3.8 MPa mix) had average slumps of only 25.4 mm.
Construction Report and Data Evaluation
Subgrade pumping occurred on several phase 1 sections due to rainy weather and a locally high water table. Pumping did not occur on the phase 2 sections. The embankment in these test sections consisted of stable fill material including pulverized concrete and asphalt capped by a fine sand layer.
Several of the PATB sections contained too many fines in the mix. This resulted in removal and replacement of the mat in section 080221.
Due to its high plasticity, the 6.2 MPa concrete mix was harder to work with than the 3.8 MPa mix.
While paving section 080218, equipment problems and concrete delivery problems (muddy haul roads) caused several work stoppages. The dowel bars and basket assembly were torn up at station 141+50 but not replaced.
No major problems occurred during construction of the DGAB and LCB layers.
The Delaware SPS-2 project site is located in the southbound lanes of U.S. 113 in central Delaware, between Milford and Georgetown. U.S. 113 is a rural principal arterial with a 1989 AADT of 10,708 and 10 percent heavy trucks. The design KESALs were calculated to be 3,048,600 for the 15-year design life of the pavement. The SPS-2 project was included in the addition of two southbound lanes to an initial two-lane roadway. The two new southbound lanes were separated by a 7.92- to 12.8-m-wide median from the existing northbound lanes. Route S-625 (State Route) and another access road bisect this SPS-2 site. The traffic from these routes is expected to have little impact on the SPS-2 site.
The roadway typically consists of two 3.66-m-wide lanes, an outside shoulder width of 3.05 m, and an inside shoulder width of 1.22 m. The SPS-2 project site lies within the Atlantic Coastal Plain. The subgrade consists of sand and silty sand. The topography is flat to gently rolling, and bedrock does not exist near the pavement surface. Test sections were constructed on shallow cuts or fills. The cut sections ranged up to 1.52 m in depth. Several wetland areas exist adjacent to the mainline pavement, where the water table is at or near the surface for an extended time period.
This test site is in the wet-freeze zone. The subgrade is classified as coarse grained. A weather monitoring station, WIM, and AVC equipment were installed onsite.
This project was completed and opened to traffic on May 1, 1996.
Test Sections Constructed
All required core sections were constructed. Supplemental State test sections constructed included the following:
Delaware DOT standards require the top 305 mm of subgrade to be Type A borrow material. This material consists of 95 to 100 percent of the material passing a 76.2-mm sieve and a maximum of 35 percent of the material passing the No. 200 sieve. Existing subgrade material in cut sections that met Type A borrow requirements was left in place. All other areas received at least 305 mm of Type A borrow.
Subgrade and embankment construction started in the spring of 1994 under wet conditions. Wet conditions persisted throughout the summer with little completion of earthwork. Most of the earthwork was completed during the fall of 1994 under dry conditions. Earthwork resumed in April 1995 and was completed by the end of the month.
The DGAB consisted of an igneous diorite (trap rock) crushed stone with approximately 90 percent passing the 25.4- screen. The DGAB was placed in 102-mm lifts with a spreader box and compacted with a DynapacTM 105 CA25 single drum vibratory compactor.
This layer was placed with a material transfer vehicle and a paver. The PATB was rolled with a 10-ton steel-wheeled static roller when the mat temperature was between 66 and 77 °C. Construction traffic was allowed on the PATB after placement and cooling. Damage to the mat did occur from turning movements of the construction traffic.
The mix design for the LCB included 144 kg of Type I low-alkali cement, 718 kg fine aggregate, 720 kg coarse aggregate, and 143 liters water. The mix had a water-cement ratio of 0.96. Fourteen-day compressive strengths ranged from 7 to 8 MPa.
The LCB was placed with a Blaw-Knox MC30 mobile conveyor and an ABG TitanTM 511 paver. A roller recompacted the subgrade before LCB placement to avoid rutting of the sandy subgrade. The LCB was laid at a 8.84-m-width for the 3.66-m-wide lane sections and at a 9.45-m-width for the 4.27-m- wide lane sections. A grooving tool was used to form the longitudinal joint in sections 100205, 100206, and 100207, while the longitudinal joint was sawn in section 100208. For sections 100205 and 100208, the longitudinal joint was offset 305 mm into the passing lane. This joint was offset 457 mm into the driving lane on sections 100206 and 100207. High spots in the LCB were milled before placement of the PCC.
The SPS-2 3.8 MPa concrete mix utilized on this project contained a Type I low-alkali cement while the 6.2 MPa mix utilized 65 percent Type I cement and 35 percent New CemTM (slag cement).
The PCC was delivered to the SPS-2 site in side-dump trucks. A Maxon TM spreader and a Gomaco GP 3500 slip-form paver were used in the paving operations. Concrete paving commenced on June 15, 1999, with the paving of the Delaware control section (section 100260) with a DelDOT Type B mix. Sections 100205, 100201, and 100209 were paved on June 16, 1999, with the SHRP 3.8 MPa mix.
Shrinkage cracks developed in the PCC in sections 100205, 100201, and 100209 placed on June 16, 1995. The cracked PCC in these test sections was removed and repaved with a Delaware DOT Type B mix on October 12, 1995. The DelDOT Type B mix is a 20.7 MPa (compressive strength) mix and has a corresponding flexural strength of approximately 4.5 MPa. Several transverse cracks again developed in these test sections that had been repaved with the DelDOT Type B mix.
Subsequent 3.8 MPa test sections that had not been paved (sections 100211, 100203, and 10207; 4.27-m-wide lane sections) were paved with the DelDOT Type B mix on June 28, 1995. Sections 100212 (PATB), 100208 (LCB), and 100204 (DGAB) are all 6.2 MPa test sections and were paved with a 7.5-bag mix without NewCemTM. Section 100212 was paved on July 17, 1995 (6:45 a.m. start). By 9:00 a.m., paving was stopped because the concrete temperature had reached 32 °C. Paving was restarted at 8:15 p.m. and was completed by 9:45 p.m. Paving continued through section 100208 and reached the midpoint of section 100208 when heavy rain was encountered. The concrete was covered with polyethylene. Paving resumed at 1:15 a.m. on July 18 and reached the end of section 100208 by 4:00 a.m. Rain resumed again at approximately 6:30 a.m. during concrete placement in section 100204. The concrete temperature was measured at 29 to 31 °C in the transition area between sections 100208 and 100204 while the air temperature was measured at 27 °C.
The 6.2 MPa SPS-2 mix was placed on June 29, 1999, in sections 100206, 100202, and 100210. Sections 10206 and 100202 developed excessive shrinkage cracks. The concrete was removed and replaced in sections 100202 and 100206 with another 6.2 MPa mix on November 21, 1995, while only patching of the transverse cracks was performed in section 100210.
Several transverse cracks were noticed in these test sections the following day. Longitudinal cracks were also noticed in section 100206, which had an underlying LCB.
