Direct answer

Record the height of a defined interface or level, h(t), during a test with documented conditions. Over an interval where the curve is approximately linear, the downward velocity can be estimated as v_s = (h_1 - h_2) / (t_2 - t_1). Report the interval, units and test conditions.

That value describes the movement of the observed interface for the sample and protocol used. It is not automatically the terminal velocity of one particle, does not summarise the entire settling curve and does not by itself predict the performance of an industrial thickener.

What result is being measured

Before calculating a slope, define the measurand. In a suspension that forms a visible interface, the boundary between the clarified zone and the suspension can be tracked. If several layers, distributed turbidity or a diffuse interface appear, the reading rule must be fixed before tests are compared.

Settling may pass through more than one regime. An initial mixing or floc-formation stage may precede a zone with approximately uniform descent. Later, the curve may change as particle interactions increase and compression begins. A single slope does not necessarily represent all of those stages.

The minimum result should include the time-height series, not just one final number. Retaining the curve makes it possible to review the fitting interval, identify slope changes and compare the complete behaviour of samples.

Minimum protocol for a comparable curve

  • identify the sampling point, origin, date and time;
  • document solids concentration and the method used to determine it;
  • record temperature and, when relevant, water chemistry or provenance;
  • describe vessel geometry, useful diameter and initial height;
  • document flocculant preparation, concentration, maturation and dosage;
  • fix the mixing sequence, energy and duration;
  • define t = 0 and the visual or instrumental interface criterion;
  • record heights frequently enough to resolve the expected change;
  • repeat the test under equivalent conditions and retain every replicate;
  • record deviations, bubbles, wall adhesion or loss of visibility.

Comparability is lost if the material, dilution, polymer preparation, mixing or reading method changes at the same time. The observed difference cannot then be attributed to one cause.

How to calculate and report velocity

First plot h(t) and identify the interval corresponding to the regime of interest. If that segment is approximately linear, estimate its slope with a linear fit and report it, for example, in millimetres per minute or metres per hour. Define the sign convention; reporting the magnitude of the downward velocity usually avoids ambiguity.

Do not automatically select the first two points. Mixing, flocculant reaction time or reading stabilisation can dominate the beginning. Do not extend the fit into a compression zone when the objective is to characterise interface descent before that transition.

Report the time and height range, number of points, exclusion rule and a measure of fit or dispersion between replicates. If the curve has no sufficiently defined segment, the correct result may be to state that a unique velocity was not obtained under that criterion.

Quality control and uncertainty

Repeatability evaluates dispersion under specified, similar conditions. It does not by itself demonstrate that the method is accurate or that the sample represents plant feed. At least three sources of variation should be separated: sampling and preparation, the measurement system and real variability of the material.

The measurement system can be checked using video or images with a known scale, a height reference, synchronised clocks and explicit rules for locating the interface. Automated readings also require evaluation of resolution, bias, blind zones and performance when contrast changes.

Uncertainty should not be hidden behind decimal places. If variation between replicates is larger than the difference between two conditions, the available evidence does not support the claim that one settles faster.

Interpretation and limits

A higher velocity can be useful for comparing test conditions, but it does not automatically mean better clarity, greater compaction, lower flocculant consumption or greater water recovery. Those responses require additional variables and a defined process objective.

Kynch's theory formalises hindered settling through flux-concentration relationships. Settling-compression models add effects that become relevant at higher concentrations. These references explain why an interface slope has a physical context and must not be treated as a universal property of the tailings.

This page does not prescribe a dosage, sampling frequency or plant acceptance criterion. Transfer to operation requires comparing the protocol with actual conditions, delays and thickener variables.

Sources and evidence status