HIGH PERFORMANCE AUDIO AMPLIFIER
The power supply circuit with a MOSFET transistor and associated sub-circuits addresses signal distortions and efficiency issues in high-power audio amplifiers, enhancing sound quality and efficiency.
Patent Information
- Application Number
- FR2022000839
- Authority / Receiving Office
- FR · FR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-01-31
- Publication Date
- 2025-07-25
- Estimated Expiration
- 2042-01-31
AI Technical Summary
High-power audio amplifiers suffer from signal distortions and efficiency issues, particularly in Class G and H amplifiers, due to insufficient supply voltage and voltage drops affecting MOSFET transistors.
A power supply circuit with a MOSFET transistor controlled by a charging assistance, discharging assistance, and voltage shift sub-circuit to manage power supply buses, reducing distortions and improving efficiency.
The solution significantly reduces signal distortions and enhances amplifier efficiency, particularly for sinusoidal inputs across the sound spectrum, improving sound reproduction quality.
Smart Images

Figure 00000025_0000 
Figure 00000025_0001 
Figure 00000026_0000
Abstract
Description
Title of the invention: HIGH PERFORMANCE AUDIO AMPLIFIER Technical field
[0001] The invention relates to the field of amplifiers. The invention relates in particular to a high-power audio amplifier intended to control at least one loudspeaker.
[0002] High-power amplifiers find applications in particular for sound reinforcement in recording studios or concert halls. They are used in particular to power electrodynamic loudspeakers.
[0003] The invention advantageously makes it possible to obtain a high-power audio amplifier with better efficiency than the amplifiers of the prior art, while guaranteeing low distortion. prior art
[0004] Conventionally, a high-power amplifier comprises at least one active component such as a transistor or a tube making it possible to amplify the power of a signal received at the input, while retaining the shape of the input signal.
[0005] The circuit of a high-power amplifier generally comprises a pre-amplification stage followed by a power amplification stage. More precisely, the output of the power amplification stage is counter-reacted to the pre-amplification stage so that the pre-amplification stage detects, over time, the difference between the output signal and the input signal, for example by means of differential pairs. This difference detected at the pre-amplification stage is then amplified in the power amplification stage to form the output signal, conventionally transmitted to at least one loudspeaker.
[0006] Thus, by way of example, [Fig.l] illustrates a high-power amplifier 100 whose input signal is applied to point SI. The output of the high-power amplifier 100 feeds a loudspeaker, represented by a resistor R44, connected between a point 3 and ground.
[0007] The diagram of the high-power amplifier 100 is symmetrical, it comprises an upper part 140a, which amplifies the positive alternation of the input signal SI and a lower part 140b, which amplifies the negative alternation of the input signal SI.
[0008] In the example of [Fig.l], the high-power amplifier 100 is powered by a single voltage level of + / - 65 V.
[0009] The pre-amplification stage 201 comprises two differential pairs 110a, 110b each comprising two transistors Q1, Q2 and Q3, Q4 mounted in mirror image to one of the other. Thus, the emitter of transistors Q1, Q2 is connected to the -65V power supply bus via a resistor R2, R3 and a first constant current source I, while the emitter of transistors Q3, Q4 is connected to the +65V power supply bus via a resistor R4, R5 and a second current source of the same value I.
[0010] The collector of transistors Q2, Q4 is respectively connected to the +65V and -65V supply bus, while the collector of transistors Q1, Q3 is respectively connected to the +65V and -65V supply bus via a resistor RI, R11. In the absence of feedback, the gain of the pre-amplification stage 201 depends on the ratio of resistors R1 / R2 and R11 / R4.
[0011] The bases of transistors Q1 and Q3 and the bases of transistors Q2 and Q4 are connected to each other. The bases of transistors Q1 and Q3 are also supplied by the input signal SI. The bases of transistors Q2 and Q4 are connected to loudspeaker R44 via a resistor R18, so as to form the feedback applied to the pre-amplification stage. The collectors of transistors Q1 and Q3 ensure the coupling of the pre-amplification stage 201 with the power amplification stage 202.
[0012] The power amplification stage 202 comprises two transistors Q8, Q9 connected by their respective bases to the pre-amplification stage 201 via a resistor R23, R57. A voltage source T1 is also arranged between the resistors R23 and R57 so as to bias the transistors Q8, Q9 in class AB. This voltage source T1 makes it possible to supply a voltage equal to the sum of the voltages necessary to turn on the transistors Q8 and Q9.
[0013] Transistors Q8, Q9 are also connected respectively to the +65 V and - 65 V power supply buses by their collector. The emitter of transistors Q8, Q9 is connected to speaker R44 via resistors R16, R17. These resistors R16 and R17 are added in order to control the quiescent current, i.e. the value of the current conducted by the amplifier when it receives no input signal. Without these resistors R16, R17, the quiescent current would be dependent on the characteristics of transistors Q8 and Q9 and the temperature, which is prohibitive.
[0014] Furthermore, the circuit of [Fig.l] has a particularity at the level of the bases of the transistors Q2 and Q4. Indeed, the latter are connected to a protection line 403 comprising a resistor R28 connected in series with a capacitor C4 connected to ground. This assembly is a voltage divider such that the voltage at point S2 is equal to R28 / (R18 + R28) multiplied by the voltage of the output signal 3 in alternating current. In practice, the capacitor C4 behaves like a short circuit when the voltage flowing through the circuit is alternating current. On the other hand, when the voltage flowing through the circuit is direct current, the capacitor C4 behaves like an open circuit. In this case, the output signal 3 of the high-power amplifier 100 is directly connected to point S2. The voltage gain of the high-power amplifier 100 is then equal to 1, which makes it possible to limit any undesirable DC component on the voltage applied to the terminals of the loudspeaker R44.
[0015] In this type of amplifier, the voltage measured with respect to ground at point SI and the voltage measured with respect to ground at point S2 are equal. In other words, the gain of the high-power amplifier 100 is equal to the ratio of the values of the resistors (R18+R28) / R28.
[0016] [Fig.l] therefore illustrates a high-power voltage amplifier. There is also another category, high-power current amplifiers, as illustrated in [Fig.2].
[0017] This category of high-power current amplifiers 101 has the same topology as for the high-power voltage amplifiers 100, namely a pre-amplification stage 201 coupled to an amplification stage 203.
[0018] Unlike the high-voltage amplifier 100 of [Fig.l], a current measuring resistor R6 is inserted between the loudspeaker R44 and ground. Furthermore, the current amplifier 101 does not have a protection line 403. The bases of the transistors Q2 and Q4 are connected to an interconnection point PI located between the resistor R6 and the loudspeaker R44.
