U.S. patent number 3,818,138 [Application Number 05/166,057] was granted by the patent office on 1974-06-18 for barrel shaped speaker enclosure.
Invention is credited to Augustine J. Sperrazza, Jr..
United States Patent |
3,818,138 |
Sperrazza, Jr. |
June 18, 1974 |
BARREL SHAPED SPEAKER ENCLOSURE
Abstract
Speaker enclosure having a general barrel-shaped profile
containing a plurality of loudspeakers. Two embodiments are shown,
one of which utilizes a plurality of full range loudspeakers, the
second of which utilizes a plurality of high frequency loudspeakers
in combination with a low frequency loudspeaker. The curved shape
of the sidewall of the speaker enclosure provides substantial
improvement in sound reproduction and, combined with the plurality
of speakers, produces an omnidirectional sound source that projects
breadth as well as depth to achieve true realism in phonetic
reproduction. The speakers are so positioned in the barrel shaped
enclosure that the sound emanating from the back of an individual
speaker is reflected downwardly by the internal surface of the
opposite sidewall to a point other than the back of the same
speaker.
Inventors: |
Sperrazza, Jr.; Augustine J.
(Troy, NY) |
Family
ID: |
22601634 |
Appl.
No.: |
05/166,057 |
Filed: |
July 26, 1971 |
Current U.S.
Class: |
381/336; 312/7.1;
381/89; 381/98; 381/99; 381/103; 381/351 |
Current CPC
Class: |
H04R
1/323 (20130101) |
Current International
Class: |
H04R
1/32 (20060101); H04r 001/28 () |
Field of
Search: |
;179/1E,16A,1G,1.12A
;181/31B,31R ;325/352 |
References Cited
[Referenced By]
U.S. Patent Documents
Foreign Patent Documents
Other References
Olson, Direct Radiator Loudspeaker Enclosures, Audio Engineering,
11/1951, pp. 34, 36, 38 and 59-64..
|
Primary Examiner: Claffy; Kathleen H.
Assistant Examiner: Leaheey; Jon Bradford
Attorney, Agent or Firm: Dorsey, Marquart, Windhorst, West
& Halladay
Claims
I claim as my invention:
1. A speaker enclosure comprising:
a planar top;
a planar bottom below said top;
a geometrically barrel-shaped sidewall connected between said top
and said bottom having a generally circular horizontal cross
section with the greatest diameter at approximately the midpoint of
the height of said enclosure and curving inwardly both upwardly and
downwardly so that the smallest diameter cross section occurs where
said sidewall joins said top and said bottom of said speaker
enclosure and having an aperture formed therein disposed between
the midpoint and the top of said speaker enclosure; and;
a speaker element having a longitudinal axis fixedly secured to the
interior of said sidewall and directed to penetrate through the
aperture formed therein so that the internal surface of the
sidewall opposite the speaker which intercepts the longitudinal
axis of the speaker is neither parallel to nor tangentially
parallel to a plane perpendicular to the longitudinal axis of the
speaker whereby sound emanating from the back of the speaker is
reflected downwardly to a point other than the back of the
speaker.
2. The speaker enclosure of claim 1 wherein the radius of curvature
of each vertical portion of the sidewall extends beyond the area
bounded by the vertical cross section.
3. The speaker enclosure of claim 1 wherein the sidewall has a
plurality of apertures formed circumferentially therein and wherein
a plurality of speaker elements having longitudinal axes are
fixedly secured to the interior of the sidewall and directed to
penetrate through the apertures formed therein so that the internal
surface of the sidewall opposite each speaker which intercepts the
longitudinal axis of each speaker is neither parallel to nor
tangentially parallel to a plane perpendicular to the longitudinal
axis of that speaker whereby sound emanating from the back of each
speaker is reflected downwardly to points other than the back of
the speaker.
4. The speaker enclosure of claim 3 wherein said speaker elements
are multi-range speaker elements.
5. The speaker enclosure of claim 3 wherein said plurality of
speaker elements are equidistantly spaced about the circumference
of said sidewall.
6. The speaker enclosure of claim 3 wherein said plurality of
speaker elements are all disposed circumferentially about said
sidewall on the same horizontal plane.
7. The speaker enclosure of claim 3 further comprising a horizontal
diaphragm dividing said speaker enclosure into two air-tight
chambers and wherein said plurality of speaker elements are high
frequency speaker elements and are located in the upper chamber and
having at least one low frequency speaker disposed in said lower
chamber.
