THE DEVELOPMENT OF A METHOD SPECIFICATION FOR COMPACTION OF CDSM SPOIL AS FILL AT SAN PABLO DAM

摘要

San Pablo Dam is a 170-foot-high, 1,200-foot-long earthfill embankment located in El Sobrante, approximately 10 miles from downtown Oakland, California, and it impounds a 38,600 acre-feet reservoir for customers in the East Bay Municipal Utility District's (EBMUD's) service area. As a result of updated studies, a seismic upgrade was designed and constructed in 2008 through 2010. Work included the removal of an existing 140,000 cubic-yard downstream buttress, which was replaced with an enlarged 300,000 cubic-yard buttress founded on alluvium and colluvium that were improved in-place by Cement Deep Soil Mixing (CDSM) technology. The CDSM generated spoil, which was comprised of soil-cement slurry that hardened into soil-cement. The CDSM process generated 160,000 cubic yards of spoil, which was processed on-site and reused as compacted fill for the lower portion of the new buttress, which, by design, will remain unsaturated. Laboratory shear strength tests showed that processed CDSM spoil had excellent engineering properties for compacted unsaturated fill. However, results from traditional laboratory compaction testing methods were found to be unrepresentative of the densities that could be achieved under field conditions. When compaction tests were completed in the lab, the material degraded and deformed in the compaction testing mold, which made the laboratory dry unit weight unrealistically high. The initial quality control tests showed that the fill materials failed to meet compaction requirements despite clear indications that compaction was being achieved, e.g., walk-out of the rolling tamping compaction equipment, difficulty in penetrating the surface with a geotechnical probe, and difficulty in excavating pits for sand cone testing. Also, the material's low compacted dry unit weight (approximately 65 to 80 pounds per cubic foot (pcf)) and high water content (approximately 30 to 45 percent) was outside the normal range for compacted materials. EBMUD had specified traditional laboratory compaction testing for earthwork quality control; ultimately, however, a method specification was found necessary. This paper describes the various tests that were performed in an effort to find methods that would be appropriate for CDSM spoil. The final method specification required additional field inspection and testing but it was necessary and appropriate for use with this material.