Mixing know-how
Brewing Water: How Hardness, TDS and Residual Chlorine Shape a Drink
bartender.com.vn · 31/07/2026 · 11 min read

Tap water that meets the standard is safe, not necessarily tasty. Hardness, TDS and chlorine thresholds, and how to pick a bar water filter.
Quick summary
Tap water that meets a national standard is SAFE to drink, not necessarily GOOD to brew with — those are two different goals. Vietnam's regulation sets health ceilings: QCVN 01-1:2024/BYT allows hardness up to 300 mg/L CaCO3 and total dissolved solids (TDS) up to 1,000 mg/L. Sensory thresholds — the point where a guest actually notices — sit far lower: the World Health Organization records that the taste threshold for the calcium ion is in the range of 100–300 mg/l and that water becomes noticeably unpalatable above roughly 1,000 mg/L TDS.
Put differently: two bars can use compliant tap water, the same recipe and the same beans, and still serve two different-tasting coffees — with neither of them breaking any rule. This article covers the four numbers a bartender should be able to read (hardness, TDS, residual chlorine, pH), the laboratory evidence on how far water minerals actually shift coffee flavour, and how to choose water treatment for a bar without wasting money.
Scope: water used for BREWING (coffee, tea, syrups, mocktails) and water used for ICE. This article does not cover alcoholic drinks.
Last updated: 31/07/2026.
| Parameter | Figure | Source |
|---|---|---|
| Hardness (CaCO3) — Vietnam legal ceiling for domestic water | 300 mg/L | QCVN 01-1:2024/BYT |
| TDS — Vietnam legal ceiling | 1,000 mg/L | QCVN 01-1:2024/BYT |
| Free residual chlorine — mandatory range in Vietnam | 0.2 – 1.0 mg/L | QCVN 01-1:2024/BYT |
| Chlorine detection threshold for the most sensitive people | as low as 0.3 mg/L | WHO, Guidelines for drinking-water quality, 2026 |
| Hardness at which scale deposition starts | above approximately 200 mg/L CaCO3 | WHO, Guidelines for drinking-water quality, 2026 |
| TDS described as tasteless in sensory testing | 25 – 50 mg/L | WHO, Nutrients in drinking water, 2005 |
What compliant water does and does not mean
Compliant tap water means the water has been tested and sits below health-based ceilings — it does not mean the water suits brewing. The regulation in force is QCVN 01-1:2024/BYT, issued with Circular 52/2024/TT-BYT dated 31 December 2024. Its foreword states plainly that it replaces the earlier national standard QCVN 01-1:2018/BYT. Any venue still quoting the 2018 version in its hygiene file should update it — this is an easy detail to miss.
The regulation splits parameters into Group A (water utilities must test these routinely: pH, turbidity, residual chlorine, arsenic, coliforms) and Group B (tested per local technical regulation: hardness, TDS, chloride, sulfate). The detail that matters for hospitality: **hardness and TDS — the two figures that most affect taste and equipment — sit in Group B**, so they are not necessarily tested everywhere. If you want to know your own numbers, you measure them or you ask the utility.
The table below lists the thresholds that bear directly on brewing, taken from Article 4 of the regulation.
| Parameter | Unit | Permitted limit | Group |
|---|---|---|---|
| pH | — | within 6.0 – 8.5 | A |
| Turbidity | NTU | 2 | A |
| Free residual chlorine | mg/L | within 0.2 – 1.0 | A |
| Hardness, as CaCO3 | mg/L | 300 | B |
| Total dissolved solids (TDS) | mg/L | 1,000 | B |
| Chloride (Cl-) | mg/L | 250 (or 300) | B |
| Sulfate | mg/L | 250 | B |

Hardness: the number that decides both flavour and equipment life
Hardness is the amount of dissolved calcium and magnesium, expressed as milligrams of calcium carbonate per litre. It is the one figure that shapes taste and also determines how soon your espresso machine needs descaling — which is why it is worth measuring first.
The most widely used classification comes from the WHO background document: concentrations below 60 mg/l is generally considered as soft; 60–120 mg/l, moderately hard; 120–180 mg/l, hard; and more than 180 mg/l, very hard. Vietnam's 300 mg/L ceiling therefore sits deep inside the 'very hard' band — perfectly legal, but a band in which any heated appliance will scale quickly.
