Steeping the perfect read...
Quality takes time. One bean at a time.
Most Indian specialty coffee buyers carefully select beans, wait for the right rest window, and calibrate grind. Water, which makes up 98% of every cup, is left almost entirely to chance. Municipal tap water in most Indian cities runs well above coffee brewing targets, and the most common household fix, RO filtration, over-corrects in the opposite direction. This is a field guide to what Indian water looks like by city, what it is doing to extraction, and how to adjust.
Indian specialty coffee buyers are careful about sourcing. They track roast dates, research origins, and adjust grind by method. Water gets almost no attention, despite being 98% of every cup.
The mineral composition of brewing water determines what extracts from the grounds, which flavor compounds reach the cup, and how much acidity the water chemically neutralizes before the cup is ready to drink. Most Indian municipal water sits 2–6 times above the Specialty Coffee Association's recommended brewing range. The most common household response, reverse osmosis filtration, drops water so far below the floor that a separate set of extraction problems begins.
Both conditions degrade cup quality. This covers what Indian tap water looks like city by city, why alkalinity is the variable that most directly shapes the cup, and what adjustments are possible without specialist equipment.
Scope: Water chemistry and cup outcomes: TDS, alkalinity, and hardness in terms of what they do to the cup. City-level mapping of Indian water against SCA brewing standards, the RO overcorrection problem, and practical solutions by water source type. Equipment scale is addressed at the end.
Three properties matter for brewing: TDS, hardness, and alkalinity. They are related but distinct, and they affect the cup in different ways.
TDS (Total Dissolved Solids) measures all dissolved minerals combined, in mg/L or ppm. The SCA's target range is 75–250 mg/L, with 150 mg/L as the ideal. Above 250 mg/L, dissolved minerals compete with coffee compounds for the water's binding capacity; less coffee character dissolves, and the cup reads flat. Below 75 mg/L, water lacks the mineral carriers needed to pull flavor compounds into solution, and cups taste harsh and thin.
Hardness refers to dissolved calcium and magnesium. Both assist in extracting flavor compounds from the grounds, with magnesium showing slightly higher extraction efficiency than calcium in controlled research (Hendon and Colonna-Dashwood, 2014). The SCA targets hardness between 50–175 mg/L. Above 175, calcium deposits build on heating elements. Below 50, cups turn dry and astringent.
Alkalinity is the most important flavor variable of the three, per Scott Rao's 2023 analysis, and the most frequently misunderstood. It measures water's resistance to acidification, its buffering capacity, determined almost entirely by bicarbonate ions (HCO3⁻) in drinking water. When bicarbonate-heavy water is used for brewing, those ions neutralize the organic acids extracted from the coffee before they reach the palate. Citric, malic, and acetic acids, which produce brightness, fruitiness, and perceived sweetness in a well-extracted cup, are chemically suppressed. The result is a flat, one-dimensional cup, not because extraction failed but because the water cancelled what was extracted. SCA target alkalinity: 40–75 mg/L. Scott Rao's preference for light roast: 30–40 mg/L.
Hardness is not alkalinity. Both can be high in Indian tap water, but they affect the cup differently. Hard water builds scale in equipment; high-alkalinity water flattens the cup by neutralizing extracted coffee acids. An RO system removes both, but removes too much. This is why simple filtration alone doesn't resolve the brewing problem.
India's major specialty coffee cities span nearly the full range of water quality for brewing. Mumbai's lake-sourced supply sits close to the target. Hyderabad's groundwater is among the hardest in the country. Within-city variation can match the between-city gap — Bangalore is the clearest case, where the water source of a given building determines whether a brewer is working in range or well outside it.
That span is wide enough that brewing guidance calibrated for one city becomes actively misleading in another. Grind charts and ratio tables travel across Indian cities. Water guidance does not.
Mumbai has the most coffee-friendly municipal tap water of any major Indian city. BMC draws from lake catchments (Tansa, Vihar, Bhatsa), all low-mineral surface water. Treated TDS typically runs 80–150 mg/L, within the SCA acceptable range. Alkalinity is generally moderate. The main issue is chlorination, which affects smell and taste rather than mineral balance. Carbon filtration, or letting water stand uncovered for 30 minutes, removes chlorine adequately. Mumbai brewers using filtered tap water are often working near brewing-optimal water without realizing it.
Bangalore is the most complex situation. Supply splits between two sources.
Cauvery water from BWSSB (treated surface water) runs 150–300 mg/L TDS, modestly hard and manageable. Central and established neighborhoods — Indiranagar, Koramangala, Jayanagar, Basavanagudi — predominantly receive this. Borewell and tanker supply, common in newer and peripheral areas, runs 500–1,000+ mg/L TDS, with hardness well above SCA targets. Marathahalli, Electronic City, Whitefield, Sarjapur, and Hennur rely substantially on these sources.
