US 3862366 A
A sound radiator comprising at least four high frequency loudspeakers with their axes intersecting in zigzag relationship, and a single low frequency loudspeaker whose center is at a distance not more than 1.5 times the diameter of the low frequency loudspeaker from the center of that high frequency loudspeaker which is farthest from the low frequency loudspeaker. The wave length of the crossover frequency of the low frequency and high frequency loudspeakers is not more than the smallest dimension of the housing for the unit measured in a plane perpendicular to the axis of the low frequency loudspeaker. A perforated obstacle is disposed in front of part of the high frequency loudspeakers.
Description (OCR text may contain errors)
United States Patent [1 1 Huszty et a1.
[ SOUND RADIATION SYSTEM  Inventors: De'nes Huszty; Andras Illnyi; Ilona Magos Ne Ne'meth; Karoly Szabados, all of Budapest, Hungary  Assignee: Elektroakusztikai Gyar, Budapest,
Hungary 22 Filed: Aug. 10,1972
21 Appl. No.1 279,515
 Foreign Application Priority Data Jan. 21, 1975 Column Loudspeaker Systems, Electronics World, p. 25, 26, 27, & 76, by Augspurger, 179-1E, 663.
Primary ExaminerWilliam C. Cooper Assistant Examiner-.Douglas W. Olms Attorney, Agent, or FirmYoung & Thompson  ABSTRACT A sound radiator comprising at least four high frequency loudspeakers with their axes intersecting in zigzag relationship, and a single low frequency loudspeaker whose center is at a distance not more than 1.5 times the diameter of the low frequency loudspeaker from the center of that high frequency loudspeaker which is farthest from the low frequency loudspeaker. The wave length of the crossover frequency of the low frequency and high frequency loudspeakers is not more than the smallest dimension of the housing for the unit measured in a plane perpendicular to the axis of the low frequency loudspeaker. A perforated obstacle is disposed in front of part of the high frequency loudspeakers.
2 Claims, 20 Drawing Figures Pmzmsmm 3.862366 saw ozuF 11 PArtmw zlms' 3.862.366
' sum mar 11 Pmmim m 3. 62.366
SHEET CBUF 11 2o so 200 500 so 100 200. 500
F g- 1 i I v m 100 200 500 1k 2k 5k PATENTEDJANZI ms SHEET near 11 Fig.15
5k k k Fig.16
5k 10k 20k SOUND RADIATION SYSTEM Subject matter of the invention is a sound radiation system built up of sound radiatingelements and elements modifying the radiation properties, further of active and passive electrical networks, constituting a functional unit whose transmission characteristics and radiated output are at the position of the listener in the room of a frequency-dependent uniformity better than that of sound radiators so far known and realized.
It is generally known to specialists engaged in the study of the problems of electroacoustic transmission that in closed spaces, such as halls, theatres, cinemas, studios, etc., wideband sound transmission of particularly good quality cannot be guaranteed with the conventional sound radiators for high outputs. Consequently in high standard systems the audio frequency band is in general split up into two bands and the sound radiating elements are supplied by a single output amplifier controlled by a common signal source (Reference l and 4), or separately by each an output amplifier (Reference 3 and 4).
By sound radiating elements preferably sound radiators (e.g., reflection box, horn loud-speaker, or any other arrangement) specially designed for each a low, medium, or high frequency transmission band should be understood. For the distribution of the signal of the complete audio frequency range active networks, i.e., such as incorporate a loudspeaker, or passive networks, i.e., such as are void of a loudspeaker, have been developed. For the improvement of the service the circuit controlling the output amplifier supplying the sound radiators is normally equipped with a circuit controlling the high and bass tones (Reference 3).
The problem of the quality of sound transmission confronts the designer of the transmission chain with difficulties hard to overcome. The design of the transmission chain involves subjective and acoustic prob lems which in their interrelations are still subjects of research work.
Experience so far accumulated shows that the problem of designing sound radiators of satisfactory quality cannot be'solved unless by a thorough study of the store of data collected of the sound radiation, room acoustic subjective acoustic and electroacoustic transmission chain in conjunction with due consideration to the knowledge on hand of the signal transmitted over the chain. In order to be able to design a sound radiation system of unexceptionable quality for indoor use the sound pressure versus frequency characteristics of the sound radiator, the directional characteristic and the properly interpreted unevennesses of the transmisa sinusoidal signal. The rather uneven characteristic of v i the sound radiator positioned at the end of the chain sion characteristics obtained at the site of listening-in as the resultant of the interaction of the room and the sound radiator are of particular significance.
