Tap water, dechlorinator and source water

Everything in an aquarium is downstream of one decision nobody thinks of as a decision: what you filled it with.

Most keepers fill from the nearest tap, add a capful of conditioner, and never look further. Usually that is fine. But a surprising share of the problems that resist every other explanation — a pH that will not hold, a nitrate reading that will not come down, fish that look wrong for a day after every water change — begin at the tap rather than in the tank, and none of them are visible without going and finding out.

This page is about knowing what you are starting with.

Chlorine, chloramine, and why the old advice stopped working

Municipal water is disinfected, and for fish the disinfectant is the problem. Both compounds used damage gill tissue at tap concentrations, which is why an unconditioned water change is a recognised way to lose several fish at once — one of the answers on the why did my goldfish die page.

Free chlorine is volatile. Left standing it gasses off over roughly 24 hours, faster with aeration or agitation. This is where the traditional advice to leave a bucket overnight comes from, and for a long time it was sound.

Chloramine is chlorine bonded to ammonia, and it is deliberately stable. It persists through long distribution networks, which is exactly why a great many utilities switched to it, and it is still there a week later. Standing does nothing.

Three consequences follow.

You cannot tell which you have by looking, so the safe assumption is chloramine.

A dechlorinator that only handles chlorine is not enough. When a conditioner breaks the chloramine bond it frees the ammonia that was part of it. Most modern conditioners also detoxify that ammonia, and their labels say so; the cheapest ones sometimes do not. Read the bottle — this is one of the few places in fishkeeping where the label genuinely matters.

Dose for the whole volume being treated, not the fraction you are replacing, unless the product specifically says otherwise. Conditioner is inexpensive and mild overdosing is not the risk here.

What a conditioner actually does

Worth knowing, because the claims on the front of the bottle range from precise to decorative.

It neutralises chlorine. The active chemistry, usually a thiosulfate, reduces chlorine to harmless chloride almost instantly. This part is real and works.

It breaks the chloramine bond and, in a good product, binds the ammonia released. The binding is temporary — it converts ammonia to a form the fish cannot be harmed by while leaving it available to the filter bacteria, which then process it normally. That is the right design, and it is why a conditioner is not a substitute for a cycled filter.

It chelates heavy metals, in most products. Copper, zinc and lead picked up from plumbing are bound into a form the fish cannot absorb — the same metals that make copper-based treatments so hard to control, as using medications safely describes. This matters more in old properties and new copper pipework than most people assume.

“Slime coat protectants” and stress-reducing claims are the soft end of the label. They do no harm and the evidence they do much good is thin. Do not pay a premium for them.

One consequence catches people out regularly. An ammonia-binding conditioner can make a test kit read alarmingly high, because several common ammonia tests measure total ammonia and cannot distinguish the bound, harmless form from the free, toxic one. After a big change with chloraminated water, a few tenths of a milligram per litre can come from the tap itself; treat a reading of 1 mg/L, or any reading with distressed fish, as real ammonia and act on it, and check nitrite too, which is not affected the same way.

The pipework between the main and your bucket

The utility’s report describes the water leaving the treatment works. What reaches your tap has spent time in your building’s plumbing.

Use the cold tap, not the hot. Hot water sits in a cylinder, often at a temperature that encourages metals to dissolve, and in many systems it has passed through parts never intended for drinking water. Filling a bucket from the hot tap to save temperature-matching is a common shortcut and a bad one — mix hot and cold in a bucket if you must, but the safer habit is cold water warmed, or water stored in advance.

Run the tap for a minute first, especially first thing in the morning. Water that has stood overnight in copper pipe has had hours to pick up metal, and a minute of flushing replaces it with water fresh from the main.

Old properties deserve more caution. Lead pipework, galvanised steel releasing zinc, and new copper that has not yet passivated all put more metal in the water than the report suggests. A conditioner that chelates heavy metals covers most of this, which is one more argument for buying the fuller-specification product.

When the water company has been working

Occasionally the supply itself changes for a few days. Mains repairs stir sediment loose; hydrant use reverses flow and discolours water; some utilities run a periodic disinfection boost with temporarily higher chlorine.

The signs are simple: water that looks cloudy or brown at the tap, smells strongly of chlorine, or tastes different. When any of that is true, do not do a water change that day. Run the tap until it clears, and postpone. If a change cannot wait, condition generously and keep it small.

This is also one of the more satisfying explanations for a tank that went wrong for no apparent reason in an otherwise unremarkable week, and it is worth checking your utility’s works notifications before concluding that something is wrong with the tank.

Storing and ageing water

Anyone with a pond, a large tank, or a difficult supply eventually arrives at the same solution: a storage container, a small pump or air stone, and a heater.

