Why do carbon monoxide alarms matter when smoke alarms already exist?
Carbon monoxide has no smell, no colour and no smoke, and it binds to your blood hundreds of times more readily than oxygen — so nothing you can sense tells you it is there.
Simple intuition
The plain reason, in everyday words
Your body has no sense for carbon monoxide. It has no smell, no taste and no colour, and it does not irritate your eyes or throat the way smoke does. Worse, what it does inside you feels like something ordinary: headache, tiredness, nausea, muddled thinking — exactly like coming down with flu. So the natural interpretation of the early symptoms is that you should go and lie down, which is the worst possible response, because it keeps you in the room. Meanwhile a smoke alarm will not help, because it is built to detect particles of smoke, and carbon monoxide comes from fuel burning badly rather than from something being on fire. A boiler with a blocked flue can fill a house with it while producing no smoke at all.
A smoke alarm will detect carbon monoxide.
Smoke alarms detect particles, by light scattering or ionisation. Carbon monoxide is an invisible gas that produces no particles, so a smoke alarm has nothing to respond to.
You would smell it or notice something wrong.
It is odourless and colourless, and the early symptoms resemble flu closely enough that the usual response — resting indoors — increases exposure.
Pressing the test button proves the alarm works.
That tests the sounder and the battery. The sensing cell degrades over years regardless of use, which is why alarms carry a replacement date, typically seven to ten years.
It sinks to the floor, so alarms belong low down.
Its density is close to that of air, so it mixes throughout a room. Guidance places alarms at around breathing height rather than at floor level.
It is the clearest case of a hazard your senses were never built to notice, where the only warning available is one you have to install and maintain deliberately. Knowing why also fixes the two things people get wrong in practice: that a smoke alarm does not cover it, and that a working button does not mean a working sensor. The underlying idea — that detection has to match the thing being detected — applies to every alarm in a building.
Who worked it out
Carbon monoxide poisoning was a routine hazard of coal gas, which contained the gas directly, and domestic supplies in many countries were poisonous until conversion to natural gas in the 1960s and 1970s.
What problem forced it
Conversion removed the gas from the supply but not the risk, since incomplete combustion in any appliance still produces it — shifting the problem from the fuel itself to appliance maintenance and ventilation.
How it changed since
Affordable electrochemical sensors made domestic alarms practical from the 1990s, and requirements to fit them, particularly in rented housing, have spread steadily since.
How smoke alarms actually detect smoke
Ionisation and photoelectric sensing detect different fires, and the difference matters for where each belongs.
Why incomplete combustion happens at all
The same chemistry explains sooty flames, yellow gas rings and why ventilation is part of appliance safety.
Written for Curio rather than collected from a forum — it is part of the curated corpus that ships with the platform. The references it draws on are listed under Sources.