What does a water hardness number actually mean? Converting ppm, °dH, °fH and grains per gallon

Key takeaways

• Hardness is a capacity, not a substance. WHO defines it as “the traditional measure of the capacity of water to react with soap,” caused “not by a single substance but by a variety of dissolved polyvalent metallic ions, predominantly calcium and magnesium cations” (WHO, Hardness in Drinking-water).
• Five units, one quantity. 1 French degree = 10 mg/L as CaCO₃; 1 grain per US gallon = 17.118 ppm; 1 German degree (°dH) = 10 mg/L CaO = 17.848 ppm; 1 English/Clark degree = 14.254 ppm; 1 mmol/L = 100.09 mg/L CaCO₃ (Wikipedia, Hard water).
• The classifications genuinely disagree. USGS calls 121–180 mg/L “hard” (USGS); the UK’s Drinking Water Inspectorate calls 150–200 mg/L “moderately hard” and reserves “hard” for 200–300 (DWI). Nobody had to reconcile them, because WHO concluded that “no health-based guideline value for water hardness is proposed.”
• “As Ca” and “as CaCO₃” differ by a factor of 2.5. Calcium carbonate is 100.09 g/mol against calcium’s 40.08, so a calcium figure must be multiplied by 2.497 before it can be compared with a hardness figure — and IS 10500 prints both conventions in the same table (IS 10500 : 2012).
• None of it transfers to blending water. Purified water for spirit reduction is specified in microsiemens per centimetre, not degrees of hardness, and conductivity-to-TDS conversion is itself an estimate: standards permit “a TDS constant of 0.55-0.7” (Fondriest).

A hardness number is a statement about how much calcium and magnesium a water carries, expressed as though all of it were calcium carbonate. The complication is that at least five conventions exist for saying that — milligrams per litre, French degrees, German degrees, English (Clark) degrees and grains per gallon — and the national classifications built on top of them do not agree with one another. Before a distillery, bottler or bar can decide whether a water report describes a problem, it has to establish which of those conventions the report is written in.

What is water hardness actually measuring?

It is a measure of behaviour, not composition. The WHO background document is explicit: “Water hardness is the traditional measure of the capacity of water to react with soap, hard water requiring considerably more soap to produce a lather. It is not caused by a single substance but by a variety of dissolved polyvalent metallic ions, predominantly calcium and magnesium cations, although other cations, e.g. barium, iron, manganese, strontium and zinc, also contribute” (WHO).

Because several different ions contribute, and because they have different atomic weights, there is no natural unit. The convention the industry settled on is to express the whole mixture as the mass of calcium carbonate that would produce the same effect: “Hardness is most commonly expressed as milligrams of calcium carbonate equivalent per litre, water containing less than 60 mg of calcium carbonate per litre generally being considered as soft” (WHO).

That single decision — report everything as CaCO₃ equivalent — is the source of nearly every misreading that follows. The water may contain no calcium carbonate at all.

Hardness also splits into two forms worth knowing by name, because they behave differently in a boiler and in a bottling hall: carbonate (temporary) hardness, associated with bicarbonate and driven off by heating, and noncarbonate (permanent) hardness, associated with chloride and sulphate, which is not.

Why does the same water get called “soft” in one country and “hard” in another?

Because no international body ever fixed the boundaries. WHO’s conclusion on hardness is unusually blunt: “Although a number of epidemiological studies have shown a statistically significant inverse relationship between the hardness of drinking-water and cardiovascular disease, the available data are inadequate to permit the conclusion that the association is causal. No health-based guideline value for water hardness is proposed” (WHO).

With no health value to anchor them, national scales grew out of local plumbing experience instead — and diverged.

Read across the 150 mg/L row and the problem is plain. The same water is hard to an American reader and moderately hard to a British one. At 100 mg/L it is moderately hard in the United States and soft in Britain — the gap is not a rounding difference but a whole category. Any procurement conversation conducted in adjectives rather than numbers is therefore unreliable across borders, and the regional water narratives attached to whisky styles deserve the same scepticism: “soft Speyside water” and “soft Kentucky water” are not claims made on the same scale.

What are the hardness units, and how do they convert?

Every unit below expresses the same physical quantity. They differ only in the reference mass and the reference volume chosen.

Definitions and factors from Wikipedia, Hard water, which compiles the standard conversions.

Two of those deserve a second look, because they are the ones that catch people out.

The two grain units are not the same grain unit. Both are defined as one grain of calcium carbonate per gallon, but the Imperial gallon is 4.55 litres against the US gallon’s 3.79. The result is a 20% discrepancy between the English degree and the American grain per gallon — two units with nearly identical names describing measurably different waters.