DelDOT personnel believed the cracking was due to late sawing. The contractor believed that the joints were sawn as soon as the surface could not be marred. DelDOT personnel decided to remove and replace the concrete in sections 100201, 100205, and 100209 (3.8 MPa mix sections) and sections 100202 and 100206 (6.2 MPa mix sections) due to excessive shrinkage cracking.
Sections 100212, 100208, and 100204 were placed on July 17 and 18, 1995, with the modified 6.2 MPa mix that did not contain NewCem. Paving had to be shut down by 8:53 a.m. on July 17 due to high concrete temperatures. Paving was resumed at 8:15 p.m. Paving on July 18, 1999, had to be stopped twice due to heavy rain.
DelDOT supplemental section 100259 was placed on July 20, 1995. This test section contained the DelDOT 4.5 MPa Type B mix with NewCem.
Sections 100201, 100205, and 100209 were repaved on October 12, 1995, with the DelDOT 4.5 MPa Type B mix with NewCem. The interval between placement and sawing was 12 hours in section 100205. Within 2 weeks, transverse cracks again developed in section 100205. The cracks were predominantly in the driving lane close to the transverse contraction joints. Additional cracks developed during the ensuing winter. All existing cracks were patched during April 1996. Seventeen cracks were patched in section 100205.
Sections 100206 and 100202 were repaved on November 21, 1995, with the SHRP 6.2 MPa mix containing NewCem.
When the paver was stopped during paving of the LCB layer, depressions were formed in the subgrade. Transverse shrinkage cracks also developed in the LCB layer prior to PCC paving, and some of these shrinkage cracks developed in the depression areas.
During removal of the cracked PCC (DelDOT mix), construction personnel noticed that some of the shrinkage cracks in the LCB in section 100205 had reflected through the PCC. Some areas of the LCB had bonded to the PCC; however, the underside of most of the slabs was smooth and clean, which indicates an unbonded condition. The second application of a curing compound immediately before PCC paving appears to have been effective in debonding the PCC, except where surface depressions and irregularities existed in the underlying LCB.
A longitudinal crack had developed by October 13, 1995, in section 100207 at 457 mm from the centerline and parallel to the centerline. This crack was near the underlying construction joint in the LCB. This crack was cored on October 26, 1995, and was found not to extend for the full depth of the concrete pavement. This crack may be attributable to late sawing of the longitudinal joint. This section was paved on June 28, but longitudinal joint sawing was not performed until July 3.
Prior to removal of the concrete in sections 100205, 100206, and 100207, coring of transverse and longitudinal shrinkage cracks was performed. These cracks were found to extend either entirely or partially through the PCC but not into the underlying LCB. No bond was found to occur between the PCC and the underlying LCB.
Removal of some of the DGAB occurred in sections 100201 and 100202 with removal of the cracked JPC. Additional DGAB was added before JPC repaving in the test sections to create a uniform mat. The DGAB was then reshaped and recompacted.
After full-depth patching was completed, several additional cracks developed during the winter of 1995-1996 in section 100205 (LCB). Two additional cracks developed in section 100201 (DGAB), but no additional cracking developed in section 100209 (PATB).
Patching of these cracks was performed from April 18 to 19, 1996. At this time, 17 fine transverse cracks were noticed in various test sections. These cracks occurred at the edge of the pavement and only extended a few feet into the slab panel.
No. 57 stone was used as the edge drain backfill instead of PATB.
Transverse joint sealant reservoirs were sawn to a 19-mm width and a 38.1-mm depth, while the longitudinal joints were sawn to a width of 6.4 mm and a depth of 13 mm. The transverse joints in all test sections except sections 100206, 100202, and 100210 were sealed with neoprene seals. The transverse joints in the remaining sections and all longitudinal joints were sealed with hot- pour rubberized asphalt material.
Eight of the 12 test sections contained partial shallow cuts, but the cut subgrades had to meet Type A borrow specifications. Those cut subgrades that did not meet the Type A specifications were excavated to receive 305 mm of Type A borrow (with prior approval) (sections 100201, 100203, 100204, 100205, 100207, 100208, 100209, and 100211).
A transverse construction joint was placed within section 100212.
The longitudinal joint was sawn 5 days after the concrete placement in sections 100211, 100203, and 100207.
For all sections:
For sections 100212, 100210, 100211, and 100209:
3.8 MPa flexural strength concrete was not used on sections 100207, 100203, and 100211. 20.7 MPa comp was used instead.
3.8 MPa flexural strength concrete was used on sections 100201, 100205, and 100209. This concrete was removed and replaced with 4.5 MPa flexural strength concrete.
Sections 100202 and 100206 were placed with 6.2 MPa flex 6.5-bag mix. This concrete was later removed and replaced with 6.2 MPa flex 7.5-bag mix.
Profile index for all sections was greater than 158 mm/km. Note that section 100205 is scheduled for diamond grinding.
The Iowa SPS-2 project site is located in the northbound lanes of U.S. 65 in central Iowa, northeast of Des Moines. U.S. 65 is an urban/principal arterial with a 1994 AADT of 17,400 and 16 percent trucks. The calculated KESALs were 9,870 for the project over the 30-year design life of the pavement. The SPS-2 project was included in the relocation of U.S. 65 in both the northbound and southbound lanes.
The roadway typically consists of two 3.66-m-wide lanes, an outside asphalt shoulder width of 3.05 m, and an inside shoulder width of 1.22 m. The SPS-2 test sections were constructed on a portion of U.S. 65 that included both tangent and superelevated sections. All sections were constructed on a tangent except sections 190215 and 190216. These sections were constructed on the high side of a horizontal curve with a superelevation rate of 2.5 percent. Vertical grades throughout the project area range from -2.6 percent to +2.2 percent. Test sections 1902215 through 190220 were constructed on fill sections ranging from near 0 to 11.58 m in height. Sections 190221 through 190224 were constructed on cut sections ranging from 0.91 to 7.01 m.
This test site is in the wet-freeze zone. The subgrade is fine grained.
An onsite weather monitoring station was not installed until 1996.
WIM and AVC equipment was installed in June 1995 on U.S. 65 approximately 1.61 km north of the junction of U.S. 65 with IA-163.
Reconstruction was completed in 1994 during a period of relatively wet weather conditions. The project was opened to traffic on December 1, 1994.
All required core sections were constructed. Supplemental State test sections constructed included a control section according to the following specifications:
Subgrade preparation was completed with a motor grader, disk, and a 10-ton sheepsfoot compactor. A CMITM SP30AST full-width soil profiler completed final grading.
Dense-Graded Aggregate Base
The DGAB was compacted with vibratory and pneumatic-tired rollers. A profiler was used to trim the base to the required thickness.