[0019] In this configuration, the loudspeaker R44 is therefore crossed by an image of the current applied to the point S2 in [Fig.3]. It follows that the transconductance, that is to say the ratio between the output current passing through the loudspeaker R44 and the input voltage applied to the point SI, of the amplifier 101 is equal to 1 / R6 A / V. For a loudspeaker of impedance Z, the voltage gain is Z / R6.
[0020] The voltage or current amplification assemblies are therefore similar, particularly at the level of the pre-amplification stage and the power amplification stage. They only differ at the level of the loudspeaker connection and the feedback.
[0021] In the remainder of the state of the art, the other classes of amplifiers will be described with reference to voltage amplifiers, although the invention is not limited to this type of amplifier.
[0022] A class system can be used to characterize the different topologies of high-power amplifiers. The class system assigns a letter based on the relationship between the shape of the input and output signals, as well as the duration for which the active components are used when amplifying the input signal.
[0023] Among the existing classes of amplifiers, class A amplifiers have a topology such that the active components conduct, in the absence of input signal, a high current of approximately 50% of the maximum output current. This is called the quiescent current. In modulation, the latter is superimposed on the output current. This type of amplifier can provide excellent sound quality, but it has the disadvantage of generating significant heat dissipation. Thus, the energy efficiency, defined by the ratio between the effective output power and the absorbed power, of this class of amplifier is approximately 10%.
[0024] Class B amplifiers have a topology such that the active components conduct over 50% of the input signal period when a sinusoidal input signal is applied at the input. For this class of amplifiers, the quiescent current is zero. The efficiency of this type of amplifier is higher than class A amplifiers, but the distortion characteristics of the output signal are significantly degraded compared to class A amplifiers. Class B amplifiers therefore produce lower quality sound. It is now quite rare for manufacturers to use this class of amplifier.
[0025] Class AB amplifiers have a topology such that the active components conduct over 100% of the input signal period but with a low quiescent current of approximately 1% of the maximum output current. This type of amplifier has higher energy efficiency than class A amplifiers, typically between 30 and 50%, but lower sound quality. Class AB amplifiers are therefore a good compromise between performance and energy efficiency.
[0026] For Class D amplifiers, a technology is used where the active components are operated like switches. The signal is then converted into pulse width modulation. This system increases energy efficiency to about 70%. On the other hand, the output signal contains more noise and distortion and it is difficult to reproduce high frequencies with this class of amplifiers.
[0027] Class G amplifiers have multiple power buses and can switch from one to the other depending on the output power required. This increases energy efficiency by reducing the power dissipated in the active components.
[0028] Class H amplifiers use a power supply bus whose supply voltage "follows", or is modulated by the input signal. Typically, they have two power supply buses, like Class G amplifiers, but only the higher supply voltage is modulated. The modulated supply is typically achieved using a Class D amplifier.
[0029] The subject of the invention relates in particular to these last two categories of amplifiers.
[0030] [Fig. 3] illustrates the upper part of a power amplification stage 204. The latter is connected to a power supply circuit 150. Of course, this power amplification stage also comprises a lower part not shown, mirroring the upper part with an equivalent power supply circuit. Similarly, the circuit also comprises a pre-amplification stage, for example connected to an independent power supply or to a strong power supply bus, also connected to the power supply circuit 150.
[0031] The power supply circuit 150 makes it possible to select a power supply level for the power amplification stage 204 adapted to the required amplification power. Thus, when the amplification voltage is greater than a threshold value, the first strong power supply bus must be used whereas, when the amplification voltage is less than this threshold value, a second weaker power supply bus can be used. The phases of use of the weaker power supply bus make it possible to improve the overall efficiency of the amplifier in comparison with class A, B, and AB amplifiers.
[0032] Furthermore, the power supply circuit 150 has a structure independent of the pre-amplification stage 201 and the power amplification stage 204. In the example of [Fig. 3], the power amplification stage 204 comprises a transistor Q10 whose base is connected to the emitter of the transistor Q8. This so-called “Darlington” configuration makes it possible to increase the current gain. The emitters of the transistors Q8 and Q10 are coupled to the loudspeaker R44 by their respective resistors R16 and R19. They have, for example, a respective quiescent current equal to 6 mA for the transistor Q8 and equal to 75 mA for the transistor Q10.
[0033] The power amplification stage 204 is connected to the power supply circuit 150 at the collectors of the transistors Q8 and Q10. This power supply circuit 150 is connected to two voltage buses having distinct levels V+, V++. The power supply circuit 150 therefore makes it possible to select one or the other of these voltage levels depending on the amplification required. Typically, a first power supply bus delivers 65V and a second power supply bus delivers 35V. The second power supply bus is intended to be used to power the power amplification stage 204 when the output signal to be generated does not have a very high voltage, typically less than 32V.
[0034] Due to the relatively low current flowing through transistor Q8, typically less than 10% of the current flowing through transistor Q10, the collector of transistor Q8 can be directly connected to the first power supply bus V++, without this causing significant additional power dissipation. This embodiment can improve the stability of the amplifier, which operation as a current amplifier tends to compromise.
[0035] In order to carry out the selection of the appropriate voltage level, a MOSFET transistor M1 is directly connected to the first power supply bus V++ by its drain and to the second power supply bus V+ by its source via a fourth diode D3. The MOSFET transistor M1 typically switches for a threshold of 27 V, then it operates linearly beyond this threshold. It lets the first power supply bus V++ pass when a control voltage greater than a threshold value is applied to it. This voltage is controlled by the association of the first diode D8, D10 connected between the gate of the MOSFET transistor M1 and the loudspeaker R44 and a second resistor R8, R27 connected between the drain and the gate of the MOSFET transistor M1.
[0036] A diode D15 is connected between the collector of transistor Q10 and speaker R44, the cathode of the diode being connected to the collector of transistor Q10.
[0037] With such an assembly, the different signals obtained are illustrated in [Fig.4].
[0038] Thus, the signal numbered 1 in Figures 3 and 4 represents the grid signal of the MOSFET transistor Ml. The output signal numbered 3 in Figures 3 and 4 represents the output signal of the amplifier, i.e. the signal at the terminals of the loudspeaker R44. The output signal numbered 2 in Figures 3 and 4 represents the output signal of the power supply circuit 150. We can thus see that the output signal 3 is distorted. Indeed, the peak of the sinusoid is flattened and dropouts occur around 27ps. This saturation of the output signal 3 is explained by the fact that the supply voltage of the MOSFET transistor Ml is insufficient for the amplifier to be able to deliver the output voltage correctly.
[0039] Similarly, signal 2 exhibits, between 40 and 47ps, an abnormal voltage peak corresponding to an overvoltage at the gate of the MOSFET transistor M1. This leads to a loss of power and a degradation of the amplifier's efficiency.