At the present state of technics for a proper solution of the problem the following points have to be satisfied (Reference 4):
1. Above 0.8 kHz the directional characteristics must be similar to one another and within a conic angle of radiation of 120 fluctuations must be kept at a minimum.
2. At the site of listening-in and within the specified transmission band the transmission characteristic should depart from the characteristic most appropriate for performance of the task in the slightest possible degree.
manifesting itself in the room (Reference 5) will not for random signals be sensed by the human organ of hearing as uneven, if within the so-called critical bands of hearing of a width of Af the number of unevennesses of the characteristics of a uniform width of Af, e.g., of the maxima, is sufficiently large, i.e., Af,,,/Af l0 (Reference 6). This postulate is met for signals of a frequency higher than Hz, e.g., in a room of V 100 metres cube (a living room of medium dimensions, or the usual dimension of the technical room in a studio). With the growth of volume this limiting frequency (cutoff frequency) is apt to slide downwords. On the other hand the unevenness of the characteristic manifesting itself between the critical bands of hearing, i.e., making itself felt when measured with a noise voltage of a bandwidth of a third of an octave, is fairly well perceptiblee. Consequently if on the ground of the statistical properties of the programme signal and the subjective perceptibility of the unevennesses, the unevennesses of the directional characteristic and the axial sound pressure vs frequency characteristic, i.e., the most characteristic, objectively measurable acoustic data of the sound radiation system are defined by a noise of a bandwidth of one third of an octave, then the task remains to design a sound radiator whose acoustic properties when measured with a signal of a bandwidth of one-third of an octave should remain as even and uniform as possible. This postulate has been ignored in all of the known solutions of the problem (Reference 1, 2, 3).
As is known there is a considerable difference between the transmission characteristics of sound radiators measured in an open space and that measured in a room. This is explained by the interaction of the sound radiator and the room. In a studio technics already some time ago attempts were made to reduce the effects of this interaction by standardizing the values of the technical room (Reference 8.). However, according to recent research this method has failed to solve the problem, because when sound radiators of the usual arrangement and properties are operated a sound pressure fluctuating between extremes wide apart will be experienced at the various points of the room (Reference 9).
In other words, even when no other postulates have to be satisfied owing to the interaction of extreme complexity between sound radiator and room no constant frequency-independent transmission characteristic could be obtained in the room even when the sound pressure versus frequency characteristic of the sound radiator measured in the axis were ideally uniform in the open sound space. On the otner hand if for the frequencies above approximately 1 kHz the directional characteristics of the sound radiator are irrespective of the frequency recirocally very much alike, then a diffuse space of large volume will come into being in the room (Reference 9).
In the course of further research it has been recognized that the guarantee of,-frequency-independent reciprocally similar directional characteristics for the sound radiator at least above 1 kHz, in both the horizontal and vertical planes is a fundamentally essential postulate to be met for the production of a possibly perfect sound transmission in the room. As a matter of fact if the presence of frequency-independent reciprocally similar directional characteristics is guaranteed, then the observer taking up alternately different positions in the room in respect of the principal directions of radia tion of the sound radiator, will owing to the constant directional characteristics sense an essentially uniform transmission characteristic in the audio frequency band. 1
Here the level drop owing to the directional characteristics preferably departing from the inaudible properties of the electoacoustic transmission chain formed between the ear of the observer and the sound radiator, referred to the critical bandwidths of the sense of hearing, and from the spherical shape may be ignored.
Furthermore it has been recognized that a set of frequency-independent constant directional characteris tics has the welcome property that the directional factor, i.e., the quotient of the square of the sound pressure excited in the axis and the radiated output, of a sound radiator of such properties (Reference 10) may be produced in the manner specified in dependence on the frequency. So, e.g., the directional factor may be made constant irrespective of the frequency. All that has to be attended to is that the level of the signal supplying the sound radiator should as a function of the frequencyvary inversely to the variations of the axial characteristic in order just to compensate the unevenness of the axial characteristic (Reference 6). This problem may be tackled with the aid of known electrical metworks.
None of the methods described in professional literature satisfy these postulates. E.g., one of the methods (Reference 1) contents itself with a constant directional factor without recognizing the fact that for subjective reasons this circumstance will not guarantee a subjectively unexceptionable sound picture at the different sites of the room unless simultaneously a similar ity of the directional characteristics of the sound radiator has been ensured within a wide sound band.
