Confirm that the change originates in the RO
Compare the online probe with an independent measurement of a fresh sample. Check cleanliness, calibration and temperature compensation. Sample before storage and before any blending point: conductivity that rises only after the tank may come from another connection or the distribution loop.
Measure feed, permeate and concentrate during stable operation. A new source water can raise permeate conductivity even when relative rejection changes very little. Carbon dioxide and acid-base equilibria also affect the reading; conductivity cannot identify a particular contaminant or establish its concentration.
Separate system-wide changes from a local defect
Collect conductivity and flow for each train and, where suitable sampling points exist, individual stages or pressure vessels. Compare equivalent units under comparable conditions. Do not combine start-up samples with readings taken after stable production. Keep a drawing that links every sample label to its physical position.
One vessel that differs from neighbouring vessels suggests a local issue. Similar changes throughout the plant may be associated with feed or operating conditions. Neither pattern proves a cause: both help select the next inspection or test and avoid replacing components without locating the problem.
| Finding | Next check | Avoids |
|---|---|---|
| Change only after storage | Blending and distribution | Unnecessary membrane replacement |
| One abnormal vessel | Seals and permeate profile | Replacing an entire train |
| Similar change in all trains | Feed and operation | Mistaking conditions for damage |
Consider the effect of leakage before replacing everything
For an idealised mixing example, 9.9 m³/h at 20 µS/cm plus 0.1 m³/h at 2,000 µS/cm gives approximately 39.8 µS/cm if conductivity behaves linearly and composition and temperature are comparable. A small bypass can therefore almost double the final reading.
This is not a real leak calculation or a substitute for an ionic balance. It explains why seals, adapters and connections deserve attention. Internal inspection or permeate profiling needs isolation and the manufacturer’s procedure. A pressure vessel must not be opened while it remains pressurised.
Read quality alongside output and chemical history
A sudden change after maintenance calls for checks of assembly and sealing. A deviation following oxidant exposure requires investigation of chemical compatibility and possible membrane damage. Pressure and output patterns help distinguish hypotheses, but cleaning does not repair a chemically degraded active layer.
Record the exact fault location and acceptance criteria for the repair. When elements are replaced, check their position and compatibility with the rest of the train. Confirm the result through both capacity and quality measurements using representative feedwater, rather than relying on a single satisfactory conductivity reading.
Common questions
Does conductivity measure boron or nitrate?
Not specifically. Those constituents require their own analyses; a satisfactory overall reading does not guarantee their concentrations.
Will cleaning always improve rejection?
No. It may help when removable deposits cause the problem. It does not correct assembly leaks or restore a damaged active layer.