16-Civ-B3 Geotechnical Design · May 2013
Nivaar worked solution (AI-drafted; not reviewed by a licensed engineer)
Paper format. National Examinations, May 2013 — 98-Civ-B3 Geotechnical Design. Three hours, open book, any non-communicating calculator. Section A holds five 7-mark questions (answer any four); Section B holds the long 24-mark design questions (answer any three). Candidates are asked to identify the source of every design chart and assumed value used. Every question is answered here, because the set is a study resource rather than a sitting.
Reference texts. B. M. Das, Principles of Foundation Engineering (9th ed.) and Principles of Geotechnical Engineering (9th ed.); Canadian Geotechnical Society, Canadian Foundation Engineering Manual (CFEM, 4th ed.) — the governing Canadian reference for foundation practice; D. P. Coduto, Foundation Design: Principles and Practices; R. F. Craig, Craig's Soil Mechanics.
Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.
Why load tests are used with the empirical methods. The three methods are regressions on databases of instrumented piles, and the scatter in those databases is large: predicted-to-measured capacity ratios between about 0.5 and 2.0 are routine. That scatter has four sources, and only a load test closes them. First, soil variability — the design profile comes from a few boreholes, while the pile samples every metre of soil it passes through. Second, installation effects — driving remoulds and then reconsolidates the clay around the shaft, generating set-up that can double capacity over weeks, while boring can soften the shaft wall and reduce it; no empirical method captures the contractor's actual technique. Third, workmanship — a load test is the only practical proof that the installed element is sound. Fourth, economics — codes and the CFEM permit a lower resistance factor (a higher allowable load) when static load tests are performed, so testing one pile in ten routinely pays for itself in shortened piles across the rest of the site.
A load test also gives the load–settlement curve, which no static formula does. Since serviceability, not collapse, usually governs a pile group, that curve is often the more valuable output. Dynamic testing with signal matching and static analysis of restrikes extend the same benefit to a larger sample of the piles at lower cost.
Which method for long-term capacity in clay. The $\beta$ (effective-stress) method is the reliable one. Long-term behaviour is by definition drained: the excess pore pressures generated by installation have dissipated, and shear resistance on the shaft is governed by the effective normal stress acting on it, so $f = \beta\,\sigma'_v$ with $\beta = K\tan\delta'$. The parameters are drained ones that do not decay with time.
The $\alpha$ method ties unit friction to the undrained strength, $f = \alpha c_u$, which describes the short-term, total-stress condition immediately after installation and says nothing about the drained state. The $\lambda$ method of Vijayvergiya and Focht is a hybrid, $f_{av} = \lambda(\bar{\sigma}'_v + 2\bar{c}_u)$, calibrated on long driven steel pipe piles in normally consolidated offshore clays; outside that population, and in the long term, it is the least transferable of the three.
