Page 16 - 2016Fall

Analog Sound Recording and Reproduction
ists as much as possible, that is, a brass band should sound like a brass band, a famous singer should be recognizable, and an orchestral accompaniment should sound convincing. Due to the various distortions and limitations of the early recording and reproducing equipment, they had to “cheat” by modifying the sound that was recorded so that the result was as purported by the record label when it was played. Still they claimed that they represented the artist in the best way. That is a tradition that has continued to this day and is the basis for the work of the modern record producer.
The Wish for a Strong Sound
The phonograph reproducing a record was competing with live instruments, and for this reason, it was desired that the reproduction be loud, particularly when performing out- doors. The first experiments used a mechanical amplifier in which a small reproducing diaphragm controlled a larger force from a rotating pulley that drove a large diaphragm and a large reproducing horn. But the real breakthrough came when Charles Parsons in the United Kingdom in 1903 developed the Auxetophone in which the reproducing dia- phragm was replaced by a valve that controlled pressurized air (Parsons, 1904). This amplifier enabled phonograph concerts in public parks. In 1916, Peter Jensen and Edwin Pridham in Oakland, CA, demonstrated the first electrically amplified public address system that was developed into the Magnavox loudspeaker, which provided limitless power to radiate any sound from an electrical signal, such as for a public address.
Use of Old Principles for
Sound Retrieval, 1887
Cylinders were simple to record and replay, but commercial manufacture that involved casting in a mold was difficult and had to rely on the contraction of the wax when it had cooled in its cylindrical mold. Emile Berliner realized that the ideas of Charles Cros were fertile and in 1887 first de- veloped a practical photo-reproducing method for etching grooves in a flat disc and then realized that it was possible to dispense with the optical step. The recording stylus only needed to scratch through a thin layer of wax to expose a metal surface below so that a spiral groove could be etched by an acid after recording. He also devised a method to make a negative impression of the groove, which he could use as a stamper to manufacture disc records. His grooves were lat- eral and not nearly as shallow as a hill-and-dale groove, and he obtained a very strong basic patent on letting the groove guide the reproducing stylus across the record. We now had
14 | Acoustics Today | Fall 2016
Figure 1. The two kinds of modulation of a groove in a record: the vertical (or hill- and-dale; top) that modulates the depth of cut and the lateral (or Berliner; bottom) in which the spiral groove is made “wob- bly” by a sideways movement of the cutting stylus. Copyright © 1983 George Brock-Nannestad.
two groove structures available: the vertical or hill-and-dale and the lateral (sometimes called “Berliner”). These are shown in Figure 1.
The Surviving Method of Mechano-Acoustic Recording,
Until 1925
The two-stage process of obtaining a groove by etching re- quired the use of strong chemicals in the recording studio, and this method produced noisy grooves. For this reason, Eldridge R. Johnson (later president of the Victor Talking Machine Company) refined a process that had originally been demonstrated by the Volta Laboratory, which was di- rectly cutting the groove into a waxlike tablet. The groove, considered as a zero line, was modulated by a stylus that was vibrated by the diaphragm.
When the recording was obtained in the wax, it was ex- tremely fragile, and even though very gentle reproduction soundboxes were used for listening to the wax recording, it became useless for further processing. To obtain test re- cords for proper evaluation, it was necessary to pass through several stages of electrolytic deposit; the wax surface was made conductive by means of very finely divided graphite and bronze powder, with the surplus brushed off by means of a very soft brush. Copper plating was performed to ob- tain a negative, and it was quickly discovered that although you could press a run of perhaps 200-300 records from the negative, it ultimately wore out and delivered a low-quality reproduction of sound. For this reason, two further plating steps were introduced: from the original negative a posi- tive was made and from that the proper tool, the stamper. When the stamper was worn out, one could produce another identical one from the positive. This complicated process is shown schematically in Figure 2 and it remains in the in- dustry and is still in use, although the materials are different. The provision of the consumer goods, a record, therefore re- lied on a large supporting industry. The large part played by industrial chemistry is not for this account.