Radio Broadcast (May 1928-Apr 1929)

Advantages and Disadvantages of THE CONDENSER-TYPE LOUD SPEAKER By JOSEPH MORGAN International Resistance Company IN 1881, Professor Dolbear announced the first condenser receiver for telephone systems. This was the fore-runner of the condenser loud speaker which is being introduced into radio to-day. While the principle is interesting and may very well yield important results, there is nothing essentially new in the idea. It is to be noted that the magnetic and condenser types of loud speakers are about equally old in principle and, therefore, we must examine the condenser loud speaker with great care before we pronounce a verdict. It must be evident, therefore, that the condenser loud speaker is no more nearly ideal in its basic principle than the magnetic. In this article the principle of the condenser-type loud speaker will be set forth together with its advantages and disadvantages. Also, a brief account of some typical loud speakers which are now being manufactured will be given together with a description of the method of application of such loud speakers. When two conductors of electricity are separated in space by a non-conductor of electricity, we have what is called an electrical condenser. If these two conductors are charged with electricity of like sign, they tend to repel each other and if they are charged with electricity of unlike sign, they tend to attract each other. Suppose that these two conductors are large, flat, metallic plates of equal area, separated by a thin film of air. (See Fig. 1a.) If a difference of potential or voltage is applied to these plates, a force will be exerted tending to draw these plates together, and the force will be proportional to the area, A, of one side of one plate; it will be proportional to the square of the voltage between the two plates; and it will be inversely proportional to the square of the distance, D, between the two plates. From the above paragraph it is seen that the greater the voltage the greater the force, the larger the size of the plates the greater the force, and the smaller the distance between the plates the greater the force. If we make one of these plates quite heavy and stationary and the second plate very light and movable (see Fig. 1b), the application of a varying voltage to these plates will tend to draw the light movable plate to the heavy stationary plate with a force which will increase as the square of the voltage. If an alternating voltage, for example, the usual 60uycle, 110-volt house current, is applied between the two plates, the movable plate will tend to move in and out at double the frequency of the applied voltage which, in this case, would amount to 120 times per second. This result would be obtained since the plates tend to pull together both on the positive and on the negative halves of the alternating-voltage cycle (see Fig. 2, diagrams a and b). Thus instead of obtaining a 60cycle tone by virtue of the motion imparted to the surrounding air by the movable plate, we would obtain a 120-cycle tone. This is a perfect instance of complete distortion, since the original tone is absent and is replaced by one of entirely different frequency. Suppose that this alternating house current Colin Kyle (left), inventor of the Kyle condenser loud speaker, is demonstrating his inven tion to three authorities on radio and acoustics. In his hand he is holding a section of the loud speaker, and standing on the floor is a completed model As has been predicted in these pages, the condenser-type loud speaker is apt to attract a great deal of attention in the industry during the coming season. What is it? How does it work? How does it compare with other types? These and other questions, Mr. Morgan, of the International Resistance Company, whose previous articles on loud speakers in this magazine have been so well received, attempts to answer. The device is not a panacea; it will not "revolutionize" the industry, Here is a straight-forward analysis of the whole question. — The Editor. be replaced by the voice current from the output of a broadcast receiver. It must be obvious that the light movable plate, which we shall henceforth call the diaphragm, would produce a hopelessly distorted sound since it would move in accordance with the square of the voice voltage and at double the voice frequencies. Minimizing Difficulties LET us see how these essential difficulties are minimized. Suppose that we place a high direct voltage, for example 500 volts, across the plates of our crude condenser loud speaker. There will be a strong constant attraction between these plates, due to this constant direct voltage. If now we superimpose a much smaller 60cycle sine-wave voltage upon these same plates, this alternating voltage will tend to increase and decrease slightly the direct potential which we have already established between the plates. In other words, the force will alt ernately become a little greater and a little less than the initial force due to the direct voltage (see Fig. 2, diagrams c and d). It can be shown mathematically, that the motion of the diaphragm under these conditions will be approximately in accordance with, and proportional to the alternating voltage applied between the plates. The smaller the ratio of the alternating voltage to the constant applied direct voltage, the more accurately the diaphragm will follow the alternating voltage variations. It is exceedingly important to note that there will always be a component of the motion which is twice the frequency of the original voltage and also that the motion will never be exactly directly proportional to the applied alternating voltage. In other words, in this type of loud speaker, as well as in the magnetic and electrodynamic types, there is always some inherent distortion. A mathematical analysis of the condenser loud speaker shows that the greatest response is obtained when the plates are as close as possible together and both the constant direct voltage and the alternating applied voltage are as great as possible. We have just shown that the alternating voltage must be a small fraction of the direct voltage in order to minimize distortion. This, therefore, is our first limitation. Second, the direct voltage, which we shall henceforth call the polarizing voltage, must not be increased beyond 500 or 600 volts because of the danger of break-down between the fixed plate and the diaphragm. Further, it is not safe nor practicable to generate much higher voltages than 600 for such a purpose. Third, the distances between the plates cannot be made ^definitely small for several reasons: (a) because the polarizing voltage would tend to puncture the insulation between the two plates (in this case, air) if the distance were too small; (b) there must be sufficient distance so that the diaphragm may move back and forth in order to impart a mechanical wave motion to the air in front of it; (c) if this distance were too small, the diaphragm might actually strike the stationary plate causing a short circuit if too great a voice voltage were applied or if resonance obtained either in the electrical circuit or in the mechanical construction of the loud speaker. Hence, it is seen that compromises must be effected throughout the design of this type of loud speaker just as in the case of the magnetic and electrodynamic loud speakers considered in previous articles. As a result of these compromises, the sensitivity and efficiency of the condenser loud speaker is, in general, low. Due to the small permissible distance between the diaphragm and the back plate the large amplitudes of motion necessary for the adequate radiation april, 1929 page 369 9