[0040] The technical problem that the invention aims to solve is therefore to obtain a high-power audio amplifier making it possible to limit the distortions identified on the signals and therefore to improve the efficiency and reduce the saturation of the amplifier. Statement of the invention
[0041] To solve this problem, the invention proposes a power supply circuit comprising a MOSFET transistor controlled by a charging assistance sub-circuit, a discharging assistance sub-circuit and a voltage offset sub-circuit making it possible to obtain better efficiency from the amplifier while limiting saturations and distortions of the amplifier output signal.
[0042] In other words, the invention relates to a high-power audio amplifier intended to control at least one loudspeaker, said amplifier comprising: - a pre-amplification stage receiving an input signal; - a power amplification stage connected to the pre-amplification stage and providing an output signal intended to power said at least one loudspeaker; the pre-amplification and power amplification stages comprising an upper part and a lower part mounted in mirror; - a feedback providing the pre-amplification stage with an image of the output signal, - an upper power supply circuit, connected to the upper part of the power amplification stage, and allowing it to be powered by a first or a second power supply bus; - a lower power supply circuit connected to the lower part of the power amplification stage and allowing it to be powered by a first or a second power supply bus; - each power supply circuit comprising a MOSFET transistor and supervision means, the MOSFET transistor being controlled by the supervision means so as to carry out switching between one or the other of the two power supply buses, the MOSFET transistor being connected to the second power supply bus via a fourth diode, a first terminal of which is connected to the source of the MOSFET transistor, and the drain of the MOSFET transistor being connected to the first power supply bus.
[0043] The invention is characterized in that each power supply circuit further comprises - a charging assistance sub-circuit of said MOSFET transistor comprising at least a first resistor, a first terminal of the first resistor being connected to the gate of the MOSFET transistor and a second terminal of the first resistor being connected to an interconnection point; - a sub-circuit for assisting the discharge of said MOSFET transistor comprising at least a second and a third resistor and a bipolar transistor; the base of the bipolar transistor being connected to a first terminal of the third resistor, its emitter being connected to the gate of the MOSFET transistor and its collector being connected to the source of the MOSFET transistor via the second resistor, the second terminal of the third resistor being connected to the interconnection point; and - a voltage shift subcircuit comprising a first diode connected in parallel with a first capacitor; a first terminal of the first diode and a first terminal of the first capacitor being connected to a first interconnection node; a second terminal of the first diode and a second terminal of the first capacitor being connected to a second interconnection node.
[0044] In particular, the charging assistance sub-circuit allows the MOSFET transistor to charge more quickly. Indeed, it first comprises the first resistor, which preferably has a high value resistor so as not to exchange too much energy with the amplifier output and generate more distortion, i.e. approximately 15 kΩ. This first resistor only allows a low current to charge or discharge the gate of the MOSFET transistor. Since the gate behaves like a capacitor, the combination of the MOSFET transistor with the first resistor increases the duration of the charge and discharge times.
[0045] The discharge assistance sub-circuit allows the MOSFET transistor to discharge more quickly. The bipolar transistor allows in particular to discharge the gate of the MOSFET transistor when the sinusoidal input signal is in its decreasing phase.
[0046] The voltage shift sub-circuit ensures that the potential of the gate of the MOSFET transistor is always higher than that of the amplifier output. Typically, the potential of the gate of the MOSFET transistor can be 15 V higher than that of the amplifier output. The voltage shift sub-circuit also compensates for voltage drops of the MOSFET transistor.
[0047] According to a second embodiment, the voltage shift sub-circuit further comprises at least a fourth resistor and a second diode, connected in parallel. A first terminal of the second diode and a terminal of the fourth resistor are connected to the first interconnection node, a second terminal of the fourth resistor and a second terminal of the second diode being connected to a third interconnection node.
[0048] The addition of these components makes it possible to reduce the distortions observed on the amplifier output signal. This improvement in distortion is effective for a sinusoidal input signal, if it is of relatively low frequency, i.e. approximately 1 kHz.
[0049] Advantageously, according to a third embodiment, the charging assistance sub-circuit further comprises a fifth resistor connected in series with a third diode, the fifth resistor and the third diode being connected in parallel with the branch of the charging assistance sub-circuit including the first resistor.
[0050] The third diode, connected in parallel with the first resistor, allows the current intended to charge the gate of the MOSFET transistor to pass and blocks the current intended to discharge the gate of the MOSFET transistor. The fifth resistor is preferably of low value, i.e. approximately 300 Q. This fifth resistor allows the gate of the MOSFET transistor to be charged much more quickly, since the charging time constant is equal to the product of the resistance by the capacitance.
[0051] The addition of these components again makes it possible to improve the distortion for a sinusoidal input signal up to a frequency of 20kHz. The distortion is therefore improved across the entire sound spectrum. The sound reproduction of a loudspeaker connected to the amplifier of the invention is therefore improved. The listener perceives less distortion compared to amplifiers of the prior art.
[0052] According to a fourth embodiment, the voltage shift sub-circuit further comprises a second capacitor and a third capacitor, the second capacitor being connected in parallel with the first capacitor and the first diode and the third capacitor being connected in parallel with the third diode and the fifth resistor. These additional components make it possible to limit disturbances, i.e. parasitic signals superimposed on the expected output signal.
[0053] In practice, each power supply circuit comprises a first protection diode, a first terminal of which is connected to the source of the MOSFET transistor and a second terminal of which is connected to the gate of the MOSFET transistor. Similarly, each power supply circuit further comprises a second protection diode connected between the source and the drain of the MOSFET transistor.
[0054] The first protection diode is added in order to protect the MOSFET transistor from overvoltages on its gate, which could damage it or even make it unusable, by breaking the insulation between the gate and the channel which can only support + / - 20V continuously and + / - 30V transiently.
[0055] The second protection diode serves to protect the MOSFET transistor from a reverse drain-source voltage that could occur if the voltage of the first power bus only appears after that of the second power bus. The different power buses each have their own transformer windings and smoothing capacitors, hence different time constants.
[0056] According to a fifth embodiment, each power supply circuit further comprises a capacitor connected in parallel with the fourth diode. This capacitor makes it possible to eliminate the parasitic peaks generated by the diode, a first terminal of which is connected to the source of the MOSFET transistor, when the latter switches. The harmonic distortion rate (THD) is also improved. This rate is a measure of the linearity of the processing carried out. It is calculated by comparing the output signal of a device to a perfectly sinusoidal input signal.
