Tuesday, June 25, 2013

Can the Audion Tube Be Used As A Hi-Fi Audio Amplifier?


Given that the iconic Lee De Forest’s first ever active electronic amplifying device initiated the 20th Century electronics revolution, will it still work today when used as the output tube of a high fidelity audio amplifier?

By: Ringo Bones

Ever since the 1970s when dedicated Japanese audiophiles were experimenting with first generation vacuum tubes that were first designed and manufactured during the 1920s and 1930s into single-ended triode zero negative feedback audio amplifiers, audio amplifier design technologies long ago abandoned by the industrial West, there are probably dedicated audio hobbyist today that are starting to wonder if the Lee De Forest designed iconic first triode vacuum tube- the Audion – would be feasible when used as an output audio tube in a single-ended triode audio amplifier design. But will such an unseemly “old school” electronic audio design still viable and work well into the second decade of the 21st Century? 

Based on established written history on the development of electronic engineering, it was the pioneering work of Nikola Tesla and Guglielmo Marconi on radio that initiated the development of the thermionic vacuum tube. Before the discovery of the germanium detector, early crystal detectors employed a piece of galena – a type of lead ore – and a catwhisker. Not all spots on the galena were sensitive and you had to hunt for a spot to touch with the catwhisker. A slight vibration on the workbench and you’ll lose the sensitive spot. Also, if the galena should become “dirty” via further exposure to atmospheric oxygen, you might never find a sensitive spot for your crystal radio set to work again. Obviously, the galena detector had serious drawbacks. 

Oddly enough, the first hint as how to improve the detector came in 1883, long before the crystal detector was first used as a radio receiver – or become a favorite of electronic hobbyists and elementary school level science projects. In that year, Thomas Alva Edison was experimenting with filaments for his new invention – the electric light bulb. He placed a filament in a glass bulb and then exhausted the air, creating a vacuum. By means of an electric current, he tested the filament until it glowed brightly and produced light. 

Edison soon observed an undesirable feature about his bulbs. After short time, a black substance was deposited on the inside of the glass, interfering with the light given out. In an attempt to eliminate this deposit on the glass, Edison inserted a metal plate. Now, this plate did not help much to solve the problem, but one day he connected a delicate electric meter between the plate and the positive end of the filament. To Edison’s amazement, the meter showed that a small electric current was flowing through the circuit. He did not know why this current should flow and he merely jotted down this strange fact in his notebook and forgot about it.
Today, we know why this current flows. When a filament is heated to incandescence (heated to when it becomes hot enough to give off light), it shoots off streams of electrons. This behavior is known as the “Edison Effect” or “Thermionic Effect”, of a filament heated to incandescence. These electrons given off by the hot filament collect on the cool plate and, if a path is furnished for them, they will flow along the path of the filament. The electric meter in that path can show that electrons are flowing. 

Since the discovery of the Edison Effect back in 1883, electron theory was still a relatively under-investigated phenomena. But in 1904, J. Ambrose Fleming, an Englishman, who understood the flow of current in terms of electrons, decided to experiment a bit. To depend upon the electrons piling up on the cool plate, thought Fleming, is too slow. Suppose we were to create an actual deficiency of electrons on the plate by placing a positive charge on it, wouldn’t that attract still more electrons from the filament? Fleming connected a battery in the circuit from the plate to the filament in such a way that the positive post of the battery was connected to the plate. He also connected another battery to the filament to heat it to incandescence. Note that this filament battery is not in the plate circuit. With such set-up, Fleming basically invented the first thermionic vacuum tube diode. 

Soon after Fleming’s thermionic vacuum tube diode appeared, in 1907, an American inventor, Lee De Forest, undertook to carry further some ideas suggested by one of Fleming’s experiments. De Forest knew that when Fleming placed a positive charge on the plate of his tube by means of a battery connected between the plate and filament, a much greater electric current flowed through the meter than when there was no such charge. Further, the greater the positive charge on the plate, the greater the flow through the meter. Actually, this did not go on forever, after the positive charge reached a certain value, placing a greater positive on the plate had no further effect. 

It was then that Lee De Forest had a stroke of genius. Since the flow of current in the plate circuit starts with the stream of electrons shot out by the heated filament, he began to experiment with that electron stream. After more experiments, De Forest eventually met the difficulty of practical amplification of weak RF signals received by the Fleming thermionic diode detector by making the new electrode in the form of a mesh of very fine wire – a grid. Since most of the grid consisted of open space, most of the electrons pulled over by a positive charge on the grid now shoot through these open spaces an d continued right on the plate. The grid was the solution to his problem. Since charges on the grid control the flow of electrons from the filament, we are able to control the large plate currents by means of a small charge on the grid – and this is what De Forest set out to do and eventually created the first ever triode thermionic vacuum tube which he called the Audion. When Dr. Lee de Forest placed a third element, the grid, between the cathode and the plate of the Fleming thermionic vacuum tube diode, he introduced the magical word – amplification.   

Given that it was from the first improved Edison light bulbs that the first thermionic vacuum tube triode were derived by Lee de Forest in 1905, every hi-fi hobbyist is probably now wondering whether those old Audion tubes could be made into single ended triode audio amplifiers. After all, during the 1990s, tubes developed and manufactured during the 1930s – as in the Western Electric 300B – or from the 1920s – like the old light bulb looking PX25 tube – were successfully used in the finest sounding hi-fi audio amplifiers during the latter half of the 1990s, would an even older triode tube – as in the turn of the century (early 1900s that is) era Audion even sound better?

Looking at a typical Audion tube – even recently constructed working replicas constructed with parts that then exist before 1910, one would wonder if this tube would be more” microphonic” in comparison to “newer” designs like the 300B or the PX25. And given that the Audion doesn’t have the much improved plate current ratings – therefore lower output impedance – of the output power tubes that were invented years after it – like the PX25 from the mid 1920s or the Western Electric 300B that cam much later in the 1930s, one would wonder if the Audion would produce a “good musical sound” if connected in parallel since most output transformers in the hi-fi DIY market today were more likely to have been designed to work with the more robust plate current capable PX25s and 300Bs. But like most DIY audiophiles, I would probably never pass the chance to hear first hand the sound of a working SET amplifier using authentic period-correct replica De Forest’s Audion tubes.