GB2577569A - Loudspeaker enclosure with slot/horn apparatus for improved polar response and low frequency output - Google Patents

Abstract

A loudspeaker enclosure has at least one forward facing speaker. Some of the acoustic energy generated from the rear of the speaker cone passes through a slot d between two baffles or panels b and c acting as a rear facing horn. The slot varies in width, Fig 2 and Fig 5 tapering in width narrow at the centre and widening toward the top and bottom. The shape and size of the slot is calculated to set up Helmholtz resonance within the cabinet creating an air spring. The cabinet may also have additional circuitry which may be a passive crossover to separate bandwidths. In an alternative embodiment the cabinet mat have more than one loudspeaker and may have side and rear facing horns or ports, Fig 6, d. The cabinet is intended for use with an electric guitar and to enhance bass response.

Description

BACKGROUND OF THE INVENTION:

Field of the Invention:

This invention relates to the field of loudspeaker systems and in these examples is focused on the amplification of electric guitar, however it could be used in any environment where the more omnidirectional or figure-8 polar pattern of a dipole loudspeaker is advantageous, whilst avoiding either the loss of low frequency output of an open baffle design or the extra cost and weight of a back to back dual driver dipole.

Historical and practical notes: In the 1950s a loudspeaker system for electric guitar differed little from a loudspeaker for home audio (hi-fi) or public address (PA); over the subsequent decades hi-fi and PA speakers have progressed significantly, with PA systems in particular exhibiting hugely improved sonic clarity, increased output and improved dispersion. The colouration (or distortion) of a guitar speaker is a significant contributor to the tone of electric guitar, therefore the developments which have applied to PA systems (e.g. splitting the frequency spectrum between drivers of different size, using horns or waveguide to control midrange and treble dispersion, etc) could not be used with guitar loudspeakers without detriment to the guitar tone. For the purposes of this document, low (i.e. bass) frequencies can be considered to be those below 200Hz, midrange frequencies between 200Hz and 1600Hz and high (i.e. treble) frequencies those above 1600Hz.

Guitar loudspeaker systems can be broadly split into three categories, of which the most common are the first two: 1. Open backed cabs -these typically use a cuboid shaped enclosure, with one or more loudspeaker drivers mounted in the front of the enclosure and either no rear panel or a large opening in the rear of the enclosure. Thus sound radiates from both the front and rear of the cones. The rearwards radiation is uncontrolled and generates some standing waves between the internal panels, especially those that are parallel, thus causing the frequency response of the rearwards radiation to differ substantially from the forwards radiation. At low frequencies the rear radiation is omnidirectional but 180 degrees out of phase with the omnidirectional low frequency front radiation, thus partially cancelling out these low frequencies. The resultant decrease in the low frequency output causes a thin tone.

2. Closed backed cabs -these typically use a cuboid shaped enclosure, with one or more drivers mounted in the front of the enclosure which fully surrounds the rear of the driver, thus sound only radiates from the front of the cones. Any rearwards radiation is contained, except for some midrange reflections escaping through the cones or low frequencies being re-radiated by enclosure panel resonances. Preventing the 180 deg out of phase rearward low frequency radiation from cancelling the forwards low frequency radiation results in greater low frequency output giving a much fatter or thicker tone.

3. Ported cabs -these typically use a cuboid shaped enclosure, with one or more drivers mounted in the front of the enclosure which fully surrounds the rear of the drivers, and one or more slots or tubes which act as Helmholtz resonators, improving efficiency, power handling and output at low frequencies. Although ported cabs are very common for bass guitar, hi-fi and PA they are relatively rare for guitar.

Due to the simple design of a guitar cab, the forwards radiation has inherently poor dispersion, with strong directionality at midrange and treble frequencies. This makes it hard to hear clearly unless the listeners' ears are on-axis, which is impossible for the majority of the band or audience. However, open-backed cabs are easier to hear because the rearwards output reflects off the floor and walls and thus reaches the listeners' ears indirectly.

