Understanding TDS and hard water: what the meter tells you

TDS stands for total dissolved solids: everything dissolved in the water that would be left behind if you evaporated a sample to dryness. In a laboratory that is exactly how it is measured — gravimetrically, by weighing the residue.

A handheld meter does nothing of the sort. It measures electrical conductivity, because dissolved ions carry current and pure water barely does, and then multiplies that conductivity by a fixed conversion factor — typically somewhere between 0.5 and 0.7 depending on the scale the manufacturer chose. The number on the display is a proxy, calibrated against a reference salt solution that is almost certainly not what is in your tap.

That matters in one practical way: two waters with genuinely identical TDS can read differently on the same meter if their ionic make-up differs, and the same water can read differently on two meters using different conversion factors. For deciding whether you need a membrane, this imprecision is irrelevant — you are looking at whether the number is 150 or 1,200, not whether it is 148 or 156. For anything finer, it is the wrong instrument.

This is where the category oversells hardest, so it is worth being blunt. The meter is blind to:

• Bacteria, viruses and protozoa. Micro-organisms are not dissolved ions and contribute essentially nothing to conductivity. A reading of 40 mg/L tells you nothing at all about whether the water is microbiologically safe.
• Which ions are present. Calcium bicarbonate, sodium chloride and nitrate all push the number up. Benign hardness and a genuine contamination problem look identical on the display.
• Heavy metals at concentrations that matter. Lead, arsenic and similar are dangerous in micrograms per litre, thousands of times below the resolution of a meter reading in milligrams per litre. They are invisible here.
• Most organic contaminants. Pesticides, solvents and many VOCs are poorly ionised or not ionised at all, so they barely move the reading.
• Chlorine taste. The residual that makes municipal water smell of a swimming pool contributes very little to conductivity, and carbon removes it without changing the TDS figure meaningfully.

The practical rule that follows: a high reading is informative, a low reading is not reassuring. High TDS tells you there is a lot dissolved and gives you a reason to act. Low TDS only tells you there is not much dissolved — which rules out one class of problem and leaves every other class exactly where it was.

Two things are commonly confused here. The WHO does not set a health-based guideline value for TDS at all: it treats TDS as an acceptability parameter, a matter of taste rather than safety. The palatability bands widely quoted from the WHO background document on TDS run roughly like this — under 300 mg/L excellent, 300 to 600 good, 600 to 900 fair, 900 to 1,200 poor, above 1,200 unacceptable. Those are descriptions of how water tastes to a panel, not thresholds for harm.

National standards do set limits. In India, BIS IS 10500:2012 gives TDS an acceptable limit of 500 mg/L and a permissible limit of 2,000 mg/L where no alternative source is available. In China, GB 5749 sets a TDS limit of 1,000 mg/L. Those are the numbers to check your reading against, and which one applies depends on where you live, not on which is more convenient.

The last row is where this catalogue lives. Twelve of the 13 machines here are rated for high-TDS feed, and the one that is not is the gravity filter. What that means for your choice is worked through in do you need reverse osmosis .

1. 1

   Run the tap first

   Let the cold tap run for thirty seconds before you sample. Water that has sat in the pipe overnight picks up whatever the pipe had to give it and is not representative of your supply.

2. 2

   Sample into a clean glass, at room temperature

   Conductivity rises with temperature. Decent meters compensate automatically, cheap ones compensate badly, and either way a sample that has sat for a couple of minutes gives a steadier reading than one straight off a cold tap in winter.

3. 3

   Read it, then read it again on another day

   Supplies vary — seasonally on groundwater, and day to day where a building tank or a tanker is involved. One reading is a data point; three readings across a fortnight are a picture.

4. 4

   Test hardness separately with a strip

   A hardness strip reports calcium and magnesium specifically, as mg/L of CaCO3 equivalent. That is the number your kettle cares about, and the meter cannot give it to you.

5. 5

   Write the numbers down with the date

   After you install a purifier, a reading of its output is your baseline. When that baseline starts drifting upward, the membrane is ageing — and you will know months before anything tastes different.

Hardness is the concentration of dissolved calcium and magnesium, conventionally reported as an equivalent concentration of calcium carbonate. It is a subset of TDS, not a synonym for it, and the difference has consequences you can see in your kitchen.

Hardness is what scales your kettle: heating the water drives calcium carbonate out of solution and it deposits on the element as a white crust. It is also what defeats soap — calcium and magnesium react with soap to form an insoluble scum, which is why hard-water households use more detergent and why glassware comes out of the machine cloudy.

Here is the case that breaks the shortcut: a brackish or coastal supply dominated by sodium chloride can have very high TDS and low hardness. It will taste salty and it will not scale your kettle at all. So "my kettle is clean, therefore my water is fine" is not a safe inference. The water finder uses the kettle-scale question because it is the one observation nearly everyone can make honestly — but it asks about your source in the same breath, for exactly this reason.

For reference: BIS IS 10500:2012 puts total hardness, as CaCO3, at an acceptable limit of 200 mg/L and a permissible limit of 600 mg/L. GB 5749 in China sets 450 mg/L. Above those you are in territory where an RO membrane is doing visible work — and where a separate softener may be the better answer for washing water, since you do not need drinking-quality water in the shower.

Dissolved minerals are a large part of what water tastes like. Strip almost all of them out and what is left is technically excellent and noticeably dull — the flat, slightly empty quality people describe after switching to RO. This is the same effect the WHO palatability bands describe at the bottom end: very low TDS is rated less palatable, not more.

That is why nearly half the machines in this catalogue carry a mineraliser or copper stage: it adds a controlled amount of mineral content back after the membrane has removed it. The sequence is admittedly odd — remove, then replace — but it is an honest response to a real complaint, and on genuinely high-TDS feed water there is no alternative, because the membrane cannot be selective about which ions it keeps.

If taste is what you care about most, the prepared shortlists split on exactly that: mineral retention on municipal water and mineral retention on borewell water under ₹15,000 .

A meter is the right tool for a screening question: is there a lot dissolved in this water, yes or no. It is the wrong tool for every question that names a specific substance. Pay for a laboratory analysis when:

• You are on borewell or other untreated groundwater — nitrate, fluoride and arsenic are all plausible, all invisible to a meter, and all worth knowing about before you choose a machine.
• The building has old plumbing, or you have any reason to suspect lead. Metals that matter at microgram levels cannot be screened with a conductivity meter.
• Someone in the household is an infant, pregnant, or immunocompromised, and you want microbiological certainty rather than an inference.
• Your TDS reading is high and you are about to spend a lot. A one-off test costs a fraction of a purifier and tells you what you are actually buying a machine to remove.
• The reading, the taste and the appearance disagree with each other. That combination usually means something the meter is not seeing.

Once you have the picture, the rest of the decision — technology, capacity, tank or tankless, upkeep — is laid out in how to choose a water purifier , with the running-cost arithmetic in water purifier running costs .