The authors of another'method (Reference 2) are though aware of that the directional characteristic and the frequency response as closely related properties are essential for sound radiators of good quality, yet fail to specify conditions for the constancy of the directional characteristics. On the other hand these authors disapprove of the electrical compensation of the transmission characteristic of the sound radiator referred to the site of listening-in. This statement may in fact hold its own if by appropriate preliminary measures no provision is made for a similarity of the directional characteristics of the sound radiator independently of the frequency above appr. 1 kHz. The only two conditions one of the loudspeakers of latest design has to meet are a high sound output in the complete audio frequency band, and, secondly, a homogenous sound space free of unevenness due to interference at the medium and high tones in the horizontal plane (Reference 5). According to another author (Reference 6) it is obvious that from subjective considerations unevennesses of the characteristic owing to interference are void of any significance, if the bandwidth is sufficiently narrow, so that this condition has not even to be stipulated. Since on the other hand this author has not postulated a sound space of a definite character in any other plane departing from the horizontal, he has guaranteed neither a frequency-independent constant directional characteristic at the site of listening-in irrespective of the room. At the same time in order to obtain a directional characteristic free of local unevennesses in the horizontal plane, i.e., to meet a by itself unjustified condition if nothing has been postulated in respect of other planes, the same author has placed the high-tome loudspeaker over a deep-tone sound radiator. By this means the longitudinal axis of the high-tone sound radiator unit coincides with the longitudinal axis of the deep-tone sound radiator and at the same time the high-tone soud radiators are turned in respect-of one another roud their longitudinal axis. As has been confirmed by the testing results published in the relevant literature (Reference 5) this method does not guarantee an even approximate uniformity of the directional characteristics in the horizontal plane above 1 kHz.
As taught by experience, in view of the circumstance that the deep-tone and high-tone loudspeakers at sites sufficiently distant from one another, for multi-channel sound radiators arranged as described above the crossover frequency has to be selected low enough in order to prevent wideband, and consequently audible, unevennesses from arising in a manner essentially independent of the direction in the transmission characteristic.
On the same consideration a crossovef frequency of 300 Hz has been selected for the above sound radiator, and accordingly at frequencies higher than 600 Hz for practical purposes only the high-tone sound radiator unit built up of loudspeakers arranged one under the other in the vertical plane will radiate. It is exactly for this reason that a critical handicap of the system, viz. that from about 1 kHz onwards the complete sound radiator does not produce a frequency-independent fairly constant directional characteristic, cannot be eliminated.
It has further been recognized that in order to produce a constant directional characteristic on an as wide as possible band, the crossover frequency has to be selected in a way that at the frequency where the sound radiation of the deep-tone and high-tone sound radiator units is overlapping, the deep-tone and high-tone sound radiator units have to present directional radiation properties already in the horizontal plane. Directional radiation is strongly fluenced by sound deflection arising on the surface of the house. As a matter of fact at low frequencies every closed loudspeaker of necessity possesses a spherical directional characteristic, and
a directional radiation will dependent on the dimensions of the loudspeaker and the house arise only about above 1 kHz. Therefore, if the usual house diemnsions are accepted, contrary to recommendations published in literature (Reference 2, 3) the crossover frequency should be selected by about two octaves higher in order to meet the postulates brought forward for the directional characteristic also in the neighborhood of the crossover frequency (Reference ll). Experience appears to conflict even with the statement made (Referfrequency-independent However, this sound radiator radiating a large sound ence 1) by a certain author that the crossover frequency has to be specified in a way that at this fre' quency the loudspeaker radiating the deep tone should hot yet present directional properties.
Furthermore it has been discovered that owing to the statistical properties of the programme signal there is somewhere between 630 and 1,250 Hz a freqency band of the width of a third of an octave, below and above which the programme signal represents an on the whole uniform signal output. Consequently for an optimum exploitation of the load carrying capacity of the deep and high sound channels in multi-channel systems a crossover frequency by at least an octave higher than the usual 300 Hz should preferably be chosen.
At translating the subjecbmatter of the invention into reality the cardinal idea was to design a sound radiation system built up of deep-tone and high-tone sound radiating elements which contrary to known methods would above 0.8 kHz guarantee frequencyindependent, reciprocally similar directional characteristics, and exploit the load carrying capacity of the deep-tone and high-tone radiating elements to the utmost possible. At the same time the sound radiator should in dependence of the frequency and, by varying the transmission characteristic of the amplifier supplying the sound radiator system, be convertible to one of a directional factor variable by a definite method, all this inorder to approximate at the site of listening-in in the room a for the given task most appropriate transmission characteristic as good as this could be done. The sound radiation system according to the invention provides a solution of the problem achieved by means of several measures applicable in'mutual independence of one another, yet in conjunction producing the best possible results.