What it buys you:

• Temperature already matched, so a change is a matter of moving water rather than blending it.
• Degassed and stabilised, so the pH is the figure your tank will actually see.
• Conditioned in advance, with the whole volume treated at once rather than dosed into the tank.
• Water available immediately when something has gone wrong, which is precisely when you do not want to be waiting for a bucket to warm up.

A food-grade barrel, a lid, a cheap air pump and an aquarium heater is the whole apparatus. For ponds it is close to standard practice, and the pond guide assumes it for large changes.

Top-ups are a different question from water changes

A distinction almost nobody makes, and it compounds quietly over months.

Evaporation removes water and leaves everything dissolved in it behind. Top up with tap water and you are adding minerals without removing any, so hardness and total dissolved solids — TDS, the figure a cheap handheld meter reads — climb steadily. Over a year in a warm room with an open-topped tank, the drift can be substantial. A TDS meter reads that figure directly. It is not routine kit — the testing guide is right that the number is rarely actionable, and nothing about a goldfish tank is decided by it — but this is the one question it answers that no other test can: take a reading from the tap and one from the tank, and a tank sitting far above its own source water has been topped up rather than changed. If you already suspect that, you do not need the meter to confirm it; change more water and see.

So:

• Top-ups, replacing evaporation, are ideally done with low-mineral water — and this is the one place where RO, distilled or clean rainwater is genuinely the right choice rather than an affectation.
• Water changes, replacing water you removed, use your normal conditioned tap water, because you removed the minerals along with the water.

If you only ever use tap water for both, the practical countermeasure is simply to do proper water changes rather than living on top-ups, which the water change frequency guide argues for on other grounds anyway. A lid reduces evaporation substantially and is worth having for several reasons.

Find out what is in your water

Every water utility publishes an annual quality report, searchable by postcode or zip code, and it is free. It will tell you:

• whether the supply is chlorinated or chloraminated
• hardness, usually as calcium carbonate, and often alkalinity with it
• nitrate, which matters more than most people realise
• typical pH
• anything unusual in your area

Five minutes with that report replaces a year of guessing. It is a cost-effective thing a new keeper can do and it is rarely mentioned.

Then test the tap water yourself, once, with your own kit: pH, KH, GH, nitrate, and ammonia. Let a sample stand for 24 hours with an air stone in it first and test that as well, because the standing figure is the one your tank will actually see.

Hardness: what your area gives you

Hardness comes from the ground the water passed through, and it is not something you chose. Chalk and limestone regions produce hard, well-buffered water; granite and upland regions produce soft water with very little to hold pH steady.

For goldfish this is mostly good news. They do well anywhere from moderately soft to quite hard (FishBase gives roughly 5 to 19 °dH), which covers most supplies. The pH, KH and GH page sets out the detail; the short version for source water is:

• Hard or moderately hard tap water is ideal for goldfish and needs nothing done to it. It is also what a healthy fish has been quietly relying on all along.
• Very soft tap water — low KH especially — means a tank with little resistance to pH movement, which is the situation where an under-maintained tank can drop sharply. It is manageable, it is a reason to keep the filter well maintained since biological filtration consumes alkalinity, and it is worth knowing about in advance rather than discovering during a crash.
• Chasing a specific pH number with chemicals is a bad trade in either case. Goldfish tolerate a wide range and do not tolerate a moving target.

The water softener problem

This one deserves its own section because it is common, invisible, and gets the reasoning exactly backwards.

A domestic ion-exchange softener does not remove minerals from water. It swaps them: calcium and magnesium out, sodium in, in roughly equivalent chemical amounts. The water feels soft, soap lathers, the kettle stops furring — and the total dissolved solids are about where they started, now composed of the wrong things.

For an aquarium that is the worst of both worlds. The calcium a fish needs for osmoregulation, scale and bone is gone. The sodium is elevated. And because the alkalinity is largely unaffected, you cannot even rely on it behaving like naturally soft water.

Avoid filling a tank from a softened tap.

The practical fix is often already installed. Many softener systems leave one tap unsoftened — commonly the kitchen cold tap, in the UK at least — because softened water is not advised for drinking or for making up infant formula either. If yours does, that is the tap to fill buckets from. If you are not sure which taps are on which side of the system, the installer’s paperwork will say, and testing GH at two taps will settle it in five minutes: the softened one will read close to zero.

Nitrate you cannot dilute away

A water change dilutes the tank toward the composition of the water going in. If that water already contains nitrate, the tank cannot go below it.

In agricultural regions tap nitrate of 20 to 40 mg/L is unremarkable, and regulatory limits for drinking water sit around 50. A keeper in such an area doing everything right will still read 30 or 40 in the tank, will assume the maintenance is failing, and will respond by changing more water — which does nothing, because the new water is the source of the nitrate.

