Design Manual › Tertiary & Reuse › Reverse Osmosis
FREE ENGINEERING TOOL · REUSE

Reverse Osmosis for Water Reuse

Recovery is the number everyone wants to push, and the number two quiet constraints will not let you push. As recovery rises the concentrate gets saltier, the osmotic pressure you must beat rises with it, and the sparingly soluble salts arrive at their limit long before the membrane does. This tool sizes the array from design flux, builds the pressure from osmotic pressure and temperature-corrected permeability, and then runs the Langelier index and silica check on the concentrate — the two that actually decide how far you can go.

1Duty & separation

Start with what the water is and what it has to become. Everything else — area, pressure, energy — follows from those two numbers and the recovery you can actually achieve.

Results — permeate, area & pressure

Permeate
—m³/d
Membrane elements
——
Specific energy
—kWh/m³
ItemValueBasis

2Membrane array

Area comes from flux; the element and vessel count comes from area. The array is then staged 2:1 so the second stage keeps the crossflow up as the volume falls.
ItemValueBasis

3Pressure & energy

Feed pressure has to beat the osmotic pressure of the concentrate, not of the feed — and the temperature changes both the membrane's permeability and its salt passage.
ItemValueBasis

4Scaling — the real recovery limit

This is where a reuse scheme gets decided. The concentrate, not the feed, is what precipitates.
ItemFeedConcentrate

5What passes, what does not

Rejection is not one number. Neutral species and small uncharged molecules cross RO membranes freely, and that catches reuse schemes out.
SpeciesRejectionConsequence for reuse

6Instrument schedule

RO is a differential-pressure and a two-conductivity process. Everything else is there to protect the membrane.
PointInstrumentWhy here

Quote the RO instruments at factory price

Feed and permeate conductivity, differential pressure, flow, pH/ORP and level — all mappable to our catalogue and supplied at the manufacturer's ex-works price (our margin is the export rebate, not a markup on the device).

Request a quote for this duty ← Granular Activated Carbon Adsorber Effluent Online-Monitoring Configurator →
Response within 24 hours · Inputs stay in your browser — nothing is uploaded.

Sources & parameter basis

[1]METCALF & EDDY / AECOM — membrane treatment: municipal reuse schemes typically treat secondary or tertiary effluent of 300–3000 mg/L TDS; single-pass brackish RO gives permeate well under 200 mg/L.
[2]Element nominal salt rejection 96–99.5 % at standard test conditions (25 °C). Salt passage roughly doubles for every 5–6 °C rise, so a warm feed gives more water and dirtier water at the same time.
[3]Recovery: 65–85 % for single-pass brackish RO on well-pretreated municipal effluent; 85–90 % only with concentrate recirculation or a second stage. Recovery is limited by concentration, not by the membrane.
[4]Design flux 13–25 LMH for wastewater RO downstream of UF/MF; 8-inch element active area 30–40 m²; 4–7 elements per pressure vessel; arrays staged 2:1 or 3:2:1.
[5]Permeability: flux = A·NDP, with A ≈ 1.5–4 LMH/bar at 25 °C, temperature-corrected by a factor of 1.03^(T−25). Osmotic pressure ≈ 0.00077 bar per mg/L TDS. Feed pressure = NDP + (π_avg − π_permeate) + ΔP/2.
[6]High-pressure pump: kW = Q(m³/h)·P(bar)/(36·η). Brackish RO on municipal reuse typically lands at 0.4–0.8 kWh per m³ of permeate.
[7]Langelier saturation index on the concentrate: pH_s = 9.3 + A + B − C − D with A = (log₁₀TDS − 1)/10, B = −13.12·log₁₀(T+273) + 34.55, C = log₁₀(Ca as CaCO₃) − 0.4, D = log₁₀(alkalinity as CaCO₃). Concentrate pH ≈ feed pH + log₁₀(CF). LSI above about +0.5 needs antiscalant and/or acid; above +1.8 reduce recovery.
[8]Amorphous silica solubility is roughly 120 mg/L at 25 °C and rises with temperature and pH. Silica scaling is effectively irreversible — it is the constraint you cannot chemically fix after the event.
Full derivation and worked example: ro-reuse.html. Push the recovery from 75 % to 85 % and watch the concentrate LSI jump — the membrane would happily do it, the calcium carbonate would not.
Questions about a duty, a specification, or transparent sourcing? Contact our application engineers →