Shared water supply flow

Why does someone flushing a toilet on another floor turn your shower scalding?

You are standing in a shower that has been comfortable for four minutes. Someone two floors away flushes a toilet, and within a couple of seconds the water on your shoulders turns scalding. Nobody touched your taps. The boiler did not suddenly work harder. Yet the temperature moved, sharply, and then drifted back.

The tempting story is that the flush "stole the cold water". That is nearly right, and being nearly right is what makes it worth pulling apart — because it does not explain why the shower goes hot rather than simply weaker, and it does not explain why the same building can produce the opposite lurch when the dishwasher fills.

Start with something that cannot be argued with. Water is very close to incompressible, and the pipes do not store any meaningful amount of it. So at every junction in the building, the flow arriving must equal the sum of the flows leaving. Nothing is created at a tee and nothing accumulates there. That is the whole of it: a conservation statement about volume per second.

Now ask what your shower actually is. It is not one stream. It is two — a hot leg and a cold leg — meeting at a mixing valve, and the temperature you feel is set by their ratio, not by either one alone. Half and half might give you forty degrees. So already we can see the shape of the answer: anything that changes the ratio changes your skin temperature, even if the total flow barely moves.

So what does a flush change? A tank-type toilet refills through a valve fed from the cold side only. It never touches the hot pipe. So the flush adds a new cold demand — a litre or two per minute for perhaps a minute — to a cold branch that was already feeding your shower.

Here is where the conservation statement does its work. The supply into that branch is not infinite and it is not free. It comes from a street main at some pressure, and the flow it delivers depends on how much pressure is left after the losses along the way. Ask yourself: if the branch must now carry the shower's cold flow plus the toilet's fill flow, what has to give?

Two things. First, the total flow through the shared upstream pipe rises. Second — and this is the part people skip — friction loss along a pipe rises steeply with flow, roughly with the square of it. So carrying more water through the same pipe costs disproportionately more pressure. The pressure arriving at your cold inlet therefore drops.

And what happens at a mixing valve when the cold side's supply pressure falls but the hot side's does not? Cold flow through its orifice falls. Hot flow does not. The ratio shifts toward hot. You are scalded.

Notice what that predicts, and check it against experience. If instead someone opened a hot tap — filling a bath, or a dishwasher drawing hot — the same argument runs with the legs swapped, and your shower should go cold. It does. The asymmetry in the story is not in the physics; it is in the fact that toilets are the most common single-temperature draw in a house, and they happen to be cold-only.

One honest caveat: the size of the effect depends on things this argument treats as fixed. A house with generous pipe diameters, short runs, or a pressurised supply at high static pressure will barely flinch, because the extra friction loss is small next to the available pressure. A long run of narrow pipe near the end of a branch is where the lurch is violent. The mechanism is the same; only the margin differs.

Two refinements are worth making explicit.

The first is that "losing pressure" and "losing flow" are not two separate misfortunes; they are the same event described from two ends. The mixing valve does not sense pressure and decide anything. It is just two openings. Lower pressure across the cold opening simply means less water per second through it, and the temperature is a bookkeeping consequence of that.

The second is that the flush is a transient. The fill valve closes after the cistern is full, the extra demand disappears, the friction loss falls back, and your cold flow recovers. That is why the scald is a lurch rather than a new steady state — and why the fix is not bigger pipes but a valve that notices.

That is precisely what a pressure-balancing shower valve does: it carries an internal piston or diaphragm that senses the hot and cold supply pressures and throttles the stronger side to track the weaker one, holding the ratio roughly constant while the total flow sags. A thermostatic mixing valve goes further and regulates on measured temperature rather than on pressure. Neither creates water; both simply refuse to let a conservation-driven imbalance reach your skin.

• Why water hammer — the bang when a valve shuts fast — is the same conservation story run in reverse, with momentum instead of volume.
• How a combi boiler's minimum flow switch can make the hot side drop out entirely mid-shower.
• Why buildings above a certain height need booster pumps and pressure-reducing valves on lower floors at the same time.

Every term the collection defines is gathered in the glossary.