Method for Load Measurement in Switching Amplifiers, Corresponding Device and Amplifier

Abstract

A method can be used to measure a load driven by a switching amplifier having a differential input, an LC output demodulator filter and a feedback network between the amplifier output and the differential input. The amplifier is AC driven in a differential and in a common mode by applying a common. The feedback network provides feedback towards the differential input from downstream the LC demodulator filter by computing the impedance of the load as a function of the differential mode output current and the common mode output current. The feedback network provides feedback towards the differential input from upstream the LC demodulator filter by measuring the impedance value of the inductor of the LC demodulator filter, and computing the impedance of the load as a function of the differential mode output current, the common mode output current and the impedance value of the inductor of the LC demodulator filter.

Claims

1. A method of determining an impedance of a speaker driven by a switching differential amplifier, the method comprising: driving the switching differential amplifier in a differential mode with a differential signal having a frequency in an upper portion of an audio frequency covered by a tweeter of the speaker; measuring amplitude and phase of a differential output current of the switching differential amplifier while driving the switching differential amplifier in the differential mode with the differential signal; driving the switching differential amplifier in a common mode with a common mode signal having a frequency in the upper portion of the audio frequency covered by the tweeter of the speaker; measuring amplitude and phase of a common mode output current of the switching differential amplifier while driving the switching differential amplifier in the common mode with the common mode signal; and computing the impedance of the speaker based on the amplitude and phase of the differential output current and the amplitude and phase of the common mode output current.

2. The method of claim 1, wherein computing the impedance of the speaker comprises using a function that comprises, $Z_0={\left[\frac{\mathrm{idiff}-\mathrm{icomm}}{\mathrm{Vo}}\right]}^{-1},$ where Z.sub.o is the impedance of the speaker, idiff is the amplitude of the differential output current, icomm is the amplitude of the common mode output current, and Vo is an output voltage of the switching amplifier.

3. The method of claim 1, wherein the upper portion of the audio frequency comprises 20 kHz.

4. The method of claim 1, wherein the speaker comprises a woofer.

5. The method of claim 1, further comprising computing an impedance of a dumping network coupled between outputs of the switching differential amplifier and the speaker based on the amplitude and phase of the differential output current and the amplitude and phase of the common mode output current.

6. The method of claim 5, wherein computing the impedance of the dumping network comprises dividing an output voltage of the switching differential amplifier by the amplitude of the common mode output current.

7. The method of claim 1, further comprising measuring a value of an inductance coupled between outputs of the switching differential amplifier and the speaker by measuring a ripple of the common mode output current, wherein computing the impedance of the speaker is further based on the ripple of the common mode output current.

8. The method of claim 7, wherein computing the impedance of the speaker comprises using a function that comprises $Z_0={\left\{{\left[\frac{\mathrm{Vo}}{\mathrm{idiff}}-Z_L\right]}^{-1}-{\left[\frac{\mathrm{Vo}}{\mathrm{icomm}}-Z_L\right]}^{-1}\right\}}^{-1},$ where Z.sub.o is the impedance of the speaker, idiff is the amplitude of the differential output current, icomm is the amplitude of the common mode output current, Vo is an output voltage of the switching amplifier, and Z.sub.L is value of the inductance.

9. The method of claim 1, wherein amplitude and phase of the differential output current are measured without using an external shunt resistor.

10. The method of claim 1, further comprising switching the switching differential amplifier to a diagnostics mode before measuring amplitude and phase of the differential output current.

11. A switching differential amplifier comprising: a first differential output terminal configured to be coupled to a first terminal of a second circuit; a second differential output terminal configured to be coupled to a second terminal of the second circuit, the second circuit comprising a first inductor coupled to the first terminal, a second inductor coupled to the second terminal, a dumping network coupled between the first and second inductors, and a load coupled between the first and second inductors; and a measurement circuit configured to: measure amplitude and phase of a differential output current of the switching differential amplifier while the switching differential amplifier is driven in a differential mode with a differential input signal, measure amplitude and phase of a common mode output current of the switching differential amplifier while the switching differential amplifier is driven in a common mode with a common mode input signal, and compute an impedance of the load based on the amplitude and phase of the differential output current and the amplitude and phase of the common mode output current.

12. The switching differential amplifier of claim 11, wherein the measurement circuit is configured to operate when the switching differential amplifier is in a diagnostics mode.