US Pat. No. 3,648,801 shows the way to the potential realization of a radiator having a nearly constant, directional characteristic.
output even at deep tones does not by itself satisfy the requirements in every respect, as the load carrying capacity of the built-in loudspeakers before long set a limit to the load carrying capacity of the sound radiator. Conseqnetly this system cannot be operated as a sound radiator ofa wide sound band and of a large load carrying capacity unless considerable concessions have been made.
On the other hand a satisfactory solution may be achieved by having recourse to the two-way arrangement according to the description referred to above where the high-tone sound radiator incorporates four loudspeakers each of a diameter of 125 millimetres, whereas the deep-tone sound radiator consists of a loudspeaker of a diameter of 300 millimetre assembled in a completely sealed, cotton-damped house.
For the achievement of a sound radiation system of a constant directional characteristic the crossover frequency of the two-way sound radiation system has been chosen so as to permit the maximum possible exploitation of the load carrying capacity of the loudspeakers constituting the sound radiation system and at the same time possibly in a way independent of the frequency to guarantee a similarity of the directional characteristics of the sound radiation system at the crossover frequency as well as in the neighbourhood of this frequency within a wide spatial angle.
The performance of the system may even be improved, when the sound radiator according to the patent referred to above is used as high-tone radiation feature. By means of this sound radiator a system of relatively moderate dimensions, a wide sound band, high output and constant directional characteristic may be built up, provided that by ch'bosing an appropriate position for the deep-tone sound radiation element care is taken that this latter feature should be sufficiently close to the high-tone sound radiators and that an adequate high crossover frequency should be selected. It has been discovered that if the centre of the extreme loudspeaker of the high-tone radiation unit is at a distance of less than 1.5 times of the diameter of the loudspeaker of the deep-tone sound radiation unit and if at the same time the crossover frequency has been selected in a way that its associated wavelength should not exceed the lowest value of propagation measured on the surface of the house incorporating the deep-tone and high-tone sound radiation units parallel to the plane of the aperture of the deep-tone radiation loudspeaker, i.e., on the surface of the house of the sound radiation system, then as confirmed by experience a constant directional characteristics may be guaranteed even in the surroundings of the crossover frequency.
Satisfactory results may also be achieved if the crossover frequency has been chosen so that its associated wavelength is below the lowest value of propagation measured on the surface of the house incorporating the loudspeaker features parallel to the plane of the aperture of the deep-tone radiating loudspeaker. By aper' ture of radiation of the sound radiator the limiting superficial element, real of fictitious, of the sound radiating features is understood where a sound space generated through it is directly transmitted to a medium out side the sound radiator.
Although a sound radiation system so designed satisfies most of the conditions specified for practical needs as formulated earlier, yet further improvements may be achieved if recourse is had to the obstacles applied according to the patent referred to above. However, experience has taught that although the obstacles referred to above beneficially influence the development of the directional characteristic, yet further improvements may be carried through in the system.
Satisfactory results may also be achieved if instead of the known rigid obstacles a sheet or more with holes or apertures made in them are placed before the loudspeakers.
Details and further advantages of the sound radiation system of the invention will be described on hand of drawings and embodiments of the invention.
FIG. I presents an embodiment of the sound radiation system according to the invention, of lower volume.
FIG. 2 presents a sound radiation system being uniform with that of FIG. 1, still for a higher output.
FIG. 3 presents directional characteristics in the neighbourhood of a crossover frequency of f, 1.25 kHz for a design according to FIG. 1.
FIGS. 4 and 5 for the sake of comparison present directional characteristics with and without means of diffractions.
FIGS 6 and 7 present the applied means of diffraction, respectively viewed from above and perspectively.
FIG. 8 presents the design of a preferred means of diffraction.
FIGS 9 and 10 present examples of the modification of the transmission characteristics in the range of high tones in response to diffraction obstacles.
FIGS. 11, 12 and 13 present the modification of sections of a sound pressure versus frequency characteristic obtained indoor for different positions of the sound radiator as the result of the interaction of the sound radiator and the room.
FIG. 14 is a layout of the electroacoustic chain of a sound radiation system.
FIGS 15 and 16 are plots of the frequency characteristics of filters used in the system.
FIG. 17 presents the sound pressure versus frequency characteristics of the sound radiation system according to the invention plotted indoor.
FIGS. 18, 19 and 20 present the for practical purposes frequency-independent directional characteristics of the sound radiation system according to the invention.
