Method for controlling the diaphragm excursion of electrodynamic loudspeakers and devices therefor

Abstract

Procedure with the following steps: when an audio input signal is applied to a coil (20) of a loudspeaker (1), detecting a current flowing through the coil (20) of the loudspeaker (1) and / or a voltage across the coil (20) of the loudspeaker (1); Estimating the diaphragm excursion of the loudspeaker (1) from an analysis of a low-frequency (LF) component of the audio input signal and using an adaptive digital model (610) of the loudspeaker (1), wherein the detected voltage and / or detected current is fed to the adaptive digital model (610); and if a value of the estimated membrane stroke exceeds a predetermined threshold: Modifying the low-frequency (LF) component of the audio input signal to provide a modified low-frequency (LF) component of the audio input signal; and Combining the modified low-frequency (LF) component of the audio input signal with a high-frequency (HF) component of the audio input signal to provide a combined signal; and Output of the combined signal to the loudspeaker (1).

Description

AREA OF INVENTION

[0001] The present invention relates to a method for controlling the diaphragm stroke of an electrodynamic loudspeaker and corresponding devices for this purpose. BACKGROUND OF THE INVENTION

[0002] The present invention relates to a method for controlling the diaphragm excursion of an electrodynamic loudspeaker. The electrodynamic loudspeaker can, for example, be mounted in a substantially sealed or vented enclosure. The substantially sealed or vented enclosure can form part of a portable communication device such as a smartphone or tablet computer. The electrodynamic loudspeaker can be used for sound reproduction purposes, e.g., as a receiver for sound generation by acoustic coupling with the user's ear, or as a loudspeaker for playing back recorded music or for speech reproduction in teleconferencing applications.

[0003] US 2012 / 0179456A1 concerns maintaining the original volume level of an audio signal for a mobile device while preserving sound quality as much as possible and protecting the speaker used in the mobile device. The volume of an audio (e.g., speech) signal can be maximized while controlling the excursion of the speaker diaphragm in a mobile device so that it remains within a permissible range. In one implementation, the peak excursion is predicted (e.g., estimated) using the input signal and an excursion transfer function. The signal can then be modified to limit the excursion and maximize the volume.

[0004] US 2013 / 0077795A1 concerns a teaching in which excessive excursion of a diaphragm in an electrodynamic transducer is reduced by attenuating the low-frequency component of an audio signal when the audio signal's power exceeds a predetermined power limit. The audio signal drives the input of an amplifier, and the amplifier's output drives the electrodynamic transducer. If the audio signal does not exceed the predetermined power limit, the low-frequency component of the input audio signal is amplified.

[0005] US 2013 / 0287203A1 concerns methods and devices for improved echo suppression. In one example, an audio signal intended for output through a loudspeaker is received and processed to generate an optimized audio signal. The optimized audio signal causes less distortion when played back through the loudspeaker. The method further includes using the optimized audio signal to reduce echo in a transmitted audio signal.

[0006] In numerous audio reproduction applications, it is crucial to accurately estimate and limit the diaphragm excursion or displacement of an electrodynamic loudspeaker, for example, to protect the loudspeaker from mechanical damage. Precise control or limitation of the diaphragm excursion is essential to prevent the diaphragm or diaphragm assembly from being driven beyond its upper excursion limit. This can occur if the maximum excursion limit of the electrodynamic loudspeaker is exceeded, potentially causing various types of temporary or permanent mechanical damage, rendering the loudspeaker temporarily or permanently inoperable.Mechanical damage can be caused by a collision between moving loudspeaker components, such as the voice coil, diaphragm, or voice coil former, and a stationary component of the loudspeaker, such as the magnetic circuitry. Since electrodynamic loudspeakers are generally nonlinear devices, especially at high sound pressure levels, with a large number of complex nonlinearities, estimating diaphragm excursion using various predictive model-based approaches has proven difficult. Furthermore, linear loudspeaker parameters, such as mechanical compliance and voice coil resistance, also vary slowly but significantly with time and temperature, posing further challenges in maintaining an accurate model of the electrodynamic loudspeaker.It is therefore of significant interest and value to provide a relatively simple methodology for correctly estimating and limiting the diaphragm excursion of the electrodynamic loudspeaker, despite these challenges.

[0007] The present method for controlling the diaphragm excursion involves dividing the audio input signal into at least one low-frequency band signal and one high-frequency band signal using a band-division network. The low-frequency band signal is applied to a diaphragm estimator to determine the instantaneous diaphragm excursion based on the low-frequency band signal. Band-division of the audio input signal before signal limiting is advantageous because the diaphragm excursion at and below the fundamental resonance frequency of the electrodynamic loudspeaker is much greater than above the fundamental resonance frequency. This relationship is inversely proportional to the sound pressure level, which is much higher above the loudspeaker's fundamental resonance frequency than below it.To provide effective excursion control and thus good speaker protection without unnecessarily limiting the speaker's maximum sound pressure level capability, the audio input signal is split into low-frequency and high-frequency band signals. A cutoff frequency of the band-splitting network can be set above the fundamental resonant frequency of the electrodynamic speaker, so that the high-frequency band signal lies above the fundamental resonant frequency. The low-frequency band signal can therefore lie at or below the fundamental resonant frequency of the electrodynamic speaker. In this way, the low-frequency band signal can be limited separately to prevent excessive diaphragm excursion and protect the speaker, while the high-frequency band signal can remain essentially unlimited. The subjective loudness of the sound produced by the electrodynamic speaker thus remains relatively unaffected by the diaphragm excursion limitation. PRESENTATION OF THE INVENTION

[0008] The invention relates to a method according to claim 1, a device according to claim 12 and a device according to claim 24. Advantageous embodiments are described in the dependent claims.