[0057] In practice, the pre-amplification stage is connected to the first power supply bus of each power supply circuit by means of a circuit for damping the power variations of said first power supply bus, said circuit for damping the power variations comprising at least one capacitor and at least one resistor mounted as a low-pass filter. In other words, the resistor is connected in series with the capacitor, which is itself connected to ground.
[0058] This assembly makes it possible to isolate noise and voltage dips. This phenomenon occurs in particular when the high-voltage amplifier delivers a high current. The The capacitor then plays the role of an energy reservoir.
[0059] Advantageously, said power variation damping circuit further comprises an additional capacitor mounted in parallel with the at least one capacitor. The capacitor mounted in parallel has a more modest value, typically lower by a factor of 103. It makes it possible to suppress high-frequency interference.
[0060] According to a specific embodiment of the invention, the feedback applied to the pre-amplification stage provides a signal proportional to the current flowing through the loudspeaker. This embodiment makes it possible to obtain a high-power current amplifier. As described with reference to [Fig.2] of the prior art, this category of high-power current amplifiers has the same topology as for high-power voltage amplifiers, namely a pre-amplification stage coupled to an amplification stage. However, the feedback applied to the pre-amplification stage differs.
[0061] By using the power supply circuit of the invention for a high-power current amplifier, the high-power current amplifier has very limited consumption.
[0062] Indeed, when the amplifier delivers a low voltage at output, the power supply circuit of the invention will not be active. Thus, the power supplied by the power supply will be equal to the product of the output current multiplied by the voltage supplied by the lowest power supply bus, instead of the maximum voltage supplied by the single power supply bus for a standard class AB amplifier. At equal power, and assuming for example that the voltage supplied by the lowest power supply bus is equal to half that of the highest power supply bus, half as much power will be supplied by the power supply. The transistors, which dissipate the difference between the power supplied by the power supply and the power supplied to the loudspeaker, will heat up significantly less. Brief description of the figures
[0063] The manner of carrying out the invention, as well as the advantages which result therefrom, will emerge clearly from the description of the embodiments which follow, with the support of the appended figures in which:
[0064] [Fig. 1] is an electrical diagram of a prior art high-power voltage amplifier,
[0065] [Fig.2] is an electrical diagram of a high-power amplifier in current of prior art,
[0066] [Fig.3] is an electrical diagram of a high-power amplifier in voltage of the prior art comprising a power supply stage,
[0067] [Fig.4] is a graph representing the evolution of the grid voltage of the MOSFET transistor, the amplifier output voltage, and the output signal of the power supply circuit for the circuit in [Fig.3].
[0068] [Fig.5] is an electrical diagram of the high-power voltage amplifier according to an embodiment of the invention,
[0069] [Fig.6] is an electrical diagram of the power supply circuit of the high- power according to a second embodiment,
[0070] [Fig.7] is a graph representing the evolution of the grid voltage of the MOSFET transistor, the output voltage of the amplifier, and the output signal of the power supply circuit for the circuit of [Fig.6] with a sinusoid of frequency 20kHz as input,
[0071] [Fig.8] is a graph representing the evolution of the gate voltage of the MOSFET transistor, the output voltage of the amplifier, and the output signal of the power supply circuit for the circuit of [Fig.6] with a sinusoid of frequency 1kHz as input,
[0072] [Fig.9] is an electrical diagram of the power supply circuit of the high- power according to a third embodiment,
[0073] [Fig. 10] is a graph representing the evolution of the MOSFET transistor gate voltage, the amplifier output voltage, and the power supply circuit output signal for the circuit of [Fig.9] with a 20kHz frequency sinusoid as input,
[0074] [Fig. 11] is a graph representing the evolution of the MOSFET transistor gate voltage, the amplifier output voltage, and the output signal of the power supply circuit for the circuit of [Fig.9] with a sinusoid of frequency 1kHz as input,
[0075] [Fig. 12] is an electrical diagram of the power supply circuit of the high-power amplifier according to a fourth embodiment,
[0076] [Fig. 13] is a graph representing the evolution of the MOSFET transistor gate voltage, the amplifier output voltage, and the power supply circuit output signal for the circuit of [Fig. 12] with a 20kHz frequency sinusoid as input,
[0077] [Fig. 14] is an electrical diagram of the power supply circuit of the high-power amplifier according to a fifth embodiment,
[0078] [Fig. 15] is a graph representing the evolution of the MOSFET transistor gate voltage, the amplifier output voltage, and the power supply circuit output signal for the circuit of [Fig. 14] with a 20kHz frequency sinusoid as input,
[0079] [Fig. 16] is an electrical diagram of the power supply circuit of the high- power according to the first embodiment of [Fig.5],
[0080] [Fig. 17] is a graph representing the evolution of the MOSFET transistor gate voltage, the amplifier output voltage, and the power supply circuit output signal for the circuit of [Fig. 16] with a 20kHz frequency sinusoid as input,
[0081] [Fig. 18] is an electrical diagram of the high-power current amplifier according to an embodiment of the invention, and
[0082] [Fig. 19] is an electrical diagram of a protection structure of a high-power current amplifier according to an embodiment of the invention. Detailed description of the embodiments
[0083] As illustrated in [Fig.5], the high-power amplifier 102 of the invention is symmetrical, it comprises an upper part, which amplifies the positive alternation of the input signal SI and a lower part, which amplifies the negative alternation of the input signal SI.
[0084] The high-power amplifier 102 comprises a pre-amplification stage receiving the input signal SI and supplying signals from the collector of the transistors Q5, Q6, via a network consisting of the resistor R23, R57, connected in parallel with a capacitor C7, C8. These signals are supplied to a power amplification stage 302 based on the transistor Q8, Q9. The latter supplies an output signal 3 intended to power a loudspeaker R44. A feedback provides the pre-amplification stage 201 with an image of the output signal 3.
[0085] The high-power amplifier 102 also includes an upper power supply circuit 155a, connected to the upper portion of the power amplifier stage 302, and a lower power supply circuit 155b connected to the lower portion of the power amplifier stage 302.
[0086] The pre-amplification stage 301 comprises two differential pairs 110a, 110b each comprising two transistors Q1, Q2 and Q3, Q4 mounted in mirror image of each other. Thus, the emitter of the transistors Q1, Q2 is connected to the -65V supply bus via a resistor R2, R3 and a first constant current source II, while the emitter of the transistors Q3, Q4 is connected to the +65V supply bus via a resistor R4, R5 and a second current source of the same value II.
[0087] The current sources II and I2 each comprise a transistor Q23, Q24 whose collector is respectively connected to the resistors R2, R3 and R4, R5. The emitter of the transistors Q23, Q24 is connected to a resistor R6, RIO. The second terminal of the resistor R6 is connected on the one hand to the -65V supply bus and on the other hand to ground, via a capacitor C11 connected in series with a resistor R46. The second terminal of the resistor RIO is connected on the one hand to the bus power supply at + 65V and on the other hand to ground, via a capacitor C12 mounted in series with a resistor R48.