Permeable Asphalt-Treated Base
This layer was placed with a Cedar RapidsTM CRS61 paver in a 3.96-m pass. The PATB was laid down at approximately 138 °C and compacted with two passes of a 10-ton tandem steel-wheeled roller. Monolithic construction transversely and into the shoulder was not obtained.
The LCB mix design included 669 kg of coarse aggregate limestone, 821 kg sand, 72 kg Type II cement, 22 kg Type C fly ash, and 120 kg water. Admixtures included a water reducer and an air-entraining agent. This mix had a water-cement ratio of 1.28, an air content of 6.3 percent, and a 38.1-mm slump. Eight-day compressive strength averaged 4.3 MPa.
The LCB was paved at a 7.92-m width with a CMI SF 450 slip-form paver. Finishing work included machine troweling and hand troweling. The LCB had a water-cement ratio of 1.28 and an average 8-day compressive strength of 4.3 MPa.
All mix designs on this project utilized Type II cement. The 3.8 MPa mix design produced an average 7-day flexural strength of 3.9 MPa and an average 14-day flexural strength of 4.2 MPa. The 6.2 MPa mix design produced an average 7-day flexural strength of 5.9 MPa and an average 14-day flexural strength of 6.0 MPa.
Underground structures were located in 6 of the 13 test sections (190213, 190214, 190215, 190217, 190219, and 190221). These ranged from a 0.61-m diameter concrete pipe at 2.44-m below profile grade to a 2.44-m by 3.05-m concrete pipe at 12.19-m below profile grade.
The contractor removed at least 0.3 m of geotextile from the longitudinal edge drains due to the low permeability of the geotextile.
The boundaries of section 190222 were relocated after construction because dowel bars with the wrong diameter were placed in the initial boundaries of this test section.
Four sections (190215, 190216, 190212, and 190223) had concrete thicknesses in excess of SPS-2 tolerances. These thicknesses ranged from 8 to 23 mm above the desired thickness (203 or 279 mm as applicable).
During the placement of the PCC pavement in test section 190222, incorrect dowel baskets were placed. This area was removed, and the test section location was shifted to avoid the replaced pavement area. This will shift the location of bulk sampling, nuclear density testing, and coring locations. Some tests will now be located outside and within the test section.
Because of misinterpretation of guidelines, the test section numbers were revised. The correct numbers should be from 13 through 24. This revision was done after most of the sampling, testing, and data collection had been completed.
The thicknesses of the following test sections deviated from the construction guidelines more than 0.012 m or 12 mm.
Section 190222 was removed after all sampling and testing and data collection was completed.
The Kansas SPS-2 project site is located in the westbound lanes of I-70 in central Kansas, east of Abilene. I-70 is a rural interstate with an AADT of 13,750 with 21.4 percent trucks. The yearly ESALs in the design lane are estimated at 1,300,678. The 20-year design ESALs is estimated at 26,013,550. The roadway typically consists of two 3.66-m- wide lanes, an outside shoulder width of 3.05-m, and an inside shoulder width of 1.22-m. The SPS-2 project was included in the reconstruction of I-70. The existing pavement was concrete. The SPS-2 test sections were constructed on a tangent section of I-70 with vertical grades ranging from -2.48 percent to +2.11 percent. All test sections were constructed on fills.
This test site is in the dry-freeze zone. An onsite weather monitoring station had not been installed before completion of the project; however, installation was scheduled to occur by 1994. A Toledo Model 9430 high-speed WIM system was installed onsite.
Construction of this SPS-2 project was completed on July 1, 1992. The project was opened to traffic on August 1, 1992.
All required core sections were constructed. Supplemental State test sections included a control section (I-70 reconstruction typical) as follows:
The subgrade soil is classified as silty clay. Since the SPS-2 project involved the reconstruction of an existing highway, the top subgrade layer was reworked and recompacted after incorporation of the existing granular subbase and shoulder material. Construction of the SPS-2 project occurred during an abnormally rainy period. KDOT added a Type C fly ash to the underlying subgrade to stabilize and dry this layer.
The contractor placed this base too thick, but subsequently trimmed the base to the required thickness.
This mix included 70 percent natural sand and 30 percent crushed limestone. The mix contained 1,256 kg of Type II cement, 976 kg fine aggregate, 418 kg coarse aggregate, and 126 kg water.
All mix designs utilized on this project contained 70 percent natural sand and 30 percent crushed stone. Crushed limestone was used in the 3.8 MPa concrete, but calcite- cemented sandstone was utilized in the 6.2 MPa concrete mix in part to help increase the flexural strength. Type II cement was used for each mix design.
The 14-day flexural strength averaged 4.2 MPa (0.34 MPa standard deviation) for the 3.8 MPa mix and 5.8 MPa (0.34 MPa standard deviation) for the 6.2 MPa mix.
The project construction report indicates that the PATB was difficult to place. The contractor placed this material too thick in several of the test sections. The excess was removed with a trimmer. During initial construction operations, the PATB deformed when compacted. This problem was resolved as the contractor gained experience with this material.
The thickness of the PCC did not meet SPS-2 tolerances for the following test sections:
Underground structures were present in sections 200204, 200208, 200209, 200210, 200211, and 200212. Median drains were present in several test sections; however, these drains were at least 1.52 m below the pavement surface.
The Michigan SPS-2 project site is located in the northbound and southbound lanes of U.S. 23 in southeastern Michigan, approximately 16 km west of Toledo. U.S. 23 is a rural principal arterial with a 1989 AADT of 35,000 and 22 percent heavy trucks. Twenty-six million ESALs were calculated for the design lane over the 20-year design life of the pavement. The SPS-2 project was included in the reconstruction of 9.7 km (6.021 mi) of U.S. 23 in both the northbound and southbound lanes. Consear Road, a low-volume county road, bisects this SPS-2 site. Traffic counts taken in the northbound lanes reveal that traffic south of this interchange is only 7 percent higher than traffic north of this interchange (AVC data only).
The roadway typically consists of two 3.66-m-wide lanes, an outside asphalt shoulder width of 3.05 m, and an inside shoulder width of 1.22 m. The SPS-2 test sections were constructed on a portion of U.S. 23 that is relatively straight and flat. Vertical grades throughout the project area range from 0.00 percent to +0.55 percent. Most of the test sections were constructed on shallow fills; however, sections 260214, 260218, and 260219 were constructed on deeper fills. All sections except 260214, 260218, and 260219 were constructed in tangent sections. Section 260214 was constructed in a superelevation transition area, while sections 260218 and 260219 were constructed on a full superelevation of 0.037 m/m.
This test site is in the wet-freeze zone. The subgrade is fine grained. An onsite weather monitoring station was not installed until the winter of 1995-1996.
WIM and AVC equipment were installed on U.S. 23 south of Consear Road. Only AVC equipment was installed north of Consear Road.