[0009] A first aspect of the invention relates to a method for controlling the diaphragm excursion of a loudspeaker, such as an electrodynamic loudspeaker. The method for controlling the diaphragm excursion comprises steps for receiving an audio input signal from an audio signal source, Splitting the audio input signal into at least one low-frequency band signal and one high-frequency band signal using a band-division network having a predetermined fundamental frequency. Applying the low-frequency band signal to a diaphragm excursion estimator, Determining or estimating, based on the low-frequency band signal, a momentary diaphragm excursion, Comparing the current membrane stroke with a stroke limit criterion, Limiting the low-frequency band signal based on a result of comparing the instantaneous diaphragm excursion and the excursion limit criterion, in order to produce a limited low-frequency band signal, Combining the limited low-frequency band signal and the high-frequency band signal to produce a stroke-limited audio signal.

[0010] The present diaphragm excursion control methodology and associated audio reproduction circuitry can be used in a wide range of electrodynamic loudspeakers in various applications, such as low-frequency or full-range loudspeakers for high-fidelity applications, automotive applications, or applications in public address systems, as well as in miniature electrodynamic loudspeakers for portable communication devices and / or music players. In the latter case, the electrodynamic loudspeaker can be integrated into a mobile phone or smartphone and housed in a sealed, vented, or open enclosure with a volume between 0.5 and 2.0 cm³. 3 , such as 1 cm 3 , be mounted. The electrodynamic loudspeaker mounted in a housing can produce a usable sound pressure level from below 100 Hz up to 15 kHz or even 20 kHz.

[0011] The present diaphragm excursion control method is preferably designed to sufficiently limit the diaphragm deflection or excursion to prevent various types of mechanical damage to the loudspeaker, as described above. This mechanical damage can be caused by a collision between moving loudspeaker components, such as the voice coil, diaphragm, or voice coil former, and a stationary component, such as the magnetic circuitry. Limiting the low-frequency signal can involve dynamic range compression or peak clipping of the low-frequency signal in response to the instantaneous diaphragm excursion being estimated to meet the excursion limit criterion. The excursion limit criterion can include an excursion threshold corresponding to a maximum excursion of the electrodynamic loudspeaker specified by the loudspeaker manufacturer or to a specific amount / percentage of that maximum excursion.It will be obvious to those skilled in the art that the maximum excursion of a given type of electrodynamic loudspeaker depends on its dimensions and numerous structural details. For the miniature loudspeaker 1 discussed above, with external dimensions of approximately 11 mm x 15 mm, the maximum diaphragm excursion is typically about ±0.45 mm. However, loudspeakers with larger dimensions, for example for home stereo applications or applications in public sound systems, can have much larger maximum diaphragm excursions, for example, over ±5.0 mm.

[0012] The audio input signal can include a real-time digital audio signal supplied from an external digital audio source, such as a digital microphone. The real-time digital audio signal can be formatted according to a standardized serial data communication protocol such as I2C or SPI, or according to a digital audio protocol such as I2C. 2 S, SPDIF, etc.

[0013] The method for controlling the membrane stroke may further include steps that include: The excursion-limited audio signal is applied to a power amplifier to produce a loudspeaker drive signal, and the loudspeaker drive signal is applied to the electrodynamic loudspeaker to produce a sound pressure level corresponding to the excursion-limited audio signal. Various types of linear or pulse-modulated power amplifiers can be used. The power amplifier, or output amplifier, can be a switching amplifier or a Class D amplifier, for example, a pulse-density modulation (PDM) or pulse-width modulation (PWM) output amplifier, both of which have high power conversion efficiency. This is a particularly advantageous feature for use in battery-powered portable communication devices. Alternatively, the output amplifier can include traditional non-switched power amplifier topologies such as Class A or Class AB.

[0014] The excursion-limited audio signal exhibits both low-frequency and high-frequency components of the audio input signal due to the combination or summation of the low-frequency and high-frequency band signals. Since the limitation of the low-frequency band signal can be essentially zero for small instantaneous diaphragm excursion values, low-level audio input signals can propagate through the limiting and combination stages without frequency response modifications and thus without subjective signal coloration. Conversely, for large instantaneous diaphragm excursion values—that is, instantaneous diaphragm excursions that meet the excursion limit criterion—the low-frequency band signal is limited, for example, by peak clipping or dynamic range compression, thus reducing the level of the low-frequency band signal.This reduction in the level of the low-frequency band signal leads, at high values ​​of the instantaneous diaphragm excursion, to a slope in the frequency response of the excursion-limited audio signal, so that in the excursion-limited audio signal the level of the high-frequency components of the audio input signal increases relatively compared to the level of the low-frequency components of the audio input signal.

[0015] The division of the audio input signal into the high-frequency band signal and the low-frequency band signal can be achieved in various ways, either in the digital domain with a digitized audio input signal or in the analog domain. In some embodiments, the audio input signal is divided by means of a band-division network that has a predetermined cutoff frequency within the audio frequency range between 20 Hz and 20 kHz, such as between 100 Hz and 10 kHz or between 400 Hz and 2 kHz. The predetermined cutoff frequency preferably depends on the fundamental resonant frequency of the electrodynamic loudspeaker to which the excursion-limited audio signal is applied, as will be discussed in more detail below.It will be obvious to those skilled in the art that the fundamental resonant frequency of a corresponding electrodynamic loudspeaker varies considerably depending on the loudspeaker's characteristics, such as the diaphragm suspension and mass, and on the specific mounting of the loudspeaker, for example, the dimensions of the loudspeaker enclosure. The fundamental resonant frequency of miniature electrodynamic loudspeakers mounted in an enclosure for portable devices typically lies somewhere between 400 Hz and 1.0 kHz.

[0016] One embodiment of the band subdivision network has an all-pass frequency response such that the sum of the low-frequency band signal and the high-frequency band signal provides a substantially flat (i.e., constant or having a magnitude of one) frequency response throughout the relevant part of the audio frequency range.

[0017] The predetermined cutoff frequency of the band division network is preferably at least 1.2 times higher, and may even be at least twice higher, than the fundamental resonant frequency of the electrodynamic loudspeaker in question, to ensure that those signal components of the audio input signal that cause a large diaphragm excursion are contained in the low-frequency band signal. In yet another embodiment, the predetermined cutoff frequency of the band division network is between 1.2 and 2.5 times the fundamental resonant frequency of the electrodynamic loudspeaker in question.

