CA1140862A

CA1140862A - High fidelity speaker system

Info

Publication number: CA1140862A
Application number: CA000348157A
Authority: CA
Inventors: Arnold Douven
Original assignee: Individual
Current assignee: Individual
Priority date: 1980-03-21
Filing date: 1980-03-21
Publication date: 1983-02-08

Abstract

ABSTRACT OF THE DISCLOSURE

The disclosure is a high fidelity sound-producing system that has two substantially similar full range drivers capable of accepting energy and producing sound with tolerable distortion over the entire frequency range and one low range driver of appreciably more sound radiating area than the two full range drivers. The low range drive and the full range drivers are in one enclosure and connected across the amplifier output.
The low range driver resonates at a similar frequency to the full range drivers and the efficiencies of said drivers are matched to provide substantially uniform sound level output over the frequency range of interest.

Description

This invention relates to an improved means for producing sounds (music, voice, etc.) by electro-mechanical means, commonly known as a speaker system to which the term "High Fidelity" or "Hi Fi" is used to describe the performance when the reproduced sound closely resembles the original.
The chain linking the final speaker system and the original sound comprises many links; e.g. microphones, amplifiers, recording means, etc., but modern technology has refined these components to the point where the original sound can be converted to electrical signals and conveyed to the final transducers with a high degree of accuracy. The final transducers are often the weakest link in this chain.
The conventional so-called "High Fidelity" speaker system usually comprises a plurality of individual drivers ~speakers) with the audio spectrum covered by drivers specifically designed to cover a particular range of frequencies. It is generally agreed that better over-all fidelity is obtained in this way since an individual driver can be designed to perform better if the range of audio frequencies handled is reduced. A typical system might comprise a mid-range driver capable of converting electrical to sound ; energy over the important middle frequencies of the audio spectrum.
Because the conventional mid-range driver becomes inefficient at the low and high frequency ends of the audio spectrum, it is ; usual to augment its response with a low-frequency driver or woofer and a high frequency driver or tweeter. It is common practice , also to use some form of frequency selection means in the electric-I al system so that low-frequency energy is applied to the woofer and high-frequency energy is applied to the tweeter only mid-~ ~;

- ' ' , ~ , "' ~ .

' , ~` ` li~U862 -range frequencies being applied to the mid-range driver. One ~methocl of accomplishing this selection is with the use of separate power amplifier-s for each type of driver, with the frequency range for each determined in a part of the system before the output stage~ The most-used system by far, however, employs passive filters between the output of the power amplifier and the speaker system. Filters to accomplish this frequency selection using passive compon~nts are called crossover networks.
Since an objective of this disclosure is to show a means whereby crossover networks are eliminated, the reason why such networks are employed should be discussed. A driver when fed with electrical energy outside the range of frequencies it is designed to handle will not produce a proper sound level but may effect ad-versely the operation of other drivers connected to the same power amplifier. A tweeter in parallel with a mid-range driver and/or woofer may impose a very low impedance across them at frequencies below those handled properly by the tweeter, thereby reducing the energy available at mid and low range frequencies, causing distor-tion in the power amplifier and even power amplifier failure.
20 - Low frequency energy present in the electrical drive to the mid-range driver and tweeter may produce little sound output at those frequencies but the presence of such frequencies may amp-litude modulate the sound output having frequencies lying in the range they are designed to produce.
One problem with crossover networks is that they inter-fere with the fidelity with which the original sound source is reproduced because of a factor known as phase distortion. If the waveform of an electrical signal produced by a single noté from a