[0088] A diode Dl, D2 is mounted between the base of the transistor Q23, Q24 and the second terminal of the resistor R6, RIO, while a resistor R45, R47 is mounted between the base of the transistor Q23, Q24 and the interconnection point between the capacitor Cil, C12 and the resistor R46, R48.
[0089] The collector of transistors Q2, Q4 may be directly connected, respectively, to the +65V and -65V power supply bus. Alternatively, the collector of transistors Q2, Q4 may be connected, respectively, to the +65V and -65V power supply bus via a power variation damping circuit 304, 305.
[0090] The power variation damping circuit 304, 305 comprises a resistor R52, R55 connected in series with a capacitor C15, C19 connected to ground. The second terminal of the resistor R52, R55 is connected to the + / -65V power supply bus. Advantageously, another capacitor C16, C20 can be connected in parallel with the capacitor C15, C19. In practice, the capacitor C15, C19 has a value between 150 and 300 pF, while the capacitor C16, C20 has a value between 150 and 300 nF.
[0091] The collector of transistors Q1, Q3 is respectively connected to the power supply bus at +65V and at -65V via a resistor RI, R11 and advantageously via the power variation damping circuit 304, 305. In the absence of feedback, the gain of the first pre-amplification stage 102 depends on the ratio of resistors R1 / R2 and R11 / R4.
[0092] The bases of transistors Q1 and Q3 are connected to each other and connected to ground via a resistor R7. The bases of transistors Q2 and Q4 are also connected to each other. The bases of transistors Q1 and Q3 are supplied by the input signal SI. The voltage source VI, connected between point SI and ground, represents the generator of the input signal SI. An intermediate filtering circuit can be interposed between the input signal SI and the bases of transistors Q1 and Q3. This circuit comprises, for example, a bandpass filter comprising a resistor R9 and a capacitor CIO connected to ground, and two capacitors C1, C6, connected in parallel.
[0093] The bases of transistors Q2 and Q4 are connected to loudspeaker R44 via a resistor R18, so as to form the feedback applied to the pre-amplification stage 301. The collectors of transistors Q1 and Q3 ensure the coupling of the pre-amplification stage 301 with circuits 306, 303, 307 and resistors R12, R13.
[0094] The power amplification stage 302 comprises two transistors Q8, Q9 connected by their respective base to the pre-amplification stage 301 via the circuit 306, 307.
[0095] The circuit 306, 307 comprises a transistor Q5, Q6 whose emitter is connected to the voltage bus at + / -65V via a resistor R12, R13 and whose base is connected to the collector of the transistors Q1, Q3. A capacitor C2, C3 is connected between the base and the collector of the transistors Q5, Q6 in order to improve the stability of the amplifier. The collector of the transistors Q5, Q6 is connected on the one hand to a resistor R57, R23 mounted in parallel with a capacitor C8, C7 and on the other hand, to a bias circuit 303. The latter comprises a transistor Q7 whose emitter is connected to the collector of the transistor Q5 and whose collector is connected to the collector of the transistor Q6. A resistor R14 is mounted between the collector and the base of the transistor Q7 and another resistor R15 is mounted between the emitter and the base of the transistor Q7.Finally, a capacitor C5, having for example a value of IpF, is connected between the emitter and the collector of the transistor Q7. This capacitor C5 improves the stability of the amplifier. Alternatively, the resistor R13, R12 can be connected, respectively, to the +65V and -65V power supply bus via a power variation damping circuit 304, 305.
[0096] Circuits 306, 307 perform a second voltage amplification. The gain of this amplification is proportional to the ratio of the resistance "seen by the collector" to that present on the emitter of Q5, Q6. When one of the transistors Q5, Q6 conducts, its dual is blocked, the transistor which is conducting therefore sees a very high resistance, hence a very significant voltage gain.
[0097] The power amplification stage 302 further comprises two transistors Q10, Q11, the bases of which are respectively connected to the emitter of the transistors Q8 and Q9. This so-called “Darlington” configuration makes it possible to increase the current gain. The emitters of the transistors Q8 and Q10 are coupled to the loudspeaker R44 by their respective resistors R16 and R19, while the emitters of the transistors Q9 and Q11 are coupled to the loudspeaker R44 by their respective resistors R17 and R20. For example, they have a respective quiescent current equal to 6 mA for the transistors Q8, Q9 and equal to 75 mA for the transistors Q10, Q11.
[0098] The power amplification stage 302 is connected to two power supply circuits 155a, 155b at the collectors of the transistors Q8 Q11. These power supply circuits 155a, 155b are connected to two voltage buses having distinct levels V+, V++. The power supply circuit 155a, 155b therefore makes it possible to select one or the other of these voltage levels depending on the amplification required. Typically, a first power supply bus delivers + / -65V and a second power supply bus delivers + / -32V. The second power supply bus is intended to be used to power the power amplification stage 302 when the output signal to be generated does not have a very high voltage, typically less than 27V.
[0099] Due to the relatively low current flowing through transistors Q8 and Q9, typically less than 10% of the current flowing through transistors Q10 and Q10, the collectors of Q8 and Q9 can be directly connected to the first power supply bus V++ without causing significant additional power dissipation. This embodiment improves the stability of the amplifier in terms of its phase margin and gain margin.
[0100] Furthermore, the bases of transistors Q2 and Q4 are connected to a protection line comprising a resistor R28 connected in series with a capacitor C4 connected to ground. This assembly is a voltage divider. In practice, capacitor C4 behaves like a short circuit when the voltage flowing through the circuit is alternating. On the other hand, when the voltage flowing through the circuit is direct, capacitor C4 behaves like an open circuit. In this case, the output of the high-power amplifier 100 is directly connected to point S2. The voltage gain of the high-power amplifier 100 is then equal to 1, which makes it possible to limit a possible undesirable direct component on the voltage applied to the terminals of the loudspeaker R44.
[0101] Several embodiments are possible for the power supply circuit 151-155, 155a, 155b.
[0102] For all the following embodiments, the signal numbered 1 represents the gate signal of the MOSFET transistor M1, M2. The output signal numbered 3 represents the output signal of the amplifier, i.e. the signal across the loudspeaker R44. The output signal numbered 2 represents the output signal of the power supply circuit 151-155,155a, 155b, i.e. the signal present on the collector of the transistors Q8 and Q10.
[0103] In the remainder of the description, only the upper power supply circuit is illustrated but [Fig.5] makes it possible to understand the positioning of the corresponding components for the lower power supply circuit.
