Water Softener Sizing Calculator

Hardness is dissolved calcium and magnesium, picked up as groundwater moves through limestone, chalk and dolomite. It is not a contaminant in any health sense — the EPA does not regulate it, and hard water is perfectly safe to drink. It is a nuisance, and an expensive one.

The nuisance comes from what those minerals do when the water is heated or evaporates. They precipitate out as scale, a hard deposit that coats the inside of pipes, kettles, dishwashers and above all water heaters. Scale is an excellent insulator, so a scaled heating element has to work harder to move the same heat into the water, and the appliance loses efficiency steadily over its life. Hard water also reacts with soap to form the insoluble curd that leaves films on glassware and a residue on skin and hair, and it is the reason soap lathers poorly in some parts of the country and readily in others.

Hardness is reported either as grains per gallon or as parts per million of calcium carbonate, and one grain per gallon is 17.1 ppm. Anything under 1 gpg is soft; 1 to 3.5 slightly hard; 3.5 to 7 moderately hard; 7 to 10.5 hard; and above 10.5 very hard. Large parts of the American midwest and southwest sit well above 10.

A conventional water softener is a tank of small polystyrene beads — the resin — each carrying a negative charge, initially loaded with sodium ions. Hard water passes through the bed, and because calcium and magnesium ions carry a double positive charge, they are held more strongly by the resin than the single-charged sodium is. The resin grabs them and releases sodium in exchange. Water leaves the tank with its hardness swapped for a small amount of sodium.

That process is finite. Eventually every site on every bead holds calcium or magnesium and the resin is exhausted — this is the capacity, measured in grains. At that point the unit regenerates: it draws a strong brine solution from the salt tank and floods the resin with it. The sheer concentration of sodium ions reverses the exchange by mass action, driving the calcium and magnesium off the beads and out to the drain, and reloading the resin with sodium.

This is why softeners consume salt, and why they discharge water. Neither is optional; both are the mechanism. And it is why the two numbers on a softener that actually matter are its capacity in grains and its efficiency in grains recovered per pound of salt.

Softeners are sold by grain capacity — a 32,000 grain unit, a 48,000 grain unit — and that number is very nearly always the capacity at maximum salt dose. The trap is that nobody should run one there.

A cubic foot of standard resin will deliver about 32,000 grains if you regenerate it with 15 pounds of salt. The same cubic foot delivers about 24,000 grains on 6 pounds. So tripling the salt buys a third more capacity, and salt efficiency collapses from roughly 4,000 grains per pound to about 2,100. Run at the high dose the unit lives up to its badge and costs nearly twice as much to operate for the same water treated.

The consequence for sizing is that the badge number is not the number to size against. If you want 31,500 grains between regenerations and intend to run efficiently, you need a badge rating around 42,000, not 32,000. Sizing to the badge is the most common reason a new softener regenerates twice as often as the dealer promised — the customer was quoted maximum-salt capacity and the installer, sensibly, set it up to run efficiently.

The general rule that falls out of this: buy a larger unit than you think and run it gently. It costs a little more upfront, uses substantially less salt, regenerates less often, and lasts longer.

Softeners do remove some dissolved iron, which is why they are often sold as a solution to orange staining on private wells. Within limits that works. Beyond those limits it fails in a way that damages the equipment.

The problem is that iron does not simply occupy exchange sites the way calcium does. Dissolved ferrous iron oxidises on contact with air or oxidising agents, and inside a resin bed it precipitates as ferric hydroxide — a rust coating that physically fouls the beads rather than sitting on their exchange sites. Regeneration with brine does not remove it. Over months the resin loses capacity permanently, and the only fix is a resin cleaner or a rebed.

The convention of adding 4 grains of apparent hardness per ppm of iron exists to account for this disproportionate impact, and it works up to a point. Above roughly 3 ppm of iron, the accepted guidance is that a softener should not be the primary treatment: an iron filter — air injection, greensand, or a chemical oxidiser depending on the water chemistry — belongs ahead of it. Manganese behaves similarly and causes black rather than orange staining.

Anyone on a private well should test before buying. Municipal water is treated for iron before it reaches the house; well water frequently is not, and the difference determines whether a softener alone is the right purchase or a component of a larger system.

Ion exchange adds sodium to the water, in proportion to the hardness it removes — approximately 7.5 milligrams per litre for each grain per gallon exchanged. On 15 gpg water that is around 113 mg/L, which for someone drinking two litres a day is roughly 225 mg of additional sodium. That is a small fraction of a typical daily intake and irrelevant for most people, but it is not nothing for someone on a medically restricted sodium diet.

Two answers exist. Potassium chloride can be substituted for sodium chloride in the brine tank and works identically, adding potassium instead of sodium — it costs several times more and is worth it where sodium is a genuine concern. Alternatively, and more commonly, the cold tap at the kitchen sink is plumbed upstream of the softener so drinking and cooking water is untreated, while everything feeding appliances and showers is softened.

A separate category of products is sold as "salt-free softeners", and the name is misleading. Template-assisted crystallisation and similar technologies do not remove hardness at all; they convert dissolved calcium into microscopic crystals that are less inclined to adhere to surfaces. Scale formation is reduced, which is a real benefit for a water heater. But the water remains chemically hard: soap will still not lather, and spotting on glassware continues. They are a scale-control device rather than a softener, and evaluating them as such avoids disappointment.

Older and cheaper softeners regenerate on a clock — every three days, say, regardless of how much water has been used. This is simple and wasteful. A household away for a week still regenerates; a household with guests exhausts the resin early and delivers hard water until the next scheduled cycle.

A demand-initiated or metered unit counts the gallons that have passed through and regenerates when capacity is genuinely used up. It cuts salt and water consumption substantially for a household with variable usage, and it is close to standard on anything but the cheapest units. It is worth insisting on.

One refinement matters even on a metered unit: most include a forced regeneration after some maximum interval, typically 7 to 14 days, even if capacity remains. This is not a design flaw. A resin bed that sits saturated and undisturbed for weeks is a warm, wet, nutrient-bearing environment, and bacterial growth in stagnant softener resin is a documented problem. The forced cycle keeps the bed exercised. It is also why sizing for a very long regeneration interval — buying a huge unit to avoid handling salt — is a false economy.