Uniform components and parts have been distinguished by uniform reference code numbers.
In the layout according to FIG. 1 the sound radiation system 1 incorporated the electronic system 2, the deep-tone radiating loudspeaker 4 encased in the deeptone radiator house 3, further four high-tone radiating loudspeakers 6 accommodated in a high-tone radiator house 5. The loudspeakers 6 have been arranged horizontally in conformity with US. Pat. No. 3,648,801. For the measurements a reference system of coordinates has been used whose axes x, y and z are righthand twisted.
For the layout according to the invention it is essential that the distance of the critical section 8, i.e., the centre of the deep-sound radiating loudspeaker 4 from the centre of the extreme loudspeaker 6 of the hightone radiation system should not exceed 1.5 times the diameter of the loudspeaker 4 incorporated in the system.
FIG. 2 represents a design of the sound radiation system according to the invention of higher output where deep-tone radiating loudspeakers 4 accommodated in deep-tone sound radiating houses 3 and such radiating high-tones 5, ther high-tone radiating loudspeakers 6 according to the above patent arranged in two rows have been built in. The references correspond to the code numbers used in FIG. 1. In the graph in FIG. 3 the measuring results obtained for an embodiment of the layout according to FIG. 1 have been plotted. For the sound radiation system a crossover frequency of f, 1.2 5 kHz has been specified and in the graph the semiplanar directional characteristics developing in the vicinity of this crossover frequency have been plotted. The crossover filters have presented a characteristic of a slope of 12 dB/octave for a frequency sufficiently departing from the crossover frequency. The directional characteristics have been measured in the plane xy in an open sound space at a distance ofZ metres from the sound radiator at the crossover frequency and in its environment.
In the sound radiation system according to the layout shown in FIG. 1 the effects of various obstacles producing a diffraction have been subjected to an analysis. The results so obtained have been plotted in the graph in FIGS. 4 and 5. Here the continuous line represents the semiplanar directional characteristics associated with the different frequencies of the sound radiation system without the use of a diffraction producing obstacles, whereas the dotted line has been drawn for results obtained for the use of diffraction bars according to the patent referred to above. For the sake of comparison the generatrices represent the characteristics obtained with the use of a diffraction feature according to the invention. The arrangement of the obstacles is shown in FIG. 6. FIGS. 4 and 5 clearly demonstrate that, e.g., at 5 kHz and at 6.3 kHz the diffraction bar produces no appreciable effect. It should be noted that the loudspeakers 6 have been arranged on the sound wall 10.
A considerably better result may be obtained when before the two extreme loudspeakers 6 of the sound radiating feature a single, lamellar obstacle 11 with holes made in it is erected. A layout of this kind is shown in FIG. 7. A satisfactory result has been achieved when the free, perforated surface of the lamellar obstacle 11 is smaller than one half of the full surface.-
A particularly good result has been produced in the given case by a sheet according to FIG. 8, where rows of holes of small diameters l2 alternate with such of holes of large diameters 13. The improvement reached in the directional characteristics are shown in genera trices plotted in the graphs of FIGS. 4 and 5. Hence a method as detailed above guarantees a better frequency-independent directional characteristic by means of a suitably shaped perforated sheet as diffraction producing obstacle in the higher frequency range than any earlier method.
However, the effect of the diffraction-producing lamellar obstacle ll manifests itself not only in the directional characteristic, but also in the sound pressure versus frequency characteristic measured in the axis of the sound radiator in a manner that where in response to the diffraction the directional characteristic tends to improve appreciably, the axial sound pressure curve will present a downwards trend, provided that the hightone sound radiation element is built up of the usual individual loudspeakers of a nearly straight-lined transmission characteristic. This decaying characteristic of the loudspeaker system shown in in FIG. 9, where as taught by experience decay sets in at and about 3 kHz. causes no inconveniences, moreover in a room the absence of sharply directed high tones normally concomitant of other methods stands for a subjectively pleasant sensation (Reference 13). In addition the transmission characteristic conforms to'the relevant recommendations of sound film technics (Reference 14.). Still in order to guarantee the transmission characteristic normally obtained in radio and TV studios (Reference 15 however, in a manner that the directional characteristics of the sound radiation system should be similar for frequencies above 0.8 kHz, preferably care should be taken that in the given instance within a wide sound band the transmission characteristic of the sound radiation system measured at the position of thelistener should be a straight line. To this end the high-tone radiating feature should preferably be built up of individual loudspeakers whose axial transmission characteristic presents an expressly rising character at high tones. Loudspeakers of such a characteristic will by themselves create an expressly unpleasant acoustic effect, however, they will lend themselves readily for use in the sound radiation system arranged in the manner explained above.
