By André Opperman, Managing Director, Rolfes Water
Water in an industrial context is not a utility in the way electricity or compressed air might be described as utilities. It is a controlled input. It goes into boilers, cooling towers, heat exchangers, and membrane systems, all of which are calibrated to operate within defined chemistry parameters. When the quality of that input shifts outside those parameters, the equipment responds. Scale forms. Corrosion accelerates. Biological growth finds its window. The responses are predictable and well understood. What is less well managed is the period of variability that creates the conditions for them.
Water supply interruptions are now part of daily operational life for many industrial facilities in South Africa. Most of the public conversation around this focuses on availability, which is understandable. But availability is only one part of the problem. The part that tends to get less attention is what happens to water quality during shutdown and restart cycles, and that is where the real technical exposure sits for process-driven operations.
I have spent decades working in industrial water treatment, and the consistent lesson across different industries and different operating environments is straightforward: when supply becomes unreliable, the chemical management inside the facility needs to become more rigorous, not less. The temptation under pressure is to treat water chemistry as a secondary concern behind getting the plant back up and running. That ordering tends to be expensive.
What Changes in the Distribution Network During a Shutdown
When municipal supply is interrupted, whether through planned maintenance, infrastructure failure, or any of the other scenarios that South African operators have become familiar with, the distribution network does not simply hold its state until flow resumes. Changes happen inside the pipework during that period.
Sediment that normal flow velocity keeps in suspension begins to settle. Biofilm, which is present in any distribution system to some degree, expands when the shear conditions that normally limit it are removed. Disinfectant residuals continue to be consumed by oxidant demand in the system but are no longer being replenished, so they fall. In extended shutdowns, they can become negligible.
The biological risk during these periods is not theoretical. National performance assessments over recent years have highlighted persistent microbiological compliance challenges in several municipal systems. Against that backdrop, biological expansion during a shutdown is not an edge case. It is a predictable response to stagnant conditions.
When the system restarts and velocity increases, accumulated material is disturbed and carried forward. The first water arriving at a facility intake after a significant interruption frequently carries elevated turbidity, reduced or inconsistent oxidant residuals, and variation in pH and alkalinity that may be outside the normal operating range. I have been in plants where that first flush was visibly discoloured. Some operators restart on that water without checking. That is a mistake.
This quality drift at the restart boundary is transient in most cases. The municipal system stabilises, the pipeline clears, and water quality returns to something closer to baseline. But transient does not mean harmless. Process equipment does not distinguish between sustained and brief excursions outside chemistry limits. The damage mechanisms respond to the condition regardless of how briefly it existed.
Why Process Equipment Does Not Tolerate Variable Inputs Well
Boilers operate within defined chemistry parameters for a reason. Hardness and alkalinity variation increase scale deposition rates on heat transfer surfaces. Even modest scale accumulation introduces thermal resistance that reduces efficiency and drives up fuel costs. Dissolved oxygen excursions during startup phases create conditions for corrosion that can go undetected until pitting has progressed to a point where intervention becomes significant. pH variation compounds both problems and tends to do so in ways that are not immediately visible during routine checks.
Cooling towers are sensitive in a different way. Elevated suspended solids in restart water contribute to fouling on fill media and exchange surfaces. Inconsistent oxidant residuals widen the opportunity for microbiological growth. Once biofilm establishes itself in a cooling system, recovering thermal efficiency and re-establishing chemical control takes considerable time and cost that would not have been incurred had the incoming water quality been properly checked and managed from the outset.
Membrane systems carry their own specific vulnerability. Turbidity spikes at restart are a routine consequence of municipal line re-pressurisation and can cause membrane fouling that is not reversible. SDI values that sit comfortably within limits under normal conditions can spike briefly during these events and still cause damage that only becomes apparent over subsequent weeks as flux declines and cleaning frequency increases.
None of this is speculative. These are known responses to known conditions. The water treatment programme needs to be structured around that variability, not around an assumption of stable influent quality that no longer reflects operating reality.
The Chemistry Problem That Bulk Storage Introduces
The practical response to intermittent supply at most South African industrial facilities has been to increase bulk storage capacity. That investment buys volume security, which is real and necessary. What it also introduces is water age, and aged water behaves differently from freshly delivered water in ways that matter for process systems.