13. The switching differential amplifier of claim 11, wherein the measurement circuit is further configured to compute an impedance of the dumping network based on the amplitude and phase of the differential output current and the amplitude and phase of the common mode output current.

14. The switching differential amplifier of claim 13, wherein computing the impedance of the dumping network comprises dividing an output voltage of the switching differential amplifier by the amplitude of the common mode output current.

15. The switching differential amplifier of claim 11, wherein the measurement circuit is further configured to: measure a value of the first and second inductors by measuring a ripple of the common mode output current; and compute the impedance of the load further based on the ripple of the common mode output current.

16. The switching differential amplifier of claim 11, wherein the measurement circuit comprises a state machine, an arithmetic unit, and a memory.

17. The switching differential amplifier of claim 11, further comprising: a third output terminal configured to be coupled to a third terminal of the second circuit; and a fourth output terminal configured to be coupled to a fourth terminal of the second circuit, wherein the first inductor is coupled between the first terminal and the third terminal, and the second inductor is coupled between the second terminal and the fourth terminal.

18. The switching differential amplifier of claim 11, wherein the dumping network comprises: a first capacitor coupled between the first inductor and a reference node; a second capacitor coupled between the second inductor and the reference node; a first resistor coupled between the first inductor and a reference node; and a second resistor coupled between the second inductor and a reference node.

19. The switching differential amplifier of claim 11, wherein the measurement circuit is configured to measure amplitude and phase of the differential output current while the switching differential amplifier is driven in a differential mode with a differential input signal having a frequency in an upper portion of an audio frequency; and measure amplitude and phase of a common mode output current of the switching differential amplifier while the switching differential amplifier is driven in a common mode with a common mode input signal having a frequency in the upper portion of an audio frequency.

20. The switching differential amplifier of claim 19, wherein the upper portion of the audio frequency comprises 20 kHz.

21. The switching differential amplifier of claim 11, further comprising the second circuit, wherein the load is a speaker having a tweeter, and wherein the measurement circuit is configured to measure amplitude and phase of the differential output current while the switching differential amplifier is driven in a differential mode with a differential input signal having a frequency covered by a tweeter of the speaker; and measure amplitude and phase of a common mode output current of the switching differential amplifier while the switching differential amplifier is driven in a common mode with a common mode input signal having a frequency covered by a tweeter of the speaker.

22. The switching differential amplifier of claim 21, wherein the speaker further comprises a woofer.

23. A class-D differential amplifier comprising: a first differential output terminal configured to be coupled to a first terminal of a second circuit; a second differential output terminal configured to be coupled to a second terminal of the second circuit, the second circuit comprising a first inductor coupled to the first terminal, a second inductor coupled to the second terminal, a dumping network coupled between the first and second inductors, and a load coupled between the first and second inductors; and a measurement circuit configured to: measure amplitude and phase of a differential output current of the class-D differential amplifier while the class-D differential amplifier is driven in a differential mode with a differential input signal, measure amplitude and phase of a common mode output current of the class-D differential amplifier while the class-D differential amplifier is driven in a common mode with a common mode input signal, measure a value of the first and second inductors by measuring a ripple of the common mode output current, and compute an impedance of the load based on the amplitude and phase of the differential output current, the amplitude and phase of the common mode output current, and the ripple of the common mode output current.

24. The class-D differential amplifier of claim 23, wherein the measurement circuit is further configured to compute an impedance of the dumping network based on the amplitude and phase of the differential output current and the amplitude and phase of the common mode output current.

25. The class-D differential amplifier of claim 23, wherein the measurement circuit comprises a state machine, an arithmetic unit, and a memory.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] One or more embodiments will be described, by way of example only, with reference to the annexed figures, wherein:

[0010] FIG. 1 is a schematic representation of an amplifier driving a load such as a 2-way speaker;

[0011] FIG. 2 is exemplary of one or more embodiments;

[0012] FIGS. 3a, 3b, 4a, 4b, 5a and 5b are exemplary of possible operation of one or more embodiments;

[0013] FIG. 6 is exemplary of one or more embodiments;

[0014] FIGS. 7a, 7b, 8a, 8b, 9a, 9b, 10a, 10b and 11 are exemplary of possible operation of one or more embodiments;

[0015] FIG. 12 is a flow chart exemplary of one or more embodiments; and

[0016] FIG. 13 is an exemplary block diagram of embodiments of a measuring system.

DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0017] In the ensuing description, one or more specific details are illustrated, aimed providing an in-depth understanding of examples of embodiments of the instant description. The embodiments may be obtained by one or more of the specific details or with other methods, components, materials, and so on. In other cases, known structures, materials or operations are not illustrated or described in detail so that certain aspects of embodiment will not be obscured.

[0018] Reference to an embodiment or one embodiment in the framework of the present description is intended to indicate a particular configuration, structure, characteristic described in relation to the embodiment is compliance in at least one embodiment. Hence, phrases such as in an embodiment or in one (or more) embodiments that may be present in one or more points in the present description do not necessarily refer to one and the same embodiment. Moreover, particular conformation, structures or characteristics as exemplified in connection with any of the figures may be combined in any other quite way in one or more embodiments as possibly exemplified in other figures.

[0019] The references used herein are provided merely for convenience and hence do not define the extent of protection or the scope of the embodiments.

[0020] FIG. 1 is a schematic representation of an amplifier A driving a load LS.

[0021] A multi-way audio loudspeaker system may be exemplary of such a load.

[0022] In the case of a two-way system as exemplified herein, such a load may include a first loudspeaker (or woofer) WF and a second loudspeaker (or tweeter) TW for reproducing the low portion and the high portion of the audio frequency range (notionally 20 Hz to 20 kHz), respectively. A wide variety of technologies (electrodynamic, piezoelectric, electrostatic, ribbon, and so on) may be applied in manufacturing the loudspeakers WF and TW.

[0023] In driving the load LS the amplifier A may feed the load with an output voltage V.sub.amp and an output current i.sub.amp. A so-called crossover network CN may partition the current i.sub.amp in a low-frequency component to the woofer WF and a high-frequency component to the tweeter TW, indicated i.sub.woofer and i.sub.tweeter respectively. A wide variety of crossover arrangements are known in the art. A crossover arrangement including an inductor L.sub.W and a capacitor CT coupled in series to the woofer WF and the tweeter T.sub.W is exemplified here for the sake of simplicity

[0024] In various applications the need may arise of discriminating (e.g., on the basis of V.sub.amp and/or i.sub.amp) between a normal tweeter condition, and an unconnected tweeter condition due to cabling being interrupted or the tweeter TW being inoperative, e.g., following the tweeter being blown by excessive power being fed thereto.

[0025] In various applications, performing such a task may take into account various factors. For example, in certain arrangements, the amplifier output current i.sub.amp may be measured (only) within the amplifier without relying on external shunt resistors. As another example, the parameters of the crossover network CN, e.g., the values for L.sub.W and C.sub.T in FIG. 1, may vary within a certain tolerance range (e.g., 10%).

[0026] Performing such a function in a conventional (linear) amplifier may rely on the fact that the output voltage and current from the amplifier A may equal the voltage and current applied to the load LS.

[0027] In a switching (e.g., class D) amplifier things may be different insofar as the presence of an LC demodulator filter at the amplifier output leads to a situation whereeven in the audio bandthe load voltage and current may not correspond to the output voltage/current from the amplifier A.

[0028] In fact the values for L and C may have a certain tolerance (e.g., 20%), which makes it hardly feasible to evaluate the load even in the presence of a precise current measurement.

[0029] For instance, even assuming that the load LS may be modeled simply as a resistor R, the following relationships may apply:

V.sub.load=V.sub.amp{(1/[1+jL/(R/2).sup.2LC]}

I.sub.R=V.sub.amp/(R/2){(1/[1+jL/(R/2).sup.2LC]}

where V.sub.load and V.sub.amp denote the voltage on the load and the amplifier output voltage, respectively, I.sub.R is the current through the load (resistor R), L and C are the inductance and capacitance value representative of the LC demodulator filter at the amplifier output, and is the pulsation, representative of the frequency f, with =2f.

[0030] In certain applications the values for L, C and the load impedance values may represent variables to be measured (for instance in order to equalize system frequency response).

[0031] U.S. Publication No. 2009/0097667 A1 discloses a method and circuit for testing an audio high-frequency loudspeaker which is a part of a loudspeaker system. Such a document discloses two implementations involving single-ended driving (by keeping one of the input terminals to a virtual DC level) and the use of feedback both before and after the LC demodulator filter). In the former case, the measurement result is inaccurate, even though more precise that in the case that no single-ended driving is adopted.

[0032] While effective, the arrangement disclosed in such prior art document may exhibit certain limitations in terms of the architectures to which such solution may apply.