The German degree is not measured as calcium carbonate at all. It is defined as 10 mg/L of calcium oxide (CaO), which converts to 17.848 mg/L CaCO₃ because CaO’s molar mass is lower. This is why °dH sits so awkwardly close to the American gpg figure of 17.118 without being equal to it — a 4% error for anyone who treats them as interchangeable.

Worked through one water, at 150 mg/L as CaCO₃:

• 15.0 French degrees
• 10.5 English (Clark) degrees
• 8.8 grains per US gallon
• 8.4 German degrees
• 1.50 mmol/L

Five numbers spanning a tenfold range, all describing one sample.

What is the difference between hardness “as CaCO₃” and calcium “as Ca”?

A factor of roughly two and a half, and it is the single most consequential conversion on this page.

Calcium carbonate has a molar mass of 100.09 g/mol; calcium is 40.08 g/mol (Wikipedia). Converting a calcium figure into calcium-carbonate equivalent therefore means multiplying by 100.09 ÷ 40.08 = 2.497. For magnesium, whose molar mass is 24.31, the factor is 4.12 — higher, because each milligram of magnesium carries proportionally more charge.

So a report showing “calcium 80 mg/L” is describing a water of roughly 200 mg/L hardness before magnesium is even counted. A buyer who reads 80 mg/L as a hardness figure has understated the water by a factor of two and a half.

The Indian drinking-water standard demonstrates how easily the two conventions sit side by side in a single table. IS 10500 : 2012 sets acceptable limits of 75 mg/L for calcium (as Ca), 30 mg/L for magnesium (as Mg), 200 mg/L for total hardness (as CaCO₃) and 200 mg/L for total alkalinity as calcium carbonate, with hardness and alkalinity both relaxable to 600 mg/L “in the absence of alternate source” (IS 10500 : 2012).

Convert the first two into the units of the third and something instructive falls out. Calcium at 75 mg/L is 187 mg/L as CaCO₃; magnesium at 30 mg/L is another 124. A water sitting exactly on both individual acceptable limits would show a total hardness near 311 mg/L — comfortably over the 200 mg/L acceptable limit for hardness, though still inside the 600 mg/L relaxation. The limits are not inconsistent; they are simply written in two different currencies, and only one of them is the currency the hardness row is quoted in.

Where does alkalinity fit in, and why is it also reported as calcium carbonate?

This is the other CaCO₃ trap. Alkalinity — the water’s acid-neutralising capacity, mostly bicarbonate — is conventionally reported as mg/L of calcium carbonate too, which is why IS 10500 lists “total alkalinity as calcium carbonate” a few rows above “total hardness (as CaCO₃)” at the identical value of 200 mg/L (IS 10500 : 2012).

They are different properties measured on a shared scale. Hardness counts cations; alkalinity counts anions. A water can be high in one and low in the other — softened water is the standard example, retaining all of its alkalinity while its hardness goes to zero, which is precisely why a hardness figure cannot distinguish softened water from purified water. Seeing “200 mg/L as CaCO₃” on a certificate tells you nothing until you have read which row it sits on.

Why do none of these units appear on a blending water certificate?

Because the numbers are too small to be useful. Degrees of hardness are a scale built for waters carrying tens or hundreds of milligrams per litre of dissolved mineral. Water made for reducing spirit to bottling strength is built to carry essentially none, so its hardness is reported as “not detected” and the specification moves to a measure with resolution at the bottom of the range: electrical conductivity, in microsiemens per centimetre.

That scale has a hard floor. Genuinely pure water measures “0.055 uS/cm, or a resistivity of 18 megohms at 25 °C,” rising to “closer to 1 uS/cm” once it has equilibrated with air (Fondriest) — which is itself a useful fact, since a supplier quoting a figure below 0.055 µS/cm is quoting something that cannot exist.

Two cautions apply to conductivity figures, and they mirror the hardness problem exactly.

A conductivity-derived TDS is an estimate, not a measurement. Meters apply a multiplier — “a common, approximated constant around 0.65” — and standards permit “a TDS constant of 0.55-0.7,” with higher constants for calcium- or sulphate-rich sources (Fondriest). The reason is structural: “solutions with the same conductivity value, but different ionic constitutions (KCl vs NaCl vs 442) will have different total dissolved solid concentrations.” Two certificates reporting the same TDS may have measured the same conductivity and divided by different numbers, a point developed further in our guide to pH and TDS specifications.

