Particle size
distribution (PSD) is one of the most important fundamental physical properties
of soils, as it determines their physical, chemical, mechanical, geotechnical,
moreover environmental behaviour. Although the measurement of PSD with
different techniques is commonly performed in soil laboratories, their
automation and continuous PSD curve generation have not been solved yet. However, there are some physical principles,
various sensors and different data storing methods for measuring the
density-time function. In the present paper a possible solution is introduced
for the measurement of the soil particle density database as a function of
settling time. The equipment used for this purpose is an areometer that is
widely used e.g. for determining the sugar content of must, or the alcohol
content of distilled spirits, etc. The device is equipped with patent pending
capacitive sensors on the neck of the areometer. It measures the changes in the
water levels nearby the neck of the areometer in 1 μm units with <10
μm accuracy. The typical water level changes are 3-5 cm, which makes
possible a very accurate determination of particle density changes due to
settling in particle size analysis. The measured signals are stored in the
equipment's memory and can be downloaded to the controller computer via a
modified USB port. Data evaluation can be carried out online or later. The
large number of measured data points led to the introduction of a new
evaluation method, the Method of FInite Tangents or shortly the “FIT Method”.
The dispersed soil particle system is considered as the aggregation of many
mono-disperse systems. From this it follows that the measured density-time
function can be divided into grain size fractions with tangent lines drawn to
finite, but optional points. These tangent lines are suitable for calculating
the settling speed of a given fraction, as the changing speed of density is
equal to the multiplication of settling speed and mass of the given grain size
fraction. The settling speed of all fractions is calculable by using the Stokes
law, so the mass of all of the floating fraction can be calculated. Because the
soil suspension is a poly-disperse system, the measured density decrease can be
considered as an integration of finite mono-disperse systems. From this, it
follows that it can be interpreted as the sum of linear density vs. time
functions. If the mass of each grain size fraction is known, the particle size
distribution is calculable. The method is relatively easily programmed
and the intervals of grain size fractions are freely adjustable, so with this
program almost all types of particle size distribution are calculable, not only
those being uniform. Using the appropriate controller and evaluation program,
soil particle size distribution can be calculated immediately after downloading
the measured data. This technique does not need more sample preparation than
past methods. The automated reading lessens the manpower required for
performing the measurement - which also reduces human error sources - and
provides very detailed PSD data that has advantages, among others, like
revealing multi-modality in the particle-size distribution.