[0033] For instance such an arrangement may be used in the case of a full bridge PWM (class D) amplifier using two distinct single-ended branches and feedback applied after the inductor filter, so that single-ended driving can be easily applied with the frequency response of the output voltage V.sub.out essentially load-independent.

[0034] Conversely, the solution disclosed in the document captioned in the foregoing may not be applicable in the case of a full-bridge PWM arrangement using a fully differential topology and feedback before the LC demodulator filter. In that case, single-ended driving of the output may be difficult to apply. Also, the frequency response of the output voltage V.sub.out may be load-dependent (not equalized).

[0035] In the diagram of FIG. 2, reference 10 denotes a (fully differential) amplifier arrangement including a basic switching (class D) amplifier circuit coupled with PWM (Pulse Width Modulation) modulators 12a, 12b with associated LC demodulator filters 14a, 16a and 14b, 16b having a common-mode (high impedance) sensing network 18a, 18b to provide a feedback signal to the amplifier 10 over a common mode feedback line 20.

[0036] Input to such a amplifier arrangement may be applied differentially (e.g., V.sub.in+, V.sub.in).

[0037] This may occur, as exemplified herein, via input resistors 22a, 22b, which may be used e.g., for systems DC-coupled with a voltage source.

[0038] In other arrangements, a high-pass voltage input may be resorted to: in that case the elements exemplified here as the resistors 22a, 22b may be an impedance including a RC network. In the case of e.g., integrated analog-to-digital converters (ADC's) the differential input may be via an array of current generators.

[0039] In one or more embodiments, impedances 24a, 24b and 26a, 26b may provide feedback towards the input of the amplifier core 10 via the impedances 24a, 24bfrom the outputs of the PWM modulators 12a, 12b, that is upstream or before the (inductor 14a, 14b of) the LC demodulator filters, and via the impedances 26a, 26bfrom the outputs voltages V.sub.out, V.sub.out+ applied (differentially) across the load L, that is downstream or after the LC demodulator filters.

[0040] Further details on such an arrangement may be derived e.g., from US 2007/0229155 A1.

[0041] The description provided in the foregoing may also apply to the basic switching (class D) amplifier circuit illustrated in FIG. 6, save thatin the case exemplified in FIG. 6feedback towards the input of the amplifier core 10 may be provided (only) via the impedances 24a, 24b, namely from the outputs of the PWM modulators 12a, 12b, that is upstream or before the LC demodulator filters.

[0042] One or more embodiments may involve the use of a measuring system MS configured to co-operate with a class D amplifier by sensing/measuring certain electrical parameters and calculating certain values as discussed in the following.

[0043] In one or more embodiments, the measuring system MS may be external to the amplifier.

[0044] In one or more embodiments the measuring system MS may be on-board the amplifier, that is at least partly incorporated to the amplifier.

[0045] In one or more embodiments, the measuring system MS may be configured for operating according to the flow-chart exemplified in FIG. 12.

[0046] A possible exemplary embodiment of such a measuring system will be discussed in the following in connection with FIG. 13.

[0047] One or more embodiments will make it possible to provide an effective AC load and LC network impedance measurement (also) for a class D amplifier system possibly having a no-single ended driving possibility: this may be the case of fully-differential system configurations wherein feedback is from upstream (namely from before) the LC demodulator filter.

[0048] In one or more embodiments, the common-mode control circuit may be resorted to support operation as discussed in the following in connections with the exemplary flow-chart of FIG. 12.

[0049] After a START step, in a step 1000 the AC output current magnitude and phase may be sensed (e.g., by the measuring system MSin a manner known per se) while driving the amplifier in default conditions, namely differentially via the inputs V.sub.in+, V.sub.in.

[0050] In a step 1002, the AC output current magnitude and phase may be sensed while AC driving the amplifier in a common mode fashion (that is by applying a same signal to both inputs V.sub.in+, V.sub.in).

[0051] If high frequency measurement/monitoring is contemplated (e.g., in the higher portion of the audio frequency range covered by a tweeter TW as exemplified in FIG. 1) AC driving may be in both cases (differential/common-mode) in the upper portion (high frequency) the audio frequency range e.g., at about 20 kHz, which may make such testing practically inaudible for the system user.

[0052] Subsequently (as exemplified by step 1004) a distinction may be made between cases where feedback after the LC filter may apply (see e.g., FIG. 2) from those cases where (standard) feedback before the LC filter may apply (see e.g., FIG. 6).