A conductivity figure without a temperature is incomplete. Conductivity rises “2-4%” per degree Celsius in ordinary waters and “approximately 5% per degree Celsius” in nearly pure water (Fondriest). In the purified range where blending water lives, the temperature sensitivity is at its worst — so a reading is only meaningful if it states the reference temperature, conventionally 25 °C.

How should a buyer read a report that mixes units?

Four habits cover almost every case.

1. Convert everything to mg/L as CaCO₃ before comparing anything. Multiply °fH by 10, gpg by 17.1, °dH by 17.8, °e by 14.3. Do this before forming any view on whether a water is hard.
2. Check the reference species on every row, not the units column. “mg/L” is not a convention; “mg/L as CaCO₃” and “mg/L as Ca” differ by 2.5, and both may appear on one page.
3. Distrust the adjective, trust the number. “Soft” means below 60 mg/L to USGS and below 100 mg/L to the DWI. Ask for the figure.
4. For water that will touch spirit, ask for conductivity at a stated temperature rather than hardness at all. Hardness has no resolution in the purified range, and it is silent on the sodium, alkalinity, chloride and silica that a full certificate of analysis reports separately.

The underlying point is not that one convention is right. It is that hardness was never standardised internationally, because — as WHO concluded — there was never a health finding solid enough to force the issue. What the trade inherited instead is a set of regional habits that look like measurements. They convert cleanly once you know the factors, and they mislead reliably until you do.

Glossary

• Hardness — the concentration of dissolved polyvalent cations, principally calcium and magnesium, conventionally expressed as the mass of calcium carbonate that would have an equivalent effect.
• As CaCO₃ — a reporting convention, not a statement of composition. The water need not contain calcium carbonate.
• Carbonate (temporary) hardness — the fraction associated with bicarbonate, precipitated by heating.
• Noncarbonate (permanent) hardness — the fraction associated with chloride and sulphate, unaffected by heating.
• French degree (°fH) — 10 mg/L as CaCO₃.
• German degree (°dH) — 10 mg/L as CaO, equal to 17.848 mg/L as CaCO₃.
• English / Clark degree (°e) — one grain of CaCO₃ per Imperial gallon, equal to 14.254 mg/L.
• Grain per gallon (gpg) — one grain of CaCO₃ per US gallon, equal to 17.118 mg/L.
• Alkalinity — acid-neutralising capacity, mostly bicarbonate; also reported as mg/L CaCO₃, and a different property from hardness.
• Conductivity — the water’s ability to carry current, in µS/cm; the working purity measure for demineralised water, temperature-dependent and conventionally quoted at 25 °C.

Frequently asked questions

How do I convert grains per gallon to ppm? Multiply by 17.118. One grain per US gallon is one grain (64.8 mg) of calcium carbonate per US gallon (3.79 litres), which works out at 17.118 ppm (Wikipedia). Note that the English or Clark degree uses the same grain in an Imperial gallon and equals 14.254 ppm instead.

What is 1 °dH in mg/L? 17.848 mg/L as CaCO₃. The German degree is defined as 10 mg/L of calcium oxide rather than calcium carbonate, and the conversion follows from the difference in molar mass (Wikipedia).

Is 150 mg/L hard water? It depends whose scale you use — which is the point. USGS puts 121–180 mg/L in the “hard” band (USGS), while the UK’s DWI classifies 150–200 mg/L as “moderately hard” and does not use “hard” below 200 (DWI).

Why is hardness reported as calcium carbonate when the water contains calcium and magnesium? Because several different ions contribute hardness and they have different atomic weights, so a common reference was needed. Expressing everything as calcium-carbonate equivalent makes the contributions additive on one scale (WHO).

Is there a legal limit on water hardness? Not a health-based one. WHO states that “no health-based guideline value for water hardness is proposed” (WHO). National standards set aesthetic or operational limits instead — India’s IS 10500 : 2012 gives an acceptable limit of 200 mg/L as CaCO₃, relaxable to 600 mg/L in the absence of an alternate source (IS 10500 : 2012).

Does hardness matter for water used to dilute whisky? It matters for the feedwater and not for the finished product, because the treatment train is designed to remove it entirely. Calcium is what drives haze in a reduced spirit, so the specification for water meeting spirit is written in conductivity rather than degrees of hardness — see demineralised water specifications for spirit dilution.

Can I compare two suppliers by their hardness figures alone? No. Hardness is silent on sodium, alkalinity, chloride, silica, organics and microbiology, and a zero-hardness claim is equally true of softened water and of purified water — two very different products, as set out in softened water is not purified water.

Blenders International · Water Science series