[0018] In this context, the fundamental resonant frequency of the electrodynamic loudspeaker is the resonant frequency determined or defined by the total compliance acting on the moving diaphragm assembly and the total moving mass of the electrodynamic loudspeaker. The total compliance acting on the moving diaphragm assembly typically comprises a parallel combination of the compliance of the loudspeaker's edge suspension and the compliance caused by the air trapped within the enclosure. The fundamental resonant frequency of an electrodynamic loudspeaker mounted in an enclosure can be determined by finding the local highest impedance, for example, by examining the impedance curves of Fig. 3 & 4, as will be discussed in more detail below.

[0019] Dynamic range compression to limit the low-frequency band signal discussed above can be implemented in various ways. According to one embodiment, the dynamic range compression involves amplifying the low-frequency band signal with a gain value dependent on the diaphragm excursion, according to a predetermined gain rule. The predetermined gain rule may, for example, include a suitable table or mathematical equation expressing a desired relationship or connection between the diaphragm excursion and the gain value. In one embodiment, the predetermined gain rule includes an excursion limit. According to the latter embodiment, the predetermined gain rule additionally features a substantially constant gain of the low-frequency band signal below the excursion limit and a decreasing gain value with increasing diaphragm excursion above the excursion limit.In the latter embodiment, the low-frequency signal can be multiplied by a constant, such as one, for all diaphragm excursions below the excursion limit. Above the excursion limit, the gain can decrease so that the level of the low-frequency signal remains essentially constant regardless of the diaphragm excursion.

[0020] Determining or estimating the instantaneous diaphragm displacement can be performed in many ways. In some embodiments of the invention, the instantaneous diaphragm displacement can be derived from a sensor signal. The sensor signal can be generated or supplied by an acceleration, velocity, or displacement sensor attached to or coupled with the loudspeaker diaphragm. In an alternative embodiment, the instantaneous diaphragm displacement is determined using a linear or nonlinear state-space model of the electrodynamic loudspeaker. According to these state-space models of the present methodology, the determination of the instantaneous diaphragm displacement further comprises steps for: Determining the voice coil voltage (Vcoil) at the voice coil, detecting the voice coil current (Icoil) in response to the voice coil voltage, Applying the detected voice coil current and the determined voice coil voltage to a linear or nonlinear state-space model of the electrodynamic loudspeaker, Applying the low-frequency band signal to the linear or non-linear state-space model of the electrodynamic loudspeaker, Determining the instantaneous membrane displacement from an output variable of the linear or nonlinear state-space model.

[0021] The present methodology may include further steps, such as applying the detected voice coil current and the determined voice coil voltage to a linear adaptive digital loudspeaker model that has several adaptive loudspeaker parameters. to calculate several parameter values ​​of the several corresponding adaptive loudspeaker parameters using the linear adaptive digital loudspeaker model, to apply the multiple parameter values ​​to the linear or non-linear state space model of the electrodynamic loudspeaker.

[0022] The linear adaptive digital loudspeaker model can include an adaptive FIR or an adaptive IIR filter. The adaptive IIR filter can be a second-order or higher-order filter. The linear adaptive digital loudspeaker model preferably has at least one fixed parameter, such as the total moving mass of the loudspeaker.

[0023] The characteristics and operation of the linear and nonlinear state-space models of the electrodynamic loudspeaker are described below with reference to Fig. The features and operation of the linear adaptive digital loudspeaker model are discussed in detail below with reference to the diagrams shown in section 6. Fig. 6 of the attached drawings are discussed in detail.

[0024] Furthermore, a sound reproduction circuit or an electrodynamic loudspeaker is revealed, featuring: an audio input for receiving an audio input signal supplied by an audio signal source, a frequency band subdivision network coupled to the audio input signal and designed to divide the audio input signal into at least one low-frequency band signal and one high-frequency band signal, a diaphragm displacement estimator coupled to the low-frequency band signal for determining or estimating an instantaneous diaphragm displacement based on the low-frequency band signal, a comparator designed to compare the instantaneous diaphragm stroke estimate with a stroke limit criterion, a signal limiter designed to produce a limited low-frequency band signal based on the result of a comparison between the current Membrane stroke and the stroke limit criterion to produce a limited low-frequency band signal, a signal combiner designed to combine the limited low-frequency band signal and the high-frequency band signal to produce a stroke-limited audio signal.

[0025] The audio playback circuit may include a power amplifier or output amplifier with an input coupled to the excursion-limited audio signal. The power amplifier or output amplifier is designed to produce a loudspeaker drive signal at a pair of loudspeaker terminals of the audio playback circuit. The characteristics of the power amplifier or output amplifier have been disclosed in detail above in connection with the corresponding diaphragm excursion control methodology. The Class-D output amplifier may include a half-bridge driver stage with a single output coupled to the electrodynamic loudspeaker, or a full-bridge / H-bridge driver stage with the pair of output terminals coupled to the corresponding sides or terminals of the electrodynamic loudspeaker.The properties of the frequency band division network were disclosed in detail above in conjunction with the corresponding diaphragm stroke control methodology and will be discussed in more detail below with reference to the accompanying drawings.

[0026] It is clear to those skilled in the art that the process steps and circuit function of each, or at least one, of the frequency band division network, the diaphragm excursion estimator, the signal limiter, and the signal combiner or summer can be implemented wholly or partially by means of a software-programmable microprocessor, such as a programmable digital signal processor, which operates according to executable program instructions. The software-programmable microprocessor can operate with digital domain signals that are applied to and output by the aforementioned circuit functions. According to such an embodiment, the frequency band division network comprises a first set of executable program instructions for the programmable microprocessor. The membrane stroke estimator has a second set of executable program instructions for the programmable microprocessor, The comparator has a third set of executable program instructions for the programmable microprocessor, The signal limiter has a fourth set of executable program instructions for the programmable microprocessor, and The signal combiner has a fifth set of executable program instructions for the programmable microprocessor.

[0027] Alternatively, the circuit function of each, or at least one, of the frequency band division network, diaphragm excursion estimator, signal limiter, and signal combiner or summer can be implemented by means of a separate hard-wired digital logic circuit that has a suitably designed sequential and combinational digital logic.