-2-. , ~ ~ .. ; . .............. . . . ....... . ......... .
' ' musical instrument is examined on an oscilloscope, a shape will be seen characteristic of the particular instrument. Fourier analysis mathematically represe~ts such a waveform as a fundamental signal of pure sinusoidal form plus a number of harmonics of the funda-mental, each having an amplitude and phase related to the fundamen-tal. The harmonics could be present in precisely the same ampli-tude ratios but with different phase relationships and a complet~ly different looking waveform would be the result. Experts in sound evaluation have determined that the phase relationship of the harmonics determines the timbre of the note. Crossover networks, because they contain inductive and capacitive components, cause a change in the phase relationship between harmonics and fundamental, resulting in the sound radiated from its speaker system being different from that originating from the sound source.
Failure of a speaker system to reproduce the original sound precisely can occur when there is a sudden sound event.
The ability of a system to respond to a sudden change is known as transient response. It is possible for a system which has a reason-ably uniform response over the entire audible range to steady state signals to have a poor transient response with resulting failure to reproduce properly original sound of a transient nature. Cross-over networks by inserting inductive and capacitive elements between the power amplifier and the speaker system worsens the transient response of the system.
; This invention avoids the use of crossover network with their inherent poor transient response. A high fidelity sound producing system according to this invention comprises at least two substantially similar full range drivers capable of accepting . ~
, energy and producing sound with tolerable distortion over the entire . . .
' frequency range of interest; audio-electrical energy source terminals;
one low range driver of appreciably more so~und radiating area than the two full range drivers; the low range driver being connected across the energy source terminals; the full range drivers being connected across the energy source terminal; the said drivers all being mounted in an enclosure; said low range driver resonating at a similar frequency to sai~ full range drivers; the efficiencies of said drlvers being matched to provide substantially uniform sound level output over the frequency range of interst. The invention will be clearly understood after reference to the following detailed specification read in conjunction with the drawings.
- In the drawings:
Figure 1 shows diagramatically the connections for one form of the invention;
Figure 2 shows a modification to Figure l;
Figure 3 shows a different modification to Figure l;
Figure 4 shows a combination of the modifications of Figures 2 and 3; and Figure 5 shows an enclosure for mounting the drivers as disclosed in the invention.
Referring to Figure 1, amplifier 2 represents the power source for operating the speaker system. To permit amplifier 2 to perform adequately it must be designed using techniques well known in the art to produce an output of adequate peak power and neg-, ligible distortion. The input from whatever signal source is utilized is applied to the input of amplifier 2 across lines 1.
The amplifier output is applied to the sound transducers over out-1 put lines 3.

;,j 4 .. .. . . . . . . .
. ' ., Sound transducers 4 and 5 must be identical to the extent possible with good manufacturing techniques. In Figure 1 the speakers are connected in series in such a way that the radiation from the cones is in phase,e.g. the cones move in the same direction at the same time. While the type of speaker used is a matter of choice by the designer, provided the characteristics are adequate, we have found speakers manufactured by the Coral Audio Corporation known as type BETA-10 to be adequate in all respects. A list of the pertinent specifications of this device is as follows: Nominal Size (Diameter) 10 inches; Frequency Response 25 to 20,000 Hz.;
Resonance Frequency 25 to 40 Hz.; Music Power Input 20 watts;
Impedance 8 ohms. These driver units achieve extended high fre-quency response by the use of a dual cone and diffuser system which avoids the requirement for the additional tweeter mentioned in the preamble.
The bass transducer 6 is chosen to have suitable charac-teristics compared to drivers 4 and 5 and a cone diameter to move ade~uately the volume of air necessary for good bass response. It must be connected as indicated in Figure 1 so that the movement of the cone is in phase with that of the full-range drivers 4 and 5, When connected as in Figure 1, the electrical impedance of woofer 6 is chosen to be equivalent to that of full-range driver 4 or 5.
The efficiency, or ratio, between applied electrical energy and sound energy output must be reasonably equal to that of the full--range drivers if the simple arrangement of Figure 1 is used.
Woofer 6 is chosen to have a resonant frequency similar to drivers 4 and 5 to maintain the same phase for the radiated sound from all three speakers. We have found a suitable woofer to go with the _5_ ,. ~ '' . ~

..

BETA-10 transducers described above to b~ the 15L-100 woofer manu-factured by the Coral Audio Corporation with pertinent specifications as follows: Nominal Size (Diameter) 15 inches; Resonant Frequency 25 Hz.; Frequency Response Resonance to 1,500 Hz.; Music Power Input 150 watts; Impedance 8 ohms.
Using the arrangement of Figure 1 and presuming a flat (uniform) frequency response over the whole audio spectrum from the power amplifier 2, the efficiencies of the speakers must be matched , in such a way as to provide a uniform sound output. For example, if all speakers had the same efficiency and the same impedance the two full-range speakers 4 and 5 together accept one-third of the total power utilized while the woofer takes two thirds. If the efficiency of all speakers was the same, the sound level of the bass frequencies handled by the woofer would be greater compared to the remainder of the spectrum. If, on the other hand, the sound power generated by the full-range speakers was 3 db greater for the same applied power as for the woofer, the speakers would be matched and the sound level accurately reproduced over the entire range. Figure 1 could be varied in arrangement by connecting the drivers4 and 5 in parallel so that all three drivers would be in parallel across the power source.
It may not always be possible to choose speakers with ;i matched efficiencies. In general, full-range drivers are consider-, ably more efficient than woofers. In such circumstance some compen-~ sation may be required as illustrated in Figure 2, Resistor 7 is ^~ added in series with full-range drivers 4 and 5 to cause a greater i~ proportion of the total energy used by the speakers to be dissipated '' 1 by the woofer 6.

