Who was the first? Regardless of what the first is, the question has always fascinated those interested in the history of technology. There rarely is a clear cut answer to the question; here’s a version regarding sound engineering based on freely combining different sources.
Who was the first to record sound? Thomas Edison? Really? Edison unveiled his glowing phonograph in 1877. Twenty years earlier, in 1857, the Frenchman Édouard-Léon Scott de Martinville (1817 – 1879) patented a sound recording device he had built, the phonautograph. It was this, not Edison’s phonograph, that the first device in world history to make the invisible sound, fleeting in the air, visible on a physical surface.
However, Edison’s phonograph was the first device to also reproduce recorded sound. Well, it depends. Namely, Edison’s contemporary, Frenchman Charles Cros (1842 – 1888), developed a working method for reproducing sounds recorded by Édouard-Léon Scott. Cros wrote down his method, depending on how one calculates it, 4 to 8 months before anyone knew anything about Edison’s phonograph. However, Cros never built the playback device he designed.
André Charlin: a hardworking engineer
Who invented the dynamic speaker driver? The same one that still produces sound in nine out of ten loudspeakers? According to the canon of Hi-Fi history, it was General Electric’s Chester W. Rice & Edward W. Kellogg, who, to simplify matters a bit, developed the modern moving-coil cone driver with an upper and bottom suspension based on the late 19th century inventions of Graham Bell, Ernst Wermer (Siemens & Halske) and Sir Oliver Lodge. They filed a patent in 1925. Around the same time, Edward Wente of Bell Labs and Paul G.A. Voigt of Edison Bell, filed their own patent applications to the same effect.
Before these gentlemen, however, there was a Frenchman who is claimed to have built a dynamic element, albeit a small one, as early as 1915, and André Charlin (1903 – 83), who already in 1922, three years before Rice & Kellogg, received a patent for his own electrodynamic element. So who in the end was the first? The British company Thomson Houston bought both Rice-Kellogg and André Charlin’s patents right away.
Who invented the electrostatic driver? The American Arthur Janszen? He received a patent for his tweeter in 1953, around the same time that Quad’s Peter Walker built his own full-range electrostatic ESL57 speaker. However, the history of electrostatics is much older. In the mid-1920s, single-ended versions were offered by several parties (C. Kyle, Eugène Reitz, etc.), and in 1927 Hans Vogt brought the first working push-pull electrostatic speaker to the Berlin Radio Exhibition. The first? Hard to say. André Charlin had received a patent for his own push-pull electrostatic speaker a year earlier.
Other inventions
Charlin holds several early loudspeaker patents related to cone materials, voice coils, acoustic lenses, crossover filters, among other things. In 1930, he was granted a patent for an acoustic labyrinth enclosure for low-frequency reproduction. For some reason, the labyrinth enclosure is also usually credited to Benjamin Olney (Stromberg-Carlson Telephone Co.), even though he filed his patent application four years after Charlin had received his.
Charlin’s technical innovations were immediately transferred to the loudspeakers that he customized for the studios of Radiovision-PTT and O.R.T.F. Later in the 1970s, Charlin utilized his inventions in his own commercial A. Charlin loudspeakers. For example, the cylindrical Colonne had two electrostatic tweeters at the top, in accordance with Charlin’s old concept, a dynamic midrange driver immediately below them, and a woofer directed downwards into the labyrinth beneath.
Overall, André Charlin left behind a long list of inventions, of which, according to one source, over 200 were granted a patent. Among these are, for example, the variable reluctance cartridge patented as early as 1926 (instead of the usual long cantilever, a permanent magnet is directly on the surface of the groove above the stylus), and a 1927 patent for the use of negative feedback for linearizing an amplifier and controlling loudspeakers. Harold Stephen Black, who is usually credited for inventing negative feedback, did not file his own patent application until the following year.
I’m not claiming that the above is absolutely accurate in all respects: To establish with certainty who exactly was the first, and in what, would require a micro-historical study of patents and patent granting practices. In one sense it would be pointless: In the early 20th century, it was common for engineers to invent similar things at the same time, independently of each other. All that said, it cannot be ruled out that André Charlin, with some of his inventions, may well have been years ahead of Rice & Kellogg, Vogt, Black or others in his inventions.
