Volume 29, Issue 7, July 1957
Index of content:
29(1957); http://dx.doi.org/10.1121/1.1909052View Description Hide Description
The acoustical design of a church auditorium is of equal importance to that of the architectural design. It is essential, then, that the acoustical design follow the principles of projecting sound from the source to the listeners with maximum fidelity both for the listeners and for the performers. Church acoustics has varied from the dead, low reverberent type at one extreme to the excessively reverberant type at the other extreme. In between these extremes there is a range of good listening conditions for music and speech, for all types of liturgy. Longer reverberant times than usual are recommended for small and medium size auditoriums. The enhancement effect of reverberation on musical tones is shown to be related to the optimum reverberation characteristic. A new design of church auditorium is submitted which is interesting in its novelty, and is still practical from a modern architectural point of view.
Organ design has passed through extremes such as the pure romantic and the baroque. Either in its extreme is undesirable for general church use. Space requirements for organs and some principles of organ design including the design of mixtures to meet any reasonable acoustic condition are given. This paper is given to stress the musical point of view of church design to the acoustician and architect so that extremes in acoustics will be avoided in the future.
29(1957); http://dx.doi.org/10.1121/1.1909054View Description Hide Description
This note describes some measurements on the network representing the basilar membrane, which were made to determine the effect of the position of excitation on the response of the cochlear partition. It was found that the response of the electrical model is insensitive to the position of excitation if the excitation is applied on the stapes side of the point of maximum response. If the excitation is applied on the helicotrema side, the response is highly localized around the point of excitation and has an over‐damped character.
29(1957); http://dx.doi.org/10.1121/1.1909056View Description Hide Description
Underwater hearing thresholds were measured at frequencies of 250, 500, 1000, 2000, and 4000 cps on four Aqualung equipped divers with normal hearing in air. In the frequency range investigated, the minimum audible fields underwater are from 44 to 60 db greater than the minimum audible fields in air, and the shape of the threshold vs frequency curve is different.
29(1957); http://dx.doi.org/10.1121/1.1909058View Description Hide Description
Several investigators have attempted indirect determination of the bone‐conduction threshold in a free sound field. It has recently become possible to attenuate sound transmitted through the auditory meatus to a level below that of a sound reaching the ear through other pathways. By varying the properties of sound‐attenuating devices, and with the help of resonance phenomena, it could be shown that the remaining hearing was actually controlled by body conduction, if not more specifically by bone conduction. According to these measurements, the “minimum audible field” for body conduction (MAFB) exceeds 80 db re 0.0002 dyne/cm2 at low frequencies and decreases gradually to a minimum of about 45 to 50 db between 1500 and 2000 cps. For higher frequencies it increases somewhat again. The difference between body conduction and air conduction seems to be maximal in the vicinity of 500 cps—approximately 65 db. At frequencies around 100 cps the difference is of the order of 50 db. This difference holds also for the frequency range between 2000 and 6000 cps. Between 1500 and 2000 cps the difference is only about 45 db.
29(1957); http://dx.doi.org/10.1121/1.1909060View Description Hide Description
The mathematical analysis of the directional characteristics of linear additive arrays is given in a polynomial representation. The result of multiplying and taking a time average of the outputs of several detectors also has directional characteristics that may be expressed as polynomials. It is shown that the same directional characteristics may be obtained from multiplicative arrays having a small number of detectors as with an additive array with a large number of elements. The length of the multiplicative array is about half the length of the additive array having the same directional characteristics.
29(1957); http://dx.doi.org/10.1121/1.1909062View Description Hide Description
The absolute calibration of a variable capacitance type displacement detector was performed by using the diaphragm of a condensermicrophone as a reference moving surface. The detector was calibrated by coupling its associated probe to the front end of the microphone to detect the variation in capacitance between the diaphragm and the probe, when the diaphragm was driven by applying an electrical signal to the microphone terminals. Computation of the displacement detector response was made in two ways: one based upon the reciprocity relationship in a set of two linear equations describing the microphone as a reversible transducer; the other, upon the acoustical impedance of a cavity with rigid walls. Measured values of the displacement detector response are plotted from 10 cps to 40 000 cps.