Reconstruction of U.S. 23 began in April 1993 with removal of the existing pavement layers. Construction of the subgrade progressed from mid-May through mid-June, and placement and compaction of the embankment was completed by mid-June. Undercuts were completed in sections 260216, 260022, and 260223 due to unstable soil conditions revealed during proofrolling. These undercuts were 11 m wide and 0.3 m deep, but only extended for a partial length of each test section. The undercuts were backfilled with embankment borrow clay. Base and subbase layer construction began by mid-June and was completed by mid-September 1993. Concrete paving commenced on September 13, 1993 (excluding control section) and was completed on September 21, 1993. The project was opened to traffic in November 1993.
Materials and Construction Methods of Individual Pavement Layers
The existing sand subbase was removed before the embankment clay. It was then graded and rolled with a 14-ton single-drum static roller and proofrolled with a 25-ton pneumatic tired device on a rigid frame. Embankment clay was then spread on grade in 229-mm lifts and compacted with a 13-ton sheepsfoot roller. The embankment was fine graded with a CMI autograder and proofrolled again.
The DGAB was end dumped on grade, leveled by a bulldozer and grader, and compacted with a 14-ton single-drum vibratory roller. A CMI autograder trimmed and fine graded the DGAB.
This layer was placed with a Barber Greene (BG)TM 225B paver. The laydown width for the paver varied between 2.74 and 4.88 m. The PATB was compacted with six passes of an 8-ton static roller. Monolithic construction transversely and into the shoulder was not obtained.
The mix design for the LCB included 722 kg coarse aggregate, 617 kg fine aggregate, 74 kg cement, and 128 kg water. The mix incorporated a water reducer and an air-entraining agent.
The LCB was delivered in end-dump trucks from adjacent haul roads and conveyed to a CMI SF 250 slip-form paver through a CMI MTP00 transport vehicle. Machine troweling and hand troweling were performed. A membrane-curing compound was applied to the LCB after placement and was reapplied before placement of the PCC. No data on placement widths of the LCB was provided.
Transverse shrinkage cracks developed in each LCB test section shortly after paving. Shrinkage crack development may have been augmented by very hot and dry conditions during LCB paving, as well as by dry embankment clay, as evidenced by the desiccation cracks.
All mix designs on this project utilized Type I cement.
The PCC was delivered to the SPS-2 site in end-dump trucks on haul roads adjacent to the mainline, placed on the belts of a CMI belt placer, and the first layer thickness was spread equal to 75 percent of the final depth. The final layer was again delivered via adjacent haul roads and deposited on belts of a CMI 450 slip-form paver. Dowel baskets were placed at the transverse joints. Finishing consisted of machine and hand troweling. The surface was burlap dragged and then tine textured. A membrane-curing compound was applied to the surface. Transverse contraction joints were sawcut as soon as the surface could not be marred.
Cracking underneath the sawcuts had not occurred under all joints for several weeks. This was attributed to the time of year the pavement was constructed. Little change in temperature occurred from day to night. The transverse joints were sealed with low-modulus silicone.
Construction and Deviation Reports
The project construction and deviation reports indicate that some deviations occurred due to site conditions (unequal traffic volumes, culvert within test section, unequal fill height geometry).
For sections 260213 through 260220, the moisture content of the subgrade was not maintained within 85 to 120 percent of the optimum moisture content. Moderate-to-severe desiccation cracks (up to 50.8 mm in width and 254 mm in depth) developed in the subgrade compacted dry of optimum since the completed embankment was exposed to hot and dry weather conditions before construction of the overlying base or subbase layers. This occurred on all sections except those constructed with PATB. Cracking did not occur on PATB sections because the DGAB was placed soon after the completion of the embankment.
Michigan DOT required the contractor to scarify the desiccated subgrade sections and recompact severely desiccated subgrade to SPS-2 requirements.
The following are some specific observations:
Longitudinal joint seal damage at the lane-shoulder joint occurred in several test sections by 1994. The entire length of this joint in all test sections (except control section) failed by 1995. No damage was evident in the control section, which was constructed with tied concrete shoulders.
Pumping at the longitudinal joint and transverse joints was observed in most of the sections constructed with a DGAB and all of the sections constructed with a LCB (undrained). No pumping was observed in PATB (drained) sections.
Low-severity transverse joint sealant damage occurred in several test sections by 1995.
Structural distresses-including pumping, transverse joint faulting, transverse cracking, longitudinal cracking, and corner breaks-had occurred in section 260218 (203 mm PCC on 152 mm LCB).
Data Collection and Materials Sampling and Testing Deviations
The Nevada SPS-2 project site is located in north central Nevada, approximately 8 km west of Battle Mountain, in the outer eastbound lane of interstate I-80. The SPS-2 sections extend from station 1596+65 to station 64+50 (milepost 223.7).
The construction work on this segment of I-80 consisted of removing the existing AC surfacing, CTB, DGAB, and embankment. The original subgrade was stabilized with lime and the embankment was replaced. The SHRP structural sections were then placed on top of the embankment.
The terrain surrounding the test sections is generally flat with minimal ground cover.
The location of the test site is in the dry-freeze zone. Based upon climatic information collected at a Battle Mountain weather station from 1961 to 1990, the average yearly high temperature was 39 °C, the average yearly low temperature was -26 °C, and the average yearly precipitation was 209 mm.
The soil in this area varied throughout the project. The Nevada SPS-2 project site fills the dry-freeze, coarse subgrade categories.
For this project, one supplemental State section was included. This section is designated as 320259 and consisted of a 38.1-mm leveling course over the existing AC and a 267-mm PCC surface layer. This was the Nevada standard design for the remainder of the project.
This project was constructed over an existing section of highway; removal of the existing AC layer was necessary. Upon this removal, there were problems that are described in the following sections. To correct these problems, a layer of lime- stabilized soil was placed and topped with a layer of granular material to produce a suitable subbase for the test sections.
Based on laboratory testing, the natural subgrade was primarily sandy silt. The percentage of clay ranged from 4.5 to 13.9.
Section 320212 had severe shrinkage cracking following paving and was removed on August 4, 1995. The PATB was also torn out, and 127 mm of CTB was placed in the excavation. The CTB was followed by a 267-mm lift of Nevada DOT PCC mix.
This project was constructed over an existing section of highway, and removal of the existing pavement structure was required. When this was performed, the subgrade (which was sandy silt) was found to be out of specifications for NDOT subgrade material. This required the lime stabilization of the top 0.3 m of subgrade material.
After this stabilization, embankment material was placed and compacted. FWD testing on the embankment showed that sections 320201, 320205, 320207, and 320209 had significantly higher deflections than the other sections.
The DGAB was placed on 8 of the 12 sections. The material was placed in either one or two lifts, depending on the design thickness. Sections 320201 and 320209 were found to have high variations in deflections during FWD testing, and section 320203 had deflections in the first 38.1 m, while the other five sections were more consistent.
As per the SPS-2 experiment design, four sections received a 102-mm PATB. Edge drains were constructed on these sections utilizing a geotextile and open graded rock placed in trenches.
