Pile testing - Piling Engineering - page 331

capacities of pile sections above and below the jack. The jack is positioned at a level
where the shaft capacity of the pile above the jacks is approximately equal to the
combined shaft and base capacity below the jacks.
The expansion of the jack is measured through the volume of oil pumped and also
by independent transducers that span across the jack. Tell-tale rods built into the pile
allow determination of the upward movement of the top of the jack, and hence also
the downward movement of the bottom. A more sophisticated arrangement of jacks
is possible, for example, by installing two levels, which allows an intermediate length
of pile shaft to be failed independently, thus giving a direct measurement of shaft
friction. An application using such a system of two levels of jacks has been described
by Randolph (2003).
For a single level of jacks, as shown in Figure 9.23, either the upper (shaft) or lower
(base) section of the pile will reach failure first, and typical results may be as indicated
in Figure 9.26. In the standard method for interpreting an Osterberg cell pile test
(assuming a single level of jacks), the load-displacement plots of the separate upper
and lower sections of the pile are first plotted. Ignoring any compression of the pile,
the total load-displacement response at the pile head is then estimated by summing
the separate loads mobilized in each section of the pile at equal displacements. This is
illustrated above for typical points A (4mm displacement) and B (7mm displacement),
where the total pile resistance is summed and plotted at the give displacement to give
the combined (rigid pile) curve.
An improved method makes approximate allowance for the compressibility of the
upper section of the pile. For each displacement, the average load that would be mobi-
lized in the upper section is first estimated (the simplest approach being to take the
load from the lower section plus half the load from the upper section), and hence the
compression of the upper section of the pile is estimated. This compression is added
to the initial displacement for which the loads were evaluated. Considering point A,
at a displacement of 4mm, the upper and lower section loads are 19.5 and 23MN,
respectively. The average load that would then occur in the upper section for those
conditions is therefore:

P average ∼ 0 . 5( P lower + ( P upper + P lower ) = P lower + 0 . 5 P upper = 32 . 75MN

The estimated compression of the upper section is then w = P average L / EA =

18 147MN). The true pile head displacement is there-

fore the sum of the original displacement, 4mm, and the pile compression, 18.9mm,
giving a total of 22.9mm. The corrected curve is shown for comparison with the
uncorrected (rigid pile) case. The limitation of this approach is that the actual displace-
ments in the upper section of the pile are much larger than the original displacement,
and thus the load mobilized will be higher. A better approach is to undertake a full
load transfer model of each section separately, and then analyze the complete pile
subsequently.
While the Osterberg cell method of pile testing has some limitations, in particular
due to the different direction of loading of the upper part of the pile, and problems
associated with creating a gap in the vicinity of the loading jacks, the system has the
potential for testing cast in situ piles of extremely high capacity.

. 9mm (taking L = 85m, EA =