UF-RO for STP Water Reuse: Turning Treated Sewage into Industrial Water

Every industrial estate in India has the same two things sitting a few hundred metres apart: a sewage treatment plant discharging treated water to a drain, and a dozen factories buying tanker water or drawing groundwater under a CGWA NOC that is getting harder to renew. Joining the two is not a new idea — the UF-RO tertiary train that does it is now a standard product — but the projects that work are the ones that understood what the STP water actually is, and what the RO downstream will and will not tolerate. Here is what separates the plants that run from the ones that foul in a season.

What you are starting with

Secondary-treated sewage from a well-run STP — an MBBR, SBR or activated sludge plant with a clarifier — typically leaves with BOD under 10–20 mg/L, TSS of 10–30 mg/L, TDS of 600–1,200 mg/L, ammonia that depends on whether the plant nitrifies, and a residual of dissolved organics, surfactants, phosphate and bacteria. It looks clear. It is not RO feed. The SDI of clarified secondary effluent is usually above 5, the organic load is exactly what biofouls a membrane, and the phosphate and calcium together form a scale that surprises designers who only modelled carbonate. From the RO's point of view, this is the most difficult feed it will ever see — harder than borewell, harder than most industrial process water — and the pre-treatment has to be designed accordingly.

Why UF is not optional here

On borewell water a multimedia filter and cartridges are often enough ahead of an RO. On STP effluent they are not. The colloidal fraction that survives a clarifier — fine biological floc, cell debris, colloidal organics — is exactly the size that a sand filter passes and an RO feed spacer traps. Plants that tried MMF-only pre-treatment on sewage reuse have generally ended up cleaning the RO every two to four weeks and replacing membranes inside two years. Ultrafiltration changes that arithmetic: it is an absolute barrier to particles and bacteria, it delivers an SDI reliably under 3 regardless of how the clarifier performed overnight, and its permeate is what an RO membrane warranty was written for. On sewage the UF itself needs care — a pressurised hollow-fibre module with 100–200 micron strainer protection, a CEB with hypochlorite every day or two, and an honest design flux of 40–60 L/m²·h rather than the 80+ that a clean-water datasheet might suggest. Where the STP is being built new, a submerged MBR folds the UF into the biology and is often the cleaner design.

Designing the RO to actually last

The RO on a sewage-reuse train has its own rules, most of them different from a borewell RO:

• Fouling-resistant elements with 34-mil spacers, not standard BWRO. The wider channel and hydrophilic surface pay for themselves in cleaning frequency alone. See RO membrane types.
• Conservative flux. Design at 14–18 L/m²·h average, not the 25+ that a clean feed allows. The whole difference between a plant that cleans monthly and one that cleans quarterly is in this number.
• Biofouling control from day one. Chloramination of the UF feed (chlorine plus ammonia, which sewage often supplies for free) gives a residual the polyamide tolerates; or dechlorinate fully and dose a non-oxidising biocide periodically. A plant that plans to "see how it goes" on biofouling will see.
• Recovery set for calcium phosphate and silica, not just carbonate. Phosphate at 3–8 mg/L, common in effluent, forms a scale antiscalants control only partially. 70–75% is a realistic single-stage recovery on most sewage; the reject needs a plan — see RO reject treatment.
• An equalisation tank between STP and UF. The STP's output swings with the estate's shifts and the UF-RO wants a steady feed. A few hours of buffer is the cheapest reliability the plant will ever buy.

Where the water goes, and what that decides

Not every reuse duty needs RO. Cooling tower make-up, toilet flushing, landscape and construction water are well served by UF permeate alone, or UF plus chlorination — and skipping the RO removes the reject problem entirely. RO earns its place when the water is going to a boiler, a process line, or a cooling system running at high cycles where TDS matters. Many estates do best with a split design: UF for the bulk low-grade demand, RO on a fraction for the high-grade users. The water reuse calculator is a quick way to see how the split changes the economics, and the "which water, for what?" question is the right place to begin.

What it takes to run the plant

A UF-RO reuse plant is an operating asset, not a civil structure. It needs daily SDI and normalised-data logging, a chemical programme that is actually followed, and a person who knows the difference between a UF backwash problem and an RO biofouling problem. The estates where reuse has succeeded are the ones that put the tertiary plant under an operations contract with performance guarantees on flux and cleaning interval — and the ones where it has failed are, nearly always, the ones that bought a UF-RO skid and left it to the STP operator. The tertiary treatment conversation should start with the water balance and end with who is going to run it.

If your estate or campus has an STP discharging water you are paying to replace, the feasibility is a short conversation to have with us — [email protected] or +91-98100 00233.