Coherer - Limitations of Coherers

Limitations of Coherers

Coherers have difficulty discriminating between the impulsive signals of spark-gap transmitters, and other impulsive electrical noise:

All was fish that came to the coherer net, and the recorder wrote down dot and dash combinations quite impartially for legitimate signals, static disturbances, a slipping trolley several blocks away, and even the turning on and off of lights in the building. Translation of the tape frequently required a brilliant imagination —Greenleaf Whittier Pickard

As a simple switch registering the presence or absence of radio waves, the coherer could detect the on-off keying of wireless telegraphy transmitters, but it could not demodulate complex waveforms of audio (AM) broadcasting, which began to be experimented with in the first years of the 20th century, which required the radio signal to be rectified. This problem was solved by the rectification capability of Reginald Fessenden's hot wire barretter and electrolytic detector. These in turn were replaced by the crystal detector around 1906, and then around 1920 by vacuum tube technologies such as John Ambrose Fleming's thermionic diode and the triode-based regenerative detector invented by Edwin Howard Armstrong.

A "ball" coherer, designed by Branly in 1899. This imperfect contact type had a series of lightly touching metal balls between two electrodes. Tripod coherer, built by Branly in 1902, another imperfect contact type. Although most coherers functioned as "switches" that turned on a DC current from a battery in the presence of radio waves, this may be one of the first rectifying (diode) detectors, because Branly reported it could produce a DC current without a battery. Another tripod detector built by Branly