In 1930, in a laboratory in Zurich, a physicist named Walter Jaeger was trying to build a sensor that would detect poison gas. His device worked on the principle of ionization: a small electrical current passing through air would be disrupted when gas molecules entered the chamber and attached to the ions carrying the current. Jaeger had what seemed like a working design. To test it, he lit a cigarette.
The cigarette smoke did not trigger the alarm. The device was, as a poison gas detector, a failure. But Jaeger noticed something: the instrument's needle had moved. The smoke particles, though not ionizing the air themselves, had disrupted the ion current in a different way — by attaching to the ions already present and reducing their mobility. Jaeger had accidentally built the sensing principle behind every ionization smoke detector sold in the following century. He published the observation and filed it away as a curiosity.
The isotope that made it practical
The physics was understood by 1930. The problem was the source material. Early ionization chambers used radium, which was expensive, difficult to obtain, and — in the quantities needed for a cheap consumer device — genuinely dangerous. The sensing principle was sound. The economics were impossible.
They remained impossible until the nuclear weapons program of the 1940s and 1950s generated, as a byproduct of plutonium production, enormous quantities of americium-241. Americium is an alpha emitter with a half-life of 432 years — long enough to function for the lifetime of any practical device, short enough to produce a steady, reliable emission. It produces no penetrating gamma radiation in useful quantities. A tiny amount, roughly one microcurie, was sufficient to ionize the air in a sensing chamber. The weapons program had created, inadvertently, the ideal isotope for a household fire alarm.
By the late 1950s, americium-241 was available at low cost to licensed manufacturers. The technical barriers had dissolved. What remained was the problem of turning laboratory physics into a device a homeowner could afford to buy.
The $125 breakthrough
In 1969, a Colorado engineer named Duane Pearsall, working with a company called Statitrol Corporation, introduced the first battery-powered ionization smoke detector designed for residential use. It contained a small americium-241 source, a nine-volt battery, and a piezoelectric buzzer. It cost $125 — roughly $1,000 in 2024 dollars. It was, by any reasonable measure, too expensive for mass adoption.
The device Pearsall had built was acquired by BRK Electronics, which began manufacturing it as the "Sta-Alert." Other manufacturers entered the market. Prices fell — not quickly, but steadily. By the mid-1970s, a residential smoke detector could be purchased for under $20. But consumer demand alone was not enough to drive meaningful adoption. The product that saves a sleeping family's life is precisely the product that most families do not buy until after a fire they wish they had been warned about.
The mandate that markets could not create
The mechanism that actually drove adoption was not consumer demand but building code. Between 1970 and 1990, state after state, and then virtually every local jurisdiction in the United States, mandated smoke detectors in new residential construction and — more consequentially — in existing housing stock. The mandates were unevenly enforced and inconsistently structured, but their cumulative effect was to transform a product that many households would not voluntarily purchase into one that virtually every occupied dwelling in the country was legally required to contain.
The results were not subtle. In the United States, residential fire deaths fell from roughly 6,000 per year in the mid-1970s to approximately 2,500 by the late 1990s — a reduction achieved during a period when the housing stock grew substantially. The devices in those houses were not sophisticated. They were cheap, battery-powered, and prone to false alarms from cooking smoke. They worked anyway, because the failure mode they were protecting against — a sleeping family overcome by smoke before the fire reached them — was one where even a ten-second warning was sufficient to change the outcome.
The nine-volt problem
The nine-volt battery requirement turned out to have its own long tail of dysfunction. The device required an annual battery replacement that an unknowable fraction of households would not perform. By the 1990s, the "chirping detector with a dead battery" had become a recognizable cultural fixture — the device that had been purchased, installed, and then rendered inoperable by the one maintenance action it required. Studies of fire deaths found that a significant fraction of the homes involved had detectors that were present but not functioning.
The solution — sealed ten-year lithium batteries, introduced commercially around 2010 — was straightforward in retrospect. A device designed to sit unattended for a decade should contain a battery rated for that duration. The delay between understanding the problem and solving it was roughly twenty years, occupied largely by battery manufacturers who preferred the annual replacement market and detector manufacturers who had optimized for the lowest possible shelf price.
The photoelectric debate
The ionization detector Pearsall built is the type that fills most American homes. It is also, for certain categories of fires, the wrong type. Ionization detectors are faster at sensing fast-flaming fires with visible flames and small combustion particles. Photoelectric detectors — which work by shining a light beam across a sensing chamber and measuring scatter from smoke particles — are faster at sensing slow, smoldering fires that produce larger particles and substantial smoke before the flames become significant.
Smoldering fires are disproportionately the ones that kill sleeping people. A mattress fire that smolders for thirty minutes before flaming is precisely the scenario where the faster warning matters most — and where the photoelectric detector has a meaningful advantage. The ionization detector that reaches the same alarm threshold will do so, on average, thirty to fifty minutes later.
The dual-sensor detector — containing both technologies in a single unit — became widely available in the 2000s and eliminates the tradeoff. Building codes in most jurisdictions have not been updated to require it. The single-technology ionization detector remains the dominant type sold, largely because the codes that drove the original adoption have not kept pace with the technology they mandated.
Three observations
The smoke detector's history contains patterns that appear repeatedly in the history of safety technology.
The first is that the sensing principle and the practical device can be separated by decades. Jaeger's 1930 observation and Pearsall's 1969 device were separated not by a failure of engineering but by the absence of a cost-effective ionization source. The invention that mattered was not the detector but the nuclear weapons program that produced americium-241 as a byproduct.
The second is that consumer markets reliably underprovide safety products. The smoke detector that protects a sleeping family is paid for by the family that does not have a fire, which is most families, most of the time. The private benefit arrives in the rare catastrophe. Building codes worked not because they were efficient but because they transferred the decision from individual households, who systematically underweighted low-probability severe outcomes, to governments that could mandate the decision uniformly.
The third is that solving a problem does not end the problem's evolution. The nine-volt battery created a maintenance failure mode that undermined the mandate's intent. The ionization-versus-photoelectric distinction created a performance gap that existing codes did not address. Each generation of the solution arrived carrying the seeds of the next round of dysfunction.
Jaeger's cigarette, lit to test a poison gas alarm, started a chain of consequences that runs through nuclear weapons development, insurance industry lobbying, state fire marshal offices, and the particular acoustics of a chirping detector at three in the morning. The device in most American homes today costs less than fifteen dollars. The infrastructure required to produce it spans the entire 20th century.
More at anethoth.com. Building something? List it on Builds — a directory for indie SaaS projects with transparent revenue and real founders.