Stereophony
But to the point. Who created the technical basis of modern stereo reproduction? Alan Blumlein in the UK in the early 1930s? Let’s first go back to where it all began.
It all began in Paris in 1881. At the International Electrotechnical Exhibition, the Frenchman Clément Ader presented his famous teatrophono. The demonstration worked by placing ten carbon telephone microphones in front of the stage of the Paris Opera (Opéra Garnier), from which a signal suitably directed to two channels was sent in real time via telephone lines to the exhibition center, where the recipients listened to the performance with a device similar to a modern headset. Strictly speaking, the performance was not stereophonic, but rather binaural.
EMI’s Alan Blumlein deservedly received a patent covering a wide range of stereophonic reproduction in June 1933. In the same year, the first two-channel stereo records, made according to Blumlein’s patent, were engraved – 25 years before the first commercial stereo records! At the same time, Bell Labs’ Harvey Fletcher also conducted stereophonic experiments with large sound walls and test discs.
André Charlin? Inspired by the possibilities of film sound, he began to manufacture equipment for synchronous sound reproduction as early as 1930. He soon went so far as to produce a stereophonic soundtrack for Abel Gance’s monumental silent film Napoléon. In 1934!
Charlin continued to work on the ‘seventh art’ until the late 1940s, patenting improvements to cinema speakers, amplifiers and projectors of the time (e.g. the Cyclope focuser from 1935, the Actua Colour colour film projection method from 1938, etc.). After World War II, the technology known as the “Charlin system” was in use in over a thousand cinemas in France and abroad.
Back to music
This continued until the late 1940s. Charlin then sold his patents related to cinema playback to Philips and returned to recording and producing music. What makes Charlin’s life-work so great is not his groundbreaking technical inventions, but the way in which he, together with world-famous artists, recorded and produced recordings of significant musical history from the 1950s to the 1980s. An engineer, yes, but also a true music lover, a melomane, and Tonmeister in the truest sense of the word.
Charlin’s early recordings include, for example, records released on the Les Discophiles Francaise label, on which Erik Satie’s favorite pianist, Marcelle Meyer, plays Debussy’s preludes, the performance of which Meyer had discussed with the composer himself in his twenties!
Charlin recorded about a dozen Ravel albums for the Ducretet-Thomson label, which established the reputation of the prominent Portuguese conductor Pedro de Freitas Branco as a great Ravel interpreter, not to mention the fact that Branco and Ravel knew each other well. In addition to Ducretet-Thomson, Charlin’s recordings from the 1950s were released by at least EMI, Westminster, Decca, l’Oiseau Lyre, Hispavox, Lumen and BAM.
Charlin’s collaboration with the Erato label was short-lived, but it resulted in an album that has been heard by even those who have never heard Charlin’s name. Erato’s first album, Charpentier’s Te Deum orchestral overture (Marche en rondeau, Erato LDE 3009, 1953), has been replayed for decades by European broadcasters as the musical theme of Eurovision!
In 1962, Charlin founded his own record label, Disques A Charlin. Over a period of more than ten years, together with the French musicologist Carl de Nys, he unearthed and recorded rare musical treasures across Europe: music by French composers, organ music, chamber music, church music, and so on. In total, he accumulated over 100 records, many of them Grand Prix du Disque winners, and among them gems such as Pierre Faragg’s piano recordings. Not all of the material from that period has been reviewed.
The recordings selected and approved by Charlin can be listened to and ordered via the Editions André Charlin website. According to Bruno Gaullier, who is in charge of editing, a considerable amount of effort has been put into digitizing the analog master tapes. According to Gaullier, Charlin was not suspicious of the emerging digital technology, although he never had the time to make use of it.
Dummy head stereo
Charlin completed his first artificial head microphone in 1954. This led to a series of excellent European mono compatible stereo-records as early as 1958. In 1963, Charlin patented the dummy head microphone known as the Tete Charlin, which became the guarantee and trademark of the Charlin sound.
The artificial head was not a new idea. Blumlein described it in his patent, Bell Labs’ K. Hammer and W. Snow were playing with their Oscar with the Philadelphia Symphony Orchestra in the early 1930s, and in the early 1940s, Philips’ de Boer and Vermeulen built their first Kunstkopf to investigate the role of the head in hearing. More serious artificial head research had to wait until the 1960s, and in that wave Charlin was closely involved.
