Motion Picture News (Oct-Dec 1930)

October 4 , 19 30 Motion Picture News 85 Cones or Horns, Which Better for Speakers? By LOUIS MALTER Of RCA Plwtophone, Inc., New York THE ultimate goal in theatre reproduction of sound pictures is the complete simulation in each part of the theatre of the sound originally impinging upon the pick-up microphone. This goal is at present far from having been achieved, one of the weakest links in the chain between sound striking the microphone and the sound impinging upon the ear of the auditor being the loudspeaker. At the present time the two chief types of loudspeakers in use in theatres are the cone speaker with directional baffle and the horn type speaker. In the following discussion the elements which measure the satisfactoriness of a loudspeaker and the influence of these elements in the reproduction obtained from the two types of loudspeakers mentioned above will be considered. The results of certain measurements will be used to explain the quality of reproduction obtained with each type of loudspeaker in theatres. The extent to which a loudspeaker can deliver satisfactory reproduction is measured by five factors, which may be classed in order of importance as follows : 1 — Frequency range. 2 — Uniformity of response. 3 — -Radiation of distribution characteristics. Loudspeaker Measurements If the pressure at a great distance from the speaker is measured at various angles with the normal to the mouth of the speaker a continuous curve is obtained. If the intensity along the normal is set equal to unity the curve obtained is defined as the radiation distribution characteristic for the particular frequency at which the measurement is obtained. The family of curves showing the radiout the frequency range is a measure of the quality of reproduction at various angles to the normal. The ideal characteristic is that in which the intensity is uniform for all frequencies throughout the angle defined by the entire audience in a theatre at the center of the speaker mouth and which then falls off to zero very sharply outside this angle. 4— Efficiency. The absolute efficiency of a loudspeaker is defined as the ratio of the total acoustic power radiated by the loudspeaker to the total acoustic power radiated by an ideal loudspeaker if fed from the same electrical source. 5 — Input power capacity. The input power capacity of a loud e2 speaker is measured by the value of ■ — , 4r where e is the maximum open circuit voltage which can be impressed upon the loudspeaker terminals without producing noticeable distortion, and r is a resistance eaual in magnitude to the impedance to which the speaker is designed to be connected. In order to obtain an accurate and absolute comparison between the performance of the "directional baffle type" loudspeaker and the horn type loudspeaker, frequency response characteristics of the most widely used type of directional baffle type loudspeaker and of the most widely used type of horn loudspeaker were obtained in the following way. Each loudspeaker was placed on the ground out-of-doors pointing directly upward and at a sufficiently great distance from buildings so that reflections from these did not affect the results obtained. A condenser microphone was suspended directly above the center of the loudspeaker mouth at a distance of 20 feet. The condenser microphone was connected to a sound amplifier. The entire sound measuring equipment was corrected electrically so as to possess a uniform over-all frequency characteristic. The frequency-response characteristics obtained in this manner are correct down to a frequency at which the radiation distribution characteristic of the speaker becomes so broad that sufficient sound is reflected from the ground to interfere with the radiation shooting straight up. In order to obtain the frequency characteristic for the lower frequencies, the speaker is placed along the ground with its mouth pointed towards the condenser microphone, which is placed close to the ground and at a distance of 20 feet from the speaker mouth. In this case the phase difference at the microphone between the direct radiation and that reflected from the ground is negligibly small at low frequencies so that the actual low frequency characteristic is obtained. This low frequency characteristic, however, must be divided by two due to reflection from the ground. (The assumption of practically complete reflection from the ground has been checked by experiment.) The fact that over a certain region (300 to 500 cycles) both methods yield the identical result is a further check on the composite method. Experimental Results A beat-frequency oscillator and high quality amplifier were used in making the measurements. These were set up indoors and leads run out to the speaker and the bullet amplifier associated with the condenser microphone. The recording mechanism shown of a rapi dand accurate frequency response characteristic being taken in a few moments. A complete description of a similar sound pressure recording mechanism has been previously published. In order to enable a fair comparison to be made between the frequency characteristics of the two types of speakers the two frequency response curves have been so placed as approximately to overlap in the center of the range, i. e.. between 300 and 800 cycles. This has Air Column Type of Speaker Controversial The Showman presents herewith *■ an authoritative discussion of one side of that hotly contested twosided question: "For power and uniformity of volume over a large area, which is superior, the directional baffle or the horn speaker?" Both types have their adherents, each supporting his claims of superiority for the type he champions with engineering data and opinion. In the belief that open discussion offers the surest road to final answer, or at least majority opinion, The Showman presents herewith the paper by Louis Malter, of RCA Photophone, Inc., New York, read at the last meeting of the Society of Motion Picture Engineers and reprinted with permission from the Society's Journal, in which advantages claimed for the directional baffle speaker are discussed. Discussions of others supporting the views and facts presented herewith or contesting them by citation of offsetting advantages of the horn type speaker, will be welcomed for consideration by readers of The Showman. Directional Baffle Type of Horn necessitated raising the curve of the directional baffle type loudspeaker. A careful study of these curves enables us to make a competent comparison between the two types of speakers as regards frequency range, uniformity of response and efficiency. It is immediately evident that the frequency range of the directional baffle type speaker is greater at both the low and high ends of the scale. The cut-off frequencies of the directional baffle type speaker are about 85 cycles and 6000 cycles at the low and high ends, respectively, whereas those of the horn type speaker are about 125 cycles and 5000 cycles, respectively. Thus at the lower end of the frequency scale the directional baffle type speaker has a half octave greater frequency range than the horn tvpe and about one-fifth octave more at the higher end. As regards uniformity of response the directional baffle type of speaker obviously has the advantage. Between 300 and 5000 cycles the response of the directional baffle type is slightly more irregular, but between these limits the variations in the response of the horn type are greater. Below 300 cycles, however, the horn type speaker is anpreciablv less responsive than the directional baffle type. As will be seen further, this deficiency in low frenuency resnonse in the (Conlimtea on page 111)