Real-time density monitoring can show changes that intermittent samples may miss, but more data does not automatically create better control. The signal must represent the correct stream, respond to the operating change of interest and be connected to a defined action.
The practical question is not simply “can density be measured?” It is “which decision will use the measurement, and what other information is needed to make that decision safely?”
A real-time signal is useful only when it supports a defined decision
Begin by naming the manipulated variable and the expected process response. In a grinding circuit, a density trend may help operators review water addition or classification behaviour. At a thickener underflow, it may support pumping and downstream handling decisions. In a transfer line, density may be combined with flow when the two measurements represent the same material and time.
The PMA mining application hub maps these duties without treating them as interchangeable.
Start with trustworthy trending
A sensible first stage is monitoring. Compare the online trend with representative samples and known process events. Confirm that the signal moves coherently during water-to-slurry transitions, changes in operating state and the normal density range. Operators should understand units, damping, update behaviour and invalid-signal indications before the measurement is used for alarms or control.
Rhosonics reports a gold-processing installation where online density information supported operational control of the classification stage. That named result is useful evidence, but it is not a guarantee for another plant. The new duty still needs its own hydraulic, material and validation review.
Move from monitoring to an alarm with documented limits
An alarm requires more than a high or low number. Define the normal operating envelope, persistence time, response expected from the operator, and conditions in which the signal should be treated as invalid. Consider pump trips, empty or partly filled pipes, flushing, start-up and maintenance.
The alarm should sit within the plant’s existing operating philosophy rather than becoming an isolated instrument alarm with no assigned action.
Advisory and automatic control need a complete signal path
For advisory control, the measurement can help an operator choose an action while preserving human judgement. Automatic control requires stronger evidence: documented range, signal quality, update behaviour, alarm strategy, manual fallback, change control and testing of the complete path from sensor to control system.
Density alone does not identify particle size, flow, pressure, level or the cause of a disturbance. The grinding-circuit and mill-overload article explains why density should be interpreted with the rest of the circuit rather than as a single-variable solution.
Combine density and flow carefully
A density signal and a volumetric-flow signal can support a mass-flow calculation when both are valid, time-aligned and representative of the same stream. Check units, signal filtering, installation locations and failure states. Validate the combined result, not only each instrument in isolation.
- Define the operational decision before selecting the measurement point.
- Validate the density trend against representative process evidence.
- Document alarm limits, delays and operator actions.
- Test manual fallback and invalid-signal handling.
- Review any density-and-flow calculation as one measurement system.
Choose the next engineering step
If the duty is in a grinding or classification circuit, start with the grinding-circuit density guide. For pumping or tailings decisions, use the thickener-underflow guide. These pages help define the evidence needed before control changes are considered.
Sources and further reading
- Rhosonics gold-processing classification case study
- Rhosonics Grupo Mexico thickener-underflow case study
- Rhosonics SDM ECO product information
Define the control objective: ask PMA to review the stream, outputs and validation plan.