8. The speaker enclosure of claim 7 wherein said low frequency
speaker is directed downwardly and having means for dispersing said
low frequency sound including apertures formed in the sidewall
enclosing the lower chamber of said speaker enclosure; and a
geometrically shaped member attached to said bottom having a
highest point in alignment with the center of said low frequency
speaker and sloping downwardly so that its lowest edge is at the
circumferential edge of said geometric shaped member.
9. The speaker enclosure of claim 7 further comprising an
electrical cross-over network connected with said plurality of high
frequency speaker elements and said low frequency speaker element.
Description
BACKGROUND OF THE INVENTION
Speaker enclosures of the past have been generally constructed with
rectangular planer sidewalls and have had a rectangular or
hexagonal horizontal cross section. As the loudspeaker system
contained therein moves to produce the sound, these structures are
characterized by vibration and standing waves produced by the
perpendicular planer walls as well as diffracted waves caused by
sharp boundaries of the sidewalls of the enclosure which lessens
the quality of the sound response. More recently, in an attempt to
improve the sound response, speaker enclosures have assumed a
generally cylindrical shape. However, as distinguished from the
present invention, the generally cylindrical shape improves sound
response in only one dimension. Since the cylindrical sidewall is
still parallel to the loudspeaker face, standing waves continue to
be produced and vibration continues to be caused by the generally
perpendicular sidewall rescting to the movement of the
speakers.
SUMMARY OF THE INVENTION
The present invention is concerned with a specific construction of
a speaker enclosure having a generally circular cross section and a
curvilinear sidewall. Such construction substantially increases the
strength of a speaker enclosure and thereby reduces vibration.
Since no planar sides are present or any structure which is
perpendicular to the emitting source of the soundwaves, standing
waves are also reduced. In addition, since the curvilinear design
produces a geometrically smooth over-all shape, frequency response
is improved due to the reduction in amplitude of diffracted
waves.
By using a generally barrel-shape, rather than a spherical form,
the speaker enclosure can be balanced and the speakers may be
positioned at an optimum level and slightly tilted to effectively
disburse the sound being reproduced. If the plurality of speakers
are utilized, this allows true omnidirectional sound
dispersion.
In addition, utilizing a plurality of speakers improves cone
excursion, lowers the resonant frequency, and improves power
capability and transient response.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a top view of my invention showing the equidistantly
spaced speakers located in the upper half of the speaker
enclosure;
FIG. 2 is a sectional front view of my invention taken along the
line 2--2 of FIG. 1 showing the generally barrel-shaped profile and
the internal positioning of the speaker elements;
FIG. 3 is a front view of a second embodiment of my invention
showing similarly spaced high frequency speakers, and also showing
apertures and a geometrically shaped member for dispersion of low
frequency sound waves;
FIG. 4 is a sectional front view showing the chamber separation of
high frequency and low frequency speakers as well as positioning of
the speakers for optimum sound dispersion; and
FIG. 5 is a cross-over network which may be utilized with the
second embodiment of my invention to effectively control the
frequency response emitting from the speaker system.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
A general understanding of my invention can be gained from FIGS. 1
and 2. The speaker enclosure 11 has a geometrically shaped sidewall
12 constructed of wood or any other suitable material. If
constructed of wood, it may be shaped and glued together as
commonly done in constructing a wooden barrel. Whatever material is
utilized in the construction of the sidewall 12, it is preferable
that each piece extends the full length from the top to the bottom
of the speaker enclosure 11. As shown in the figures, the material
should be shaped to produce a curvilinear, smooth sidewall having a
general circular horizontal cross section with the greatest
diameter approximately at the midpoint of the heighth of the
enclosure and curving inwardly both upwardly and downwardly so that
the smallest diameter cross section occurs at the top and the
bottom of the speaker enclosure.
Conventional loudspeaker enclosures having planar sides will tend
to vibrate. The vibrations are caused by a change in air pressure
within the enclosure as the loudspeakers move. As the loudspeakers
move inward, the enclosure moves outward, and as the loudspeaker
moves outward, the enclosure moves inward. The enclosure vibration
is, therefore, out of phase with the loudspeaker cone and thus,
less speaker output can be delivered to a listener due to the power
expended to vibrate the enclosure. In addition, the enclosure walls
do not follow the exact movement of the cones, but instead create
their own sound which is not present in the original material and
represents distortion.
