3 ms·

We Rediscover Spark Gap Radio by Accident

peter_d_sherman 6y ago
Related: https://en.wikipedia.org/wiki/Coherer https://en.wikipedia.org/wiki/Coherer >"The coherer was a primitive form of radio signal detector used in the first radio receivers during the wireless telegraphy era at the beginning of the 20th century. Its use in radio was based on the 1890 findings of French physicist Édouard Branly and adapted by other physicists and inventors over the next ten years. The device consists of a tube or capsule containing two electrodes spaced a small distance apart with loose metal filings in the space between. When a radio frequency signal is applied [PDS: Correction, should read "in the vicinity"] to the device, the metal particles would cling together or "cohere", reducing the initial high resistance of the device, thereby allowing a much greater direct current to flow through it [PDS: Or some at all, as the video shows]. In a receiver, the current would activate a bell, or a Morse paper tape recorder to make a record of the received signal. The metal filings in the coherer remained conductive [PDS: In a low resistance to DC state] after the signal (pulse) ended so that the coherer had to be "decohered" by tapping it with a clapper actuated by an electromagnet, each time a signal was received, thereby restoring the coherer to its original state." Anyone see a pattern with coherers? Don't they seem similar, patternwise, to N-P-N transistors, or P-N-P transistors? Except for the following things: 1) Control gate (base) acts on RF, not voltage. 2) Coherer remains conductive even after the RF source is switched off. It would be interesting if a transistor could be constructed in a similar manner to a Coherer, that is, if metal filings (well, "metal oxide filings", because the metal filings are typically exposed to oxygen!) between two other metals in some configuration -- would yield a working transistor... Now, I don't know that this could actually be done -- I'm only speculating aloud... <g>

Related: https://en.wikipedia.org/wiki/Coherer https://en.wikipedia.org/wiki/Coherer >"The coherer was a primitive form of radio signal detector used in the first radio receivers during the wireless telegraphy era at the beginning of the 20th century. Its use in radio was based on the 1890 findings of French physicist Édouard Branly and adapted by other physicists and inventors over the next ten years. The device consists of a tube or capsule containing two electrodes spaced a small distance apart with loose metal filings in the space between. When a radio frequency signal is applied [PDS: Correction, should read "in the vicinity"] to the device, the metal particles would cling together or "cohere", reducing the initial high resistance of the device, thereby allowing a much greater direct current to flow through it [PDS: Or some at all, as the video shows]. In a receiver, the current would activate a bell, or a Morse paper tape recorder to make a record of the received signal. The metal filings in the coherer remained conductive [PDS: In a low resistance to DC state] after the signal (pulse) ended so that the coherer had to be "decohered" by tapping it with a clapper actuated by an electromagnet, each time a signal was received, thereby restoring the coherer to its original state." Anyone see a pattern with coherers? Don't they seem similar, patternwise, to N-P-N transistors, or P-N-P transistors? Except for the following things: 1) Control gate (base) acts on RF, not voltage. 2) Coherer remains conductive even after the RF source is switched off. It would be interesting if a transistor could be constructed in a similar manner to a Coherer, that is, if metal filings (well, "metal oxide filings", because the metal filings are typically exposed to oxygen!) between two other metals in some configuration -- would yield a working transistor... Now, I don't know that this could actually be done -- I'm only speculating aloud... <g>