[0028] Furthermore, a semiconductor substrate or chip is disclosed on which an audio playback circuit according to one of its embodiments described above is integrated. The semiconductor substrate can be manufactured using a suitable CMOS or DMOS semiconductor process.

[0029] Furthermore, an audio reproduction arrangement is disclosed, comprising: an electrodynamic loudspeaker having a movable diaphragm assembly for generating audible sound in response to actuation of the diaphragm assembly, and an audio reproduction circuit according to one of its embodiments described above. The pair of loudspeaker terminals of the audio reproduction circuit is electrically connected to the voice coil of the movable diaphragm assembly of the electrodynamic loudspeaker. An audio signal source is functionally connected to the audio signal input of the audio reproduction circuit. The present audio reproduction arrangement can advantageously function as a self-contained audio output system with integrated loudspeaker excursion control, which can operate independently of an application processor of the portable communication terminal to provide reliable and convenient protection against excursion-induced mechanical damage to the electrodynamic loudspeaker. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Preferred embodiments of the invention are discussed in more detail with reference to the accompanying drawings. These show: Fig. 1A) a schematic cross-sectional view of an electrodynamic miniature loudspeaker for various portable sound reproduction applications for use in the present invention, Fig. 1B) a schematic cross-sectional view of the miniature electrodynamic loudspeaker mounted in a closed enclosure or housing, Fig. 2 a curve of an experimentally measured loudspeaker impedance versus the frequency of an exemplary electrodynamic miniature loudspeaker, Fig. 3 a graph 300 of the loudspeaker impedance curve and a diaphragm excursion curve of the exemplary electrodynamic miniature loudspeaker 1, Fig. 4 a graph with curves of a sound pressure level and diaphragm excursion of the exemplary electrodynamic miniature loudspeaker, Fig. 5 a simplified schematic block diagram of a sound reproduction circuit for an electrodynamic loudspeaker according to a first embodiment of the invention, and Fig. 6 A simplified schematic block diagram of an exemplary diaphragm displacement estimator for use in the in Fig. 5 shown sound playback circuit. DETAILED DESCRIPTION OF PREFERRED EXECUTION FORMS

[0031] Fig. Figure 1A) is a schematic cross-sectional view of a typical miniature electrodynamic loudspeaker 1 for mounting in a sealed box and use in portable audio applications such as mobile phones and smartphones, where the loudspeaker 1 provides sound reproduction for various application types such as hands-free calling and music playback. It is clear to those skilled in the art that electrodynamic loudspeakers exist in numerous shapes and sizes, depending on the intended application. The electrodynamic loudspeaker 1 used in the following described methodology for detecting enclosure leaks and the corresponding arrangements for detecting enclosure leaks has a rectangular shape with a maximum external dimension, D, of approximately 15 mm and a transverse external dimension of approximately 11 mm.However, it is clear to the expert that the existing methods for controlling the diaphragm movement and the corresponding circuits are applicable to almost all electrodynamic loudspeaker types mounted in a housing or box.

[0032] The electrodynamic miniature loudspeaker 1 has a diaphragm 10 attached to an upper edge surface of a voice coil. The diaphragm 10 is also mechanically coupled to a loudspeaker frame 22 by a flexible edge or outer suspension 12. An annular permanent magnet assembly 18 generates a magnetic flux that is guided through a magnetically permeable structure 16 with a circular air gap 24 arranged therein. A circular ventilation channel 14 is provided in the frame assembly 22 and can be used to dissipate heat from a chamber assembly formed below the diaphragm 10, which is otherwise sealed. The elastic edge suspension 12 provides a relatively well-defined compliance of the movable diaphragm assembly (voice coil 20 and diaphragm 10).The compliance of the elastic rim suspension 12 and a moving mass of the diaphragm 10 determines the free-air fundamental resonance frequency of the miniature loudspeaker. The elastic rim suspension 12 can be designed to limit a maximum excursion or a maximum displacement of the movable diaphragm assembly.

[0033] In operation of the miniature loudspeaker 1, a voice coil voltage or drive voltage is applied to the voice coil 20 of the loudspeaker 100 via a pair of loudspeaker terminals (not shown) electrically connected to a suitable output amplifier or power amplifier. In response, a corresponding voice coil current flows through the voice coil 20, resulting in a substantially uniform vibration of the diaphragm assembly in the direction indicated by the velocity arrow V within a piston-like region of the loudspeaker. This generates a corresponding sound pressure level from the loudspeaker 1. The vibration of the voice coil 20 and the diaphragm 10 in response to the flow of the voice coil current is caused by the presence of a radially directed magnetic field in the air gap 24.The applied voice coil current and voltage result in power dissipation in the voice coil 20, causing it to heat up during operation. Prolonged application of excessively high drive voltage and current can therefore lead to overheating of the voice coil 20, which is another common cause of failure in electrodynamic loudspeakers.

[0034] Applying excessive voice coil currents, which force the movable diaphragm assembly beyond its maximum permissible excursion limit, is another common failure mechanism in electrodynamic loudspeakers, leading to various types of irreversible mechanical damage. One type of mechanical damage can occur, for example, from a collision between the lower edge of the voice coil 20 and an annular part 17 of the magnetically permeable structure 16 facing it. Another type of mechanical damage can be caused, for example, by overstretching the elastic edge suspension 12 beyond its tensile strength.

[0035] Fig. Figure 1B) is a schematic cross-sectional view of the miniature loudspeaker 1 mounted in a housing, box, or chamber 31 with a predetermined internal volume 30. The housing or chamber 31 is located below the diaphragm 10 of the loudspeaker 1. An outer circumferential wall of the frame structure 22 of the loudspeaker 1 is fixedly attached to a contact wall surface of the sealed box 31 to form an essentially airtight coupling, thereby isolating the air trapped in the volume 30 from the environment. The enclosed volume 30 can be between 0.5 and 2.0 cm³. 3 amount to approximately 1 cm 3for typical applications in portable devices such as mobile phones and smartphones. Mounting the loudspeaker 1 in the sealed enclosure 30 results in a higher fundamental resonance frequency of the miniature loudspeaker than its free-air fundamental resonance frequency discussed above, due to the compliance of the air trapped in the chamber 30. The compliance of the air trapped in the chamber 30 acts in conjunction with the compliance of the elastic edge suspension 12 to reduce (i.e., increase) the total compliance acting on the moving mass of the loudspeaker. Therefore, the fundamental resonance frequency of the loudspeaker 1 mounted in an enclosure is higher than the free-air resonance. The magnitude of the increase in the fundamental resonance frequency depends on the volume of the enclosure 30. The wall structure surrounding the sealed enclosure 31 can be formed by an elastomeric mass with limited impact resistance.

