Geoscience Reference
In-Depth Information
5.1.3
Relative Quality of Each Drill Hole
or Sample Type
It may not be appropriate to use samples from a percussion
rig and from a diamond drill hole simultaneously to obtain
a resource estimate. One set of samples may be biased with
respect to the other. When more than one sample type is
available, it is necessary to make comparisons of each set of
samples and their statistical properties. Ideally, it is better to
compare sets of twins or duplicate samples, but they are not
always available. It is common that channel or chip samples
will also show significant differences with nearby drill hole
data. Data from biased drill hole or channel samples should
be discarded or used cautiously solely for the construction of
a geological model. In some cases secondary poorer quality
data could be used in some form of cokriging (Journel and
Huijbregts
1978
).
Samples from percussion drilling commonly suffer from
significant loss of material and little control during the drill-
ing operation; high or low grades may be preferentially lost.
Also, significant mixing of material occurs as the samples
come up the hole; thus the exact location of the sample is
uncertain. In most cases, data from percussion drilling is not
acceptable for resource estimation.
Reverse circulation drilling is cheaper than diamond
drilling, and so, for a given budget, may provide more in-
formation. If done carefully and under good sampling con-
ditions, it can provide good samples. Often, reverse circula-
tion drill holes are of larger diameter than common diamond
drill holes. It may be difficult to obtain good geological de-
scriptions, since the material is recovered as broken rock
chips.
Diamond drilling is more expensive, although if core re-
covery is good, it has the advantage of bringing intact rock
to the surface. This allows for better geologic mapping, and,
after splitting the core in half, provides a representative sam-
ple for preparation and assaying. The down-the-hole loca-
tion is better known than other types of drilling. Diamond
drilling is generally considered to provide the best sample
quality. Figure
5.2
shows a partial view of the very large core
farm for BHP-Billiton's Olympic Dam multimetal deposit in
South Australia.
Fig. 5.2
Partial view of the core farm for Olympic Dam's mine and
expansion project, courtesy of BHP Billiton, Roxby Downs, South
Australia
Reverse circulation drilling and sampling could be par-
ticularly difficult below the water table or in the presence of
significant amounts of water. Down-the-hole contamination,
washing, and cave-ins are concerns, as well as loss of min-
eralization in the slimes produced. In these situations, and to
avoid losing fines, the output water from the hole can be re-
directed to a large decantation barrel before final discharge.
The fine material that decants in the barrel can be collected
and analyzed, providing an indication of whether the loss of
fines results in a mineral grade bias. In practice, the amount
of material that can be decanted is limited and it is difficult
to assign an exact down-the-hole location to the analyzed
fines.
Diamond drill holes can also have problems with ex-
cess water during drilling. For example when the mineral is
lodged in veinlets that can be washed away before the core
is recovered. A multiple-tubing system is sometimes used to
achieve better core recoveries in weak, fractured rock. Core
cutting is sometimes a source of concern, particularly when
a diamond-blade saw cutter is used in the presence of schis-
tose or friable material; a hydraulic press may be preferable
in these cases.
5.1.5
Core and Weight Sample Recoveries
5.1.4
Sampling Conditions
For drill types where the samples are taken from drill cut-
tings, it is useful to record the total weight of the material
extracted from the hole for the given sampling interval. This
total weight should be compared to the theoretical weight of
the sample:
The quality of samples also depends on the material being
sampled and the conditions under which the samples are
taken. For example, the presence of underground water or
very fractured rock requires careful and sometimes much
slower and expensive sampling methods to minimize pos-
sible biases.
Sample Weight =
π
∗∗ ∗
d
2
length
δ