Also according to the SPS-2 experiment design, four sections had a 152-mm LCB placed directly on the embankment. The LCB was placed in one 12.19-m-wide pass and there were no joints sawed. All sections except 320206 exhibited extensive cracking within two 2 weeks of paving.
The PCC consisted of three different mixes. Section 320259 was the State standard mix, while the other 12 sections had 6 sections of a 3.3 MPa mix and 6 sections of a 5.2 MPa mix. The typical SPS-2 project has six 3.8 MPa and six 6.2 MPa mixes, but it wasn't possible to reach the 6.2 MPa target using local materials, so the target strengths were revised.
A number of problems were encountered during PCC paving. Section 320201 had sections areas that needed to be hand-finished, and shrinkage cracks appeared shortly after paving. Section 320202 had several areas of tearing in the last 91 m. Within a day of paving section 320203, shrinkage cracks appeared. There was tearing of the PCC in the areas around the dowel bar inserter (DBI) on section 320204. Transverse tie bars had to be pounded in by hand for the first 88 m of section 320205 due to an equipment failure, and there were a number of areas that required hand finishing. Shrinkage cracks also developed in sections 320205 and 320208. In section 320209 had to have, approximately every other tie bar was pounded in with a hand mallet due to equipment problems. It rained for about 15 minutes, which resulted in dimples in about 61 m of the surface of section 320209. After the fifth load of PCC on section 3202011, the 19-mm aggregate was reduced by 2 percent and the fines were increased by 2 percent. Section 320212 exhibited such severe cracking after paving that it was removed and replaced with nonconforming materials, thereby removing it from the study.
The majority of the problems with the PCC paving came as a result of the mixes being significantly different from those typically used by the paving crew. This was especially true for the 5.2 MPa mix. Proof of this fact is that section 320259, which was the State standard mix, had none of the problems with shrinkage cracks and tearing that were so common for the majority of the project. The primary conclusion that can be drawn is that trying to perform nonstandard construction can cause significant problems. It is highly unlikely that the majority of the test sections will last anywhere close to their design lives.
The North Carolina SPS-2 project site is located in the southbound lanes of U.S. 23 near Lexington, NC. U.S. 52 is a rural principal arterial with an AADT of 23,500 to 26,100 (1994) and 13 percent heavy trucks. The 18-kip design KESALs was calculated as 10,784,326 for the 20-year design life of the pavement. The SPS-2 project was included in the construction of 7.8 km of U.S. 52 in both the northbound and southbound lanes. U.S. 64 bisects this SPS-2 site. All test sections except section 370204 are located north of the U.S. 64 interchange. This section will be monitored with AVC equipment to determine if the traffic south of U.S. 64 is different than traffic to the north of U.S. 64.
The roadway typically consists of two 3.66-m-wide lanes, an outside shoulder width of 3.05 m (10 ft), and an inside width of 1.22 m. The shoulders were constructed with econocrete instead of the SPS-2 required flexible bituminous material. The majority of SPS-2 test sections were constructed on tangent sections with slight grades. Five test sections were completely located within horizontal curves, and one section was partially located within a horizontal curve. Sections that include a 203-mm PCC slab were constructed as add-on lanes adjacent to the mainline travel lane. This parallel roadway section was constructed through some deep cuts and high embankments.
This test site is in the wet no-freeze zone. The subgrade consists of fine-grained soils. An onsite weather monitoring station was installed in August 1994.
No information is available on the installation of WIM and AVC equipment.
Seasonal monitoring sensors, strain gauges, and linear variable differential transducers were installed on several test sections.
Reconstruction began in 1992 with earthwork grading.
Materials and Construction Methods of Individual Pavement Layers
The existing subgrade is silt, which retained a large percentage of moisture and lost its bearing capacity when wet. Soil stabilization was performed with lime slurry in all test sections except three (320204, 320207, and 320260), which were treated with cement. The subgrade was ripped to a depth of 203 mm, mixed with the applied lime, and recompacted with vibratory sheepsfoot and vibratory steel-wheeled rollers. This process was repeated the following day, and the soil was then fine-graded to the required profile grade. The top 76 mm of the stabilized subgrade in sections 370209 and 370210 (add-on lanes) had to be removed to meet the required profile grade.
Sections 370204, 370207, and 370260 were stabilized to a depth of 178 mm with cement and granular base material. Mixing and recompaction was similar to that performed on the other test sections. The subgrade in all test sections was sealed with a CRS-1 emulsion.
The dense graded aggregate base in sections 370209 and 370210 was not treated with a prime coat before the construction of the overlying PATB.
This layer was placed with a track mounted Blaw-Knox spreader and was rolled (one pass) with a 10-ton steel wheeled tandem roller when the mat temperature was between 66 and 77 °C. Construction traffic was allowed on the PATB after placement and cooling. Damage to the mat occurred from turning movements of the construction traffic.
The mix design for the LCB included 222 kg Type I cement, 67 kg Class 1 fly ash, 553 kg fine aggregate, 866 kg coarse aggregate, and 115 liters water per cubic meter. Admixtures included an air-entraining agent and a water reducer. The LCB had a 14-day compressive strength of 24 MPa. The LCB was delivered to the SPS-2 site in side-dump trucks. A Maxon spreader and a Gomaco GP 3500 slip-form paver were used in the paving operations. The LCB was paved at widths up to 7.92 m.
All concrete mixes utilized on this project contained a Type I cement. The 14-day flexural strength of the 3.8 MPa mix ranged from 4.0 to 5.1 MPa, while the 6.2 MPa mix had a 14-day flexural strength of 6.1 MPa.
The PCC was delivered to the SPS-2 site in side-dump trucks. The concrete was dumped on grade ahead of the paver and spread with a front-end loader. A Maxon spreader and a Gomaco GP 3500 slip-form paver were used in the paving operations. Paving of the test sections located in the add-on lanes was accomplished by dumping the concrete from the previously constructed adjacent mainline lanes. Concrete paving commenced on October 24, 1994, and was completed on November 26, 1994. Construction joints were formed in sections 370204 and 370260. Air temperatures during concrete placement ranged from 3.3 to 20 °C.
LTPP SPS Construction Reports
Edge drains were located at a 0.61-m offset from the pavement edge, rather than the SPS-2 required 2.4-m offset. Stone was used instead of PATB as trench backfill.
Econocrete shoulders were approved for use instead of asphalt shoulders.
The DGAB extended only 0.61 m into the shoulder from the pavement edge.
Dowel bars (25.4 mm in diameter) were utilized on sections that included a 203-mm thick PCC. The LCB was constructed to extend only 0.61 m into the shoulder from the pavement edge.
Cracks developed in the LCB layer in several sections before construction of the PCC. These cracks were covered with tar paper prior to PCC paving. Several of these cracks reflected through the PCC. Consequently, some of these slabs were repaired.