A key note regarding guitar loudspeaker drivers is that they are not designed for accurate or hi-fi reproduction of sound and thus not only do they exhibit uneven frequency response and large amounts of cone break-up and associated distortion, the voice coil / magnet motor system tends to have little or no voice coil overhang. This lack of overhang means that the motor exhibits additional distortion at relatively low power levels, and this too is a key contributor to guitar tone, with players often preferring to drive their speaker until it reaches particular amount of cone excursion and thus a corresponding amount of motor distortion.

Horns have been used to improve the output of sound production devices since before the moving coil loudspeaker was invented. These were designed empirically for many years but over the past few decades the physics of horns has well been established and software packages written which allow the accurate modelling of horn loudspeakers. As will be shown by this document, although the invention may appear superficially to be a rear-loaded horn loudspeaker with a slot throat, it does not behave according to accepted horn theory and should not be considered a horn but a unique invention.

SUMMARY OF THE INVENTION

Embodiments of the present invention comprise a loudspeaker enclosure accommodating at least one speaker. Sound waves emanating from the rear of the speaker exit through a slot and then pass through a horn before fully exiting the enclosure and dispersing around the room. This slot and horn act like a conventional loudspeaker bass reflex port at low frequencies, improving efficiency, power handling and output in the lowest register. At midrange and treble frequencies this slot and horn have similar but superior benefits to an open backed loudspeaker cabinet, with the combined apparatus of slot and horn increasing efficiency and dispersion at midrange and treble frequencies.

This invention works by passing the rear radiation through a narrow vertical slot causing the sound to be be diffracted laterally across a wide bandwidth, thus improving the consistency of reflected sound reaching the listener. If this slot is too wide it fails to provide significant diffraction whilst if this slow is too narrow it has a significant low pass filter effect, blocking the rearward radiation of higher frequencies. The slot does not have to be uniform width but can be varied to suit the physical dimensions of the magnet structure, the rear radiation pattern of the driver and the desired frequency response.

Coupling the diffraction slot with a horn both increases the rearward output through horn loading and further improves the dispersion. If the horn is too narrow in angle it excessively accentuates a certain midrange bandwidth whilst if it is too wide in angle it fails to provide significant output and dispersion benefits.

The internal faces of the horn panels also act as part of the internal structure of the enclosure, thus changing it from being essentially cuboidal to a shape made up of trapezoidal and/or pentagonal prisms. This reduces the colouration caused by standing waves, as these now occur across a wide range of frequencies rather than the two or three frequencies and their overtones whose wavelengths correspond with the internal width, depth and height.

EXAMPLE OF THE INVENTION

This example is a loudspeaker cabinet designed for electric guitar, containing a single loudspeaker (a) of 10" nominal diameter mounted on the front panel (b), as shown in Fig. 1, with the side panel (d) and its opposite panel and the top panel (c) and its opposite panel forming the solid sides. The enclosure is cuboidal but the rear panel has a large cut-out as shown in Fig. 2, which illustrates the remaining rear panel (a), the slanted horn panels (b) and (c), the loudspeaker frame (d) and the loudspeaker cone (e) and loudspeaker magnet (f) Within the enclosure are mounted two panels, shaped as shown in Fig. 3, both identical, where (a) is the top edge of the horn panel which is fixed to the underside of the enclosure's top panel, (b) is the bottom edge of the horn panel which is fixed to the topside of the enclosure's base panel. (c) is the rear edge of the panel which is fixed to the inside of the enclosure's rear panel to the outside of the cut-out area. (d) is the curved front edge of the panel -these are mounted adjacent to each other, with their spacing determining the slot width.

These panels are mounted within the enclosure, perpendicular to the top and base panels, and at a 48 degree angle to the rear panel, as shown in Fig 4. (a) is the 10" loudspeaker. (b) and (c) are the two panels the form the horn and also the slot (d). (e) and (f) are the remaining parts of the rear panel that ensure the loudspeaker is fully enclosed so a Helmholtz resonance can be generated in the slot/horn apparatus.