[0036] Fig. Figure 2 shows in a graph 200 an experimentally measured curve 201 of an impedance against the frequency of the electrodynamic miniature loudspeaker 1 mounted in a closed housing with a volume of about 1 cm³. 3 The x-axis of graph 200 shows the measurement frequency on a logarithmic scale over a frequency range from 5 Hz to approximately 5 kHz, and the y-axis shows the magnitude of the measured electrical impedance of the miniature loudspeaker 1 on a linear scale ranging from approximately 6 Ω to 15 Ω. The measured fundamental resonant frequency of the loudspeaker 1 is approximately 838 Hz, as indicated by reference numeral 203, and the corresponding peak impedance of the loudspeaker 1 is approximately 14 Ω. At very low frequencies, for example below approximately 50 Hz, the impedance curve 201 remains essentially flat, reflecting the fact that the voice coil impedance gradually approaches a DC resistance of the voice coil of approximately 8 Ω.

[0037] Fig. Figure 3 shows, in a graph 300, the loudspeaker impedance curve 301 and a corresponding diaphragm excursion curve 305 of the exemplary electrodynamic miniature loudspeaker 1 discussed above. The x-axis of graph 300 shows the measurement frequency on a logarithmic scale over a frequency range from 5 Hz to approximately 5 kHz. The y-axis shows the magnitude of the measured electrical impedance and the diaphragm excursion of the miniature loudspeaker 1 on a dimensionless scale for comparison purposes. The diaphragm excursion exhibits a local maximum at approximately 838 Hz at the fundamental resonance frequency of the loudspeaker 1, as indicated by the reference symbol 303. The frequency of this local maximum of the diaphragm excursion therefore coincides with the peak impedance frequency of the loudspeaker 1.At frequencies far below the fundamental resonance frequency, the diaphragm excursion increases monotonically with decreasing frequency, so that at frequencies below approximately 25 Hz, the diaphragm excursion exceeds the excursion at the fundamental resonance frequency. Since a large portion of the signal energy of ordinary speech and music signals is concentrated in the low-frequency range, the sharp increase in diaphragm excursion in this frequency range can lead to irreversible mechanical damage to the loudspeaker if no suitable precautions are taken to limit the diaphragm excursion.

[0038] Fig. Figure 4 shows the diaphragm excursion curve 405 again in a graph 400, this time together with a corresponding sound pressure level curve 407 of the exemplary electrodynamic miniature loudspeaker 1 discussed above. The x-axis of graph 400 shows the measurement frequency on a logarithmic scale over a frequency range from 5 Hz to approximately 5 kHz. The y-axis shows the diaphragm excursion and the sound pressure level of the miniature loudspeaker 1 on a dimensionless scale for comparison purposes. As expected with a sealed loudspeaker enclosure, the sound pressure level reaches a maximum around the fundamental resonance frequency of the loudspeaker 1, and thereafter the sound pressure level decreases monotonically by approximately 12 dB per octave for decreasing frequencies.

[0039] Fig. Figure 5 shows a simplified schematic block diagram of a 500-series audio playback circuit for electrodynamic loudspeakers, for example the one described in Fig. 1B) above, miniature loudspeaker 1 mounted in a housing according to a first embodiment of the invention. It is clear to those skilled in the art that the sound reproduction circuit 500 can easily be adapted to numerous other types of electrodynamic loudspeakers mounted in a housing or on an open baffle. This loudspeaker-specific adaptation of the sound reproduction circuit 500 can include setting a cutoff frequency f c to adjust a frequency band subdivision network 502 depending on the electroacoustic characteristics of the loudspeaker and / or the intended application.

[0040] The audio playback circuit 500 produces a limited-excursion audio signal at the terminal or wire 512 to an input of a power amplifier 510. The power amplifier 510 is coupled or connected to the miniature electrodynamic loudspeaker 1 by a pair of externally accessible speaker terminals 511a, 511b. The power amplifier 510 may include a pulse-modulated class-D output amplifier or any other suitable power amplifier to supply sufficient drive voltage and current for a voice coil of the miniature electrodynamic loudspeaker 1.

