I-90 Demonstration Project: Long-Term Performance of Tiebacks Installed in Glacially Overridden Clays

This report provides the results of a Federal Highway Administration demonstration program conducted in the 1980’s to investigate the long-term performance of tiebacks installed in heavily overconsolidated glacial clays.  The primary focus of this study was to help establish the relationship between long-term tieback creep performance and short-term creep behavior.  To accomplish this, a long-term test site was established on the northeast side of Beacon Hill above I-90 just east of the I-5/I-90 interchange. There, nine low pressure grouted, straight shafted tiebacks were installed to conduct two pullout tests and conduct long-term monitoring (up to 4.3 years) of seven tiebacks locked off at various percentages of the ultimate tieback capacity. Six additional tieback pullout tests for a subsequent construction project were also evaluated, extending the study to larger diameter and longer tieback bond lengths as well as to disturbed glacial clay (paleolandslide) deposits.
The results of this study indicate that the “standard” creep rate criterion used in the USA of 2 mm/log cycle (0.08 inches/log cycle) of time is too high for the clays evaluated, and that a creep rate of 1.0 to 1.5 mm/log cycle (0.04 to 0.06 inches/log cycle) of time should be used instead.  For tieback bond zone sizing and installation procedure verification purposes, pullout tests should be conducted to verify the assumed bond zone design adhesion used.  The typical incremental loading procedure for pullout testing used in which each load increment is held for one hour to measure the creep rate can be used to confirm the relationship between load level and creep rate, but recognizing that the long-term creep rate for the tiebacks is likely to be twice as high as the one hour creep rate; hence the limitation of 1 mm/log cycle (0.04 inches/log cycle) of time if only one hour creep tests are conducted.  The long-term test program also demonstrated that tieback stressing history can significantly affect the creep rate.  Complicated tieback loading sequences should be avoided when assessing the creep rate is an objective, and retesting of tiebacks that failed the creep test criterion should be prohibited, as either of these loading scenarios can make the tieback creep behavior look much better than it really is.


Publication Date: 
Monday, February 24, 2020
Publication Number: 
WA-RD 823.1
Last modified: 
05/27/2020 - 07:33
Tony M. Allen
Washington State Department of Transportation. State Materials Laboratory
Number of Pages: 
Tiebacks, Anchors (Structural connectors), Creep tests, Creep properties, Glacial soils, Clay soils, Performance tests, Monitoring, Demonstration projects.