Knowing the tap figure changes the whole approach:

• Judge by the trend, not the absolute. If tank nitrate sits a little above tap nitrate and stays there, the tank is in balance.
• Reduce production rather than increasing dilution. Lighter stocking, lighter feeding, and live plants, which consume nitrate directly.
• Consider blending, in a bad case, with RO or distilled water to bring the incoming figure down. This is one of the few situations where RO genuinely earns its place in goldfish keeping.

Well and borehole water

Private supplies are common and workable, with a different set of things to check.

No chlorine, usually, so no conditioner is needed — though it is worth confirming rather than assuming, particularly if there is any treatment system in the line.

Variable and often very hard. Test it rather than assuming; hardness from a borehole can be considerably higher than anything a municipal supply delivers.

Often supersaturated with carbon dioxide and other gases. Straight from the ground, water can carry gas loads that come out of solution once it reaches atmospheric pressure. Two consequences: a pH reading taken immediately is artificially low, and water used immediately can release gas inside the tank and, in extreme cases, inside the fish, which bubbles on the fins, skin or eyes describes. Aerate it for 24 hours before use and before testing. This single habit removes both problems.

Iron, manganese and sulphur. Water that stains fixtures orange or smells of sulphur is carrying something worth identifying. Neither is necessarily fatal to fish, but both point to water that deserves a proper analysis rather than a hobby test kit. Iron in particular can irritate gills, which is one of the things behind unexplained red streaking.

Nitrate from surrounding land use. A well in farmland can read higher than the mains, and the same arithmetic as above applies.

Low oxygen. Groundwater arrives with very little dissolved oxygen. Aeration deals with this too, which is another reason the 24-hour standing habit is worth building.

RO, rainwater and bottled water

Source	Use for goldfish	Why
Municipal tap	The default, and usually correct	Predictable, tested, published, buffered
Softened tap	Avoid	Sodium for calcium; no benefit, and test GH before relying on it
Well or borehole	Good, after aeration and testing	Variable; needs knowing rather than assuming
RO, neat	No	No buffering; pH becomes unstable
RO, blended or remineralised	For a specific problem	Sensible for very high source nitrate
Rainwater	Pond top-up at most	No buffering; collects whatever is on the roof
Bottled or “spring” water	No	Expensive, inconsistent, not formulated for this

Reverse osmosis strips almost everything, including the carbonate hardness that holds pH steady. Neat RO in a goldfish tank is not a purer version of tap water; it is water with nothing left to resist change, and the result is a pH that moves with every shift in the tank. Where RO is the right answer — high source nitrate, an unusual contaminant — it is blended back with tap water or remineralised with a proper salt mix, and the target is ordinary moderately hard water. Which is, for most people, what the tap was already delivering.

Rainwater has the same buffering problem, plus a collection problem. A butt fed from a roof gathers dust, bird droppings, moss, pollen and whatever the roof and gutters are made of. For topping up a pond in a dry spell it is defensible. For an aquarium it is a variable you do not need.

Bottled water is expensive per litre, varies between brands and between batches, and its mineral profile was chosen for taste.

Temperature, which is the other half of a water change

Source water is a chemistry question and a temperature question, and the temperature half causes more visible distress.

Goldfish tolerate a wide range and tolerate sudden change badly. A tank at 21 °C refilled with water at 12 °C — which is what a cold tap delivers in winter — puts every fish through a shock that a thermometer would have prevented. Match to within a couple of degrees, use a thermometer rather than your wrist, and for large changes go slower rather than colder. The water change guide covers the method and the temperature guide covers what the fish can actually absorb.

For a pond, the same point scales up: a large volume of cold tap water into a pond in early spring is a temperature event as much as a chemical one.

A practical routine

Do this once, then stop thinking about it:

1. Read your utility’s water quality report. Chlorine or chloramine, hardness, nitrate, pH.
2. Test your own tap, and test a sample that has stood aerated for 24 hours. Record both.
3. Identify which taps are unsoftened, if you have a softener.
4. Buy a conditioner that handles chloramine and ammonia, and dose it for the volume as the label directs.
5. Write the figures on a card taped inside the cabinet, next to your tank’s real volume.

Then the only recurring job is matching temperature and adding conditioner, which takes fifteen seconds.

When the tap is the answer

What you are seeing	Check the source water for
Fish distressed shortly after every water change	Chloramine, under-dosed conditioner, temperature mismatch
Nitrate that will not go below 30–40 mg/L	Nitrate in the tap
pH that drifts down between water changes	Low KH in the source water
Tank GH near zero and fish not thriving	Filling from a softened tap
Tiny bubbles on glass and fish after filling	Supersaturated gas; aerate the water before use
Readings that change between tap and tank	Degassing on standing — test the stood sample

If none of those fit, the source water is probably not the problem, and the testing guide is the better place to continue.