Version: July 11, 2002
by Geoffrey I Brown, 1999
I built a machine (which I call a transcription deck) at the Lowie Museum of Anthropology in the early 1970's, based on a variety of modified components. The mandrel (with end-gate) and base were a modified Dictaphone cylinder lathe, intended for 6" (152 mm) long cylinders. I had a new fairly massive flywheel pulley machined, which was driven by a Hitachi variable-speed instrument motor with integral tachometer-feedback speed-control. The speed was regulated by a 10-turn precision potentiometer fitted with a large diameter knob. The operating range was ~50 - 300 RPM.
A large-face microampere-meter was scaled to show RPM directly. As a final touch, strobe dots were painted on the flywheel and illuminated by a small neon lamp. When this lamp was switched on, the dots appeared stationary (with US 60 Hz current) when the mandrel was adjusted for 180 RPM. A screwdriver-adjust potentiometer allowed calibration of the RPM meter to this built-in strobe.
The motor was supported on vibration absorbing mounts consisting of Buna-N rubber diaphragms clamped into ring-brackets supported on long-excursion springs. The drive belt was a small-section "O" ring under very low tension. The massive cast-iron base of the cylinder lathe was mounted on a plywood deck that was, in turn, supported on thick sponge rubber around the perimeter. Vibration isolation was excellent, with no rumble other than that recorded on the cylinders themselves.
Tracking of the grooves was accomplished by a modified Rabco SL-8E parallel tracking arm. The reducing gear-box on the DC drive motor was replaced with one of lower ratio to increase the tracking speed in order to accomodate any groove pitch, and the polarity of the battery which powered the tiny motor was reversed to drive the arm from left-to-right instead of right-to-left. In this arm the lateral advance was controlled by two hair-fine platinum wires which acted as a switch when the arm would swing from the center position. These wires were re-aligned to work with the reversed tracking direction and adjusted to allow virtually no swing of the arm.
A Dictaphone crystal cartridge (from the late 1930's or 1940's) was found to be the ideal pick-up cartridge, as it inherently limited the frequency range and screened out spurious high and low frequency noise. Adapted magnetic cartridges were not found to be as satisfactory. Spherical quartz styli were salvaged from old Ediphone and Dictaphone acoustic players and mounted on the crystal cartridge. Stylus pressure was adjusted at approximately 3 grams.
This apparatus was used to transcribe some 3000 ethnographic cylinders in the collection at the Lowie Museum of Anthropology of the University of California at Berkeley (since renamed the P.A. Hearst Museum). The stylus was checked periodically with a microscope but showed no wear at any time, nor did it collect wax. As far as I could determine, the cylinder grooves were not affected despite the fact that most were of very soft brown wax. Note that the cylinders were carefully cleaned of dirt, grit, and fungal debris before any attempt was made to play them.
The apparatus was then borrowed by the U.S. Library of Congress where it was used to transcribe over 9000 cylinders of ethnographic and commercial recordings. The machine was then transported to the Indiana Universiy Archive of Recorded Sound where it was used to transcribe their collection of 7200 cylinders.
As far as I know, the same stylus was used for all 19,200+ of these cylinders which had been recorded on a wide variety of Edison, Ediphone and Dictaphone machines, both electrical and spring-wound, from about 1903-1933.
When I left the P.A. Hearst Museum in 1990, the machine was in storage.