technology and services. These were lofty goals for a company with a staff of two and a one-room office. The name of the new company did not come easily. After much brainstorming, it boiled down to two: ShaftTest Inc. and Loadtest Inc. The stalemate ended when the author suggested that most people would rather get “loaded” than “shafted.” The next three years became the period of “preaching in the wilderness.” Although generally supportive of innovations, the engineering profession has not always embraced them easily. The fear of creative disruption of acquired “standard of practice” often makes engineers reluctant to try new technologies, especially if perceived as costly. Indeed, the cost of O- cell testing was not competitive with conventional tests until test loads exceeded 1,000 tons (8.8 MN). At this time, agencies in the U.S. did not require or specify load tests exceeding 1,000 tons (8.8 MN), the practical limit for conventional top down tests. A market survey had suggested a potential annual demand of about ten 1,000 ton (8.8 MN) load tests. In its second year of operation, Loadtest carried out 8 O-cell tests, most well in excess of 1,000 tons (8.8 MN). Things literally got shaken up, however, at the L.A. Coliseum during the 1994 Northridge earthquake. The foundation retrofit after this quake ended up requiring twenty- eight 2,000 ton (17.7 MN) O-cell tests. Since we tested every drilled shaft on the project, it became a landmark in the foundation industry: the first example of a drilled shaft foundation designed with a 0.9 resistance factor. This project marked a turning point in the quest to create a market for the O-cell test. No more wandering in the wilderness. Dramatic Improvements The very busy years after 1995 resulted in dramatic improvements in the O-cell test process. Displacement transducers replaced dial gages; every shaft had vibrating wire strain devices attached to the re-bar cage; manual controls gave way to automated systems run by software. This led to an observation by Dr. Bengt Fellenius that Loadtest carried out O-cell tests to a “research level quality.” This outcome was 46 • DEEP FOUNDATIONS • NOV/DEC 2012 The Osterberg Cell: A Truly Great Contribution I first heard about the O-cell test in the mid-1980s when Jorj Osterberg told me about his innovation, which consisted of placing a sacrificial jack at the toe of a pile and letting it simultaneously push upward and downward to a load equal to the intended working load on the pile head. He thus obtained a factor of safety of 2.0 in the test, eliminating the need to supply and build-up reaction weight for the test. I soon realized that Jorj’s innovation is one of the truly great contributions to geotechnical engineering. Indeed, working with O-cell results has significantly developed the understanding and knowledge about the static loading test and the response of piles to load from a structure—not just for me, but also for many others in the field. Here are just a few of the key lessons we have learned. 1. Any before-the-test locked-in load (residual load) is measured, eliminating the need for its “guesstimation.” 2. The routinely incorporated strain-gage instrumentation has shown many previously held beliefs about pile-soil response to load to be incorrect. One of them being that a pile toe would have an ultimate resistance. The wealth of O- cell tests performed have shown that a pile toe responds to increasing load by a more or less smooth load-movement curve that does not trend to an ultimate value. 3. For bored piles and drilled shafts, the test provides the ability to prove whether or not clean-up and removal of debris from the bottom of the shaft has been successful. 4. It is entirely feasible to use the O-cell as a construction device to prestress the pile toe and reduce the settlement for the foundation placed on that pile—and reduce construction time and foundation costs. 5. While the test can be carried out per any desired schedule of testing, it has been very instrumental in teaching the geotech community that the best test data for analysis are obtained by a test incorporating many short increments applied at same-length time intervals and excluding unloading/reloading ‘cycles’ during the test. 6. If the test would not engage the shaft fully, but only the pile toe, then, it is simple and economical to arrange for a supplemental head-down test to push the shaft with the O-cell open. This will ensure that shaft and toe responses are not just separately engaged, and that both responses are determined at large movements. 7. The positive measurements of the pile toe response enable designers to determine the long-term foundation settlement due to the pile working load and to the influence of the settlement of the surrounding ground. Therefore, the test is exceptionally useful for assessing the long-term settlement of the piled foundation represented by the test pile. The O-cell test is now an established tool for the geotechnical engineering industry, and major projects around the world have applied it for the design of large and small foundations. There is rarely a geotechnical conference that does not include at least one case history paper describing results and lessons learned from O-cell tests. Bengt H. Fellenius, Consulting Engineer
