Environmental Engineering Reference
In-Depth Information
5.10
SAMPLING
5.10.1
Soil samples
Soil samples recovered from the subsurface site investigation are described as either dis-
turbed or undisturbed.
Disturbed samples are collected from pits, trenches and from auger flights as represen-
tative of different material types or units. They are identified by location and depth and
stored in sealed containers (usually plastic bags) for laboratory testing. For cohesive soils
about 3 kg of sample is required for classification testing and 30 kg for compaction tests
to evaluate probable performance as engineered fill in an embankment.
Undisturbed samples consist of material which is extracted from the site and trans-
ported to the laboratory with a minimum of disturbance. The ideal sample is a cube of
approximately 0.3 m sides, hand cut from a test pit, carefully packed and sealed on site
and transported to the laboratory without delay.
It is more usual, due to economic factors, and the limitations of depth of test pits, to
use thin-walled steel tubes (“Shelby” tubes) to obtain samples of cohesive soils from
boreholes. The tubes are pushed into the soil using an adaptor connected to the drill rods.
Care should be taken to ensure that the drill hole is cleaned out before sampling and that
the drilling water or mud level is maintained during sampling to avoid “blowing” of
material into the hole. Choice of sample diameter depends on the hole size but, in general,
the larger the sample diameter the less the disturbance. Common thin walled tube sizes are
50 mm, 63 mm and 75 mm. The wall thickness and cutting edge shape are defined by
codes to limit sample disturbance. Samples should be identified and sealed against mois-
ture loss using either a sample tube sealing device or several layers of molten wax as soon
as the sample is recovered from the hole. Before sealing, loose/disturbed soil at the top of
the tube should be removed, so moisture does not migrate from that area to the less dis-
turbed part of the sample. Undisturbed samples should preferably be tested within two
weeks of sampling as they rust into the tube, or despite all efforts, dry out. On extrusion
in the laboratory a proportion of “undisturbed” samples often prove to have been partly
disturbed by the sampling process and it is prudent to take enough samples to allow
for this.
In any case, even apparently undisturbed samples are affected by stress changes during
sampling, and this needs to be recognised when analysing the results.
5.10.2
Rock samples
Samples of rock exposed in the sides/floor of pits/trenches can be taken for testing of sub-
stance strength and mineralogy. Samples should be individually numbered and located.
Storage in plastic bags prevents loss of field moisture.
The most common form of rock sample is core. A drill hole with full core recovery
should present a complete linear profile through the rock mass below the ground surface.
It is important that the core is systematically stored, properly logged, photographed and
sampled as soon as practicable after drilling.
In most cases the core needs to be kept for the following purposes:
-To enable the site investigator to make an accurate, clear and concise log of those char-
acteristics of the core which are significant to the project and to use this log in the com-
pilation of the geotechnical model;
-To provide samples for testing;
-
For inspection by designers and those preparing bids to build the dam;