The axial transmission characteristic of a sound radiation system built up of loudspeakers of a rising transmission characteristic, as shown in FIG. 1, 7 and 8, is
presented in FIG. 10 by a continuous curve. The dotted curve in FIG. 9 is the transmission characteristic of the high-tone radiating feature of a nearly straight-lined transmission characteristic. Obviously the change is considerable. According to experience the change occurring in response to the use of a diffraction sheet is for a given type of loudspeakers fairly independent of the scattering among the specimens of the loudspeak ers actually used, provided that the loudspeakers are well within the usual factory tolerances.
A solution producing yet further improvement has been achieved by an electrical correctional network executed in conformity with known designs, connected to the amplifiers or circuits of the sound radiation system, preferably before the output amplifiers and having transmission characteristics of a variable slope of i 6 dB/octave in the region above I kHz. In this manner at the frequencies above 0.8 kHz a frequencyindepende'nt sound radiator of a for practical purposes constant directional characteristic has been designed which in the room at the site of listening-in, dependent on the specification, presents a directional factor and transmission characteristic adjustable within extremes wide apart.
The sound radiation system designed on this understanding produces rather satisfactory results provided it is placed in the room sufficiently far away from the reflecting surfaces, e.g., walls. On the other hand if the sound radiation system is moved close to a reflecting surface then in response to the effects of the reflecting surface the sound pressure will tend to rise emphatically at the deep tones (Reference 7). Still design engineers are notwithstanding these experiences even today eager (References I, 2, 3, to expand and compensate the transmission characteristic of the sound radiator measured in an open sound space in the direction of the low frequencies within the possible widest band. As taught by experience a satisfactory solution may be achieved if the transmission characteristic of a sound radiator ope rated in the environments of reflecting surfaces measured in an open sound space presents at low frequencies, somewhere below 100 Hz, a decay of about Ie dB/octave. Asa matter of fact as confirmed by experience in this case the development of the sound pressure versus frequency characteristic measured under actual operating conditions, i.e., in a room, will be influenced to a lesser degree by reflection arising on the surrounding surfaces. Nevertheless again as taught by experience the effect of the surfaces limiting the sound radiator cannot be eliminated altogether. This is indicated also by FIG. 11 presenting the deepfrequency section of thesound pressure versus frequency characteristic of the earlier mentioned sound radiation system of a decay of 12 dB/octave at below 100 Hz in an open space, measured in a room of a volume of I metres cube, where there is a reverberation of about 0.5 msec, at a distance of 3 metres from the sound readiator. For the test the sound radiator was, with the position of the testing microphone unchanged, placed, first, on the ground (full line) and then at a height of 2 metres (broken line) close to the wall. Similar unevennesses are shown in FIG. 12, where the measuring results have been obtained with the rear edge of the sound radiator at a distance of 0.75 metre from the wall.
Although the unevenness in consequence of the characteristic decaying in an open space at below 100 Hz is far from critical, still further improvement could be achieved by connecting an electrical correctional network before the output amplifier supplying the sound radiation system. For the purpose of the test preferably a network should be used the slope of whose characteristic should be continuously adjustable between slopes of at least :6 dB/octave at frequencies below 300 Hz.
What has been set forth above clearly indicates that yet further improvements may be achieved in the system by correcting the transmission characteristic of the sound radiation system at the deep tones, preferably in a manner dependent on the site of installation of the sound radiator.
Experience has further taught that even in the following process the effects of the room cannot be ignored. These effects are particularly considerable in sound ra' diation systems used in the TV service. As a matter of fact here a sound radiator of a constant directional characteristic otherwise presenting a uniform transmission characteristic in open space in the frequency band between 200 and 2,000 Hz, presents in the room a transmission characteristic locally varying according to the frequency, i.e., an emphasis and the bandwidth associated with the unevenness are in like way functions of the site. This is indicated by the graphs in FIG. 13 where the characteristics have been plotted, by way of example, of a sound radiation system according to FIG. 1 and placed at a distance of L5 metre from the wall. The testing microphone has been installed at a distance of two metres, and the test has been carried out, first, with the sound radiator on the ground (full line), and then with it at a height of two metres from the ground (broken line). The unevenness of the former transmission characteristic has been compensated by a passive electrical network inserted between the hot spot of a known, characteristic regulating electrical circuit and its cold spot earth point, and built up of a resistance, a condenser and an inductance in series. The resonance frequency of the network and its Q-factor have preferably been made variable. In most of the practical instances a circuit of this type is needed only when the sound radiation system has to satisfy extremely critical requirements. In this case, e.g., for studio operation, a permanent accommodation of the sound radiation system may be taken for guaranteed. In such and similar cases in order to prevent incompetent persons from modifying the appropriately set transmission characteristics, the network has preferably been installed inside the amplifier in a way that the unit will become accessible only after the bolts fixing the amplifier have been removed or loosened.
