Blog Contact

Maintaining Flotation Analysers for Reliable Metallurgical Balance

A flotation circuit can appear stable while its measurements quietly drift. A small bias in pulp density, feed grade, reagent concentration, air rate, or tailings assay can distort the daily metallurgical balance and lead operators towards the wrong process decision. Online analysers are valuable because they provide frequent measurements, but only when their sampling, calibration, cleaning, and data handling remain dependable.

For a concentrator, maintenance is therefore part of process control rather than a separate instrumentation task. The analyser must represent the stream, operate within its design limits, and produce data that metallurgists can reconcile with laboratory assays and plant mass flow. When those conditions are met, the circuit becomes easier to optimise and unexplained recovery losses can be isolated sooner.

Australian operations often work across remote locations, FIFO rosters, variable ore blends, and strict work health and safety requirements. A practical maintenance programme needs to account for long supply chains from Perth or Brisbane, high temperatures in the Pilbara and Western Australia’s Goldfields, and the need to coordinate instrumentation work with production, laboratory, and maintenance teams.

Why Online Analysis Matters

Online analysers support a near-real-time view of flotation performance. Depending on the equipment, they may measure elemental concentrations in feed, concentrate, middlings, or tailings, while associated instruments track density, flow, pH, conductivity, particle size, or froth characteristics. Together, these measurements help identify whether a change in recovery is caused by mineralogy, operating conditions, sampling error, or instrument failure.

A metallurgical balance compares measured feed, products, grades, recoveries, and flow rates. If an analyser reports an incorrect tailings grade because its sample line is partially blocked, the apparent recovery may rise or fall without any real change in the ore. The result can be an unnecessary reagent adjustment, a misleading shift report, or an incorrect assessment of a plant trial.

The most useful maintenance objective is not simply maximum analyser availability. It is trustworthy information at the times when operators and metallurgists need it. A short, documented outage with a validated laboratory result is safer than continuous data that have not been checked.

Signals That Support Metallurgical Balance

Maintenance begins with understanding the complete measurement chain. The primary sample must be representative, the transport line must remain open, the conditioning system must prevent settling, and the analyser must be correctly aligned with the process stream. A clean detector cannot compensate for a sample that has segregated or arrived intermittently.

Trend review is as important as alarm review. Gradual divergence between online results and laboratory assays may indicate calibration drift, changes in mineral composition, scale on a probe, worn pump components, or altered slurry density. Sudden spikes often point towards air entrainment, electrical interference, sample interruption, or a valve cycling problem.

Mass-balance software should retain instrument status alongside the value itself. A result recorded during a flushing cycle, low-flow condition, calibration check, or known blockage should be marked accordingly. This prevents suspect readings from being treated as valid production data during daily reconciliation.

Maintaining Sample Lines And Slurry Handling

Sample lines are frequent sources of poor analyser performance. Fine particles can settle in horizontal pipework, coarse particles can accumulate at bends, and sticky mineral or reagent deposits can restrict the sample aperture. Operators should inspect pressure, flow, return behaviour, and flush effectiveness rather than relying only on the analyser’s internal status screen.

Routine cleaning should follow the equipment manufacturer’s chemical and mechanical limits. Water flushing may remove loose solids, while approved cleaning agents can address scale or reagent films. Any chemical cleaning must be controlled under the site’s safety management system, with isolation, compatible materials, correct personal protective equipment, and clear disposal arrangements.

Pump condition also affects data quality. A worn diaphragm, blocked strainer, leaking seal, or pulsation problem can create an uneven sample stream. The maintenance record should connect pump inspections with analyser validation, because replacing a pump without checking the resulting sample flow can leave the original measurement problem unresolved.

Calibration And Laboratory Verification

Calibration should use appropriate reference materials, current process samples, and a documented method. The frequency depends on the analyser technology, ore variability, sample condition, and historical drift. A stable circuit may require less frequent adjustment than a plant processing rapidly changing blends or highly variable polymetallic ore.

Laboratory comparison must use matched samples wherever possible. Online and laboratory results should represent the same stream and time window, with sample identification, collection time, preparation method, and assay method recorded. Differences may arise from sampling intervals or laboratory preparation, so a single mismatch should not automatically trigger an instrument recalibration.

Metallurgical teams should review bias, repeatability, and response time together. A sensor that is precise but consistently biased can often be corrected, while an instrument with erratic repeatability may require mechanical inspection. For gold circuits, process chemistry also matters; operational teams can compare flotation data with related guidance on cyanide consumption reduction when downstream leaching performance influences the overall plant balance.

Data Integrity And Maintenance Records

A defensible balance requires a clear history of what the analyser measured and what maintenance occurred. Records should include calibration certificates, zero and span checks, cleaning dates, sample-line inspections, replaced components, software changes, alarm events, and laboratory comparison results. Time synchronisation between the plant control system, historian, laboratory, and maintenance platform is essential.

Data gaps should be identified rather than silently filled. If an analyser was offline for four hours, the balance should show the outage and use an approved substitute method, such as laboratory assays or a validated estimation procedure. Back-calculated values should be labelled so that future technical reviews can distinguish measured data from inferred data.

Maintenance planners can use performance indicators such as availability, mean time between failures, calibration bias, invalid sample percentage, and average time to restore a valid reading. These measures help determine whether recurring issues require a spare part, a design change, improved operator checks, or a revised sampling location.

Australian Operating Conditions

Remote Australian mines may face long travel times and limited access to specialist technicians. A concentrator near Kalgoorlie, Mount Isa, or the Pilbara can lose valuable production time while waiting for a specific pump, probe, valve, or electronic module. Critical spares should be selected from failure history and held locally where the consequence of an outage justifies the inventory cost.

Heat, dust, vibration, and water restrictions also influence maintenance. Instrument enclosures, cooling systems, sample cabinets, and air supplies need inspection suited to site conditions. In Western Australia, FIFO rosters and contractor changeovers make handover quality especially important; a concise status report should identify the fault, temporary controls, pending parts, and next validation step.

Work must align with Australian WHS duties, site isolation procedures, and applicable mining safety legislation. Queensland operations, for example, work within the state’s mine safety framework, while Western Australian sites follow their own mining safety and health requirements. Environmental approvals and water-management controls can also affect flushing, cleaning chemicals, and discharge arrangements. Procurement should consider Australian electrical compliance, vendor support, and NATA-accredited laboratory capability where appropriate.

Practical Maintenance Actions

A maintenance programme should combine operator checks, planned instrumentation work, laboratory verification, and engineering review. The following actions provide a useful baseline for a flotation plant:

Grinding conditions can influence flotation feed size, liberation, and reagent response, so analyser maintenance should be reviewed alongside upstream operating costs and process stability. A useful reference for that discussion is the guidance on steel ball consumption, particularly when changes in grinding media affect downstream assay trends.

Reliable analyser maintenance is a shared responsibility between operators, instrument technicians, metallurgists, laboratory staff, and engineering contractors. Clear ownership prevents a common failure mode in which each team assumes another group is responsible for validating the result.

For mining companies planning a new circuit, expansion, or plant upgrade, analyser locations and sample systems should be included during process design rather than added after commissioning. Lozova.org supports mineral processing projects with equipment selection, engineering, ore testing, commissioning, and operational assistance. Engage the project team early to design a sampling and online measurement system that can deliver a traceable metallurgical balance from commissioning through routine production.