Clyde E. Ingalls
Ithaca, New York
From the Cornell University, School of Electrical Engineering and Cornell Aeronautical Laboratory, Inc.
A series of experiments is described in which radar transmitters operating at 1.3 and 10 GHz were "heard." Apparently, the process of hearing did not involve the ear, but included only the brain and nervous system in the vicinity of the brain.
The effect takes place at energy levels that are considered safe for exposure all day. The effect is suggested as a means of aiding in the location of hearing difficulties in persons. It is also discussed in connection with reports of the hearing of meteors and auroras.
An interest in problems connected with re-entry bodies in the atmosphere led to an interest in reports of "hearing" meteors and auroras. The hearing of meteors was supposedly under conditions requiring sound to travel at a velocity far exceeding the velocity of sound in air at 343 meters per second, probably approaching the velocity of electromagnetic wave propagation. At this time there was a report of someone hearing a radar at an installation in Turkey. On investigation, this proved to be true.
A like radar was found in the United States and a meter secured for field strength measurements to avoid overexposure and possible damage to the eyes, brain, or other parts of the body. Although there was considerable ambient noise, the radar could be heard by a person who immersed himself in the edge of the beam, the center of the beam being strong enough to be hazardous. The sound was something like that of a bee buzzing on a window, but with, perhaps, more high frequencies.
Possibility of the effect being noise by sound waves from the radar was eliminated by placing a large (about 3 by 3 foot) square of window screening between the observer and the radar, close to the observer. With the screen shield in place, the radar sound disappeared. A hole was cut in the screening, large enough to put the ear through. When the ear was put through the hole, there still was no sound. The only part of the body which allowed the observer to hear the radar was a place on the head above the forehead.
From this, it appears that the electromagnetic wave effects the nervous system at the brain directly and does not use the normal auditory channels. No disturbance in the visual senses was found, although a search was made. Possibly the like visual senses are shielded more by the head.
The sound seemed to come from about a meter or two above the head. This varied somewhat with individuals. Placing the fingers to cut out ambient sound made the source seem to come down to the very top of the head. This is the same spot on the head at which the source seems to be when two well-separated loud-speakers with identical excitation are used and the observer is located at equal distances from the two speakers and facing them. Placing the fingers in the ears in reasonable ambient noise does not seem greatly to affect the threshold value at which the radar is heard.
Persons with defective hearing were taken to the radar location. Some of them could hear the radar and some could not. It seemed to depend on the type of hearing loss and the frequencies involved. All who could hear high frequencies could hear the radar. A person who, apparently, had normal hearing, could not hear the radar. By taking an audiogram, he was found to have deficient hearing above 1,500 Hertz, seriously so above 3,000 Hertz per second.
Other radars were used, and it was found that it was possible to "hear" radars at approximately 1, 3, and 10 GHz. Measurements of the threshold of "hearing" of the radars at 1 GHz showed the free field strength to be 0.3 milliwatts per square centimetre at a peek voltage gradient of 12 volts per centimetre. At 3 GHz, the corresponding threshold values were 0.18 milliwatts per square centimetre and 18 volts per centimetre. No measurements were made at 10 GHz.
The apparent lack of correlation of watts and volts is due to differing pulse lengths and repetition rates. The effect is seen to cover a very broad radio-frequency band.
An electrostatic field was produced between two plates and the head placed between them in various positions. Even much higher than the threshold values mentioned failed to produce effects which could be attributed to other than normal aural paths from 20 to 20,000 Hz.
Bracing the plates essentially eliminated their vibration, but the skin on the face could be heard to vibrate. No coil was available to produce magnetic fields when the coil itself did not make too much noise for proper discernment. It appears that tests must be made at modulated inaudible frequencies.
It appears that the "hearing" of electromagnetic waves is a very broad band audio-frequency effect, that is, the audio frequencies which are "heard" from a pulsed radar seem to extend to a higher frequency that can be heard normally from a noise source and a loud speaker. Experiments in matching the sound from a radar indicated that a noise source should be used for best matching, but the sound from the noise source still seemed to lack something in the high frequency region.
If the effect does indeed bypass the ear, it would seem that the effect should be valuable in determining where certain hearing defects occure physically. The possibility exists of modulating a device, such as a radar, to sample audio intelligence and communicate with an individual with defective hearing. It appears that pulse operation is necessary to have low average but high peak power. A disadvantage is that the range of power between the threshold of "hearing" radar and the level at which bodily harm can occur, with prolonged exposure, is not as great as would be desirable. The threshold of "hearing" occurs at 200 to 300 microwatts per square centimetre, and the "safe" level for working all day in radar fields is about 10 milliwatts per square centimetre, using probably at least a 10 to 1 safety factor; that is, the level of bodily damage is probably somewhat above 100 milliwatts per square centimetre.
Sommer and Von Gierke have done a great deal of work with electric fields, showing that the skin on the head can be vibrated by an electric field and that the sound reaches the ear by bone conduction. Likewise, the eardrum can be vibrated directly by an electric field. They infer that the "hearing" which occurs in radar beams is caused by the pressure exerted by the electromagnetic wave, that is, twice the power density divided by the velocity being the maximum pressure when the reflection from the head is complete.
However, they have done no work with electromagnetic waves and so can only speculate.
It is very difficult to use electrical instruments for measurements of nerves to determine what is occurring within the head, since the measuring instruments are affected directly by the electromagnetic waves.
It appears that indirect methods will be needed to determine the exact method by which "hearing" of electromagnetic waves takes place.
Certain evidence points to the fact that the effect is not due to either air or bone conduction. First, the best sensitivity of the ear, according to Sommer and Von Gierke, is to air conduction, by two orders of magnitude. The maximum peak pressure during a radio frequency cycle of a radar, even with 100 per cent reflection, is about equal to the sensitivity of the ear to the root mean square pressure of a continuous sine wave at 1000 cycles per second. The average pressure of the radar wave at the threshold of "hearing" is roughly three orders of magnitude less than the average pressure of a sine wave in air at the threshold of hearing air waves. This results from the low duty cycle of the radar wave.
Second, the location of the most sensitive area for "hearing" radar is remote from the ears, on top of the head. Third, the subjective frequency spectrum seems to include higher frequencies for radar "hearing" than for normal hearing of air waves. Forth, the direction from which sound seems to come does not change as the head is turned about in the radar field.
From these considerations, it appears that neither air conduction nor bone conduction gives a satisfactory explanation of the "hearing" of radar waves. A direct involvement of the nervous system appears to furnish a more satisfactory explanation or at least a more fruitful avenue of investigation. However, much more work is needed to determine either the mechanism by which the phenomenon takes place or its practical use.
The "hearing" of electromagnetic waves is an established fact. It appears that this takes place by direct stimulation of the nervous system, perhaps in the brain, thus bypassing the ear and much of the associated hearing system. It is a possible, perhaps the most probable, explanation of the reports of hearing meteors and auroras. It is also a possible tool in the investigation and treatment of problems in hearing.
Much more work in this field is needed.