Gregory's Serverit started in the late 1800s, people were trying to build generators to power high voltage arc lights. if you tried to use AC (alternating current) to drive an arc lamp, it made a very loud noise at 60hz (or whatever you were using)
|
34 comments
 see, a alternator's output frequency depends on 2 things: how fast it turns and how many *poles* you have (basically groups of windings.) to bump up the frequency, you have to put on as many poles as you can physically fit and then run it as fast as you can! the Thomson/Tesla design was a bit brute force in that way, so it quickly ran into physical limits--push it any faster, and the rotor would just explode. then a guy named Alexanderson came along. he had some experience with the very early radio transmitters made with these special alternators, and he figured out a way to dramatically increase the output power. he lightened up the rotor (the rotating part of the alternator), streamlining it to a thin disk. here's a cross section of the concept. the rotor shaft (not shown) would be along the bottom of the page. there are two sets of windings. windings "A" generate a large magnetic field, which loops around frame "D" and laminations "H". the disk has slots in it that act like a magnetic switch, modulating the field received by pickup coil "E" so now the maximum frequency is controlled mainly by the speed and the number of slots in the spinning rotor. and those can be packed together pretty tightly! it's very similar to how a Hammond organ works. the next big problem that Alexanderson had to solve was controlling the speed. any variation caused the output frequency to drift out of the range that the antenna could support. this circuit uses a closed-loop design with a resonant filter to control "saturation coils" (aka magnetic amplifiers) which regulate the power going into the motor that turns the alternator! the saturation coil works by using a DC control current to push what is basically an inductor into saturation--meaning it can't carry any more of a magnetic field--so the effective inductance decreases. since inductance is kinda like an AC resistance, this allows the main drive current to flow through it. the next problem was to *modulate* the signal with morse code. for that, Alexanderson used another saturation coil (mag amp). this one *detunes* the giant transformer that collects the current from all the pole windings and matches the impedance to the antenna array. turning off the key causes the frequency to shift enough to reduce the output power by over 90%. the resulting transmit power was 200 kilowatts! even by modern standards this is very high power. in the United States, the most powerful AM stations were only a bit more than twice as powerful.  @tubetime I thought even at 100 kW, you start getting spooky effects like nearby metal fences having an appreciable voltage :D! @cr1901 @tubetime I remember back in the late 80s my Telecom Australia apprentice intake went to this AM broadcasting station and antenna https://maps.app.goo.gl/EM6Luyo98ve2NUeB9 (the frequency was quite a bit below standard AM broadcast, typically around 15-20KHz.) incredibly, there is still an *operational* Alexanderson transmitter! it is called SAQ Grimeton and they fire it up twice a year. there are so many cool details to this technology i had to leave out. but watch this video. you'll see amazing things like high voltage relays quenched with compressed air, water cooled load resistors, synchronous motors, and really scary 400V switches. the facility also has an informative website: https://grimeton.org/ and if you can't make it out to Sweden, there are virtual tours available: https://alexander.n.se/en/virtual-visit-to-saq-grimeton/ for further reading, here's a useful paper. https://www.navy-radio.com/xmtrs/vlf/alexanderson-mayes-1975.pdf  @tubetime OMG it runs off 2-phase because it's from a very particular time and place in the US. @tubetime OMG the air-quenched relays are scary, and they're just right out there. @tubetime Thanks for posting this; on my list for my next visit to Scandinavia. @tubetime apparently someone made a modern receiver out of relatively equivalent-level technology http://www.wireless.org.uk/mechrx.htm  @tubetime There are 400 kW AM transmitters now that are completely solid-state, no tubes even. https://www.nautel.com/content/user_files/2020/01/NX400-spec-sheet.pdf @tubetime (and those 500 kW AM stations only operated on an experimental basis; none were ever licensed for commercial broadcasting, because small radio operators had powerful friends in Congress and they worried that they would lose their audience and network contracts if the networks could cover the entire country with just the six stations they owned) @tubetime And when Fessenden broadcast voice in 1906, he just wired a carbon microphone in series with the antenna.  @tubetime Does that mean it accidentally uses frequency-shift keying?   @tubetime Image borrowed from https://permies.com/t/18586/Smart-Drive-wind-generator @tubetime Also *that* Thomson. The Thomson-Houston Electric Company was one of the ancestors of GE, and thus the British Thomson-Houston Company, which later formed AEI with Metropolitan-Vickers, was acquired by GEC, absorbed Marconi, and merged with British Aerospace to form BAE Systems, and also thus la Compagnie Francaise Thomson-Houston, which absorbed CSF and turned into Thomson-CSF, which later turned into Alstom, Thales, Technicolor, and STMicroelectronics, among numerous others. @wollman ahhh i had a tickle of memory that he was a famous thomson but i forgot to check into it  |
but Professor Elihu Thomson and a certain Mr. Tesla (yes, *that* Tesla) figured out a way to build an AC alternator that could produce 10 amps at up to 12,000 Hz!