[0053] In the former case (FIG. 2) evaluation of the load current and its impedance may take place in a step 1006 already on the basis of the differential and common-mode AC driving of steps 1000 and 1002.

[0054] In the latter case (FIG. 6) the evaluation of the load current and its impedance may include a further step 1008a involving measuring the value of an inductance by a ripple current measurement, namely a higher frequency AC common mode measurement before the load impedance may be evaluated in a step 1008b.

[0055] In one or more embodiments, an evaluation of the load current and its impedance may be performed even in the most general case of feedback before the LC demodulator filter (e.g., FIG. 6).

[0056] In one or more embodiments, the impedance value of other passive filter components (e.g., a dumping network 30 coupled parallel to the capacitor 16a, 16b of the LC demodulator filter) may be obtained from the measurements.

[0057] In fact, such a dumping network 30 may include a parallel connection of a resistor and two capacitors. While the individual values of such components may not be determined, evaluating the overall impedance may by itself represent a satisfactory result.

[0058] In the following, possible operation in the case where feedback after the LC demodulation filter may apply (impedances 26a, 26b in FIG. 2) will be exemplified with reference of the sequence of steps 1000, 1002, 1004, 1006 of FIG. 12.

[0059] Then possible operation in the case where feedback before the LC demodulation filter may apply (impedances 24a, 24b in FIG. 6) will be exemplified with reference of the sequence of steps 1000, 1002, 1004, 1008a, 1008b of FIG. 12.

[0060] For the sake of simplicity, as exemplified in FIG. 3, the RC dumping network 30 cascaded to the LC demodulator filter 14a, 16a; 14b, 16b of the class D amplifier may be represented as a symmetrical network including two impedances Z.sub.c having an intermediate point set to ground, thus leading to an arrangement where one half bridge is equal to the other. Similarly, the load LS may be modeled in the form of two identical impedances Z.sub.o(LS=2.Z.sub.o).

[0061] Also, the (high impedance) common-mode feedback circuit 18a, 18b can be driven by an AC signal (in the place of a DC fixed signal), which is possible in the case of a fully differential architecture as exemplified in FIGS. 2 and 6. Also, one may assume that the output load voltage (differential and common mode) may be AC controlled and made known at, e.g., 20 kHz (high frequency within the context of the present invention, when applied to audio systems).

[0062] In these conditions, one may assume that the load LS is essentially constituted by the tweeter TW (see FIG. 1) in so far as the crossover network CN operates in such way that no appreciable current i.sub.woofer flows through the inductor L.sub.W and the woofer loudspeaker WF.

[0063] In one or more embodiments, proper feedback signal management (according to known principles) may render the output load voltage v.sub.O (differential and common-mode) AC controllable and known at a high frequency, thus reducing frequency response dependence on load impedance.

[0064] FIGS. 4 and 5 are exemplary of the arrangement of FIG. 2 being AC driven in a differential mode (FIG. 4) and in a common mode (FIG. 5), with differential driving being representative of a default driving mode of the amplifier between the input terminal V.sub.in+ and V.sub.in.

[0065] In the case of differential AC driving (see FIG. 4), the following relationship applies:

i.sub.diff/v.sub.o=1/Z.sub.c+1/Z.sub.o

where v.sub.o is the differential output voltage (+v.sub.o, v.sub.o), and i.sub.diff is output current sensed.

[0066] In the case of common-mode AC driving, the following relationship applies:

i.sub.comm/v.sub.o=1/Z.sub.c, that is Z.sub.c=v.sub.o/i.sub.comm

where v.sub.o is the differential output voltage (+v.sub.o, v.sub.o), and i.sub.comm is output current sensed.

[0067] By combining the two previous relationships, as derived during AC differential and common-mode driving (steps 1000 and 1002 in FIG. 12) the following relationship may be derived:

Z.sub.o=[(i.sub.diffi.sub.comm)/v.sub.o].sup.1.

[0068] In that way, both the load impedance Z.sub.o(specifically LS=2.Z.sub.o) and the impedance Z.sub.c of the dumping network 30 may be derived and the result(s) made available in a step 1010.

[0069] While complex number notation is avoided here for the sake of simplicity, it will be appreciated that the electrical parameters discussed above are complex parameters.