The North Dakota SPS-2 project site is located in the eastbound lanes of I-94 in eastern North Dakota, west of Fargo. I-94 is a rural interstate with a 1996 AADT of 8,310 and 12 percent trucks. The yearly ESALs in the design lane are estimated at 900,000. The design ESALs (30-year design life) is estimated at 2,150,000. The roadway typically consists of two 3.66 m-wide lanes, an outside shoulder width of 3.05 m, and an inside shoulder width of 1.22 m. The SPS-2 project was included in the reconstruction of a concrete pavement that included 229 mm of concrete on 76 mm of aggregate base on 152 to 229 mm of aggregate subbase. The SPS-2 test sections were constructed on a portion of I-94 that is very flat and relatively straight. All sections except North Dakota supplemental sections 380260 and 360261 were constructed on tangent sections.
This test site is in the dry-freeze zone. An onsite weather monitoring station was installed in 1994.
WIM and AVC equipment were installed onsite.
Several delays were encountered during subgrade preparation due to the presence of extremely wet clayey soils. Construction of individual test sections was completed on October 1, 1994, and the pavements were opened to traffic on November 1, 1994.
The subgrade soil can be classified as a fine-grained clay. The project is located on the remnants of glaciated Lake Agassiz. Since the SPS-2 project involved the reconstruction of an existing highway, the top 457 mm were reworked and recompacted after removal of the existing concrete pavement, base layers, and top 305 mm of subgrade. The subgrade was then loosened with a plow and recompacted with sheepsfoot rollers. Sensors were placed in unbound base layers and in the subgrade to monitor moisture conditions after construction.
The DGAB was placed with a motor grader, rolled with an 18-ton pneumatic-tired roller, and trimmed with a CMI profiler.
This layer was drum plant mixed, paved with a Barber Greene 146 paver, and compacted with a 10-ton double-drum vibratory roller. The PATB was hard to roll and lost its form and shape once rolled. This layer was rolled at a temperature of 93 °C. NDDOT typically places PATB beneath concrete interstate pavements and has not experienced deformation of this layer (NDDOT specifications) during construction.
The LCB mix contained 36.7 kg Type C fly ash, 85.3 kg Type I cement, 128.3 kg (water, 1017 kg fine aggregate, and 448 kg coarse aggregate. The LCB was paved with a REX Town and CountryTM slip-form paver and a Curb MasterTM paver. Initial construction problems associated with the LCB included sloughing of the edges. This was attributed to too many fines in the mix and migration of mix water to the outside edge of the LCB during finishing. This problem was corrected by adjusting the mix design to a strength higher than that specified by SPS-2.
All mix designs utilized on this project contained gravel coarse aggregate. Type C fly ash was utilized in each mix design instead of Type F, as required by the LTPP protocol.
The project deviation report indicates that the LCB was difficult to place until the mix design was changed to increase the strength of this layer. The thickness tolerances on four core SPS-2 sections were not met (sections 380217, 380218, 380219, and 380220). Transverse cracks developed in section 380217. These cracks reflected through the 203-mm PCC within 5 days after construction of the PCC.
The PATB deformed when compacted.
The subgrade in section 380218 was unstable and should have been undercut. This caused some initial frost heave, but the condition has corrected itself.
None known due to the involvement of Steve Pflipson with his computerized tracking approach.
The layer thickness for the following sections contained deviations.
LCB was difficult to place, so the mix was designed stronger than the guidelines.
PATB was difficult to roll due to its fluid-like characteristics and its short lengths required.
In test section 380217, the transverse cracks in the LCB reflected through to the 203 mm of PCC pavement.
The first two test sections-380260 and 380261-were built on slight superelevations just after the on-ramp from Casselton.
The Ohio SPS-2 project site is located in the northbound lanes of U.S. 23 in central Ohio, approximately 48 km north of Columbus. U.S. 23 is a rural principal arterial with a 1994 AADT of 20,210 with 12 percent trucks. The roadway typically consists of two 3.66-m- wide lanes, an outside asphalt shoulder width of 3.05 m (10 ft), and an inside asphalt shoulder width of 1.22 m. The SPS-2 test sections were constructed on a portion of U.S. 23 that is relatively straight and flat.
This site is in the wet-freeze zone. The subgrade is fine grained. An onsite weather monitoring station was installed in 1995.
Permanent WIM equipment consisting of weigh plates was mounted in each lane of U.S. 23. Additional instrumentation was installed in the SPS-2 experiment area to collect environmental data, including temperatures of individual pavement layers and moisture freeze/thaw conditions of the subbase and subgrade layers. Load-response monitoring instrumentation installed included strain, deflection, and pressure gauges.
Construction started in the fall of 1994 with the subgrade preparation. Individual test sections were completed by October 1995, and the project was open to traffic on October 1, 1996.
The subgrade was compacted in 305-mm lifts with a 22.1-ton sheepsfoot roller. No stabilizing agents were used.
The DGAB was placed in 229-mm lifts and was compacted with a 16.5-ton single-drum vibratory roller to obtain a 152-mm layer thickness. The 152-mm lift thicknesses were similarly compacted to a 102-mm layer thickness. A CMI trimming machine was used to trim excess material; however, final constructed DGAB layer thicknesses were generally in excess of design thicknesses.
This layer was placed at a 127- to 152-mm lift thickness with a Blaw-Knox PH200B paver and compacted to 102 mm with 15 passes from a 7-ton steel-wheeled tandem roller. The single-pass laydown width for this paver was 3.66 m; therefore; monolithic construction into the shoulder was not obtained.
The mix design for the LCB base utilized a water reducer but no fly ash. This material was paved with a Gomaco GP2500 slip-form paver and spreader, which had a laydown width of 7.92 m. Finishing included screeding and the application of a membrane-curing compound to the LCB. Only partial monolithic construction into the shoulder area occurred.
The mix design for this material included 1136 kg of coarse aggregate, 113 kg of cement, and 38 kg of water. This material was paved with a CMI slip-form paver with a laydown width of 9.45 m. Finishing was performed by screeding, and white polyethylene was applied to the surface.
All mix designs utilized on this project contained Type I cement, 100 percent crushed stone for the coarse aggregate, and 100 percent manufactured sand for the fine aggregate. Type C fly ash was utilized in each mix design instead of Type F, as required by the LTPP protocol.
Dowel baskets were placed at the transverse joints. Finishing was performed by screeding and the surface was tine textured. A membrane-curing compound was applied to the surface. Transverse joint sawcut depths averaged 64 mm for the 203-mm JPC and 89 mm for the 279-mm JPC. Surface profiles were corrected by diamond grinding. An additional sawcut produced a transverse joint sealant reservoir 12.7 mm wide by 25.4 mm in depth. The transverse joints were sealed with low-modulus silicone.