Two Bangalore brewers can be using the same bag from the same roaster and working with water profiles that differ by 400–500 mg/L. A persistently flat cup in a Whitefield kitchen is not the same problem as a flat cup in Indiranagar.
Delhi NCR runs hard. Delhi Jal Board supply from river and canal sources runs 300–600 mg/L TDS; groundwater in Gurgaon, NOIDA, and outer NCR reaches 600–800 mg/L. The flat character common in Delhi-brewed specialty coffee is often a chemistry effect rather than an extraction failure.
Chennai shows high variability. Metro CMC water, sourced from reservoirs, runs 300–500 mg/L. Coastal zones and outer areas, including Tambaram, OMR, and Perungudi, face 600–1,000+ mg/L TDS from heavy borewell dependency. Among major Indian cities, Chennai presents one of the more difficult tap-water situations for specialty brewing.
Hyderabad groundwater averages approximately 1,207 mg/L TDS across the basin, per government survey data. GHMC municipal supply is treated and lower, but large areas of Cyberabad, Gachibowli, Kondapur, and Hitech City draw on borewell or tanker supply at much higher mineral loads.
Pune municipal water from dam sources runs 200–500 mg/L. Borewell-dependent areas like Hadapsar and fringe zones reach 200–450 mg/L hardness — better than Delhi or Hyderabad, more variable than Mumbai.
How to check without a lab: A basic TDS meter costs Rs 300–500 and tests water in seconds. Under 75 mg/L means too soft, likely over-filtered. Between 75–250 mg/L is the target range. Above 250 mg/L is above target; suppression likely. Alkalinity testing needs a separate kit (hardness test strips or alkalinity drops, ~Rs 200–400) for a fuller picture, but TDS is enough to identify which of the two problems applies.
Reverse osmosis filtration is the dominant home water treatment in Indian cities, driven by legitimate concerns about fluoride, arsenic, heavy metals, and bacterial contamination in groundwater and ageing distribution pipes. Most units produce water at 20–50 mg/L TDS, well below the SCA's minimum of 75 mg/L.
At 20–50 mg/L TDS, extraction behaves differently than expected. Water this depleted lacks the calcium and magnesium ions that carry aromatic compounds from ground coffee into solution. The cup comes out harsh, astringent, and dry — thin in a different way from under-extraction, but equally unsatisfying. With light roast coffees, the effect is more pronounced: the delicate organic acids and fruit-forward compounds that define Indian specialty light roasts require mineral presence to solubilize and hold in the cup.
The result is a household that has invested in premium water filtration ending up with water worse for coffee than the untreated municipal supply it replaced. This is probably more common in Indian specialty coffee circles than acknowledged, because the problem is invisible. RO output looks clean, tastes neutral, and passes every health measure fine. The buyer who has improved every other brewing variable — sourcing, freshness, grind, temperature — is quietly working with miscalibrated water.
The timing compounds it. The specialty coffee buyer demographic, urban and upper-middle-class, overlaps heavily with the household demographic that adopted RO filtration as kitchen infrastructure through the 2010s. Both categories grew in parallel, without the water-coffee interaction entering the content either community received.
The buyer most likely to invest in specialty beans is the same buyer most likely to already have an RO unit installed. Both industries scaled through the same decade; neither handed the other a reason to address the interaction.
Light roast coffees are more sensitive to water quality than dark roast.
The flavor compounds at stake are more chemically fragile. The organic acids, florals, and fruit-forward notes that light roast is valued for react readily with bicarbonate alkalinity. Dark roast flavor compounds (melanoidins, caramelized sugars) are more robust and less suppressed by high-alkalinity water.
Light roast also needs more extraction energy to begin with (denser beans, less porous structure). Water that is too hard suppresses fruit and brightness at the end of the extraction; water that is too soft fails to carry the extraction forward fully. The margin for error is narrower on both sides.
With dark or medium-dark roasts, untreated Indian tap water in the 300–500 mg/L range may produce acceptable results — the robustness of the roast masks the water deficit to a degree. With light roast, the same water produces a noticeably muted, flat cup. Roughly 30% of the ICB catalog (354 of 1,165 coffees) is light or light-medium roast. For this portion of the catalog, water quality is typically the first variable worth addressing before further grind or temperature refinement.
For nearly a third of ICB's catalog, water calibration carries more cup impact than grind adjustment at typical home brewing precision. It just isn't framed that way in the content this audience receives.
Most Indian specialty light roasts are natural or anaerobic processed: expressive, fruit-forward, and extraction-sensitive. A natural Chikmagalur light roast brewed with high-alkalinity tap water will taste substantially more suppressed and flat than a washed light roast under the same conditions, because there is more character to suppress.