[0104] As illustrated in [Fig.6], in a first embodiment, the power supply circuit 151 comprises a MOSFET transistor M1, M2 directly connected to the first power supply bus V++ by its drain and to the second power supply bus V+ by its source via a fourth diode D3, DU, which can be a conventional diode or a Schottky diode. In practice, the anode of the fourth diode D3 is connected to the second power supply bus V+ and its cathode to the output signal numbered 2 representing the output signal of the power supply circuit 151-155,155a, 155b; the cathode of the fourth diode DU is connected to the second power supply bus -V+. The MOSFET transistor M1, M2 is blocked for a voltage lower than a threshold, typically 35 V, then switches and operates linearly beyond this threshold. The MOSFET transistor Ml, M2 passes the first power bus V++ when a control voltage greater than the threshold value is applied to it. This voltage is controlled by the combination of a charge assistance subcircuit 131, a discharge assistance subcircuit 141 and a voltage shift subcircuit 161.
[0105] The charging assistance sub-circuit 131 comprises a first resistor R24, R31 connected between the gate of the MOSFET transistor M1, M2 and an interconnection point A1, A2.
[0106] The discharge assistance sub-circuit 141 comprises a transistor Q12, Q13 whose emitter is connected on the one hand to the gate of the MOSFET transistor M1, M2 and on the other hand to the charge assistance sub-circuit 131. The collector of the transistor Q12, Q13 is connected to the output signal 2 of the power supply circuit 151 via the second resistor R8, R27. The base of the transistor Q12, Q13 is connected to the interconnection point A1, A2 via a third resistor R21, R39.
[0107] The voltage shift sub-circuit 161 comprises a first capacitor C18, C23 connected in parallel with a first diode D8, D10. The cathode of the first diode D8 is connected to a first interconnection node NI and the anode of the first diode D8 is connected to a second interconnection node N2. The diode D10 is connected in the opposite direction, i.e., its cathode is connected to the interconnection node N3 and its anode is connected to the interconnection node N4, as illustrated in [Fig.5]. The first interconnection node is connected to the interconnection point A1 and the second interconnection node is connected to the loudspeaker R44.
[0108] The voltage shift sub-circuit ensures that the gate potential of the MOSFET transistor M1 is always 15 V higher than that of the amplifier output.
[0109] Furthermore, the Darlington type circuit consisting of transistors Q8 and Q10 requires 5 V of dropout voltage, i.e. the voltage equal to the difference between the input voltage on the collector of transistor Q8 and the output voltage of the emitter of transistor Q10. MOSFET M1, for its part, requires 10 V of dropout voltage; i.e. the voltage equal to the difference between the input voltage on its gate and the output voltage on its source in the case where the current is maximum and being saturated. The voltage shift sub-circuit must therefore compensate for the voltage drops of the two bipolar transistors Q8 and Q10, as well as of MOSFET M1, i.e. 5 + 10 = 15 V.
[0110] The power supply circuit 151 advantageously comprises a first protection diode D5, D13. The anode of the diode D5 is connected to the output signal 2 of the power supply circuit 151 and its cathode is connected to the gate of the MOSFET transistor M1. The cathode of the diode D13 is connected to the output of the power supply circuit 155b and its anode is connected to the gate of the MOSFET transistor M2. Similarly, the Power supply circuit 151 includes a second protection diode D4, D12. The cathode of diode D4 is connected to the drain of MOSFET transistor M1 and its anode is connected to the source of MOSFET transistor M1. The cathode of diode D12 is connected to the source of MOSFET transistor M2 and its anode is connected to the drain of MOSFET transistor M2.
[0111] With such an assembly, the different signals obtained are illustrated in figures 7 and 8.
[0112] The signals illustrated in [Fig.7] correspond to the signals obtained with a sinusoid of frequency 20kHz supplied at the input of the high-power amplifier 102
[0113] We thus note that the output signal 3 is distorted at the peaks of the sinusoids which take on a triangular appearance. Thus, it is not possible to correctly reproduce a sinusoid at 20 kHz. On the other hand, we do not observe abnormal voltage peaks. Power losses are therefore limited.
[0114] The signals illustrated in [Fig.8] correspond to the signals obtained with a sinusoid of frequency 1kHz supplied at the input of the high-power amplifier 102.
[0115] We thus note that the output signal 3 is less distorted than at 20kHz. To the naked eye, the sinusoid of the output signal 3 even seems perfectly reproduced. However, it turns out that the harmonic distortion rate (THD) is greater than 0.1%.
[0116] As illustrated in [Fig.9], in a second embodiment, the voltage shift sub-circuit 162 of the power supply stage 152 further comprises a second diode D6, D29 connected in parallel with a fourth resistor R22, R26. The cathode of the second diode D6 is connected to the interconnection point NI and the anode of the second diode D6 is connected to the third interconnection node N10. The diode D29 is connected in the opposite direction, i.e., its cathode is connected to the interconnection node N20 and its anode is connected to the interconnection node N4, as illustrated in [Fig.5].
[0117] With such an assembly, the different signals obtained are illustrated in Figures 10 and 11. The signals illustrated in [Fig. 10] correspond to the signals obtained with a sinusoid of frequency 20kHz supplied at the input of the high-power amplifier 102
[0118] It is still observed that the output signal 3 is distorted at the peaks of the sinusoids, which take on a triangular appearance. Thus, it is still not possible to correctly reproduce a sinusoid at 20 kHz.
[0119] The signals illustrated in [Fig.l 1] correspond to the signals obtained with a sinusoid of frequency 1kHz supplied at the input of the high-power amplifier 102.
[0120] To the naked eye, the sinusoid of the output signal 3 appears perfectly reproduced on the positive alternation and slightly distorted on the negative alternation. The distortion by harmonics THD is equal to 0.45% because the observed deformation corresponds to the production of harmonics.
[0121] The addition of these components therefore does not reduce the distortions observed on the amplifier output signal. The addition of the second diode D6, D29 and the fourth resistor R22, R26 degrades the dynamic performance of the circuit. More complex circuitry is required for these components to be beneficial, i.e. to improve the degradations at low frequencies and high frequencies. To obtain a performance gain, it is possible to add other elements around the transistor Q12, Q13.
[0122] As illustrated in [Fig.12], in a third embodiment, the charging assistance sub-circuit 132 of the power supply stage 153 further comprises a circuit branch comprising a third diode D9, D14 in series with a fifth resistor R29, R30, connected in parallel with the first resistor R24, R31. The cathode of the third diode D9 is for example connected to the gate of the MOSFET transistor M1. The anode of the third diode D14 is connected to the gate of the MOSFET transistor M2. Alternatively, the components D9, R29 and D14, R30 can be reversed, such that the third diode D9, D14 is connected to the gate of the MOSFET transistor M1, M2 via the resistor R29, R30.
