The basic problem
A telegraph only needs to tell whether a current is on or off. Speech is far more demanding. A voice makes air pressure rise and fall in a complicated pattern, and for the listener to recognize the words, an electrical signal must rise and fall in the same pattern, then be turned back into sound at the far end. That requires two devices: a transmitter, which turns sound into a varying current, and a receiver, which does the reverse.
The question of who first reached this goal is a separate subject. This article is about the parts and how they developed.
Bell’s first designs
Alexander Graham Bell’s early experiments used thin membranes or metal diaphragms connected to electromagnets. In a magnetic design, a voice makes an iron diaphragm vibrate near a magnet wound with a coil of wire. The moving iron changes the magnetic field slightly, which induces a tiny alternating current in the coil. At the other end, the same arrangement works in reverse: the incoming current changes the strength of an electromagnet, which pulls on a second diaphragm and reproduces the sound.
This is simple and needs no battery, and the same device can serve as both transmitter and receiver. The weakness is that the signal is very small, so the sound is faint and the range limited.
Bell’s patent also described a different way of producing a varying current, by letting a diaphragm move a needle in and out of a liquid, which changed the electrical resistance in the circuit as the voice changed. A battery supplied the current, and variations in resistance made the current vary with the sound. This liquid transmitter is what is usually linked to the famous first words transmitted in March 1876. It was not practical for everyday use, since the liquid spilled and the quality was unreliable, but it demonstrated the principle: vary the resistance, and the current follows.
The variable-resistance idea becomes practical
The route to a better transmitter was to find a solid material whose resistance changes when pressure changes. Several inventors worked on this at nearly the same time.
- Emile Berliner, working in the United States in 1877, filed for a transmitter that used loose contact between metal parts. The telephone company acquired his rights.
- Thomas Edison, working on behalf of a competitor, developed a transmitter around 1877 that used carbon. A button of compressed carbon, pressed by the diaphragm, changed its resistance as the pressure changed. Edison also drew on his earlier experience with carbon in his other work, and the same fascination with the material appeared in the carbon filament of his light bulbs.
- David Hughes, a British-American scientist, described a loose-contact “microphone” in 1878, and the word itself came into use around then.
- Francis Blake produced a carbon transmitter in 1878 that the Bell company adopted, and Henry Hunnings introduced the use of carbon granules in a similar period.
Precisely who deserves priority for what is debated, and the late 1870s were a period of overlapping patents and lawsuits. What is clear is that by the end of the decade a carbon-based transmitter had become the standard, because it produced a much stronger signal than the magnetic type.
The induction coil and the battery
A carbon transmitter works by modulating the current from a battery in the line. To make the signal travel farther, designers added an induction coil, a small transformer that steps up the voltage of the varying current before it is sent down the line. This allowed telephone lines to extend beyond the few miles of the very earliest circuits.
With a battery in each home or a shared battery at the exchange, the device now had power to spare, and the receiving side could be a separate, specialized unit. The magnetic receiver, with a permanent magnet and a coil, stayed in service for decades, partly because it needed no power and was reliable.
Granular carbon, a long run
The granular carbon transmitter, which packs tiny carbon grains between two electrodes behind a diaphragm, became the common design and stayed in telephone handsets for roughly a century. It was cheap, loud and sturdy, though it had known drawbacks: the granules could pack down over time, and the sound had a characteristic hiss and limited frequency range. Tapping the handset to loosen the carbon was a familiar household remedy.
Designs of this type shaped the sound of telephone speech: a narrow band of frequencies, generally enough to recognize a voice and understand words, but not to reproduce music well. The narrow range was a deliberate trade-off, because lines and later electronic equipment could carry more calls if each call used less bandwidth.
Moving beyond carbon
In the twentieth century, other microphones took over. Condenser and moving-coil types appeared for broadcasting, recording and public address. For telephones, a major step came in the 1960s with the electret microphone, developed at Bell Laboratories, which holds a permanent electric charge and needs little power. It is small and cheap, and became the standard in many handsets and mobile phones. Today, tiny silicon microphones built with integrated circuits sit inside smartphones, hearing aids and many other devices.
Why the details matter
Much of the story of the telephone is hidden in these small parts. The invention of the telephone as a concept took a flash of insight about varying current; the development of the telephone as a service depended on the less glamorous work of making a transmitter loud, durable and cheap. A thin diaphragm, a handful of carbon grains and a coil of wire made it possible for ordinary people to talk across a town, and later across a continent.