On perception, the WHO Guidelines for drinking-water quality (fourth edition incorporating three addenda, 2026) state that the taste threshold for the calcium ion falls between 100 and 300 mg/l depending on the associated anion, that the magnesium threshold is probably lower than calcium's, and that in some instances consumers tolerate hardness in excess of 500 mg/l. In other words, 'noticeable' depends heavily on what your guests are used to locally.
The other direction has its own trap. The same guidelines note that soft water below 100 mg/l has low buffering capacity and can be more corrosive to pipework, while the hardness background document says demineralized water tends to have a flat taste. That is precisely why many venues install reverse osmosis and then find their coffee thin and their tea hollow.
| Hardness (mg/L CaCO3) | Class (WHO) | Operational meaning for a bar |
|---|---|---|
| Below 60 | Soft | Little scale, but low buffering; flavour can read flat if too low |
| 60 – 120 | Moderately hard | The most comfortable band for both flavour and equipment |
| 120 – 180 | Hard | Descaling needs to go on a schedule |
| Above 180 | Very hard | Scales fast; treat before the water reaches the machine |
| Above approximately 200 | (separate WHO GDWQ marker) | WHO notes this is where scale deposition in systems and pipework may occur |
| Up to 300 | (Vietnam legal ceiling) | Still compliant — but already in the very hard band above |
TDS: water can also be too pure
TDS measures total dissolved solids, and for brewing both extremes are a problem. At the high end, WHO states that the palatability of water with a TDS level of less than about 600 mg/l is generally considered to be good, and that drinking-water becomes significantly and increasingly unpalatable at TDS levels greater than about 1000 mg/l. Compare that with Vietnam's 1,000 mg/L ceiling: water that is compliant right at the limit is exactly water at the point WHO describes as becoming unpalatable.
At the low end the data is even more specific. The WHO report Nutrients in drinking water (2005) cites sensory work in which water with a TDS of 25–50 mg/L was described as tasteless. The same report records that the WHO expert group of 1980 recommended a minimum TDS of 100 mg/L and an optimum of about 200–400 mg/L for chloride-sulphate waters and 250–500 mg/L for bicarbonate waters. These are general health and palatability recommendations rather than a coffee-specific standard — but they give a far more sensible reference frame than the 'lower is better' story reverse osmosis marketing tends to tell.
That report also gives mineral floors worth noting: a minimum of 20 mg/L calcium with an optimum around 50 mg/L (roughly 40–80), and a minimum of 10 mg/L magnesium with an optimum around 20–30 mg/L. For a bar running bare reverse osmosis water, that is the technical argument for adding a remineralisation stage after the membrane rather than using the permeate directly.
One safety note that is rarely known: WHO warns that drinking-water containing both magnesium and sulfate at high concentrations (above approximately 250 mg/l each) can have a laxative effect. For a venue serving at volume, that is reason enough to know your source rather than assume tap water is tap water.
| TDS range (mg/L) | Description in source | Consequence for a bar |
|---|---|---|
| 25 – 50 | Described as tasteless (WHO 2005) | Bare RO permeate often lands here — coffee and tea lose depth |
| Below 100 | Below the minimum TDS recommended in the 1980 WHO report | Remineralise before using for brewing |
| 200 – 400 / 250 – 500 | Optimum range per the 1980 WHO recommendation (by water type) | A sensible reference target when designing treatment |
| Below about 600 | Palatability generally considered good (WHO GDWQ) | Sensorially safe |
| Above about 1,000 | Significantly and increasingly unpalatable (WHO GDWQ) | Exactly the Vietnam legal ceiling — treatment needed |
Residual chlorine: guests smell it before they taste anything
Chlorine is the cheapest problem to fix and the one that damages flavour hardest, because the nose catches it before the palate gets to work. Vietnam's regulation sets not just a ceiling but a floor: free residual chlorine must stay **within 0.2 – 1.0 mg/L**, because below that the water loses its disinfecting capacity along the distribution network. That is an important distinction — tap water arriving at your venue is REQUIRED to contain chlorine; that is design, not a fault.
The problem is that the perception threshold sits inside that mandatory band. WHO writes that most individuals are able to taste or smell chlorine in drinking-water at concentrations well below 5 mg/l, and some at levels as low as 0.3 mg/l. So a fully compliant sample at 0.3–0.5 mg/L is still enough for a share of guests to notice chlorine in an iced tea or a long black.
