Water softening is the largest single use of ion exchange resin, and it is also the one most often designed by rule of thumb. The chemistry is simple – a strong acid cation resin in the sodium form swaps calcium and magnesium for sodium – but the sizing, the salt consumption and the failure modes are where plants get into trouble. This page covers the practical design basis.
What softening does, and does not, do
| Aspect | Softening (SAC, Na form) | Demineralisation (SAC H + SBA OH) |
|---|---|---|
| Removes | Calcium and magnesium | All cations and all anions |
| Adds to the water | Sodium, in equivalent amount | Nothing – the ions become water |
| Total dissolved solids | Essentially unchanged | Reduced to near zero |
| Alkalinity | Unchanged | Removed |
| Regenerant | Sodium chloride brine | Hydrochloric or sulphuric acid, plus sodium hydroxide |
| Typical use | Boiler feed, cooling make-up, laundry, food processing | High-pressure boiler feed, electronics, pharmaceutical |
The common misconception worth stating: softened water is not purified water. It has the same total dissolved solids as the feed, with calcium and magnesium replaced by sodium. If the objective is lower conductivity, softening is the wrong tool.
Design basis
| Parameter | Typical basis | Why |
|---|---|---|
| Service flow rate | Of the order of 10-40 bed volumes per hour, depending on bed depth and water quality | Too fast and the exchange zone broadens, so hardness leaks before the bed is exhausted |
| Bed depth | Generally not less than about 0.8-1.0 m | Shallow beds channel and give poor utilisation of capacity |
| Working capacity | Typically 60-80% of the total exchange capacity | The total figure assumes complete exhaustion, which no real cycle achieves |
| Regenerant dose | Salt consumption well above the stoichiometric minimum | Efficiency falls as you push towards full regeneration; the optimum is an economic balance, not a maximum |
| Backwash rate | Set from the wet true density of the resin, typically enough for 50-100% bed expansion | Too low and the bed compacts; too high and resin washes out to drain |
| Hardness leakage | Often specified as a percentage of feed hardness | The realistic acceptance criterion; zero leakage is not achievable in a simple cycle |
Sizing it, in the order the calculation actually runs
| Step | What to calculate | Common error |
|---|---|---|
| 1. Hardness load per cycle | Flow rate x hardness (as CaCO3 equivalent) x hours between regenerations | Using mg/L as Ca rather than as CaCO3 – the two differ by a factor of 2.5 |
| 2. Working capacity | Take the resin’s total capacity and apply a utilisation factor | Designing on the total exchange capacity, which overstates what a real cycle delivers |
| 3. Resin volume | Load divided by working capacity | Forgetting that the bed must also give an adequate depth at the chosen vessel diameter |
| 4. Salt per regeneration | Resin volume x dose, then check against the plant’s salt budget | Sizing the vessel without checking the salt bill, which is the dominant operating cost |
| 5. Backwash and rinse water | Volumes per cycle, added to the water balance | Ignoring rinse water, which can be 5-10% of treated volume |
Where softening plants fail
| Failure | Cause | What to do |
|---|---|---|
| Hardness leaking early in the cycle | Flow too high, bed too shallow, or channelling from a fouled or compacted bed | Check flow, examine the bed for channelling, backwash properly |
| Salt consumption creeping up | Resin fouling, so more salt is needed for the same capacity; or brine quality has changed | Test the resin, inspect the brine system, check injector performance |
| Iron fouling | Iron in the feed, or corrosion products from upstream pipework | Remove iron upstream; clean with reducing agent and acid; consider a macroporous resin |
| Chlorine attack | Free chlorine in the feed oxidising the polymer | Bisulphite dosing or carbon upstream – capacity loss from oxidation is not recoverable |
| Organic fouling | Humic substances in surface water | Macroporous resin, caustic-brine cleaning, or organics removal upstream |
| Resin loss to drain | Backwash rate above the settling velocity, or a broken distributor | Recalculate the backwash from wet true density; inspect nozzles |
Choosing the grade
| Situation | Reasonable choice | Reason |
|---|---|---|
| Clean municipal or well water | Gel type SAC, standard particle size | Highest capacity, lowest cost; the feed will not foul it |
| Surface water with organics | Macroporous SAC | Tolerates fouling and can be cleaned |
| High hardness with high alkalinity | Consider a weak acid resin instead of, or ahead of, SAC | Much higher capacity and far less acid; see the dealkalisation route |
| High flow, tight pressure drop | Uniform particle size SAC | Lower pressure drop and more even flow, at higher cost |
| Potable or food use | Grade with the relevant approval | The approval is part of the specification, not an optional extra |
For the resin itself see strong acid cation exchange resin; for the alkalinity-driven alternative, weak acid cation exchange resin; and for the overall framework, the resin selection guide.