[0070] Sensing/measurement by the measuring system MS (involving voltmetric and amperometric modulesof a known typenot visible in the figures) will thus involve both magnitude (modulus) and phase (angle), this being the case also for the associated calculation steps.

[0071] FIGS. 7 to 11 are exemplary of corresponding measuring steps which may be performed in the case of a circuit arrangement as exemplified in FIG. 6, that is with feedback obtained via the impedances 24a, 24b (only) before, that is upstream, of the LC demodulator filter 14a, 16a; 14b, 16b.

[0072] In such a case the frequency response of the output voltage may be load-dependent. Even in the case of FIG. 6, however, the AC voltage upstream of the LC demodulator filter may be known both in the case of differential driving and in the case of common mode driving.

[0073] As schematically represented in FIG. 7, the same symmetrical modeling of the dumping network 30 and the load LS may be adopted e.g., by modeling the dumping network and the load as two symmetric impedances Z.sub.c and Z.sub.o.

[0074] Also, for the sake of the following discussion, the inductor/inductance 14a, 14b of the LC demodulator filter may be assumed to have an impedance value Z.sub.L=jL+R.sub.L, including a (small) resistive component R.sub.L. This resistive component may be in the range of a few tenth mOhm, and may thus be neglected or assumed to have a (very small) constant value.

[0075] Again, FIG. 8 is exemplary of differentially driving (default driving) the amplifier arrangement of FIG. 6 while sensing the output current i.sub.diff (both in magnitude and phase).

[0076] Under these conditions (step 1000 of FIG. 12) the following relationship will apply:

v.sub.o/i.sub.diff=Z.sub.L+Z.sub.cZ.sub.o,

that is

[(1/Z.sub.c)+(1/Z.sub.o)].sup.1=(v.sub.o/i.sub.diff)Z.sub.L

where Z.sub.cZ.sub.o denotes the parallel connection of Z.sub.C and Z.sub.o.

[0077] Then (see e.g., step 1002 in FIG. 12) the amplifier arrangement of FIG. 6 may be driven in a common mode while sensing the output current i.sub.comm. The following relationship will thus apply:

v.sub.o/i.sub.comm=Z.sub.L+Z.sub.c, that is Z.sub.c=v.sub.o/i.sub.commZ.sub.L.

[0078] In the case of the arrangement of FIG. 6, the inductive value (that is, L) of the inductor 14a, 14b of the LC demodulator filter may be largely dominant at a high frequency (e.g., 20 kHz) and may be evaluated with an AC frequency measurement. In one or more embodiments this may turn out to be somehow critical in view of the PWM limited high frequency response.

[0079] In one or more embodiments, this issuenamely measuring Lmay be addressed by noting that in a switching (e.g., class D) amplifier driven with in-phase modulation, the current ripple at steady state, with no input signal (duty cycle set to 50%) may admit a first order approximation as a triangular wave having an amplitude I.sub.R which is (approximately) equal to:

VCC/8.Math.L.Math.f.sub.PWM

where, as schematically shown in the upper portion of FIG. 11, VCC indicates the PWM output voltage (e.g., the output voltage of the PWM modulators 12a, 12b driven with 50% duty cycle) at a frequency f.sub.PWM.

[0080] Under these circumstances, as schematically shown in the lower portion of FIG. 11, the inductance L may be measured by sampling the difference I between the values taken of the current entering the PWM power stages at two different instants separated by a time t, namely:

L=(VCC.Math.t)/(8.Math.I),

so that the value for Z.sub.L=jL+R.sub.L may be calculated (under the assumptions on R.sub.L recalled previously).

[0081] Such a measurement may be performed, e.g., in the step exemplified as 1008a in FIG. 12.

[0082] Consequently, the various measurements performed in the foregoing may be combined (e.g., in a step 1008b) as exemplified in the following, so that both Z.sub.c and Z.sub.o may be evaluated as

Z.sub.c=v.sub.o/i.sub.commZ.sub.L and

Z.sub.o={[(v.sub.o/i.sub.diff)Z.sub.L)].sup.1[(v.sub.o/i.sub.comm)Z.sub.L)]}.sup.1

[0083] In that way, both the load impedance Z.sub.o(specifically LS=2.Z.sub.o) and the impedance Z.sub.c of the dumping network 30 may be derived and the result(s) may be made available in a step 1010.

