The question
Pretreatment was designed correctly. Why is the membrane still fouling?
Start by assuming nothing has stayed the same
An RO plant that fouled from the day it was commissioned has a design problem. An RO plant that ran well and then began fouling has a change problem — and the useful question is what changed, not what to dose.
Establishing the timeline is usually more productive than any single test. When did the trend turn, and what else happened around that date — a new supply, a shutdown, a maintenance job, a staff change?
What usually turns out to be the cause
- A stage was bypassed and never restored. A filter isolated during a shutdown, a media vessel taken offline for a repair. The bypass works, the plant keeps running, and nobody records it.
- Dechlorination is failing. Exhausted carbon or underdosed bisulphite lets oxidant reach the membrane, which damages it permanently. Overdosed bisulphite is the opposite problem — it removes the residual so early that the pretreatment train itself becomes a biological growth environment.
- Iron and manganese are oxidising before the membrane. Raw levels look acceptable, but exposure to air between the well and the plant converts dissolved iron into a particulate that arrives as an oxide deposit.
- Backwash is carrying over. A media filter returned to service too quickly after backwash passes a slug of solids straight to the cartridge filters and then to the membrane.
- SDI is not being measured. Without it there is no way to know whether pretreatment is achieving what it was designed to achieve — only whether the membrane has already suffered.
- The source water changed. A new well, a different blend, a seasonal shift. The pretreatment was correct for the water it was designed for and is now treating something else.
- Antiscalant is not actually being delivered. A pump running dry, a blocked injection point or a suction lost to degassing. The controller reports a call for dose; nothing confirms the dose arrived.
The one nobody looks at: stagnation
An RO plant that runs intermittently spends most of its life full of stationary water at ambient temperature, with the oxidant residual already removed by the dechlorination step. That is close to an ideal culture environment.
Biological fouling in a plant with a good SDI and clean pretreatment is very often a stagnation problem rather than a feed problem. Standby flushing regimes and preservation procedures exist for exactly this reason, and they are among the first things to lapse.
What to check
- 01Walk the pretreatment train physically and confirm every stage is in service and not bypassed.
- 02Test for oxidant residual immediately upstream of the membrane.
- 03Measure SDI at the membrane feed and record it, not just the plant inlet turbidity.
- 04Check cartridge element life against its historic value.
- 05Verify antiscalant is physically leaving the drum — check level against calculated consumption.
- 06Confirm standby flushing is programmed and actually running.
- 07Compare the current feed analysis with the one the plant was designed against.
When to get technical support
Oxidant exposure permanently damages a polyamide membrane, and no cleaning reverses it — a rising salt passage after a dechlorination fault usually means replacement, and confirming the cause matters before the new elements are exposed to the same condition. Where the source water has genuinely changed, the pretreatment train needs re-evaluating against the new analysis rather than adjusting.
Talk to an EngineerFrequently asked
Fouling can usually be cleaned. Oxidative damage to the polyamide layer cannot — the separation performance is permanently changed, and cleaning will not restore it. Distinguishing between the two before ordering replacements is worth the diagnostic effort.