In 1995, the Florida DOT (FDOT) began to place the Superpave mix on one of its busy interstate projects, I-75, in Columbia County. Because FDOT did not have any experience in Superpave asphalt pavement, they relied heavily on what they read, counsel from those who had developed the theories and their native intelligence. With these as tools, they would attempt to produce a suitable product in a whole new arena of road construction technology.
As with the application of any new multi-step process, problems can and will arise and this project was no exception. The regrettable factor was that the problems were not realized at the project’s inception.
Looking beyond the surface
From the beginning of the paving operation, FDOT measured the density with a nuclear density gauge in the backscatter mode and it was their mistaken idea that the pavement was compacted to its required density.
It was not until later that the District Materials Office, Bituminous section, began a periodic check of the in-place air voids on cores taken from the pavement resulting in the painful realization that the air voids were between 10 and 13% Because density is directly related to the air voids, it also told them the density was actually between 87 and 900of the laboratory theoretical maximum density, (Dmax) while the plans required the pavement be compacted to at least 949of Dmax.
Even though the nuclear density gauge readings had indicated the pavement was meeting the required density, in fact, density was failing according to the core data. While the nuclear density gauge has been relied on for years to determine density on Marshall mix asphalt, all indications were showing the gauge is an unreliable means of testing density on the Superpave mix.
At this point, FDOT knew that low density would cause a problem sooner or later, but nothing they had read or heard would prepare them for what was about to occur.
By late 1996, after an extended period of rain, water started “spurting” out of the pavement at the joint of the roadway and the paved shoulder on the low side of the pavement and water continued to weep for several days. A study of how the project was built explained what had happened.
The mainline roadway had been milled 5 1/2 in., while the very dense Marshall mix on the shoulders had been left in place. The top of the milled surface was then covered with an asphalt-rubber membrane interlayer (ARMI) which serves as a crack relief layer. This formed a sort of sloped box which was then refilled with a very porous, permeable and under-compacted Superpave mixture. The result was, water penetrated the roadway, but could not go through the rubberized membrane below nor the impermeable shoulder on either side. When there was enough rain to fill the porous roadway, the water had nowhere to go except up between the joint of the roadway and shoulder.
Searching for an answer
One thing was certain, something had to be done quickly. This was a very unstable condition and the possibility of stripping was a major concern.
Two conditions are necessary for stripping to occur: 1) a fluid medium such as water must be present in the near proximity of the asphalt-covered aggregates and 2) relatively high pressure must act upon the fluid.
For example, when high-pressure truck tires impose an impact loading on an element of pavement at about 70 mph, the fluid medium immediately below the surface of the pavement will experience a pressure build-up from 0 psi to more than 100 psi in about 0.015 seconds. This would have the effect of water blasting the asphalt-coated aggregates which would cause the liquid asphalt to strip away from the aggregates. Once the situation was assessed, FDOT realized they needed a solution before stripping began.
A great notion
A District II bituminous engineer, Gordon Eugene Pettyjohn, had an idea and decided to try an experiment to see if it would work. He went to another interstate project on I-10, which was under construction at the time. With the use of a small, portable masonry dry-blade saw, he made a cut in the roadway from the edge of the Superpave out into the shoulder at five different locations. He believed that the cut might be adequate to let some of the water escape. Since the roadway is already sloped, nature provided the driving head needed to push the water out. The equipment Pettyjohn used was inadequate for high work volume, however, the idea was successful. The process or technique, which has come to be known as “drainage seams” immediately produced a steady discharge of trapped water.
Seeing this would work, Pettyjohn coordinated with the contractor and drainage seams were placed in April 1997. An aluminum template was made for a router and saw to roll on to maintain the proper depth and length of the seam from the roadway edge out into the shoulder.
A router was used first because it was less costly to set the initial cut, then a saw cut the seam to 3/4 in. wide. The seams were placed at 15-ft intervals on I-75 and I-10 both of which tested less than the required density. The seam was cut 3 to 4 in. deep, beginning 2 in. inside the edge of the roadway and extended 8 ft into the shoulder where the depth tapered to zero.
This technique allowed usage of the original Superpave when they feared it would need to be milled and replaced. The drainage seams provided an outlet for the water to escape thereby preventing stripping.
Finally, time has given the pavement a chance to densify under the very high average daily truck traffic. Water no longer “spurts” out of the pavement at the joint and there is less water escaping from the drainage seams following rains, indicating the pavement is becoming more dense.
FDOT attributes that phenomena to the heavy weight and high volume of traffic combined with the high temperatures in the Sunshine State.
Since those early projects, the contractors have had time to fine-tune their techniques and they are able to achieve the required density. The Superpave is less porous, therefore, the water is no longer a problem.
Superpave is being included in the plans of numerous upcoming projects. FDOT has put its faith and confidence in Superpave and with the initial problems resolved, they are expecting Superpave asphaltic pavement will greatly improve and/or reduce Florida’s rutting problems. -