[0041] The audio playback circuit 500 has an audio input 501 for receiving an audio input signal supplied by an audio signal source (not shown). In the present embodiment, the audio playback circuit 500 operates mainly or entirely in the digital domain, handling and processing digital signals; however, other embodiments of the invention may instead use analog signals or a mixture of analog and digital signals. Accordingly, the audio input 501 receives a digital audio input signal supplied by an external digital audio source, such as an application processor of a portable communication device in which the present audio playback circuit 500 is integrated.The externally generated digital audio input signal can be formatted according to a standardized serial data communication protocol, such as IIC or SPI, or according to a digital audio protocol, such as IIS, SPDIF, etc. The audio playback circuit 500 further comprises the frequency band division network 502 mentioned above, which divides the digital audio input signal into a low-frequency band signal or component and a high-frequency band signal or component. The low-frequency band signal is supplied at a first output LF of the network 502, and the second high-frequency band signal is supplied at a second output HF of the network 502. The frequency band division network 502 has a predetermined cutoff frequency f. c on, which preferably lies at least 1.2 times the fundamental resonance frequency of the miniature loudspeaker 1 mounted in a housing. The latter is at approximately 829 Hz, as discussed above. However, f ceven more favorably at or above 2.0 times the fundamental resonance frequency of 829 Hz of the miniature loudspeaker 1 mounted in a housing, such as at or above 1628 Hz. In general, f c typically somewhere in between. To the expert, when considering the in Fig. The membrane stroke curve shown in 305 makes it clear that with the latter value of f cIt is ensured that all frequency components of the digital audio input signal that contribute significantly to the diaphragm excursion of the miniature loudspeaker 1 are contained in the low-frequency band signal and are thus subject to signal limiting, as discussed in detail below. The frequency response characteristics of the frequency band subdivision network 502 can vary considerably depending on application-specific requirements. In the present embodiment, the frequency band subdivision network 502 operates exclusively in the digital domain, so that the generated high-frequency and low-frequency band signals are each digital signals. Operating the frequency band subdivision network 502 in the digital domain provides considerable flexibility in selecting the appropriate frequency response characteristics of a low-pass filter element and a high-pass filter element of the network 502, which each process the low-frequency band signal LF and LF, respectively.The high-frequency band signal HF is provided. The low-pass and high-pass filter elements of the 502 network can each comprise an FIR filter or an IIR filter. The 502 band-division network can, for example, have an all-pass frequency response such that summing the low-frequency band signal LF and the high-frequency band signal HF provides an essentially flat (constantly high) frequency response. Thus, low-level digital audio input signals can propagate through the 502 band-division network without frequency response modifications and therefore without subjective signal coloration. A slope in the frequency response of the low-pass filter element above the predetermined cutoff frequency f. cThe slope can vary depending on the specific application. It is preferably greater than 12 dB per octave, such as greater than 24 dB or 30 dB per octave, to reduce the leakage of high-frequency signal components into the low-frequency band signal LF. A slope in the frequency response of the high-pass filter below the predetermined cutoff frequency f c The slope of the high-pass filter element can also vary depending on the specific application. It can be essentially the same as, or different from, the slope of the low-pass filter element. The slope of the high-pass filter element is preferably greater than 12 dB per octave, such as greater than 24 dB or 30 dB per octave, to reduce the leakage of low-frequency signal components into the high-frequency band signal HF. It will be apparent to those skilled in the art that the low-pass filter element is a high-pass filter with a cutoff frequency far below the predetermined cutoff frequency f. c, for example, at a subsonic frequency below 20 Hz, the cutoff frequency can be included, so that this embodiment converts the low-pass filter element into a band-pass filter element. The high-pass filter of the latter band-pass filter element can be advantageous for suppressing potentially harmful ultra-low frequency components of the digital audio input signal. The miniature loudspeaker can, for example, be used in the Fig. The frequency response curve 407 shown in the diagram will be unable to reproduce these ultra-low components at a usable or acoustically perceptible sound pressure level.

[0042] The digital low-frequency band signal LF is applied in parallel to an input of a diaphragm excursion estimator or calculator 504 and to an input of a signal limiter 506. The diaphragm excursion estimator 504 is configured to produce an estimate x of the instantaneous diaphragm excursion based on the digital low-frequency band signal LF and to apply this estimate x to the signal limiter 506 via the line or connection 507. The signal limiter 506 reads or receives the estimate x of the instantaneous diaphragm excursion and uses it to produce a limited low-frequency band signal at output LF_C by comparing the estimate x of the instantaneous diaphragm excursion with an excursion limit criterion. The limited low-frequency band signal at output LF_C of the signal limiter 506 is attenuated or limited in level or amplitude by the operation of the signal limiter 506 relative to the low-frequency band signal LF at its input.Preferably, the signal limiting incorporates dynamic range compression of the low-frequency band signal to reduce the amount of nonlinear signal distortion introduced by the limiting process. Alternatively, the limiting of the low-frequency band signal LF can be performed using a simple peak-clipping process, in which positive and negative signal amplitudes of the low-frequency band signal LF that exceed upper or lower threshold values ​​are clipped.

[0043] Dynamic range compression to limit the low-frequency band signal LF can include amplifying the low-frequency band signal with a diaphragm-excursion-dependent gain value according to a predetermined gain rule. The predetermined gain rule can, for example, include a suitable table or mathematical equation that expresses a relationship or connection between the specified diaphragm excursion and the gain value. In one embodiment, the predetermined gain rule has an excursion limit. The predetermined gain rule of the latter embodiment additionally features a substantially constant gain of the low-frequency band signal below the excursion limit and a decreasing gain value with increasing diaphragm excursion above the excursion limit. In the latter embodiment, the low-frequency band signal LF can be multiplied by a constant, such as one, for all diaphragm excursion values ​​below the excursion limit.Thus, the low-frequency band signal LF can be transmitted through the signal limiter 506 essentially unchanged, so that the limited low-frequency band signal at output LF_C is essentially the same as the low-frequency band signal LF at the input of the signal limiter 506. Above the excursion limit, the gain can decrease monotonically at various rates. In some embodiments, the rate of decrease of the gain can ensure that the level of the low-frequency band signal at output LF_C is essentially constant for all diaphragm excursion values ​​above the excursion threshold. This corresponds to an essentially infinite compression ratio of the limited low-frequency band signal.

[0044] It will be apparent to those skilled in the art that the excursion threshold values ​​in certain embodiments can be set using numerous criteria, for example, depending on the type of application. In some embodiments, the excursion threshold value can be set to a maximum excursion of the electrodynamic loudspeaker specified by the loudspeaker manufacturer or to a specific amount / percentage of this maximum excursion. It will be apparent to those skilled in the art that the maximum excursion of a corresponding type of electrodynamic loudspeaker depends on its dimensions and numerous structural details of the loudspeaker and its housing. For the miniature loudspeaker 1 discussed above, with external dimensions of approximately 11 mm x 15 mm, the maximum diaphragm excursion is typically about ±0.45 mm.However, larger loudspeaker types for home stereo applications or applications in public loudspeaker systems can have much larger maximum diaphragm strokes, for example over + / - 5.0 mm.

[0045] The limited low-frequency band signal at output LF_C is applied to or transferred to a first input of a summing amplifier 508, as shown in the schematic block diagram of the audio playback circuit 500. The high-frequency band signal HF is applied to or transferred to a second input of the summing amplifier 508. The high-frequency band signal HF and the limited low-frequency band signal are combined to produce a limited-excursion audio signal at output OUT of the summing amplifier 508. The limited-excursion audio signal is applied to the input of the power amplifier 510, which in turn generates a loudspeaker drive signal at the pair of loudspeaker terminals 511a and 511b, as previously discussed.

