Ion exchange is the stage where dissolved ions are retained by the bed and replaced by an equivalent quantity of ions the bed gives up in return. The choice of resin sets the task itself: softening, alkalinity reduction, nitrate removal or demineralisation. An exhausted exchange capacity is restored by regeneration on site, so the bed lasts for years rather than being consumed.
Principle
The unit is a fibreglass vessel filled with ion exchange resin; beneath the bed lies a drainage and distribution system, a Runxin control valve is mounted on the neck, and a brine tank with the regenerant solution stands alongside. In service the water flows downwards through the bed and leaves through the lower drain. There are no moving parts inside the vessel: the only mechanism in the unit is the valve that switches the flows between modes.
The separation works on the charge of the ion and its affinity for the bed, not on the size of a particle. The resin beads carry fixed functional groups, each with a mobile counter-ion. As it passes through the bed the water gives up dissolved ions and receives back a strictly equivalent quantity — as many charges as were taken in. Hence the defining property of the stage: the composition of the water changes, the total quantity of dissolved salts does not.
What exactly is removed is set by the resin, not by the vessel or the valve. A strong-acid cation resin in the sodium form removes hardness; a weak-acid one in the hydrogen form removes carbonate hardness together with alkalinity; a chloride-form anion resin selective for nitrates removes nitrates; a chelating resin removes boron; a cation and an anion resin in the hydrogen and hydroxide forms together give demineralisation. The vessel, the valve and the pipework stay the same throughout, and the solution is carried from one task to another by the choice of bed and of the reagent that restores it.
Exchange capacity is finite: sooner or later no mobile counter-ions are left in the bed, and the parameter being removed starts to pass straight through. The bed is restored by regeneration — passing a concentrated solution through it that displaces what has accumulated back to drain. For a sodium-form cation resin that solution is common salt from the brine tank; for the hydrogen form it is acid, for the hydroxide form alkali. The cycle consists of loosening the bed with an upward flow, drawing the regenerant, rinsing and refilling the tank; for the duration of the cycle the unit is taken out of service.
When regeneration happens is decided by the control valve. A valve with a water meter counts by the volume actually passed, a valve without one counts by time. A single unit stops the supply for the length of the cycle; a twin unit works with two vessels of which one is always filtering while the other regenerates or waits — the set capacity is held without interruption, and a TWIN-version valve controls both vessels at once. In the manual version the operator starts and switches the cycle: the valve has neither a timer nor a meter.
Principle
A softener does not desalinate water. It takes calcium and magnesium out, but gives back an equivalent quantity of sodium, and the total salinity downstream of the stage stays as it was at the inlet; conductivity changes little, while hardness falls to practically nothing. Hence a consequence that matters for drinking water. Hardness is measured in milligram-equivalents per litre — one meq/l corresponds to about 2.8 °dH — and every meq/l removed adds some 23 mg/l of sodium: on water of 8 meq/l the increase will be of the order of 180 mg/l. The scale goes, the mineralisation stays.
The salt used for one regeneration is set not by the hardness of the water but by the volume of resin and the specific dose chosen: softening calculations work with the order of 100–150 g of salt per litre of resin, a lower dose giving a lower working capacity and a higher dose a higher one. Hardness governs not the dose but the frequency: the higher it is, the sooner the capacity is used up and the more often the cycle repeats. Annual salt consumption is therefore reduced not by trimming the dose — that brings on an early hardness breakthrough — but by increasing the volume of the bed, that is, by a larger vessel.
What puts resin out of service is the water at the inlet, not the number of cycles. In each cycle it gives back exactly what it took in, and in that sense it is not consumed at all. Dissolved iron is held by the resin as a cation but is not released again by brine, and the deposit stays on the beads; free chlorine from a mains supply destroys the structure of the bead irreversibly; dissolved organics are adsorbed and are not washed off. The outcome is the same in all three cases — capacity falls, regenerations become more frequent, hardness breaks through — and a bed designed to last years is used up in a season with the valve in perfect order.
Practice
Hardness is not reduced in part but taken down to close to nothing, and it stays there until breakthrough. The stage has no intermediate state: while the resin is not exhausted the residual hardness in the filtrate is close to zero, and once the capacity is gone the original figure returns at once. That makes monitoring simpler — measuring hardness at the outlet is enough.
The bed is restored on site. The annual cost is made up of salt, regeneration water and the electricity for the valve, while the resin itself is replaced once every few years. Against membrane stages, where the element is replaced in full once its life is over, that is a markedly different cost structure.
The task is carried across by the choice of resin. Softening, alkalinity reduction, nitrate removal and demineralisation are all built on the same vessel with the same valve; what changes is the bed, the regeneration programme and the reagent. For a site this means that refining the task after the analysis does not entail replacing the equipment as a whole.
Uninterrupted supply is a matter of version, not of capacity. A single unit stops the supply during regeneration, and on sites with round-the-clock demand this is settled either with a storage tank or with a twin version, where one vessel is always filtering. The manual version removes the dependence on a power supply but shifts responsibility for the timing of the cycle onto a person.
The unit is assembled from standard parts — vessel, resin, Runxin valve, brine tank — and supplied in a version matched to a particular flow and connection. We do not manufacture it: our part of the work is the analysis, the calculation of resin volume and regeneration regime, the choice of vessel, valve and version, supply, commissioning and subsequent service.
Limits
Practice
The cases are typical examples, not site reports.
