Water supply test kit

Measures what the building actually delivers — hardness, alkalinity, TDS and residual chlorine — so that every treatment decision downstream is made against a reading instead of an assumption.

This is the first entry in the category because it is the first step in commissioning, and the rule it carries governs everything after it: filtration is specified AFTER the supply is tested, never before. Specifying from a machine manual or a supplier's default bundle instead of from a test of the actual supply is the most expensive avoidable mistake in this section — scale is cumulative, it forms invisibly, and it eventually becomes a service call rather than a clean. The instruments are simple, and the confusion is never about how they work but about what each one is actually telling you. A TDS meter reads electrical conductivity and converts it to a parts-per-million figure using an assumed factor: it tells you HOW MUCH is dissolved and nothing whatever about WHAT. A hardness test titrates calcium and magnesium against an indicator that changes colour once they are all bound, and the drop count is the reading. An alkalinity test does the same for the carbonate buffering that resists a change in pH. A chlorine test shows whether a disinfectant residual is present, which decides whether carbon has a job to do and whether the line beyond it is now unprotected. Those are four different numbers and none of them substitutes for another: a supply can sit comfortably inside the TDS band and still be hard enough to scale a boiler within a year of opening, and a soft supply can be heavily buffered and make every coffee taste flat. The second thing the test has to establish is that the supply is not a constant. Network blends change with season and source, a building's storage tank changes what arrives at the tap, and a long unused branch changes it again. One reading at commissioning is a baseline, not a specification for the life of the site.

• Test BEFORE you specify anything, and treat that as the non-negotiable step it is. Every other decision in this category — carbon, softener, inhibitor, RO, blend — is downstream of a reading, and a package specified without one is a package specified for a supply you have never met. This is the sentence that prevents the most expensive failure on the site.
• Sample at the tap the equipment will actually be plumbed to, not at the nearest sink. A building can carry a softened supply for a boiler somewhere else, an unsoftened branch for the kitchen, a storage tank that changes everything, and a long unused run that adds its own character. Two taps in one venue can genuinely need different treatment, and only sampling both reveals it.
• Measure hardness AND alkalinity AND TDS, because they drive three different decisions. Hardness decides whether you need removal, inhibition or nothing. Alkalinity decides how much of the coffee's acidity survives the cup, and whether a plain sodium softener will fix your taste problem or leave it exactly where it was. TDS decides whether the water is inside the target band at all. Reading one and calling it a water test is the most common version of this mistake.
• Test more than once before you commit. A supply that is blended seasonally, or fed from a tank that turns over slowly, gives a different answer in a different month, and a treatment sized on the softest reading of the year is a treatment that fails in the hardest.
• Decide who owns repeat testing before opening, not after the first fault. A logged reading at each cartridge change is what turns a change interval from a calendar guess into a measured decision, and it is the only thing that makes a slow drift visible while it is still cheap.
• Buy the instrument the team will actually use. A drop test that lives in a drawer and needs five quiet minutes measures nothing on a bar that never has five quiet minutes; a strip that gets used weekly beats a laboratory-grade kit that gets used never. Match the instrument to the routine you can honestly commit to.
• Where the site is fed from a stored tank, treat the tank as part of the supply and sample downstream of it. Storage changes temperature, residual disinfectant and everything that grows in the absence of one, and a treatment chain specified from the incoming main is specified from water the bar never receives.
• Send one sample to an accredited laboratory at commissioning even if you never do it again. The bar-side kit tells you about scale; the laboratory tells you about corrosion, and the two lead to opposite treatment choices. Discovering an aggressive supply after the fittings have started staining is discovering it in the most expensive possible way.