This research involved field investigations and laboratory testing to evaluate the properties and overall performance of geotextile separators exhumed from the roadway at eight sites in eastern and central Washington (Phase I), and 14 sites in western Washington (Phase II). Both nonwoven and woven geotextile separators of different in-service ages were examined in detail, and specimens were tested in the laboratory for strength and hydraulic characteristics. The subgrade condition and geotechnical properties of the base course aggregate and subgrade soils were also evaluated.
Although all of the geotextile separators performed their intended separation function adequately, the geotextiles experienced very different levels of damage during construction. Base aggregate type, rather than initial aggregate lift thickness, appeared to have the most influence on the level of damage. All of the recovered geotextiles installed under an angular base aggregate sustained some damage, while geotextiles installed under sub-rounded to rounded aggregate experienced minor damage, if any. The woven slit-films and needle-punched nonwoven geotextiles experienced similar reductions in strength, and both survived the installation conditions reasonably well (except for one lightweight, needle-punched nonwoven, which was over stressed during installation and which may have been installed under an excessively thin pavement section). Although the heat-bonded nonwovens were heavily damaged during installation, they were installed under some of the more severe site survivability conditions.
Test results indicated that the permittivity of the woven slit-films and the needle-punched nonwovens both increased by similar percentages after being washed. The heat-bonded nonwovens had the highest percentage increases in permittivity after washing; this finding suggests that they clog more than other geotextiles. There was evidence that the woven slit-films experienced much more binding than did the other geotextiles, and that iron staining and caking may also have affected their drainage performance adversely. Most woven slit-film geotextiles did not meet the filtration requirements set forth by Task Force 25 (1) and Christopher and Holtz (2) when they were placed on fine-grained subgrade soils.
The unwashed (i.e. "undisturbed") permittivity results also indicated that most woven slit-film geotextile permeabilities fell well below the Washington State Department of Transportation (WSDOT) required value. The presence of caked fines on the upper surface of the three woven slit-films could indicate that their port openings were too large for the intended filtration function, and that they might be subject to fines migration. However, the evidence on this point was inconclusive. There was no other evidence of fines migration at any of the sites.
All of the pavements examined were in good condition, and damage to the geotextile separators did not appear to have had any negative impact on the pavements' long-term performance. Although one pavement surface showed signs of premature failure, this could not be attributed to the performance of the geotextile separator.