The curved edges of the panels are positioned 30mm apart from each other at the top and bottom, and 20mm apart at the centre, as shown in Fig. 5. (a) is the narrow central part of the slot, 20mm wide. (b) and (c) are the two horn panels. (d) is the wider top part of the slot, 30mm wide -the bottom of the slot is the same width as the top because the slot has both vertical and horizontal symmetry about the central point..

In this way, we have formed a slot and horn apparatus, which acts as reflex port or Helmholtz resonator at low frequencies, and as a diffraction slot and horn/waveguide at midrange and treble frequencies. The Helmholtz resonance of this example is at 100Hz. This example also incorporates a switchable passive highpass filter to reduce low frequency output for guitarists who prefer the thinner tone from traditional open-backed cabinets which also gives a second motor distortion "sweet spot", similar to that achieved with the more complex design in Fig. 6.

An alternate example of the invention uses two separate sub-enclosures, each with its own loudspeaker, within the one loudspeaker cabinet, as shown in Fig 6. One sub-enclosure uses the slot and horn apparatus in a non-sealed enclosure, foregoing the Helmholtz resonance assistance at low frequencies, with the other sub-enclosure containing a high-excursion woofer and bass reflex port, using a switchable passive crossover to distribute frequencies as appropriate. (a) is the diffraction slot. (b) and (c) are the horn panels. (d) is the bass reflex port in the woofer (0 sub-enclosure. (e) is the guitar loudspeaker. (g) is the open rear 'panel' of the guitar loudspeaker sub-enclosure. (h) is the fully enclosed back of the woofer sub-enclosure. This design gives the user much greater tonal control as the "break-up point" loudness can be varied by removing low frequency demands from the guitar loudspeaker and passing them to the high-excursion woofer. In this case, "break-up" is a term used by guitarists to describe the motor distortion caused predominantly by loudspeaker non-linearity with increasing voice coil excursion, and as the main cause of excursion is low frequency signal content a switchable passive high pass filter can change the amount of excursion at a given SPL. This is not the cone break-up that occurs as the cone moves from pistonic to resonant behaviour at higher frequencies.

HYPOTHESIS AND MEASUREMENTS OF THE EXAMPLE OF THE INVENTION

According to established horn theory, the large compression chamber (also know as throat chamber or coupling chamber) of this invention should act as a parallel acoustic capacitance, causing a first order low pass filter effect. So for this particular example: Fhc = 2.Qts.Fs.(Vas/Vf) = 2 x 0.64 x 101 x (24.6 / 18.5) = 172Hz This is a 6dB/octave low pass filter with the -3dB point at 172Hz. This can also be seen in the following simulation, where Fig. 7 is the input parameters and Fig. 8 the schematic diagram. Note that this schematic shows a simplified version of the invention, where the compression chamber is cylindrical and the horn conical and mounted externally -according to horn theory, as long as the compression chamber volume, throat area, horn length and mouth area are the same then the shape should not affect the fundamental behaviour of the loudspeaker.

Fig. 9 shows the acoustical power of the loudspeaker according to this model. Note that it shows an output peak at 180Hz and above this frequency the output decreases by approximately 6dB/octave, broadly in agreement with the above formula.

Initial experiments with early prototypes of this invention showed that although this invention may appear to be a rear-loaded horn, both empirical testing and acoustical and electrical measurements showed that it does not behave as such a horn should.

The following measurements were taken using a Liberty Instruments Praxis measurement system with calibrated omnidirectional microphone and reference probes with the Audpod, and a QSC PLX 3002 power amplifier. The acoustical measurements are quasi-anechoic measurements, using a gated chirp to prevent room echoes from affecting the data, which for both legibility and analysis were then smoothed with a 1/3 octave 2' order filter. For this final comparative testing we used three loudspeaker cabinets, the final prototype version of this invention, an open-backed cabinet of equal size and a closed cabinet of equal size. The 10" loudspeaker used was an Eminence Ramrod guitar loudspeaker and one single example was swapped between the cabinets when testing to remove any variability between samples.