[0123] With such an assembly, the different signals obtained are illustrated in [Fig. 13]. These correspond to the signals obtained with a sinusoid of frequency 20kHz supplied at the input of the high-power amplifier 102.
[0124] It is observed that the output sinusoid corresponding to the output signal numbered 3 is well restored. On the other hand, the output signal 2, representing the output voltage of the supply circuit 153, has distortions between 0 and 5ps and between 18 and 20ps. The THD rate is equal to 0.26%. The addition of these components here makes it possible to improve the distortion for an input sinusoidal signal going up to a frequency of 20kHz.
[0125] As illustrated in [Fig.14], in a fourth embodiment, the voltage shift sub-circuit of the power supply stage 154 further comprises a third capacitor C21, C24 connected in parallel with the fourth resistor R22, R26 and the second diode D6, D29, as well as a second capacitor C17, C22, connected in parallel with the first diode D8, D10 and the first capacitor C18, C23.
[0126] With such an assembly, the different signals obtained are illustrated in [Fig. 15]. These correspond to the signals obtained with a sinusoid of frequency 20kHz supplied at the input of the high-power amplifier 102.
[0127] We thus observe that signal 2 is less disturbed but the switching of the fourth diode D3 still generates interference, in particular between 45 and 50ps.
[0128] Indeed, the second capacitor C17, C22, for example using electrochemical technology, has a significant value, typically between 5 and 15pF, constitutes an energy reservoir, while the first capacitor C18, C23, for example using plastic film technology, has a lower value, typically between 50 and 150nF, allows to smooth out the parasites at high frequencies. The THD is reduced to 0.21%
[0129] This association makes it possible to make the circuit more efficient in the transient phases of signal growth and decay.
[0130] As illustrated in Figures 5 and 16, in a fifth embodiment, the power supply stage 155, 155a, 155b further comprises a fourth capacitor C14, C25 connected in parallel with the fourth diode D3, DU.
[0131] With such an assembly, the different signals obtained are illustrated in [Fig. 17]. These correspond to the signals obtained with a sinusoid of frequency 20kHz supplied at the input of the high-power amplifier 102.
[0132] It is observed that the parasites generated by the fourth diode D3, DU have disappeared between 45 and 50ps. This results in a slight decrease in THD from 0.21 to 0.2%.
[0133] Although the invention is previously described with reference to figures 5 to 17 for a high power audio amplifier 102 for supplying a loudspeaker R44 with voltage, it is also possible to use the power supply circuit of the invention for a high power audio amplifier with current, that is to say for supplying a loudspeaker with current.
[0134] To do this, as illustrated in [Fig. 18] a current measuring resistor R61 is inserted between the loudspeaker R44 and ground. In addition, the current amplifier 103 does not have a protection line 403. The bases of the transistors Q2 and Q4 are connected to an interconnection point located between the resistor R61 and the loudspeaker R44 via a capacitor C40. This capacitor C40 therefore only allows the alternating component of the signal to pass.
[0135] In this configuration, the loudspeaker R44 is therefore crossed by an alternating current, an image of which is applied to point S2. It follows that the transconductance of the amplifier is equal to 1 / R61 for an alternating signal.
[0136] Alternatively, in order to protect the speaker R44 and the amplifier itself, a resistor can be added to the feedback, between the point S2 and the output of the amplifier 3. In the case of a DC component across the speaker R44, the DC component is applied to the bases of the transistors Q2 and Q4 via a resistor R18. The feedback will tend to cancel this DC component. In addition, the resistor R18 makes it possible to limit the gain of the amplifier in the absence of the speaker R44, also preventing it from oscillating.
[0137] When the current is alternating, the capacitor C40 has a negligible impedance compared to the value of the resistor R18. In normal operation, that is to say, when there is no direct component and when a loudspeaker R44 is effected tively present at the amplifier output, resistor R18 has almost no effect. On the other hand, in the absence of speaker R44 at the amplifier output, we can consider that point S2 is connected, on the one hand, to the amplifier output via resistor R18, and, on the other hand, to ground via resistor R61 since capacitor C40 behaves like an AC short circuit. The voltage amplification will therefore be limited to (R18 + R61) / R61 since the amplifier output voltage multiplied by R61 / (R18 + R61) is compared to the input voltage applied to point SI by the differential pairs formed by transistors Q1 / Q2 and Q3 / Q4. The amplifier will then not provide its maximum output voltage, which could have been dangerous.Similarly, in the presence of a DC component and in the presence or absence of loudspeaker R44 at the amplifier output, the capacitor behaves like an open circuit, the DC component is thus reinjected at point S2 via resistor R18. The voltage gain is thus limited to 1 for DC voltages, which does not risk damaging loudspeaker R44.
[0138] In another variant, a current amplifier protection circuit can be added. To do this, as illustrated in [Fig.19], the interconnection point PI located between the resistor R61 and the loudspeaker R44 is connected to a first terminal of a resistor R70. In this embodiment, the second terminal of the resistor R70 is connected to the collector of a transistor Q16. The collector of the transistor Q16 is also connected to the base of a second transistor Q15. The voltage divider formed by the resistors R71 and R72 makes it possible to adapt the current threshold from which the protection circuit acts. The first terminal of the resistor R71 is connected to the second power supply bus V+ with a value of 32 V. The second terminal of the resistor R71 is connected to the base of the transistor Q15 and to the collector of the transistor Q16.The first terminal of resistor R72 is connected to the second terminal of resistor R70 and the second terminal of resistor R72 is connected to ground. The conduction threshold voltage of transistor Q15 is thus shifted by: V+*R72 / (R71 + R72). This increases the sensitivity, i.e. the current threshold from which the protection acts is lowered.
[0139] Preferably, transistor Q15 is a darlington. This makes it possible to limit the distortion due to the circuit by drawing a lower current from the terminals of the measuring resistor R61 while maintaining sufficient sensitivity thanks to the divider network R71 / R72. Another divider bridge consisting of resistors R73 and R74 is inserted between point 3 and ground. The first terminal of resistor R73 is connected to point 3. The second terminal of resistor R73 is connected to the base of transistor Q16 and to the first terminal of resistor R74. The second terminal of the latter is connected to ground. Thus, the voltage threshold from which the current protection is neutralized. The emitters of transistors Q16 and Q15 are connected to each other and to ground, and the collector of transistor Q15 is connected to the base of transistor Q8 via a diode D21, whose cathode is connected to the collector of transistor Q15.