[0084] While complex number notation is again avoided here for the sake of simplicity, it will once more be appreciated that the electrical parameters discussed above are complex parameters. Sensing/measurement by the measuring system MS (involving voltmetric and amperometric modulesof a known typenot visible in the figures) will thus involve both magnitude (modulus) and phase (angle), this being the case also for the associated calculation steps.

[0085] In more embodiments, a measuring system MS as exemplified e.g., in FIGS. 2 and 6 may be implemented according to the general layout exemplified by the block diagram of FIG. 13, where the following block designations apply: [0086] 10Class D amplifier core (see also FIGS. 2 and 6) [0087] 100State machine (main blocks handling) [0088] 102analog-to-digital converter (ADC) of the current i(t) from the amplifier [0089] 104sinusoid (sine-wave) generator [0090] 106digital-to-analog converter (DAC) of the sine-wave to the amplifier 10 [0091] 108magnitude (i.sub.mag) and phase (i.sub.phase) evaluation block [0092] 110state machine for impedance evaluation based on i.sub.mag and i.sub.phase [0093] 112arithmetic unit (including e.g., a multiplier X*Y, an inverse calculation module 1/X and a magnitude-phase<->real(Re)-imaginary(Im) conversion module). [0094] 114memory

[0095] In one or more embodiments, various options may be adopted for signal sensing/detection.

[0096] For instance, since a switching (class D) amplifier is involved one may rely on the on resistance RdsON of the power stage: the voltage Vds may be ADC converted when the power stage is on, so that VdsRdsON*Iload, by neglecting minor secondary effects.

[0097] Another option may be direct measurement of Iout by Hall effect in the silicon.

[0098] Sensing external to the amplifier may rely on the series resistance of the inductor by accessing the voltage before and after the filter.

[0099] Arranging a sensing resistor Rsens in series to the load LS may reduce load swing and increase power dissipation, the same applying also to other current probes measuring the current on the load, which would result in a more complex system (including external components and/or additional pins in the IC).

[0100] In a possible (real world) arrangement, the following values may be assumed: [0101] inductors 14a, 14b of the LC demodulator filter: L=10 H10% [0102] capacitors 16a, 16b of the LC demodulator filter: C=3.3 F10% [0103] resistor of the dumping network 30: R.sub.d=10 Ohm [0104] capacitor of the dumping network 30: C.sub.d=1 F [0105] nominal (normal) load=4 ohm [0106] open load=4 Ohm+600 H=75 Ohm 87 deg, when measurement is performed at 20 kHz.

[0107] In a standard method, where only current magnitude is measured, values for the load current may be, e.g., i=0.88 A2% and i=0.96 A30%, in normal load condition and open load condition, respectively.

[0108] The tolerances indicated may only consider a component maximum variation without considering measurement error. It will be appreciated that, under such conditions, open load may not be detectable and might be done only to LC filter tolerance. For instance, at e.g., 20 kHz with connected load, L and C at nominal value (no error), a current measured may be e.g., 0.88A. A same measured value may be obtained with a disconnected tweeter and a lower filter capacitance C: when the tweeter is disconnected, the load measurement may be largely influenced by the values of L and C which may vary extensively forma production item to another.

[0109] By using the solution of one or more embodiments, under the conditions outlined, and by assuming an even more critical situation (that is by assuming measurements of i and v affected by a 10% tolerance/error) a normal load having a module |Z|=4 Ohm15% and a phase=09 degrees may be detected.

[0110] Similarly, in an open load condition, a module |Z|=75 Ohm16% and a phase=872 degrees may be detected by assuming as a tolerance 10% error on current and inductance (this being likely to represent a worst case correlation).

[0111] Both typical values for L and C may be considered in a measurement which is not dependent on L and C tolerances as these are included in the quantities measured: that is, no a priori knowledge of L and C is needed since these may be measured in one or more embodiments. Open load conditions may thus be detected while also providing impedance evaluation.

[0112] As indicated, in one or more embodiments, the measurement system MS may be incorporated to the amplifier circuitry, with the whole system adapted to be switched to diagnostic mode. In response to being switched to such diagnostic mode, the amplifier will be configured in order to operate, in addition to an otherwise standard AC differential mode (default modestep 1000 in FIG. 12) to an AC common mode in order to perform the step indicated as 1002 in FIG. 12.

[0113] Without prejudice to the underlying principles, the details and the embodiments may vary, even significant, with respect to what has been disclosed by way of example only in the foregoing, without departing from the extent of protection.

[0114] The extent of protection is defined by the annexed claims.