. I ~

~ 3l362 In the arrangements indicated in Figure 1 and Figure 2, the impedance of the woofer is assumed to be~relatively constant over its frequency range and increase with frequency beyond the point where the speaker can produce appreciable sound. This s~tua-tion is, in general, true because the large voice coil used in good woofers has the required value of inductance. Within the frequency range of the device, the back EMF induced in the voice coil due to its movement in the magnetic field produced by the speaker magnet provides the major component of the electrical impedance. At fre-quencies above the range of the woofer, movement of the coil isrestricted due to the relatively large mass of the voice coil and cone assembly which would cause the impedance to fall were it not for the inductance of the voice coil itself. This inductive reactance increases with frequency and normally would be expected to achieve a value such that no dip in impedance occurs at the high end of the woofer frequency range. If, however, a woofer is used which has too low a voice coil inductance to prevent such a dip in impedance, the arrangement indicated in Figure 3 can be used.
In this case, inductor 8 is connected in series with the woofer 6.
The value of the inductance is such as to have negligible effect at bass frequencies but, due to the linear increase of impedance with frequency, attains sufficient impedance at the lower mid-range frequencies to prevent the woofer circuit from absorbing the output of amplifier-2 which should go to the full-range drivers 4 and 5.
~ Figure4 ~lustrates the situation where a resistor 7 is ¦ connected in series with the full-range speakers 4 and 5 to compen-I sate for efficiency difference between full-range drivers and woofer ! - 7 ~' 11~0862 and inductor 8 is inserted in series with woofer 6 to prevent power loss through the woofer outside its frequency range.
While not forming any par~t of th~ claims for this inven-tion, some mention must be made of the enclosure in which the drivers are mounted. There are a number of good enclosure designs well known in the art which will operate adequately with the driver arrangement specified in this disclosure. One such design which has been used successfully is known as "bass relex", illustrated in Figure 5. The enclosure 10 mounts the drivers 4, 5 and 6. An additional port 9 permits radiation af sound energy prod-uced from the backs of the cones of the drivers to issue from port 9 in the forward direction. Another suitable form of enclosure is a sealed enclosure of adequate volume and containing sound absorb-ing material to eliminate resonances. In general, the sealed enclosure requires a greater volume for the same low-frequency response than does the bass reflex design.
The invention provides correct spatiRl perspectiye and correct proportional dynamic range potential regardless of frequency in the original signal source. No signal regardless of frequency is reproduced out of dynamic proportion to any others. ~he full SoAic impa~t poten ial aS contained in the ~ig~al source is produced.

. .

;', . ~
, ' ,

Claims

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. In a high fidelity sound-producing system i) at least two substantially similar full range drivers capable of accepting energy and producing sound with tolerable dis-tortion over the entire frequency range of interest;
ii) audio-electrical energy source terminals;
iii) one low range driver of appreciably more sound radiat-ing area than the two full range drivers;
iv) the low range driver being connected across the energy source terminals;
v) the full range drivers being connected across the energy source terminals;
vi) the said drivers all being mounted in an enclosure;
vii) said low range driver resonating at a similar frequency to said full range drivers;
viii) the efficiencies of said drivers being matched to provide substantially uniform sound level output over the frequen-cy range of interest.

2. In a high fidelity sound producing system as claimed in Claim 1 wherein said full-range drivers are series connected across the energy source terminals.

3. In a high fidelity sound producing system as claimed in Claim 1 having an inductor in series with the low-range driver to increase the impedance of the low-range driver at frequencies above the low-range driver range.