Disinfectant residuals decay in storage as oxidant demand is satisfied and replenishment stops. Dissolved oxygen levels shift. In warm conditions, and South African storage environments are frequently warm, biological growth risk increases in tanks where turnover is slow and circulation is limited. Stratification can develop in tall tanks, creating zones where temperature and chemistry are meaningfully different from the bulk average.
Water that met all process specifications when it entered storage may not meet those same specifications after three or four days. If it is drawn directly into the boiler or cooling circuits without any verification, the system is running on uncontrolled chemistry. The operators will not necessarily know this because the water came from their own treated storage, and the working assumption is that treated means compliant. That assumption needs to be tested regularly, not taken for granted.
Managing the chemistry of stored water is part of managing the water treatment system. It is not a secondary concern or something to address when there is capacity to do so. Under intermittent supply conditions, it becomes a primary responsibility.
The Response That Sits Within Plant Control
The municipal network, the condition of the reticulation infrastructure, and the maintenance scheduling of the local authority: none of that is available to an industrial operator to manage directly. The chemistry inside the plant boundary is a different matter. That is where the practical focus has to sit.
During restart cycles, monitoring intensity needs to increase. Turbidity, conductivity, pH, and oxidant residuals should be checked at intake before systems are loaded back up after any significant interruption. The operational pressure to restore production quickly is real, but the restart window is precisely when the system is most vulnerable. It is the worst possible time to be flying blind on your chemistry.
Treatment setpoints calibrated for stable influent conditions need to be revisited when variability increases. Alkalinity control chemistry may need adjustment to handle the hardness variation that the original programme was not designed for. Scale inhibitor dosing should respond to measured hardness rather than assumed baseline values. Biocide programmes in cooling systems need to be driven by measured residuals, not by historical dosing rates built around incoming water quality that no longer applies.
Stored water needs a preservation protocol. Maintaining oxidant residuals, circulating where the tank configuration allows it, and checking chemistry before drawing stored water back into production circuits are straightforward measures that sit within the scope of any properly managed water treatment programme. They require consistent attention rather than significant capital outlay.
For facilities that go through idle periods during supply interruptions, lay-up of boilers and heat exchangers deserves proper attention. Internal metal surfaces without appropriate lay-up chemistry will experience corrosion during idle periods that continuous operation would not have produced. The cost of a structured lay-up programme is consistently small relative to what is spent addressing the corrosion damage that accumulates when it is not in place.
What the Broader Picture Looks Like
South Africa is not the only place where industrial operators have had to adapt to intermittent supply and variable water quality. Across several parts of Africa, and in other regions where infrastructure has not kept pace with demand, this has been an operational reality for years. The facilities that manage water chemistry as a controlled process variable consistently perform better on equipment longevity and operational continuity than those that treat water treatment as a background function requiring minimal active management.
The water treatment industry has moved toward proactive management over the course of my career. Facilities that monitor consistently, that adjust chemical programmes in response to changing influent conditions rather than in response to equipment problems, and that understand what the chemistry is actually doing inside their systems perform better over time. When supply is stable, that discipline shows up in modest efficiency gains and extended equipment life. When supply becomes intermittent, it becomes considerably more important because the cost of inattention rises and the window for catching problems before they develop into something more serious narrows.
Managing Within the Boundary
Supply volatility is the current operating environment for much of the South African industry, and it is not resolving quickly. The facilities adjusting their water chemistry management to account for it are in a better position than those waiting for external conditions to improve before they act.
At Rolfes Water, we support industrial facilities with both essential chemistry, including Hydrochloric Acid, Sulphuric Acid, Caustic Lye, and Sodium Hypochlorite, and the application-specific speciality chemistry that process systems require. The two are not separate product categories. They are part of the same technical discipline, and getting both right under variable supply conditions is what effective water management looks like in practice.
If you are seeing quality issues after supply interruptions, or if your current treatment programme was calibrated for conditions that no longer reflect your operating reality, it is worth reviewing that sooner rather than later. The problems that intermittent supply introduces do not present themselves loudly or all at once. They build gradually, and by the time they are obvious, they have usually been developing for some time.