A network locally correcting the transmission characteristic has been in use in sound radiation systems operated in studios for a long time already (Reference 2). However, it was not known that the network had preferably to be a variable one and that the optimum adjustment had to be made dependent on the site of accommodation of the sound radiation system. In addi tion the resonance frequency of this network had to be selected somewhere in the neighborhood of 300 Hz, in the environment of the crossover frequency of the deep-tone and high-tone sound radiator units.
It is a matter of experience that for a sound radiation system of a constant directional characteristic this correctional network should preferably be positioned at a frequency lower than the crossover frequency, in order to obtain the best possible result in the given environment. The network might as well be built up of a parallel oscillating circuit, or several of them, inserted in series in the regulating branch, and incorporating a resistance, condenser and inductance.
The diffraction causing element shown in FIG. 8 and accommodated on the front panel of the sound radiation system before the high-tone radiating feature has been designed so as to be at the same time part of the sheet covering the front surface of the sound radiation system. The sheet itself is covered on its outer surface with a sound transmitting texture or loudspeaker silk.
With the combined application of all these features an optimum result may be achieved. A block schematic of the circuitry used in the sound radiation system so designed is shown in FIG. 14. In the diagram a correctional network 16 is attached through the level control to the input unit 14. After the correctional network at branching point 17 the circuitry by means of appropriate crossover filters bifurcates to a deep-tone radiating channel branch 18 and to a high-tone radiating channel branch 19. In both branches the deep-tone radiating loudspeakers 4 and the high-tone radiating loudspeakers 6 join the crossover filters 20 via level controls 21 and audio frequency output amplifiers 22.
A characteristic of one of the correctional networks has by way of example been plotted in the graph in FIG. 15. The dotted area represents the range of regulation actually realized.
In the graph in FIG. 16 the transmission characteristics of the low-pass and high-pass filters used in the sound switches 20 have been plotted. The curve A plotted in the graph in FIG. 17 is the uniform sound pressure versus frequency characteristic measured at a distance of 2 metres from the radiator in a room of a volume of V 120 metres cube, of a permanent reverberation of T 0.5 s, when all measures discussed above have been taken simultaneously. The sound pressure versus frequency characteristic satisfies the conditions specified for the operation and monitoring of radio and TV studios. Curve 8", which differs from the former in that with the high-tone controlling potentiometer appropriately adjusted the preamplifier will present a straightlined transmission characteristic, satisfies the postulate of pleasant-listening, whereas curve C satisfies the transmission characteristic recommended in sound film technics. All transmission characteristics have been obtained in a way that the potentiometer for high-tone control has been turned from its highest position to its lowest until the specified characteristic has appeared. The uniform transmission characteristic appearing in the low-frequency band has been obtained in the mid-position of the deep-tone controlling potentiometer, when also the correction plotted in the graph in FIG. 13 required for the compensation of the characteristic of a sound radiation sys tem positioned on the ground and at a distance of 1.5 metre from the wall has been applied.
For a characteristic measured under operating circumstances conforming to the different requirements and postulates in both the horizontal and vertical planes the sound radiation system presents a for practical purposes frequency-independent directional characteristic in a conic angle of about i 60. This is confirmed by the curves plotted in the graphs in FIG. 18
(plane xy) and FIGS. 1-9 and 20 (plane xz). The directional characteristics have been measured in an open sound space, at a distance of 2 metres from the sound radiator and then plotted for semiplanes.
REFERENCES l. D. E. L. Shorter: A survey of Performance Criteria and Design Considerations for High-Quality Monitoring Loudspeakers. Journ. Audio Eng. Soc. Vol. 7
2. F. K. Schrdder, K.. Bertram: Studioellenc'ierzt'ie hangszork koevetelmer'iyei (Requirements of Studio Monitoring Loudspeakers) IInd Acoustic Conference, Budapest, 1961,22/1-5.
