Every RO plant runs at a recovery — the fraction of the feed water it turns into clean permeate, with the rest leaving as reject. Push recovery up and you waste less water and shrink your reject problem. Push it too far and you scale and foul the membranes into an early grave. Recovery is the number where the economics and the chemistry of an RO plant collide, and getting it right is less about ambition than about knowing your water.
What recovery actually is
Recovery is simply permeate flow divided by feed flow, as a percentage. At 75% recovery, 100 litres of feed becomes 75 litres of permeate and 25 litres of reject. It sounds like a number you would always want as high as possible — more product, less waste, a smaller and cheaper reject stream to dispose of. And for the water balance, that is true. The catch is on the other side of the membrane.
Why you can't just turn it up
Here is the physics that limits it. RO removes salts by leaving them behind — so as you recover more permeate, the salts you rejected are packed into an ever-smaller volume of reject. Run at 50% recovery and the reject is roughly twice as concentrated as the feed; run at 90% and it is many times more concentrated. Two things go wrong as that concentration climbs:
- Scaling. Sparingly soluble salts — calcium carbonate, calcium sulphate, silica, barium — hit their solubility limit and precipitate onto the membrane as hard scale. Scale is the classic recovery-driven failure, and silica in particular sets a hard ceiling on many Indian groundwaters.
- Osmotic pressure. The saltier the reject, the higher its osmotic pressure, so you need more and more applied pressure — more energy — to keep pushing water across. Eventually the membrane is working against a wall.
So recovery is a balance: too low and you waste water and oversize your reject handling; too high and you scale the membrane, spike your energy, and shorten membrane life. The right recovery is the highest one your feed's chemistry safely allows — and not one point more.
The number is set by your limiting salt
The single most important idea here is that recovery is not a target you choose; it is a limit your feed hands you. Whichever sparingly soluble salt hits saturation first — often calcium carbonate or silica — is your limiting salt, and it sets the ceiling. A softened, low-silica feed might safely run at 80%+ recovery; a hard, high-silica borewell might top out far lower. Design the plant around the wrong limiting salt and you either leave recovery (and water) on the table, or you scale within weeks.
This is exactly why feed characterisation comes before recovery selection, and why a scaling projection — modelling each salt's saturation across the array — is standard practice before committing a design.
The levers that let you push higher
Once you know the limiting salt, several levers raise the safe ceiling:
- Antiscalant dosing — a small, continuous dose of scale inhibitor keeps sparingly soluble salts in solution past their normal saturation point, and is the most common way to buy extra recovery. The right antiscalant depends on which salt is limiting.
- Softening or pH adjustment — removing hardness upstream, or trimming pH, directly tackles carbonate scaling and lets recovery climb.
- Consistent, low-fouling feed — a stable low-SDI feed, ideally from UF pre-treatment, lets you run harder without fouling stacking on top of scaling.
- Multi-stage (concentrate staging) arrays — arranging pressure vessels so the reject of one stage feeds the next keeps crossflow velocity up as volume shrinks, squeezing out more recovery without dead zones where scale forms.
- Two-pass or concentrate RO — for very high recovery, a second RO on the reject pushes the overall figure up, at the cost of more plant.
Don't chase the last few points
The temptation, especially when water is expensive or discharge is restricted, is to drive recovery as high as it will go. Resist doing it blindly. The last few percentage points of recovery are the most dangerous — that is where the reject is most concentrated, scaling risk is highest, and a small miscalculation costs you a membrane set. If your goal is genuinely to eliminate the reject, the honest route is not to torture a single RO to 95%, but to design a proper high-recovery train — RO followed by further concentration and, where it makes sense, zero liquid discharge — so each stage runs within its safe limits.
Recovery, in the end, is a conversation between what you want (more water, less reject) and what your feed will tolerate (before it scales). The plants that get it right are the ones that measured the feed, found the limiting salt, and set recovery there — then used antiscalant and staging to move the limit, not to ignore it. If you want help finding the safe recovery for your water, or fixing a plant that is scaling, that is a short conversation to have with us, or reach us at [email protected] or +91-98100 00233.