The dummy head recordings were originally made for binaural playback with headphones. However, André Charlin designed his artificial head microphone for stereo or quasi-binaural listening with speakers. According to Charlin, in such use, the artificial head microphone should not be a reconstruction of the human head with its auricle and canals, but two microphones placed symmetrically at close range from each other, in Charlin’s case Schoeps’s omnis, with a special acoustic barrier between them. The barrier is not like that of the Jecklin disk, but a kind of flattened sphere, from both sides of which microphones peer.
In this arrangement, the microphones record the stereo effect in an appropriate ratio based on both amplitude and time/phase differences. The function of the barrier is to increase the separation between the microphones and improve the localization of of sound sources based on the time difference. Carlin’s barrier neither reflects nor absorbs sound, but acts as a shade for sounds on the opposite side. There is no crosstalk, as would happen if the microphones in the microphone pair were further apart, and on the other hand, due to the obstacle between them, the stereo image and the impression of depth are better than with XY or Blumlein microphone pairs.
Tete-Charlin was based on years of testing, as was the skill of optimally positioning the head in a concert hall with respect to acoustics, instruments and other influencing factors. At the Champs-Élysées Theatre in Paris, which Charlin often used for his recordings, the head microphone was in a standard position near the conductor’s podium.
Charlin did not post-mix the sound recorded by the head, but during and immediately after the recording, he reportedly engaged in equalization to compensate for the acoustic deficiencies of the recording space and, according to his patent, to correct the head’s nonlinearities, especially between 3 and 4 kHz. It is this equalization process that many consider to be the source of Charlin’s rich stereo sound and legendary tonal balance.
My own experiments
Let me finish with some personal experiences. Inspired by Charlin’s story, I decided to do a couple of small listening experiments, the idea of which was to prevent the crosstalk from the right speaker to the left ear and vice versa that occurs when playing standard stereo recordings. I wanted to hear how increasing the binaurality of the stereo playback would affect the stereo image and sound coherence more generally.
First, I built a temporary wall between the speakers (this is a very old idea). The speakers are side by side with only a partition in between, and listening takes place the nose pressed to the other end of the wall. The result? It worked! The mechanical blocking of the crosstalk created, as expected, virtual sources on both sides, and the sound spread more widely than it would have done with two speakers in a stereo triangle. The shortcomings in the treble or bass were not too alarming.
Less surprisingly, the result depended a lot on the recording. On classical music compilation albums, two to three samples out of ten benefited from the arrangement. Orchestral music seemed to work best, a good example being the Chandos album of Albert Roussel’s Suite in F major (BBC Philharmonic Orchestra conducted by Yan Pascal Tortelier). I don’t know how the sound was recorded (Stephen Rinker), but the Birmingham Symphony Hall opened up to the front across its entire width, and the individual sound sources were separated from both sides of the stage and from the back.
I then replaced the wall with two side speakers, to which the channel separation signal is fed out of phase, i.e. I tried out a version of a surround sound. According to a source (Pekka Tuomela, Introduction to Hi-Fi, 1994), this kind of surround sound was invented by the Finnish Unto V. Somerikko in 1964. However, the additional speakers were not used as rear speakers (as in the plan), but more or less in line with the main speakers near the side walls, so that the listening angle was about 110 degrees. This corresponds to the stereo recording technique developed by ORTF (Radio France’s Office de Radiodiffusion Télévision Française) in the 1960s, where two cardioid microphones, spaced 17 cm apart, are spread out at an angle of 110° towards the stage. The ear is said to be most sensitive at this angle and, on the other hand, least sensitive to other directions of sound.
The result? It too worked! Ambiphony reduces the effect of crosstalk and thus widens the sound image, adds spatiality to the sound, frees up the rigid midrange image and makes the sound breathe better. In addition, the end result depended less on the recording and the musical genre than with the parting wall. The highlight of this experiment was Ravel’s Valse (DECCA Eloquence, stereo, 1958) performed by Ernest Ansermet and the Orchestra of the Suisse-Romand of Switzerland. What ease and effortlessness of sound, what a presentation of the distant orchestral parts. Makes one wonder. Warmly recommended to try.