3. E. Kammerer: Studiolautsprecher Europhon: Siemens Zeitschrift 44 (1970) H.10, pp. 642647.
4. D. l-Iuszty, A. lllnyi, Mrs. H. Magos, T. Szele: Some Problems of the Studio Monitoring Loudspeakers, VIIth Int. Congr. Acoustics, Budapest, 1971. 23 E 9.
5. E. Wente: The Characteristics of Sound Transmission in Rooms. Journ. Acoust. Soc. Am. Vol. 7 (1935). pp. 123-126.
6. D. Huszty: The Response of the Electroacoustical Chain and the Subjective Sensation. VIIth Int. Congr. Acoustics, Budapest, 1971. 26 G 2.
7. T. S. Korn, J. Hougardy: Measures des hautparleurs dans les lieux dutilisation Acoustica 9 1959) PP 2ll26.
8. Techn. Comm. OIRT Recomm. 22 TK-XVII-l4 (I963) Moscow.
9. E. Sesztak: The sound Field in a Room at Different Excitation. VIIth Int. Congr. Acoustics, Budapest, 1971. 21 A ll.
10. L. Branek: Acoustic Measurements. John Wiley, New York. 1949, p 682.
l 1. Mrs. H. Magos: On the Crossover Frequency of Multiway Loudspeakers. VIIth Int. Congr. Acoustics, Budapest, I971. 20 E 9.
12. D. Huszty: A roevid indoere atlagolt m'tiesorjel statisztikai tulajdonsagai. Elt'iezetes kutatasi jelents (Statistical Properties of a Short-Term averaged Programme Signal. Preliminary Research Report. Elektroakusztikai Gyar, May, 1970.
13. W. Kuhl: Die Verbesserung von Presenz und Natuerlichkeit bei Lautsprechern mit raeumlichen Klangbild. Frequenz, Bd. 15 (1962) PP- 89-90.
14. Draft Proposal for Standardizing Acoustic Response of a Monitoring Chain in Motion-Picture Con trol Rooms. Journal of the S. M. P. T. E. Vol. 78 (1969) p 1045.
15. D. Orth: Physikalische Eigenschaften einer Studioabhoereinrichtung mit optimaler Wiedergabequalitaet. RFT Mitt. der Nachrichten und Messtechnik, Bd. 3(1965) pp 22-30.
What we claim is:
1. In a sound radiator comprising a low frequency and high frequency sound radiating unit housed in a common housing and an electronic circuit for said unit comprising an output amplifier and low-pass and highpass filters, said high frequency sound radiating unit comprising at least three loudspeakers excited by the same signal source and each having a flat front panel perpendicular to the radiating axis of the speaker, the panels being on a common level and meeting each other at angles substantially different from l80 along lines that are parallel to each other, said axes of adjacent speakers intersecting each other alternately before and behind saidspeakers; the improvement in which said high frequency sound radiating unit contains at least four loudspeakers, said low frequency sound radiating unit comprising a single loudspeaker whose center is at a distance of not more than 1.5 times the diameter of the low frequency loudspeaker from the center of the high frequency loudspeaker which is farthest from the low frequency loudspeaker, the wave length of the crossover frequency of the low frequency and high frequency sound radiating units being not more than the smallest dimension of said housing measured in a plane perpendicular to said axis of the low frequency loudspeaker, and a multiperforate barrier sheet disposed in front of at least a portion of said high frequency loudspeakers, the perforations through said sheet occupying less than half the area of said sheet.
2. In a sound radiator comprising a low frequency and high frequency sound radiating unit housed in a common housing and an electronic circuit for said unit comprising an output'amplifier and low-pass and highpass filters, said high frequency sound radiating unit comprising at least three loudspeakers excited by the same signal source and each having a flat front panel perpendicular to the radiating axis of the speaker, the
panels being on a common level and meeting each other at angles substantially different from along lines that are parallel to each other, said axes of adjacent speakers intersecting each other alternately before and behind said speakers; the improvement in which said high frequency sound radiating unit contains at least four loudspeakers, said low frequency sound radiating unit comprising a single loudspeaker whose center is at a distance of not more than 1.5 times the diameter of the low frequency loudspeaker from the center of the high frequency loudspeaker which is farthest from the low frequency loudspeaker, the wave length of the crossover frequency of the low frequency and high frequency sound radiating units being not more than the smallest dimension of said housing measured in a plane perpendicular to said axis of the low frequency loudspeaker, and a multiperforate barrier sheet disposed in front of at least a portion of said high frequency loudspeakers, said high frequency loudspeakers being disposed in a row, there being a said sheet disposed in front of only the end loudspeakers in said row.