It has been found that the curvilinear sidewall 12 design
represented in FIGS. 1 thru 4 has a much higher resistance to
vibration. This is because when pressure or vacuum is applied, a
portion of the forces are exerted along the surface of the
material, and not perpendicular to the thickness, and therefore, a
portion of the force exerted creates tension or compression in the
sidewall adding strength and thereby reducing vibration.
The curvilinear sidewall 12 also reduces standing waves which are
normally produced by the reflection of a plane wave on any surface
which is normal to the direction of propagation of the sound
source. Standing waves occur when a reflected wave meets the
incident wave in phase and at an equal sound level. As is known in
the art, the result is additive, causing a node or peak at the
frequency of the incident wave. Since the shape of the present
speaker enclosure 11 has no normal angles and no parallel sides,
standing waves are substantially reduced. In addition, as is
conventionally done, standing waves can be further reduced by the
use of sound absorbing material (not shown), such as fiberglass
insulation, which also provides damping for the loudspeakers.
Attached to the sidewall 12 at the top and bottom thereof should be
a top 14 and a bottom 15, which can be constructed out of any
suitable material and shaped in a circular cross section to fit the
ends of the sidewall 11. As shown in FIG. 2, both the top and
bottom may be secured to annular rings 17, 18 fixedly secured to
the sidewall 12. The top 14 and bottom 15, being parallel, can
produce standing waves. However, due to the relatively small
surface area of the top 14 and bottom 15 relative to the over-all
surface area of the speaker enclosure 11, the results are
negligible.
Finally, conventional loudspeaker responses may vary as much as
10db due to the configuration of the speaker enclosure 11. The
detrimental effects of the sound diffraction is substantially
reduced by eliminating the sharp boundaries of the enclosure 11
thereby reducing diffracted waves in amplitude. In addition, since
all edges vary radially from the speaker diaphragm, this produces
random phase relationships between the primary source and the
diffracted sound waves, also reducing the effect thereof.
Since higher audible frequencies travel in a straight line and do
not have the capability to bend around obstacles in their path, it
is important to raise high frequency drivers to ear level and above
the usual accumulation of furniture found in a home. Therefore, a
plurality of speakers 20-25 should be mounted by any suitable
method in the upper half of the speaker enclosure 12 in apertures
27-32 formed therein. It is preferable that the edges of the
apertures 27-32 be perpendicular to the tangent of the sidewall 11
and therefore, due to the curvature of the sidewall 12, the
speakers 20-25 will be tilted slightly upward.
Preferably, loudspeakers 20-26 should be spaced at equidistant
intervals along the circumference of the sidewall 12. This
increases stability of the direction of the emanating sound. Since
direct sound is received a fraction of a second earlier than
indirect sound, the omnidirectional speakers maintain directional
clues while providing the listener with a broad sound source.
Further, utilizing the plurality of speakers 20-26 instead of a
single speaker improves speaker capability and produces a superior
transient response. Transient response of a speaker is inversely
related to the moving mass of the speaker. To improve transient
response, the moving mass of the speaker is reduced by decreasing
the physical size of the cone. However, reduction in cone size
decreases power output capability in low frequency response.
Therefore, by utilizing a plurality of smaller speaker elements,
power output capability is maintained and transient response is
improved.
In addition, the resonant frequency of the speakers is lowered due
to mutual coupling by connecting the loudspeakers in phase and in
relatively close physical relationship. This increases the reactive
and resistive component of the air loading at lower frequencies
even though the radiation resistance of the loudspeaker, which is
the air loading to which the loudspeaker can deliver acoustic
power, is increased. The increased effect of the reactive and
resistive components of air loading provides a substantial
improvement in the low frequency response. Furthermore, cone
excursion is reduced because for the same input power, the multiple
loudspeakers will move less than one loudspeaker and due to the
reduction of cone excursion, Doppler distortion is decreased or
reduced to an inaudible level.
Since the major benefit obtained from this invention is the general
shape of the speaker enclosure 11, and more particularly, the
sidewall 12, two embodiments of the invention are shown, one in
FIGS. 1 and 2 which utilizes multi-range speakers 20-25
circumferentially spaced about the upper half of the speaker
enclosure 11; and FIGS. 3 and 4, which utilizes high frequency
speakers 35-40, conventionally referred to as tweeters,
circumferentially spaced about the upper half of the speaker
enclosure 11, and a low frequency speaker 42, conventionally
referred to as a woofer, positioned in the lower half of the
speaker enclosure.