The LTPP SPS project deviation report indicates that some of the DGAB cracked during compaction (sections 390259 and 390204). Contaminated PATB was removed and replaced due to an oil spill in section 390260.
Monolithic construction of base layers would have ensured that a layer of uniform thickness and material quality was constructed transversely across the typical pavement section. This would have resulted in the highest support conditions at the pavement edge, which is often the most critical stress area (edge stresses and positive curling stresses) for a doweled JPCP. Only the CTPB width can be considered monolithic.
Individual pavement layer thicknesses are often in excess of LTPP tolerances. Variability of a single layer depth occurs both within an individual test section and from section to section for those test sections that have common layer depth requirements. The constructed depth of the JPC may have the largest effect on pavement performance if subgrade strength and subsurface drainage quality are relatively uniform throughout all test sections. The allowable KESALs were developed from the AASHTO Supplemental Design Guide for Rigid Pavements and a computer program with environmental data from the nearest SHRP LTPP data site, design thickness, and strength values from SPS-2 protocols (average material values were used when these values are not defined in the SHRP SPS-2 protocol).
Sections 390259 and 390204: Some surface aggregate cracked due to compaction.
Section 390260: Oil spilled on PATB. Contaminated sections were removed and replaced.
All sections: Unbound aggregate base layers cut to grade using a CMI trimming machine.
Several test locations were moved due to obstructions. These are noted on testing log sheets.
Several other minor sampling deviations are noted on sampling data sheets.
The Washington SPS-2 project site is located in the northbound lanes of S.R. 395 in eastern Washington, 4.8 km south of Ritzville. S.R. 395 is an urban principal arterial with a 1993 AADT of 18,000. The designs ESALs for this project are 35 million for a 40-year design life. The SPS-2 project includes construction of two new northbound lanes and the upgrade of S.R. 395 to a four-lane divided highway. The new lanes were constructed uphill from the existing lanes. Two sections were located in a cut (section 530203 and 530259), while all other test sections were located on fills. The roadway design for this project consists of two 3.66-m-wide lanes, an outside shoulder width of 3.05 m, and an inside shoulder width of 1.22 m.
The initial SPS-2 test sections were constructed on a horizontal curve to the left from the beginning of the SPS-2 project to station 2050+00. Section 530201 is partially located within a horizontal curve and partially located within a superelevation runout area. Sections 530205, 530206, 530207, and 530208 are on tangent, while the remainder of the test sections were constructed on a curve to the left. The maximum superelevation rate for this curve is 3 percent. Vertical grades range from 0.14 percent to 3 percent.
This test site is in the dry-freeze zone. The average high temperature is 37 °C, while the average low temperature is -14.6 °C. It is not known if an onsite weather monitoring station was ever installed.
Construction of the test sections started in June 1993 with the removal of the existing pavement. Construction of individual test sections was completed by November 1, 1995.
All required core sections were constructed. The supplemental State test sections constructed included the following:
The existing subgrade is classified as fine-grained sandy silt. Borrow excavation for embankment construction came from three sources within the project area. All three sources were sandy silt of very low plasticity. Completed embankment depths ranged from 0.84 to 1.75 m in depth.
The existing ground from station 2004+00 to the end of the SPS-2 project was saturated and required removal before subgrade and embankment construction. Undercutting of the existing subgrade was performed from station 2004+00 to station 2085+50. After overexcavation was completed, the contractor refilled these areas with shot rock. This material was a volcanic stone with a 457-mm top size. It was placed to act as a drainage layer and to provide a stable platform for roadway construction. Borrow excavation consisted of silt and was constructed on top of the shot rock to the finished subgrade elevation. Compaction of the embankment was completed with a vibrating steel-wheeled roller and a 17-ton steel-wheeled roller. This layer was compacted at 100 percent compaction, but at a moisture level of 5.89 percent below optimum. The embankment was trimmed to final grade with a laser-controlled grader.
The DGAB consisted of a crushed stone with a 31.75-mm top size. A 7-ton vibrating steel-wheeled roller was used to compact the DGAB. Construction traffic was allowed on the completed DGAB, and no significant damage was noticed. DGAB layers, which would be overlaid with PATB, were prime-coated. Construction traffic was again allowed to run on the prime-coated DGAB. Minor tracking and bleeding of the prime coat occurred. Average compaction achieved on the DGAB layer was 97.7 percent of the optimum density.
The PATB contained an AR400W binder at 4.5 percent and an antistrip agent. An average of 87 percent of the crushed basalt utilized for the PATB passed the 0.5-inch sieve, while an average of 7.5 percent passed the No. 200 sieve. The PATB had a density of 2.56 g/cm3 (159.5 pcf) and 7.5 percent voids. This layer was placed at 77 to 82 °C with a Blaw-Knox PF150 paver. Profile grade control was maintained with a wire guide line and/or paving skis. The PATB was compacted at 66 °C by two passes each of a 17-ton steel-wheeled vibratory roller and a 10.5-ton static steel-wheeled roller. Compacted densities averaged 84 percent of maximum density (82 to 87 percent ranges). No appreciable construction traffic was allowed on the PATB.
The mix design for this material included 10.1 kg Type II cement, 1.13 kg Type F fly ash, 77.5 kg coarse aggregate (crushed basalt), 81.1 kg fine aggregate, 11.6 kg water, a water reducer, and an air-entraining agent. The water-cement ratio was 1.03. This mix design produced an average 7-day compressive strength of 4.1 MPa. Cored 14-day compressive strengths varied between 3.7 and 6.4 MPa, while companion cylinder breaks at 14 days varied between 2.0 and 6.4 MPa.
The LCB was paved with a Guntert ZimmermanTM slip-form paver with a hydraulic spreader and trowel. Laydown widths varied between 5.08 and 6.50 m. The LCB construction joint was located 1.62 m to the right of the PCC joint for both 4.27-m- and 3.66-m-wide lanes. Minor transverse and longitudinal cracking occurred, with most of the longitudinal cracking occurring in the outside shoulder.
The coarse aggregate for both SPS-2 mixes was a crushed basalt, and the fine aggregate consisted of 88 percent crushed basalt and 12 percent natural sand. All mix designs on this project utilized Type II cement. The 3.8 MPa mix included a Class F fly ash.
The PCC paving train consisted of a track-mounted Guntert Zimmerman slip-form paver, a track-mounted spreader with side-loading conveyor belts, a float finisher, tining machine, and a curing machine. Joint sawing was performed at least 12 hours after concrete placement.
JPC paving began on September 25, 1995, and was completed on October 3, 1995. The project was opened to traffic on November 1, 1995. Average daily temperatures ranged between 5.3 and 19.9 °C, and the relative humidity ranged from 30.8 to 95.9 percent. The maximum relative humidity exceeded 90 percent in the morning for each day of paving, but decreased to less than 42 percent by 5:00 p.m. for each day except September 28. The maximum relative humidity on this day stayed above 75 percent, while average temperatures ranged between 9.4 and 26. 2 °C.