29(1957); http://dx.doi.org/10.1121/1.1909064View Description Hide Description
Response curves, important both for steady sound and transients, give a far‐reaching insight into the objective characteristics of violins: good ones exhibit large amplitudes at low frequencies and small ones at high frequencies, a broad minimum near about 1500 cps, and larger amplitudes between about 2000 and 3000 cps. The musical subjective significance of these physical properties is mentioned briefly. In general, the sound pressure radiated from a violin follows the inverse‐distance law, being independent of frequency. The influence of wood thickness is very important, that of the varnish is comparably small. Pine has a greater damping at high frequencies than at low frequencies. This seems to be a good acoustical reason for making important parts of stringed instruments of pine. Sapwood is better than heartwood. Similarly, some kinds of varnish produce more damping at high frequencies than at low frequencies. If a resonance curve is to be imitated in detail, it is necessary to change carefully the wood thickness of certain parts of the violin body. The applicability of present‐day scientific knowledge to the construction of violins is here outlined.
29(1957); http://dx.doi.org/10.1121/1.1909066View Description Hide Description
Following a suggestion by Markham, Beyer, and Lindsay, the Stefan‐Boltzmann radiation law is applied to estimate the attenuation of approximately plane compression waves resulting from heat radiation. The attenuation is found to be independent of the frequency of the compression waves in the entire relevant frequency range and inversely proportional to a linear dimension of the wave fronts. At ordinary temperatures, this attenuation is negligible compared to that due to thermal conduction and viscosity, verifying the results of Rocard and Rayleigh. In gases at temperatures of 1000°K or more, however, the attenuation due to heat radiation is not entirely negligible under certain conditions.
29(1957); http://dx.doi.org/10.1121/1.1909068View Description Hide Description
The applicability of the self‐reciprocity technique to the calibration of a reversible compressional wave transducer in a solid medium is demonstrated, and a method is described of correcting for the electrical loading by the source signal generator of the transducer being calibrated.
29(1957); http://dx.doi.org/10.1121/1.1909070View Description Hide Description
This paper considers changes in the shock spectrum that may arise from minor variations in the form of an applied pulse. In addition to the rise and decay times the nature and symmetry of the discontinuities has a profound effect on the high‐frequency components. The selection of parameters in the design of a square wave accelerator is discussed. The effects of shock excited resonances in a machine are investigated together with methods of minimizing their effect.
29(1957); http://dx.doi.org/10.1121/1.1909072View Description Hide Description
Linear elastic systems which have a single input point and a single output point can be characterized by a pair of simple linear equations, namely, where the F's are forces, the V's are velocities, and the α's are the four‐pole parameters for the system. The use of four‐pole parameters permits combinations of mechanical elements to be handled as a single entity, and, as the parameters belong only to the elastic system for which they are defined, they do not depend upon what precedes that particular system or follows it in a mechanical network. This permits the statement of results in a rather general form. These parameters have been used for many years in the analysis of electrical circuits, and it is the purpose of this paper to show their utility in the field of mechanical vibrations. In particular, the four‐pole parameters for the basic mechanical elements of mass, spring, and resistance, will be obtained. Also, a description of mechanical sources will be given. The rules for the connection of four poles and a method of experimentally measuring four‐pole parameters will be presented. A few specific problems illustrating the use of four poles will be set forth, namely: a generalized vibration isolation problem, the response of an elastically mounted mass on an electrodynamic shake table, shock excitation analysis, and some distributed parameter systems (i.e., helical springs and rubber in shear).
29(1957); http://dx.doi.org/10.1121/1.1909074View Description Hide Description
This paper considers the response of thin elastic plates to sinusoidal acoustic excitation. A theoretical method for obtaining the dynamic deflection and stress in the nonlinear region is given. The theory is compared with test results showing rather good agreement. The application of the theory to the design of panels to withstand acoustic fatigue is discussed in the latter part of the paper.
29(1957); http://dx.doi.org/10.1121/1.1909076View Description Hide Description
The acoustic transmission of a sandwich consisting of a pair of plane parallel solid plates immersed in a liquid has been investigated theoretically and experimentally. Not only the dilatational waves, but also the shear waves, are considered. Results show greatly improved transmission characteristics over those of single thick metal plates with respect to dependence on both frequency and angle of incidence. Theoretical transmission curves for a sandwich of two steel plates have been prepared to exhibit the dependence of transmission on frequency, on ratio of plate thickness to wavelength and separation to wavelength, and on angle of incidence, respectively. Calculated transmission curves for single steel plates, including half‐wave plates, are shown for comparison. An experimental test was made on a steel sandwich with lateral dimensions of the order of two wavelengths and plate thickness‐to‐wavelength (in the steel) ratio equal to 0.05. The measured transmission losses are within ±4 db of the calculated values.