Removing chlorine is the cheapest link in the chain: an activated carbon stage ahead of the machine is the industry-standard answer. What matters is remembering that activated carbon barely touches hardness — it captures chlorine and odour, not calcium. Plenty of venues fit carbon, assume water is 'handled', and keep accumulating scale inside the machine.
The last figure worth a glance is turbidity: the regulation permits up to 2 NTU, while WHO notes that below 4 NTU turbidity can be detected only using instruments. Water that looks crystal clear therefore tells you nothing about minerals or chlorine — the eye is not a water test.

How far do water minerals really shift coffee flavour
The short answer from the most recent research: at mineral levels typical of drinking water the effect on coffee's organic acids is SMALL, and only at ten times those levels does the effect become pronounced. More surprising still, most of the change appears to happen AFTER brewing rather than during extraction.
A study published in Heliyon in February 2024 by Bratthäll, Figueira and Nording at Umeå University, Sweden added magnesium chloride and calcium chloride to coffee at 100 ppm and 1,000 ppm, both pre-brew and post-brew, then measured four organic acids (citric, malic, lactic, quinic) by gas chromatography–mass spectrometry and nuclear magnetic resonance. The authors' own highlights are worth reading literally: regular levels of divalent ions displayed small effects on acid content in coffee; ten-fold higher levels displayed larger effects; extraction of acids in coffee seems to proceed independent of divalent ions; and perceived flavour is most likely due to interactions taking place post-brew.
The specific figures: at 1,000 ppm, magnesium chloride significantly reduced citric, malic and quinic acid by 40–60%, while calcium chloride cut citric acid by about 70% and malic acid by 25–40%. At 100 ppm — the level typical of drinking water — the samples showed acid contents close to the control brewed with pure water.
The practical takeaway for a bar is a sober one: do not expect a few dozen ppm of minerals to rescue a badly dosed or badly roasted coffee. Water minerals are a background variable, not a flavour dial. The variables you control far more directly — ratio, temperature, contact time, ice quality — deserve your attention first. If your bar is running cold brew or matcha latte, the order of priority is still recipe first, water second.
| Condition | Citric acid | Malic acid | Quinic acid |
|---|---|---|---|
| 100 ppm MgCl2 or CaCl2 (typical drinking-water level) | limited variation | limited variation | limited variation |
| 1,000 ppm MgCl2 | down 40 – 60% | down 40 – 60% | down 40 – 60% |
| 1,000 ppm CaCl2 | down about 70% | down 25 – 40% | slightly reduced |
| Authors' conclusion | acid extraction seems to proceed independent of divalent ions | — | — |
Choosing a water treatment setup for your bar
The selection rule is compact: treat what is actually wrong, leave alone what is already fine. Measure first, buy second — doing it the other way round is the fastest way to pay for a machine that solves someone else's problem.
Four common configurations and what they genuinely do are summarised below. The 'relative cost' column is an ESTIMATE based on common hospitality practice with NO published source — use it to compare options, not to build a budget.
For most venues in Vietnam the setup worth considering first is activated carbon plus one hardness-control step, rather than jumping straight to reverse osmosis. The reasoning: chlorine is almost certainly present (the regulation requires it), while hardness depends on the source — and if hardness already sits in the 60–120 mg/L band, installing RO simply creates a new problem, namely having to remineralise afterwards.
Mobile bars at events face a different problem entirely: the water source changes with every venue, so nothing can be plumbed in permanently. The practical answer is to bring bottled water of known composition for the flavour-sensitive drinks and use the on-site supply only for washing. Volume planning per guest is covered in our guide to building an event drinks menu.
| Configuration | Removes chlorine/odour | Reduces hardness | Reduces TDS | Main risk | Relative cost (estimate) |
|---|---|---|---|---|---|
| No treatment (tap water direct) | No | No | No | Chlorine in the glass; scale building inside machines | Lowest |
| Activated carbon stage | Yes | No | Almost none | False sense that water is handled while scale accumulates | Low |
| Ion exchange softener (cation) | No | Yes | No (calcium swapped for sodium) | Higher sodium; unstabilised soft water can be corrosive (WHO) | Medium |
| Bare reverse osmosis (RO) | Yes | Yes | Yes, strongly | TDS falls into the band described as tasteless (WHO 2005) | High |
| RO plus remineralisation stage | Yes | Controllable | Controllable | More complex; needs disciplined maintenance | Highest |

A 30-minute water check for a new venue
The first thing to do when you take on a new site is measure, not shop. The three checks below take about half an hour and settle every decision that follows.