By utilizing the full range speakers 25-30 as shown in FIGS. 1 and
2, the need for cross-over networks to differentiate between high
and low frequency signals are eliminated. When utilizing cross-over
networks, problems may be present, primarily at the cross-over
frequency. For example, when operating at the cross-over frequency,
power capability is usually reduced to approximately one-half of
the total available power. In addition, transient distortion is
usually present at or near the cross-over frequency due to the
sharp attenuation rate in the network. This is caused by the phase
shift created by the components of the electrical cross-over
network through which the signal passes. As is known, the phase
shift becomes greater as the number of components is increased. In
addition, some cross-overs utilize resonant networks which may be
shock-excited at certain frequencies causing momentary oscillation
or ringing after the original signal has been removed. Also, the
increased number of components increases the time and expense in
design for components, construction, and test, and increases the
possibility of error and failure.
The second embodiment is shown in FIGS. 3, 4, and 5 wherein high
frequency speakers 35-40 are utilized along with a low frequency
speaker 42. The utilization of this combination of special purpose
loudspeakers can create a greater degree of efficiency and sound
reproduction by reducing intermodulation distortion and providing
more reserve power handling capabilities. In addition, it permits
adjustment of sound levels, thereby creating a more balanced sound
output.
Intermodulation distortion occurs when a single vibrating
loudspeaker attempts to produce a high frequency note while
simultaneously producing a low frequency note. Any irregularity of
the loudspeaker while attempting to produce a low frequency sound
will influence the tonal quality of the high frequency sound. The
isolation of the high frequency sounds to their own loudspeakers
renders them mechanically unrestricted by low frequency sounds,
thereby eliminating intermodulation distortion. To further
effectuate this as far as air loading, a rigid diaphragm 46, which
may also be mounted on an annular ring 47, should be inserted in
the speaker enclosure 11 dividing the speaker enclosure 11 into two
cahmbers, 48 and 49.
Since low frequency sounds have a longer wave length, they do not
have the propensity to form the narrow beam which characterizes
high frequency sounds. Therefore, the low frequency speaker 42 may
be positioned in the enclosure 11 directed downwardly and may be
mounted in any suitable manner, such as on the annular ring 50
fixedly secured to the sidewall 12 as shown in FIG. 4. To assure
360.degree. radial sound dispersion in an horizontal plane, a
geometric member 51 should be utilized at the bottom of the speaker
enclosure 11 and apertures 54-57 of any suitable shape should be
formed around the circumference of the geometric member, to allow
sound emmission to the surrounding environment. The geometric
member can be of any suitable and preferred shape with its high
point at the center of the low frequency speaker and its low point
at the outer circumference of the speaker. Suitable members could
be the cone, as shown in FIG. 4, an eliptical horn, etc.
A conventional cross-over network is shown in FIG. 5, which may be
utilized with the present invention. Speakers 35-40, collectively
shown as a single speaker 55, are connected in a series parallel
arrangement to give an over-all nominal impedance of a single
speaker utilized alone. The speakers 35-40 are serially connected
to the input terminals 62 through a variable potentiometer 64 and
capacitor 66. The capacitor 66 passes high frequency signals to the
high frequency speakers and prevents low frequency signals. Low
frequency signals are fed to the low frequency speaker from the
input 62 through inductor 68 which passes low frequency signals to
the speaker and prevents high frequency signals. Representative
values of capacitance and inductance resistance are 4.2
microfarads, 0.28 millihenries, and 60 ohms, respectively. This
will provide an input impedance of 8 ohms, a cross-over frequency
of 350 Hz and an attenuation rate of 6db per octave.
Variable potentiometer 64 is a brilliance control which can be
adjusted to compensate for irregularities in response attributable
to the environment in which the speaker enclosure is utilized. In
rooms containing rugs, drapes and soft chairs, etc., sound
emmission will generally be dull and muted, which requires more
brilliance. In a listening room containing hard floors, walls,
ceilings, etc., it will generally be necessary to reduce the
brilliance. In each of these situations, the sound responses from
the woofer or low frequency speaker should be utilized as a
reference sound level, and the brilliance control should be
adjusted to vary the volume of the tweeters or high frequency
section of the speaker system.
In general, while I have described two specific embodiments of my
invention, it is to be understood that this is for the purposes of
illustration only, and many variants can be made within the scope
of my invention. For example, the top, bottom and the rigid
diaphragm can be of curvilinear design to further reduce parallel
sides and consequent effect of standing waves. It should also be
obvious that a more highly specialized speaker network could be
used including high frequency, low frequency and mid-range
frequency loudspeakers.
* * * * *