[0140] This circuit must be mirrored to manage the current protection during the negative alternation, the NPN type transistors are then replaced by PNP type transistors.
[0141] This circuit limits the risk of power failure of the current amplifier, particularly when it is overloaded or when its output is short-circuited. In fact, in these cases, the product of the output current by the voltage of the power supply buses will be entirely dissipated by the transistors and these could be damaged.
[0142] With the circuit of [Fig. 19], if the current in the current measuring resistor R61 is sufficient to turn on transistor Q15, it can, via diode D21, evacuate the signal coming from the base of transistor Q8 to ground, so as to block it. On the other hand, if the voltage across the terminals of the loudspeaker R44 is sufficient to turn on transistor Q16, the signal coming from the base of transistor Q15 will be diverted to ground and it is the latter which will be blocked. The protection circuit is thus neutralized in the presence of a sufficient voltage at the output of the amplifier.
[0143] To conclude, the invention makes it possible to obtain a high-power audio amplifier making it possible to limit the distortions identified on the signals and therefore to improve the efficiency and reduce the saturation of the amplifier.
Claims
Claims
1. High power audio amplifier (102) for driving at least one loudspeaker (R44), said amplifier comprising: - a pre-amplification stage (201, 301) receiving an input signal (SI); - a power amplification stage (202-203, 302) connected to the pre-amplification stage (201, 301) and providing an output signal (3) intended to power said at least one loudspeaker (R44); the pre-amplification (201, 301) and power amplification (202-203, 302) stages comprising an upper part and a lower part mounted in mirror; - a feedback providing the pre-amplification stage (201, 301) with an image of the output signal (3), - an upper power supply circuit (151-155, 155a), connected to the upper part of the power amplification stage (202-203, 302), and allowing it to be supplied by a first or a second power supply bus (V++, V+); - a lower power supply circuit (155b) connected to the lower part of the power amplification stage (202-203, 302) and allowing it to be supplied by a first or a second power supply bus; - each power supply circuit (151-155, 155a, 155b) comprising a MOSFET transistor (Ml, M2) and supervision means, the MOSFET transistor (Ml, M2) being controlled by the supervision means so as to carry out switching between one or the other of the two power supply buses (V++, V+), the MOSFET transistor (Ml, M2) being connected to the second power supply bus (V+) via a fourth diode (D3, DI 1) a first terminal of which is connected to the source of the MOSFET transistor (Ml, M2), and the drain of the MOSFET transistor (Ml, M2) being connected to the first power supply bus (V++), characterized in that each power supply circuit (151-155, 155a, 155b) further comprises: - a charging assistance sub-circuit (131-133) of said MOSFET transistor (Ml, M2) comprising at least a first resistor (R24, R31), a first terminal of the first resistor (R24, R31) being connected to the gate of the MOSFET transistor (Ml, M2) and a second terminal of the first resistor (R24, R31) being connected to a point interconnection point (A1); - a discharge assistance sub-circuit (141) of said MOSFET transistor (M1, M2) comprising at least a second and a third resistor (R8, R27, R21, R39) and a bipolar transistor (Q12, Q13); the base of the bipolar transistor (Q12, Q13) being connected to a first terminal of the third resistor (R21, R39), its emitter being connected to the gate of the MOSFET transistor (M1, M2) and its collector being connected to the source of the MOSFET transistor (M1, M2) via the second resistor (R8), the second terminal of the third resistor (R21, R39) being connected to the interconnection point (A1, A2); and - a voltage shift sub-circuit (161-163) comprising a first diode (D8, D10) connected in parallel with a first capacitor (C18, C23);a first terminal of the first diode (D8, D10) and a first terminal of the first capacitor (Cl8, C23) being connected to a first interconnection node (NI, N4), itself connected to said interconnection point (Al, A2) of said discharge assistance sub-circuit (141); a second terminal of the first diode (D8, D10) and a second terminal of the first capacitor (C18, C23) being connected to a second interconnection node (N2, N3), itself connected to said output signal (3).;
2. Amplifier according to claim 1, characterized in that the voltage shift sub-circuit (161-163) further comprises at least a fourth resistor (R22, R26) and a second diode (D6, D29) connected in parallel, a first terminal of the second diode (D6, D29) and a terminal of the fourth resistor (R22, R26) being connected to the interconnection point (A1, A2), a second terminal of the fourth resistor (R22, R26) and a second terminal of the second diode (D6, D29) being connected to a third interconnection node (N10).
3. An amplifier according to claim 1 or 2, characterized in that the charging assistance sub-circuit (131-133) further comprises a fifth resistor (R29, R30) connected in series with a third diode (D9, D14), the fifth resistor (R29, R30) and the third diode (D9, D14) being connected in parallel with the branch of the charging assistance sub-circuit (131-133) including the first resistor (R24, R31).
4. Amplifier according to one of claims 1 to 3, characterized in that the voltage shift sub-circuit (161-163) further comprises a second capacitor (C17, C22) and a third capacitor (C21, C24), the second capacitor (C17, C22) being connected in parallel with the first capacitor (C18, C23) and the first diode (D8, D10) and the third capacitor (C21, C24) being connected in parallel with the second diode (D6, D29) and the fifth resistor (R22, R26).
5. Amplifier according to one of claims 1 to 4, characterized in that each power supply circuit (151-155, 155a, 155b) comprises a first protection diode (D5, D13) a first terminal of which is connected to the source of the MOSFET transistor (Ml, M2) and a second terminal of which is connected to the gate of the MOSFET transistor (Ml, M2).
6. Amplifier according to one of claims 1 to 5, characterized in that each power supply circuit (151-155, 155a, 155b) further comprises a second protection diode (D4, D12) connected between the source and the drain of the MOSFET transistor (M1, M2).
7. Amplifier according to one of claims 1 to 6, characterized in that each power supply circuit (151-155, 155a, 155b) further comprises a fourth capacitor (C14, C25) connected in parallel with the fourth diode (D3, DU).
8. Amplifier according to one of claims 1 to 7, characterized in that the pre-amplification stage (201, 301) is connected to the first power supply bus (V++) of each power supply circuit (151-155, 155a, 155b) by means of a power variation damping circuit (304, 305) of said first power supply bus (V++), said power variation damping circuit (304, 305) comprising at least one capacitor (Cl5, Cl9) and at least one resistor (R52, R55) mounted as a low-pass filter.
9. Amplifier according to claim 8, characterized in that said power variation damping circuit (304, 305) further comprises an additional capacitor mounted in parallel with the at least one capacitor (C16, C20).
10. Amplifier according to one of claims 1 to 9, characterized in that the feedback applied to the pre-amplification stage provides a signal proportional to the current passing through the loudspeaker.