Step one, request the latest water test report from your utility and find exactly four lines: pH, free residual chlorine, hardness as CaCO3, and TDS. Remember that hardness and TDS sit in Group B, so they may be missing from the report — if they are, you measure them yourself.
Step two, measure at the tap with a TDS pen and a drop-test hardness kit. Take readings twice in one day: early morning after the water has stood in the pipes overnight, and again at peak service. The gap between the two tells you how much your building's own plumbing contributes to the final number.
Step three, run a controlled sensory test: brew the same recipe with three waters — tap water, tap water through activated carbon, and a bottled water of known composition. Taste blind and write it down. If the three cups are indistinguishable, water is not your venue's problem and the money should go elsewhere.
Finally, log the results in your operations file with the date, and repeat every quarter or whenever the water supplier changes. Water is the only ingredient in the bar whose specification can change without anyone telling you.
The bottom line: compliant water is the condition for safety, but the taste of a drink is decided in the gap between the legal ceiling and the sensory threshold — and that gap closes only by measuring, never by trusting a filter's label.
FAQ
Is bottled water automatically better than tap water for brewing?
No. Bottled water is only better in that its composition is stable and it has usually lost its chlorine, but the TDS of some purified waters lands in the band WHO records as being described tasteless (25–50 mg/L). What matters is knowing your numbers, not the word 'purified' on the label.
Does boiling soften water?
Boiling only handles temporary hardness, the part caused by bicarbonates, because heat precipitates calcium carbonate. Permanent hardness from sulfates and chlorides stays in the water. That white crust inside your kettle is the calcium that just came out.
Is reverse osmosis the best answer for a coffee shop?
Not necessarily. RO cuts TDS hard and can take water below the 100 mg/L minimum WHO recommends for drinking water, leaving drinks hollow. If you use RO, add a remineralisation stage after it and aim back toward the 200–400 mg/L reference band.
Why must tap water contain chlorine at all?
Because residual chlorine is the disinfection barrier along the pipes after water leaves the treatment plant. QCVN 01-1:2024/BYT requires free residual chlorine within 0.2 – 1.0 mg/L, so there is a floor as well as a ceiling. Removing it is the venue's job, at the point of use.
Does clear-looking water mean good brewing water?
No. WHO notes that below 4 NTU turbidity can be detected only using instruments. Hardness, TDS and chlorine are all invisible, so looking at water is no substitute for measuring it.
How often should water be retested?
Quarterly is reasonable for a fixed venue, plus an immediate retest whenever the supplier changes, after a mains break in the area, or when filter cartridges are replaced. Mobile bars have to test or bring their own water for each site.
Sources
QCVN 01-1:2024/BYT — National Technical Regulation on Domestic Water Quality, issued with Circular 52/2024/TT-BYT dated 31 December 2024, replacing QCVN 01-1:2018/BYT. Full-text PDF read directly: view document (the government portals vanban.chinhphu.vn and vbpl.vn block automated access, so an accessible PDF copy was used; every figure quoted comes from Article 4 and the foreword of that document).
World Health Organization — Guidelines for drinking-water quality: fourth edition incorporating the first, second and third addenda. Geneva: WHO; 2026. ISBN 978-92-4-012122-5: iris.who.int.
World Health Organization — Hardness in drinking-water. Background document for development of WHO Guidelines for drinking-water quality (WHO/HSE/WSH/10.01/10/Rev/1), 2011: iris.who.int.
World Health Organization — Nutrients in drinking water. Geneva: WHO; 2005 (chapter III: Desirable mineral content of demineralised drinking water): iris.who.int.
Bratthäll T, Figueira J, Nording ML — Influence of divalent cations on the extraction of organic acids in coffee determined by GC-MS and NMR. Heliyon. 2024 Feb 17;10(5):e26625. doi:10.1016/j.heliyon.2024.e26625: pmc.ncbi.nlm.nih.gov.
Planning an event?
Get a free quote
Tell us about your event and our team will recommend the right package and send a detailed quote — no commitment.