Innovations
Scheme
Ion exchange is rarely a stage in its own right. A coarse strainer ahead of it is obligatory, and the rest of the pretreatment is determined by the analysis: reagent-free filtration where there is iron, manganese or turbidity, aeration ahead of that where oxygen is short, adsorption on activated carbon where mains water carries free chlorine. All of this protects the resin rather than the consumer.
Downstream of ion exchange come the stages for which softened water is a condition of operation. Ahead of a membrane unit, softening serves a protective function: it removes the risk of calcium carbonate deposition in the concentrate path at elevated recovery. In deep demineralisation schemes the order is the other way round — ion exchange and electrodeionization stand behind the osmosis unit and work on permeate.
The regenerant discharge line is designed together with the unit rather than run to wherever is convenient. It needs pipework of the appropriate diameter, a discharge point able to take the slug flow, and agreement on chloride with the organisation that operates the sewerage network.
Continuity of supply is settled at the scheme stage. A single unit is taken out of service while it regenerates, and that is either bridged by a storage tank downstream of it or removed by the twin version. The choice between them is a question of flow, space, cost and whether a break is acceptable — not of the capacity of the equipment.
Scheme
Operation
Energy
The unit draws little power of its own: electricity is used only by the control valve drive, which runs for minutes a day, and the manual version needs no supply at all. There is no pump in the scheme — the regenerant is drawn by an injector on the pressure in the line itself, and from that follows a requirement that is most often left out of the calculation: with an inlet pressure below the rated figure (for valves of this class the order of 2 bar, with the exact value set by the data sheet) the injector does not lift the brine, the cycle goes through in form only, and the exchange capacity is not restored. The running costs that matter here are different — salt and regeneration water.
Consumables
What goes to drain
The regenerant effluent is the principal operating and consenting feature of the technology. What goes to drain is a mixture of unreacted brine, the calcium and magnesium chlorides displaced from the resin, and rinse water: in volume it amounts to a few per cent of the throughput, but in composition it is a saline slug discharge in which the chloride content is many times the original. Three things follow. The receiving point has to withstand the slug flow, so it is calculated rather than assigned to whatever is nearest. Local treatment works and septic tanks with a biological stage are suppressed by such an effluent — salt inhibits the microflora, and the consequences show up not at the softener but in the drainage system. Discharge to the network is agreed on chloride before the equipment is ordered: where the limit is strict, the scheme is either moved to a membrane one or supplemented with a balancing tank that spreads the discharge over time.
Sizing
What we account for
What a wrong choice costs
Sizing
Questions
There is one limitation here, and it concerns sodium. Sodium-form ion exchange removes hardness by replacing calcium and magnesium with sodium, and the increase is proportional to the hardness removed: every milligram-equivalent per litre taken out adds some 23 mg/l of sodium. On moderately hard water that is immaterial; on water of 8–10 meq/l the increase is already around 200 mg/l, and where medical sodium limits apply such water is not used for drinking. The usual answer is to soften the domestic line as a whole and to take the drinking point off separately through reverse osmosis.
Sodium-form ion exchange reduces hardness by replacing calcium and magnesium with sodium; total salinity is retained and the sodium figure rises somewhat. Reverse osmosis reduces salinity as a whole, sodium, nitrates and sulphates included, but requires pretreatment, electrical energy and a concentrate discharge. If the task is limited to scale and no sodium limit applies, softening is the more economical answer, in capital and in running costs alike. If total salinity is above the limit or a sodium limit applies, it does not address the task at all: osmosis becomes the main stage, and softening ahead of it serves a protective function.
On the volume of resin and the specific dose chosen, not directly on the hardness of the water. For one regeneration, softening calculations work with the order of 100–150 g of salt per litre of resin: a lower dose gives a lower working capacity, a higher dose a higher one, and the choice is a trade-off between salt consumption and the volume of the bed. Hardness governs something else — how often the cycle repeats. Annual salt consumption is therefore reduced by increasing the resin volume, that is, by a larger vessel, and not by trimming the dose, which brings on an early hardness breakthrough.
Its life is measured in years and is not set by the number of regenerations: in each cycle the resin gives back exactly what it took in, and in that sense it is not consumed. What puts it out of service is the water at the inlet. Dissolved iron is held by the resin as a cation but is not released again by brine; free chlorine from a mains supply destroys the structure of the bead irreversibly; dissolved organics are adsorbed and are not washed off. The signs are the same in every case: regenerations grow more frequent, and hardness appears at the outlet earlier than calculated. This is why pretreatment ahead of ion exchange is not an improvement to the scheme but the condition on which the bed lasts its design life.
In control and in result. In the twin version both vessels are run by a single TWIN-version valve: one vessel is always filtering while the other regenerates or waits, and the changeover happens when the capacity of the working vessel is used up rather than to a schedule. The set capacity is held without interruption, with no stop to restore the bed — which a single unit cannot provide in principle. Two independent single units do not give that continuity by themselves: they would have to be switched by hand or given external logic, and the capacity margin would be used less well.
Yes, but not on the resin that removes hardness. Nitrates are removed on a chloride-form anion resin selective for the nitrate ion; an equivalent quantity of chloride goes into the water in return, and such a bed is regenerated with the same common salt solution. Two conditions matter. Sulphates in the raw water compete with nitrates for the resin, so selectivity is chosen from the analysis rather than from the general name of the bed. And regeneration is run with a margin on time rather than on the appearance of nitrates at the outlet: on a non-selective resin the accumulated nitrate is displaced by sulphate, and the filtrate can then carry more of it than the raw water did.
Tell us about your site — our engineer will prepare a process design, a specification and a quotation within one business day.