The water-cement ratio averaged 0.455 for the 3.8 MPa mix and 0.286 for the 6.2 MPa mix.
The 3.8 MPa mix had an average 14-day flexural strength of 3.3 MPa (0.38 standard deviation) and a 28-day flexural strength of 4.3 MPa (0.52 standard deviation). The 6.2 MPa mix had an average 14-day flexural strength of 5.7 MPa (0.24 standard deviation) and a 28-day flexural strength of 6.5 MPa (0.59 standard deviation).
For each SPS mix design (3.8 MPa or 6.2 MPa), beam flexural strengths, cylinder and core compressive strength, and splitting tensile strengths showed consistency with each design strength.
Construction traffic helped to further consolidate the DGAB, as evidenced by an average density of 2,106 kg/m3 for those DGAB sections receiving construction traffic and an average density for the control section (section 530259, which did not receive construction traffic) of 1,867 kg/m3.
Six of the eight test sections constructed with DGAB had average thicknesses between 10 and 23 mm greater than SPS-2 specifications.
The average ATB thickness was 66 mm with a 10-mm standard deviation. The SPS-2 specified thickness was 76 mm, ±6.4 mm.
The average LCB thickness was either 155 or 157 mm for each test section paved with LCB. The SPS-2 specified thickness was 152 mm, ±6.4 mm.
The 203-mm PCC test sections had average thicknesses ranging from 211 to 206 mm. Test sections 530201, 530206, and 530209 had thicknesses of 221, 218, and 216 mm, respectively.
All 279-mm PCC test sections had PCC thicknesses within 7.6 mm of the specified depth.
The 14-day core compressive strengths for three of the four LCB test sections were within SHRP tolerances of 3.4 to 5.2 MPa. Section 530207 had compressive strengths up to 2.5 times as high as other LCB test sections. This was attributed to a water-cement ratio lower than the mix design.
All but one PATB test section had an average thickness of either 97 or 99 mm. Section 530212 had an average PATB thickness of 89 mm.
The 3.8 MPa mix had hairline cracks below the sawn transverse joint and 6.4-mm joint widths several days after paving for the DGAB and LCB sections. The 203-mm PCC on LCB (section 530205) had not cracked at the transverse joints by October 2, 1995.
The 6.2 MPa mix had larger cracked joint widths than the 3.8 MPa mix for corresponding sections.
The 3.8 MPa mix had cracked joints up to 7.9 mm in the PATB sections, while the 6.2 MPa mix had transverse joint crack widths averaging 13 mm on PATB sections.
Section 530206 developed shrinkage cracks from 1.6 to 3.2 mm in width. All but 1 slab was cracked, and 19 of the 32 slabs had more than 5 cracks per slab.
Transverse and longitudinal joints were sealed with a hot- poured material.
FWD testing revealed that those sections constructed in cut areas had the most variability in support (0.4 to 1.4 mm), while those test sections constructed on embankments had more uniform support.
The Wisconsin SPS-2 project site is located on the westbound and eastbound Wisconsin State Highway 29 (S.H. 29) in Marathon County, WI. This site is roughly 5.6 km east of Hatley, WI. The site is located on a 0.3 percent downgrade with four curves in between. The maximum curve does not exceed 2 degrees with a superelevation equal to 0.055 l/l. The lanes are 3.66 m and 4.27 m wide, with an outside shoulder of 3.05 m and an inside shoulder of 1.83 m.
The project is located on a four-lane section of S.H. 29, which is classified as a rural arterial. As of 1995, the current average daily traffic was 6,650 vehicles with a truck distribution of 29.5 percent. This site had a design life of 20 years.
This SPS-2 project was planned for the wet-freeze environmental zone and on a coarse-grained subgrade. An AWS unit was installed in June 1997. A WIM system was installed on August 29, 1997. The WIM equipment used was a DAW-1000 bending plate unit.
The subgrade preparation for this project began in early June 1997, and paving operations were completed by mid-October 1997.
All required core sections were constructed. This SPS-2 project also incorporates eight Wisconsin DOT supplemental sections. Their designs are described below:
Scrapers, bulldozers, and pushcarts were used to compact the subgrade. The lift thickness was typically 203 mm. Remnants of old PCC pavement were found in the subgrade when sampling using shelby tubes. The PCC slabs were removed, and the subgrade was reworked to bring it back to the required elevation.
The DGAB and CSOGB (cement stabilized open graded base) thicknesses were 102 and 152 mm (4 and 6 in), respectively. Compaction was achieved using scrapers, bulldozers, and pushcarts. Typically, a 203-mm lift thickness was used for 152-mm layers, and a 152-mm lift was used for 102-mm layers. This procedure frequently resulted in a layer that was too thick. Therefore, a CMI trimming machine was used to achieve the proper layer thickness.
All PATB base layers were 102 mm thick. A Rex R28 was used for paving. The paver had a single- pass laydown width of 3.66 m and, typically, a first- lift placement thickness of 127 to 152 mm. The asphalt was obtained from a local plant located 3.2 km from the test site with a hauling time of 10 minutes.
The LCB were paved with a Rex R28 slip-form paver. This machine has a 7.3-m-wide laydown width. The concrete was obtained from a local concrete plant located 1.61 km from the test sections. Vibrating screeds were used to consolidate materials. Finishing was done by screeding, and a membrane-curing compound was placed on the LCB.
Two different mix designs were used in this SPS-2 project. All used a La FargeTM Type II cement. The coarse aggregate was made of 100 percent crushed aggregate, and the fine aggregate was composed of 100 percent manufactured sand.
A Rex R28 slip-form paver paved the PCC layer. The width paved in one pass varied from 6.1 to 7.9 m. The cement mixture was consolidated using internal vibrators. Vibrators were placed 152 mm below the surface approximately 610 mm apart. Finishing was done by screeding, and a membrane-curing compound was used. The surface was textured using a tine.
All sections: Unbound aggregate base layers were cut to grade using a CMI trimming machine.
During the splitspoon testing, a number of areas were found to have existing concrete slabs beneath the old pavement structure. These areas of concrete were removed and fill was placed in these areas.
Because of the process used to remove the existing pavement, it was not possible to obtain undisturbed samples of the existing base or subbase material.
Soil boring records were provided that made it unnecessary to perform shoulder probes. The depth to rigid layer exceeded 6.1 m.
Several other minor sampling deviations are noted on sampling data sheets.
Topics: research, infrastructure, pavements and materials
Keywords: research, infrastructure, pavements and materials,, Design factors, experimental design, JPCP, LTPP, performance trends, SPS-2.
TRT Terms: research, facilities, transportation, highway facilities, roads, parts of roads, pavements, Concrete--Joints--Performance--Evaluation, Pavements, Concrete--Joints--Design and construction--Evaluation, Pavement design, Pavement performance, Traffic loads, Experimental roads