Fig. 10 is the electrical impedance measurements of all three cabinets -the invention in blue, open-backed in green and closed in red. Note that the invention shows two peaks, 46 ohms at 140Hz and 72 ohms at 58Hz, with a dip between these centred at 100Hz. In contrast, Fig. 11 shows the theoretical electrical impedance of this invention according to this model. Note that it shows two peaks, 16 ohms at 180Hz and 59 ohms at 67Hz, with a dip between these centred at 129Hz. The discrepancy between the measured and modelled impedance curves for the invention confirms the belief that this is not behaving as conventional rear-loaded horn whilst the impedance curve shown in Fig. 10 does confirm that we have a Helmholtz resonance occurring at 100Hz. Also, the near identical impedance curves above 300Hz for the invention, the open-backed and the closed cabinets confirms that the slot-horn apparatus is not changing the acoustic impedance load on the driver above this frequency and will thus not change its forwards frequency response or tone to a significant degree.

Fig. 12 shows the front and rear on-axis frequency responses of the invention, front in blue and rear in red. According to Fig. 9 and the preceding formula, if this invention was acting like a true horn the rear frequency response (through the slot/horn apparatus) should diverge from the front frequency response by -6dB per octave above -180Hz. Fig. 12 shows this is clearly not the case the front response shows that the loudspeaker's output drops by 12dB/oct above 2.5kHz with the rear response merely lowering the corner frequency by 500Hz to 2kHz, nothing like a 6dB/oct lowpass filter from 180Hz.

Below 200Hz Fig. 12 shows that the front response is diminished as the slot/horn apparatus acts as a Helmholtz resonator, reducing cone excursion (and thus front output) but hugely increasing rear output. At these low frequencies the front and rear output are omnidirectional so will combine and either sum or subtract according to their relative phase. As with any Helmholtz resonator (aka bass reflex or ported) loudspeaker, the front speaker output and Helmholtz resonator output are in phase above the tuning frequency (100Hz in this case) and then invert gradually below this point, so around and above the tuning frequency this causes a significant increase in efficiency, excursion limited power handling and maximum output.

Fig. 13 shows the on and off-axis rear frequency response of the invention and the open-backed cabinets. The dark green line (highest amplitude at 2kHz) is the on-axis invention frequency response. The light green line (second highest amplitude at 2kHz) is the 45 deg off-axis invention frequency response. The red line (third highest amplitude at 2kHz) is the on-axis open-backed frequency response. The orange line (lowest amplitude at 2kHz) is the 45 deg off-axis open-backed frequency response.

Below 200Hz we can observe the increased output of the invention due to the Helmholtz resonance -according to established theory the low frequency rearward output of an open-backed cabinet will be 180 deg out of phase with the low frequency forward output and as output at these low frequencies is omnidirectional the summed front and rear low frequency output of the invention will be over 10dB greater than that of the open-backed cabinet.

At 200-250Hz and 400-600Hz we can observe the peak and corresponding dip caused by resonances within the open-backed cabinet. The invention exhibits significantly more even frequency response across this range, as hypothesised would occur due to the slanted horn panels making the internal shape non-cuboidal and spreading internal resonances across many more frequencies rather than the two dominant frequencies of an open-backed cab.

Above 1kHz we can observe many clear beneficial effects being caused by the slot-horn apparatus. From 1-2.5kHz the invention's frequency response is significantly stronger (up to 10dB) both on and off-axis than the open-backed cabinet. From 1.5kHz upwards the open-backed cabinet's off-axis response is around 6dB weaker than its on-axis response, whilst the invention's off-axis response stays very close to the on-axis response up to just over 3kHz and above that averages only -4dB less. Further analysis of these plots shows that from 1.2kHz upwards the invention's frequency response is between 1 and 8dB stronger on-axis than the open-backed cabinet, and from 800Hz upwards the invention's frequency response is between 1 and 9dB stronger off-axis than the open-backed cabinet.

All this data supports the hypothesis that the slot/horn apparatus increases both efficiency and dispersion of the rearwards radiation at midrange and treble frequencies whilst also increasing total (front and rear radiation) efficiency and power-handling at low frequencies, without detriment to the tone of the front radiation. Extensive qualitative testing by experienced guitarists further supports the hypothesis.