The right adjustment depends on which water problem applies.
Hard tap water: 300–800+ mg/L TDS
This is the dominant situation in Delhi, Chennai, peripheral Bangalore, and Hyderabad.
Hard tap water benefits from dilution with RO or bottled water. Mixing 3–4 parts RO output with 1 part unfiltered tap typically brings the blend into the 75–150 mg/L range. A TDS meter confirms the result. If mixing is impractical, Bisleri (~116–150 mg/L TDS) is the most reliably available bottled option at or near the SCA target.
Pure RO output: 20–50 mg/L TDS
The situation in any Indian city where the household uses a standard home RO unit.
The simplest fix: mix RO output with unfiltered tap at 4:1, check with a TDS meter, and adjust the ratio until it reads 75–150 mg/L.
If mixing is impractical, use bottled water. Bisleri (~116–150 mg/L) works as a consistent substitute. Kinley (<50 mg/L TDS) is nearly as stripped as RO output and does not help. Himalayan and Qua natural mineral waters (sourced from springs, TDS ~150–250 mg/L) work well for light roast, carrying more magnesium than treated packaged water. The FSSAI label distinction matters here: "packaged drinking water" (processed, TDS 10–250 mg/L) and "natural mineral water" (from a declared source) behave differently in the cup.
For more precision, remineralize: dissolve 0.15g Epsom salt (magnesium sulfate, available at pharmacies) plus 0.025g baking soda (sodium bicarbonate) into 1 litre of RO water. This adds approximately 18 mg/L magnesium and 15 mg/L alkalinity, a rough approximation of the Barista Hustle light-roast recipe with ingredients available across India. A kitchen scale accurate to 0.1g is needed.
Mumbai soft tap: 80–150 mg/L TDS
The simplest situation. TDS is within or close to the SCA target range. Chlorination is the main issue. Let water stand uncovered for 30 minutes before brewing, or use a carbon block filter. RO is not needed and would over-strip. If buying bottled water for consistency, Bisleri works; Himalayan adds mineral character well suited for light roast.
Borewell or tanker water: 500–1,500+ mg/L TDS
Common in peripheral Bangalore (Whitefield, Sarjapur, Marathahalli), parts of Hyderabad, and outer Chennai.
Standard RO processing will strip borewell water to 20–50 mg/L, converting a hard-water problem into a soft-water problem. RO filtration is necessary first — borewell water above 500 mg/L is too hard to dilute practically — but the output then needs remineralization or tap mixing. Use the pure-RO approach above after filtering. Do not brew with unfiltered borewell water above 500 mg/L; flavor suppression at that mineral load affects even dark roast noticeably.
| Water situation | Typical TDS | Primary problem | Fix | Target after fix |
|---|---|---|---|---|
| Hard tap | 300–800+ mg/L | High alkalinity | Mix with RO (4:1 or 3:1) or use Bisleri | 75–150 mg/L |
| RO output | 20–50 mg/L | Too soft | Mix with tap (4:1) or remineralize | 75–150 mg/L |
| Mumbai soft tap | 80–150 mg/L | Chlorine | Carbon filter or 30-min rest | Already in range |
| Borewell/tanker | 500–1,500+ mg/L | Hard + contaminants | RO then remineralize | 75–150 mg/L |
Starting protocol for a new water situation: Measure TDS before and after any filter. If pre-filter TDS is under 250 mg/L and the water is from a municipal supply, start by brewing directly with chlorine removal and adjust from taste. If TDS is 250–500 mg/L, mix with RO and test. Above 500 mg/L, RO followed by remineralization is the practical path. One test, one variable at a time.
Calcium carbonate, the compound responsible for hard water's flat cups, precipitates when water is heated, depositing as white scale on kettle elements, moka pot chambers, and espresso machine boilers and group heads. In cities with water above 250 mg/L TDS, scale accumulates faster than standard manufacturer descaling schedules anticipate (most are written for European water at 100–200 mg/L).
Practical descaling frequency with daily use: under 150 mg/L TDS, follow the manufacturer schedule, roughly every 2–3 months; 150–300 mg/L, monthly; above 300 mg/L, every 2–3 weeks. For kettles, the white deposit on the heating element is a reliable indicator — descale when it's visible.
Citric acid (food-grade, available at pharmacies and grocery stores, Rs 50–100 per 250g) is the most practical descaling agent in India. Dissolve 30–50g in 1 litre of water, run through the machine or steep in the kettle for 30–60 minutes, then flush twice with clean water. White vinegar works but leaves residual odor. Commercial descaling tablets cost more with no quality advantage over citric acid for standard limescale.
Fixing brewing water for flavor and protecting equipment from scale are the same intervention.
