Dr. Miller's interest in sound, and the influence that various materials had on the tone quality of a flute, necessitated his developing an instrument which could measure and record sound waves. The analysis of such records would then enable a study in a scientific manner of sound phenomena.
The desire to investigate the physical nature of musical sounds, and the sound-producing characteristics of musical instruments, led to a study of all available methods for recording the forms of sound waves. No device was found which was sophistically sensitive and free from disturbing influences for the proposed investigations, and a new instrument, the 'Phonodeik' (to show or exhibit sound) was developed. 60
With his phonodeik, Miller could actually photograph and then study sound waves of musical instruments and the human voice. Briefly, this new instrument consisted of a diaphragm of thin glass, 0.003 of an inch thick, mounted on the small end of a resonating horn. A lens, to focus a ray of light from a pin hole, and a small mirror so attached to the diaphragm that when the latter vibrated in response to a sound wave, the light ray also vibrated and was focused by a lens and reflected by the mirror to a moving film in a special camera. When the diaphragm moves under the action of a sound wave, the small spot of light traces the record of the sound wave on the film. The phonodeik was capable of responding to 10,000 complete vibrations per second.
The finished product would look to the layman like a vertical graph-like line which reflected the instrument or voice in a waving fashion much like an electrocardiogram. However, Miller found that the curves and waves represented not only sound, but also other physical phenomena. To analyze the photographic research of tones produced, he developed harmonic analyzers and systhesizers of great precision. A harmonic systhesizer is a machine which will generate separate simple harmonic motions of various specified frequencies, amplitudes and phrases, and will combine these into one composite motion which is recorded graphically.
In his investigations of the flute's tone he took about a thousand photographs, including every note of the instrument, each in several degrees of loudness. In checking flutes made of wood, silver, gold, and glass, he found that all had the same general characteristics except that gold produced overtones which were greater in number and strength.
The phonodeik was completed in 1908, and was first exhibited at the Baltimore meeting of the American Physical Society and the American Association for the Advancement of Science in December, 1908, and in Winnipeg at the British Association for the Advancement of Science in August, 1909.
A modified form of the phonodeik was arranged for demonstrations before a large audience. The sound from a voice or an instrument is produced in front of the resonating horn and the movements of the diaphragm will be thrown on a screen by the use of a motor-driven revolving mirror. The movements of the diaphragm are magnified 40,000 times, producing a wave which may be as long as 10 feet wide and 40 feet long. Thus, Miller was able to show his audience sound waves as he was lecturing about them. The first time this new innovation was put to use was at the Boston meeting of the American Physical Society in December, 1909. 61
The name 'phonodeik' was first suggested by the late professor Edward W. Morley, and was first applied to the device at the Boston meeting. 62 As to the purpose of this instrument, Miller himself had the following to say:
The ultimate purpose of the phonodeik is to produce a photographic record of a sound wave of such a size and of such definiteness and accuracy of outline as shall permit the quantitative study of waves having components of small magnitude and high frequency. 63
In 1917, photographic records were made of the sounds from a foghorn at Father Point, Quebec, on the St. Lawrence River. The laboratory phonodeik was used, with some drawbacks because of size, both on the land and on the boats which were boarded for this experiment. As a result of the difficulties and problems experienced in Canada, Miller developed a new portable phonodeik. It was this new model, as well as the larger phonodeik, that he took with him to the Sandy Hook Proving grounds in April of 1918.
During the first world war, Col. Robert A. Millikan, then Cheif of the Department of Science and Research of the Council of National Defense, requested Miller to study the physical characteristics of the pressure waves produced in the atmosphere near large guns in action. This was to be an investigation into the causes of "shell shock." Miller took a leave of absence from his teaching duties at Case from April, 1918, to November, 1919. (Millikan, it should be mentioned, won the Nobel Prize in 1923 for his work in the isolation and measurement of the electron.)
All conditions at the proving ground proved to be an extraordinary experience for the study of sound effects of various kinds. He used twelve microphones specially constructed for this experiment along with the phonodeik. Six microphones would be placed at various points on a circle with the muzzle of the gun as center or along any radius of the wave front. In some cases one was placed on the muzzle of the gun itself and the others would be placed anywhere from 50 feet to 200 feet. The final results included a better understanding of: 1.) the pressure effects in the air around large guns in action; 2.) the velocity of an explosion wave of great intensity; 3.) the form and physical characteristics of the sound waves from large and small guns; 4.) the normal velocity of sound in free air. He also obtained a measurement for the speed of sound in normal air to be 331.47 ±0.10 meters per second, equivalent to 1087.51 ±0.34 feet per second, which was very close to the presently accepted value.
How this knowledge was directly used in his investigations pertaining to musical instruments cannot be measured, but surely one study complemented the other.
A discussion is given of the tone production and of the relations of the nodes and loops of the vibrating air-column to the mouth-piece and the tone-holes in the several types of instruments. The varying qualities of tone are due to the prominence of certain overtones which are characteristic of each instrument; the general nature of these peculiarities have been determined by analysis of photographic records of the tones. 64
Wallace Clement Sabine (1868-1919) was a scientist who gave the first solution to one of the oldest and most important problems relating to sound, the acoustics of auditoriums. He laid the foundation for acoustics with his paper on "Reverberation," published in 1900. Miller's friendship with Sabine led Miller to further investigate this new field of acoustics. As a result, Miller became a recognized authority in acoustics, and he helped to acoustically design many buildings, including the Rockfeller Memorial Chapel of the University of Chicago; the chapels of Princeton and Denison Universities; Nebraska state capitol; National Academy of Science in Washington, D. C.; Church of the Heavenly Rest and Temple Emanu-El in New York City; and severance Hall in Cleveland, to name but a few. 65
In one account referring to the National Academy in Washington, the writer mentions a speaker in the Academy who dropped a paper clip during his lecture. The paper clip was heard at the back of the hall and "sounded an engineering triumph in the long-neglected science of acoustics." 66 Miller's work in acoustics was undoubtedly successful.