[0046] It is clear to those skilled in the art that the functionality of the frequency band division network 502, the diaphragm excursion estimator 504, the signal limiter 506, and the summing amplifier 508 can each be implemented, in whole or in part, by means of a software-programmable microprocessor, such as a programmable digital signal processor, as a set of executable program instructions operating with digital domain signals. Alternatively, the functionality of the frequency band division network 502, the diaphragm excursion estimator 504, the signal limiter 506, and the summing amplifier 508 can each be implemented by means of a separate, hard-wired digital logic circuit that incorporates suitably designed sequential and combinational digital logic.The hard-wired digital logic circuits can be integrated on an application-specific integrated circuit (ASIC), or implemented using programmable logic, or a combination thereof. Various variable and constant values ​​used by the 500 audio playback circuit, such as the predetermined cutoff frequency f. c of the frequency band division network, filter coefficients of the low-pass and high-pass filter elements of the 502 network, etc., can be stored in predetermined data memory addresses of a data memory accessible to the software-programmable microprocessor or the hard-wired digital logic circuits.

[0047] The diaphragm displacement estimator 504 can be implemented in many different ways. In some embodiments, the diaphragm displacement estimator 504 can be based on a sensor signal generated or supplied by an acceleration, velocity, or displacement sensor attached to or coupled with the diaphragm of the miniature loudspeaker 1 discussed above. A preferred embodiment of the diaphragm displacement estimator 504 is based on a linear or nonlinear state-space model of the electrodynamic loudspeaker as discussed below.

[0048] Fig. Figure 6 shows a simplified schematic block diagram of a preferred embodiment of the device described in Figure 6. Fig.The diaphragm displacement estimator 504 is illustrated in Figure 5. The diaphragm displacement estimator 504 incorporates a nonlinear state-space loudspeaker model 614 of the miniature loudspeaker 1. The low-frequency band signal LF is applied to an input of the nonlinear state-space loudspeaker model 614. The nonlinear state-space loudspeaker model calculates or determines an estimate of the instantaneous diaphragm displacement, x, and delivers it at the output 613. The nonlinear state-space loudspeaker model 614 may be the same as the nonlinear state-space loudspeaker model described in the applicant's concurrently pending US patent application No. 14 / 073,324 (published as US 2015 / 0124982A1), which is incorporated herein by reference in its entirety. The operation and functionality of the nonlinear state-space loudspeaker model are described in this simultaneously pending patent application.Therefore, a brief summary of the function of the nonlinear state-space loudspeaker model will be discussed below in order to demonstrate how the nonlinear state-space loudspeaker model 614 can be integrated into the present diaphragm displacement estimator 504.

[0049] The diaphragm excursion estimator 504 features a linear adaptive digital model 610 of the electrodynamic loudspeaker 1 with several adaptable or adaptive model parameters that are set in response to a digital voice coil current signal Im[n] and a digital voice coil voltage Vm[n]. The adaptive linear digital model 610 of the loudspeaker preferably includes an adaptive filter that models a frequency-dependent impedance of the miniature loudspeaker 1 over a predetermined audio frequency range, for example, between 10 Hz and 10 kHz, based on the detected or measured voice coil current and voltage represented by the digital voice coil current signal Im[n] and the digital voice coil voltage Vm[n]. The linear adaptive digital loudspeaker model 610 features several adaptive loudspeaker parameters.The linear adaptive digital loudspeaker model 610 is designed to calculate several corresponding parameter values ​​of the linear loudspeaker parameters. The details and functionality of the adaptive linear digital loudspeaker model 610 are also described in the aforementioned concurrently pending US patent application No. 14 / 073,324 of the applicant. The outputs of the linear adaptive digital loudspeaker model 610 include several parameter values ​​of the corresponding linear, albeit time-varying, adaptive loudspeaker parameters, for example, a force factor or a suspension compliance of the miniature loudspeaker 1 mounted in a housing. The several parameter values ​​output by the linear adaptive digital loudspeaker model 610 are supplied to a nonlinear functional block 612, which is part of the nonlinear state-space model 614 of the electrodynamic loudspeaker.The parameter values ​​of the adaptive loudspeaker parameters represent, or estimate, the corresponding current parameter values ​​of the respective adaptive loudspeaker parameter, so that their time- and temperature-varying properties discussed above are appropriately tracked. It will be apparent to those skilled in the art that the nonlinear functional block 612 is an optional feature of the present diaphragm excursion estimator 504. In other embodiments of the present diaphragm excursion estimator 504, the nonlinear functional block 612 is not present, and the parameter values ​​of the adaptive loudspeaker parameters are directly fed to a linear state-space model 616 of the electrodynamic loudspeaker 1.While the nonlinear functional block 612 can significantly improve the accuracy of the estimation of the instantaneous diaphragm excursion x, especially for large diaphragm excursions, its absence often saves computing resources and power consumption. Therefore, both alternatives can be highly useful depending on the requirements of a particular application of the present audio playback circuit 500. To supply the digital voice coil current signal Im[n] and the digital voice coil voltage signal Vm[n] to the adaptive linear digital loudspeaker model 610, the diaphragm excursion estimator 504 includes at least one A / D converter 608, which generates the digital voice coil current signal Im[n] and a digital voice coil voltage signal Vm[n] by sampling and digitizing the instantaneous voice coil voltage at the loudspeaker terminals 511a, 511b.The A / D converter 608 also has a second input designed to sample and digitize an analog voice coil current supplied at a second input, coil, of the converter 608.

