
| Stages | Pre-treatment, RO, disinfection |
| Pre-treatment | Multi-media filter, carbon filter, softener |
| RO | Reverse osmosis with a storage and buffer tank |
| Disinfection | UV, ozone, or both |
| Key control | Raw water analysis, recovery rate, conductivity |
| Key decision | Which stages the raw water actually needs |
The raw water analysis decides it. Surface water usually needs multi-media filtration, carbon and RO; hard well water needs softening before the RO; low-hardness municipal water may only need carbon and disinfection. A package bought without an analysis is either oversized or short of a stage.
Water stewardship — cutting litres per litre of product — is now a buyer and regulator expectation, especially under EU ESG pressure.
RO produces permeate and rejects concentrate. The recovery rate sets how much of the feed becomes product water, and the buffer tank size sets how long the line can run when the RO is in flush or maintenance. A filler that outruns the RO starves at shift start, which looks like a filler problem and is not.
Conductivity-controlled final rinse stops over-rinsing; counter-current recovery reuses rinse water for the next pre-rinse.
| Stage | Removes | Driven by |
|---|---|---|
| Multi-media filter | Suspended solids | Turbidity of raw water |
| Carbon filter | Chlorine and organics | Municipal dosing and taste |
| Softener | Hardness | Calcium and magnesium in the analysis |
| RO | Dissolved salts | Conductivity target of the product |
| UV or ozone | Micro-organisms | Storage time and final barrier |
Reference unit for this duty: the Water Treatment.
Meter the intake and the reject stream separately, then divide by production. The ratio between them is the number that moves when recovery improves. Most water-saving projects start with a leak survey and a reject re-use review, not with a new membrane.
30–50% versus an unmanaged legacy line, with most savings from CIP control and recovery.
Throughput and change-part lists are on the Pre-treatment page rather than in this note.
Conductivity at the RO outlet, turbidity before and after the filters, and a microbial sample on a fixed schedule. Continuous sensors catch drift; the microbial sample catches what the sensors cannot see.
The same question is worked through from another angle in How to Integrate Water Treatment Systems in Medium Bottling.
On condition, not on a calendar. Track permeate conductivity, pressure drop and recovery; replacement follows a trend, and a sudden change usually means a pre-treatment failure rather than membrane age.
Pre-treatment removes suspended solids, chlorine and hardness that would foul or damage the membrane. RO removes dissolved salts. RO without correct pre-treatment fails early and expensively.
RO removes minerals as well as contaminants, and very low-mineral water tastes flat. Some plants blend or re-mineralise. The target is a specification set with the product, not the lowest possible conductivity.
It is a controlled circuit that circulates cleaning solution at a set flow, temperature, concentration and time. Any line filling a product other than water needs one, and its capacity follows the largest circuit.
Country: Maldives
Product: PET Bottled Mineralized Water
Capacity: 24,000BPH (500ml); 2 Lines
Solution:Water Production Line
Country:Kazakhstan
Product: PET Bottled Sparkling Water, Canned Soft Drink, Canned Tea
Capacity: 25000CPH (250ml)
Solution:Three Lines,7000BPH(1.5L), 15000CPH(330ml),15000CPH(450ml)