Claims

[1] Procedure with the steps: when an audio input signal is applied to a coil (20) of a loudspeaker (1), detecting a current flowing through the coil (20) of the loudspeaker (1) and / or a voltage across the coil (20) of the loudspeaker (1); Estimating the diaphragm excursion of the loudspeaker (1) from an analysis of a low-frequency (LF) component of the audio input signal and using an adaptive digital model (610) of the loudspeaker (1), wherein the detected voltage and / or detected current is fed to the adaptive digital model (610); and if a value of the estimated membrane stroke exceeds a predetermined threshold: Modifying the low-frequency (LF) component of the audio input signal to provide a modified low-frequency (LF) component of the audio input signal; and Combining the modified low-frequency (LF) component of the audio input signal with a high-frequency (HF) component of the audio input signal to provide a combined signal; and Output of the combined signal to the loudspeaker (1). [2] Method according to claim 1, wherein the modification comprises performing a dynamic range compression of the low-frequency component (LF). [3] Method according to claim 1 or 2, wherein the modification comprises cutting off a portion of the low-frequency component (LF) that exceeds the threshold. [4] Method according to one of the preceding claims, wherein the predetermined frequency band includes all frequencies below and including the expected resonance frequency of the loudspeaker (1). [5] Method according to any of the preceding claims, wherein the estimation comprises estimating the diaphragm stroke from a sensor signal fed back from the loudspeaker (1). [6] A method according to any of the preceding claims, further comprising, prior to estimating, subdividing the audio input signal according to a cutoff frequency which is higher than the expected resonance frequency of the loudspeaker (1) into the low-frequency component (LF) and the high-frequency component (HF). [7] Method according to claim 6, wherein the cutoff frequency is between 1.2 and 2.5 times the expected resonance frequency of the loudspeaker (1). [8] Method according to claim 6 or 7, further comprising adjusting the cutoff frequency according to the characteristics of the loudspeaker (1). [9] Method according to one of the preceding claims, further comprising, prior to combining, suppressing low-frequency subcomponents of the low-frequency component (LF) below the sound reproduction capabilities of the loudspeaker (1). [10] Method according to any of the preceding claims, wherein the adaptive digital model (610) comprises a plurality of adaptive loudspeaker parameters which are adjusted in response to the detected current and / or voltage, and wherein the estimation of a diaphragm excursion of the loudspeaker (1) further comprises the calculation of a plurality of parameter values ​​of the plurality of respective adaptive loudspeaker parameters based on the adaptive digital model (610). [11] Method according to claim 10, wherein the estimation of a diaphragm stroke of the loudspeaker (1) further comprises: Applying the low-frequency (LF) component of the audio input signal to a linear or non-linear state-space model (614) of the electrodynamic loudspeaker (1); Applying the multitude of parameter values ​​to the linear or nonlinear state-space model (614) of the electrodynamic loudspeaker (1); and Determining the instantaneous membrane displacement from an output of the linear or nonlinear state-space model (614). [12] Device, with: a band division system (502) to divide an audio input signal into a corresponding high-frequency (HF) component and a low-frequency (LF) component, a processor circuit configured to estimate a diaphragm excursion of a loudspeaker (1) in response to the audio input signal based on the low-frequency component (LF) and an adaptive digital model (610) of the loudspeaker (1), wherein the processor circuit is configured to supply a measured voltage and / or a measured current of a voice coil (20) of the loudspeaker (1) to the adaptive digital model (610); a comparator to compare the estimated diaphragm stroke with a threshold value, a signal limiter (506) to modify the low-frequency component signal (LF) in response to an output variable of the comparator, and a signal combiner (508) to combine an output signal from the signal limiter (506) and the high-frequency (HF) component signal from the band division system (502) to produce an output signal. [13] Device according to claim 12, wherein the band division system (502) has a cutoff frequency which is higher than a resonance frequency of the loudspeaker (1). [14] Device according to claim 13, wherein the cutoff frequency is between 1.2 and 2.5 times the expected resonance frequency of the loudspeaker (1). [15] Device according to one of claims 12 to 14, wherein the signal limiter (506) performs dynamic range compression of the low-frequency component (LF). [16] Device according to any one of claims 12 to 15, wherein the signal limiter (506) cuts off a part of the low-frequency component (LF) that exceeds the threshold value. [17] Device according to one of claims 12 to 16, which further comprises an amplifier (510) having an input coupled to the signal combiner (508) and an output for coupling to a loudspeaker (1). [18] Device according to claim 17, wherein the amplifier (510) is a pulse density modulation amplifier. [19] Device according to claim 17, wherein the amplifier (510) is a pulse width modulation amplifier. [20] Device according to one of claims 12 to 19, wherein the tape division system (502), the comparator and the signal limiter (506) are implemented by means of the processor circuit. [21] Device according to one of claims 12 to 20, wherein the band division system (502) comprises a high-pass filter element and a low-pass filter element, each receiving the audio input signal. [22] Device according to any one of claims 12 to 21, wherein the adaptive digital model (610) has a plurality of adaptive loudspeaker parameters which are adjusted in response to the detected current and / or voltage, and wherein the processor circuit is further configured to calculate a plurality of parameter values ​​of the plurality of respective adaptive loudspeaker parameters based on the adaptive digital model (610). [23] Device according to claim 22, wherein the processor circuit is further configured to: Applying the low-frequency component (LF) to a linear or non-linear state-space model (614) of the electrodynamic loudspeaker (1); Applying the multitude of parameter values ​​to the linear or nonlinear state-space model (614) of the electrodynamic loudspeaker (1); and Determining the instantaneous membrane displacement from an output of the linear or nonlinear state-space model (614). [24] Device, with: a processor which executes program instructions that cause the processor to operate as: Band division system (502) to divide an audio input signal into a corresponding high-frequency (HF) component and a low-frequency (LF) component, Estimator (504) to estimate the excursion of a loudspeaker diaphragm (10) based on the low-frequency component (LF) and an adaptive digital model (610) of the loudspeaker (1), wherein a measured voltage and / or a measured current of a voice coil (20) of the loudspeaker (1) is supplied to the adaptive digital model (610), Signal limiter (506) to modify the low-frequency (LF) component when the estimated excursion exceeds a predetermined limit, and Signal combiner (508) to combine an output signal from the signal limiter (506) and the high-frequency (HF) component signal from the band division system (502) to produce an output signal, and an amplifier (510) to generate a speaker control signal from the output signal.

Images