Amplifier comprising two parallel coupled amplifier units

Abstract

An amplifier with two parallel coupled amplifier units with inverse characteristics and in particular to the parallel coupling of a sourcing limited amplifier unit and a sinking limited amplifier unit.

Claims

1. An amplifier comprising an input and an output, the amplifier comprising: a first amplifier unit having a first input connected to the input and a first output connected to the output, and a second amplifier unit having a second input connected to the input and a second output connected to the output, where the first and second amplifier units are configured so that: the first amplifier unit is configured to receive, at the first output, a first maximum current and of delivering, from the first output, a second maximum current exceeding 1.2 times the first maximum current, the second amplifier unit is configured to deliver, from the second output, a third maximum current and of receiving at the second output, a fourth maximum current exceeding 1.2 times the third maximum current, when a first current is received by the output, a first portion of the first current received by the second output exceeds a second portion of the first current received by the first output and when a second current is delivered from the output, a third portion of the second current delivered from the first output exceeds a fourth portion of the second current delivered from the second output.

2. An amplifier according to claim 1, wherein the first amplifier unit comprises a first power input and a first limiter limiting a current fed from the first output to the first power input to the first maximum current.

3. An amplifier according to claim 1, wherein the second amplifier unit comprises a second power input and a second limiter limiting a current fed from the second power input to the second output to the third maximum current.

4. An amplifier according to claim 1, wherein the first maximum current is between 0.2 and 5 times the third maximum current.

5. An amplifier according to claim 1, wherein: the first amplifier unit has as first output impedance at the first output, the second amplifier unit has as second output impedance at the second output, and the first output impedance is between 0.2 and 5 times the second output impedance.

6. An amplifier according to claim 1, wherein the amplifier, at the input, has an input impedance which exceeds 1 GOhm.

7. An amplifier according to claim 1, wherein the amplifier, at the input, has an input capacitance at input lower than 2 pF.

8. An amplifier according to claim 1, further comprising: a first power input conductor connected to both the first and the second amplifier units, a second power input conductor connected to both the first and the second amplifier units, wherein the first and second amplifier units are configured to output voltages of at least 80% of a voltage applied between the first and second power input conductors.

9. An amplifier according to claim 1, wherein: when the first amplifier unit outputs, from the first output, a current equal to the third maximum current, the second amplifier outputs, from the second output, a current exceeding 0.2 times the third maximum current and when the second amplifier unit receives, on the second output, a current equal to the first maximum current, the first amplifier receives, on the first output, a current exceeding 0.2 times the first maximum current.

10. An amplifier according to claim 1, wherein a (voltage) gain of the first amplifier unit is between 0.5 and 2 times a gain of the second amplifier unit.

11. An amplifier according to claim 1, having a constant gain for all input voltages and output voltages below a threshold voltage.

12. An assembly comprising an amplifier according to claim 1 and a load connected to the output.

13. An assembly according to claim 12, wherein the amplifier is configured to, for each voltage, between 0V and a predetermined maximum voltage, provided on the output, feed a current to the load from the output between 0.8 and 1.2 times a current received by the output when the predetermined maximum voltage, negative, is provided on the output.

14. An assembly of a transducer and an amplifier according to claim 1, wherein the transducer is connected to the input.

15. A personal audio device comprising an amplifier according to claim 1.

16. A personal audio device comprising an assembly according to claim 12.

17. An amplifier according to claim 2, wherein the amplifier, at the input, has an input impedance which exceeds 1 GOhm.

18. An amplifier according to claim 3, wherein the amplifier, at the input, has an input impedance which exceeds 1 GOhm.

19. An amplifier according to claim 4, wherein the amplifier, at the input, has an input impedance which exceeds 1 GOhm.

20. An amplifier according to claim 5, wherein the amplifier, at the input, has an input impedance which exceeds 1 GOhm.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In the following, preferred embodiments will be described with reference to the drawing, wherein:

(2) FIG. 1 illustrates the output current capabilities of two amplifier units,

(3) FIG. 2 illustrates a result of a parallel coupling of two amplifier units,

(4) FIG. 3 illustrates the supply current of parallel coupled amplifier units,

(5) FIG. 4 illustrates a first embodiment of an amplifier according to the invention,

(6) FIG. 5 illustrates an amplifier unit with a sourcing limitation,

(7) FIG. 6 illustrates an amplifier unit with a sinking limitation, and

(8) FIG. 7 illustrates a combination of the amplifier units from FIGS. 5 and 6.

DETAILED DESCRIPTION OF THE INVENTION

(9) FIG. 4 illustrates a generic embodiment of the invention where two amplifier units Buf1 and Buf2 are both connected to an input In and an output out. Both amplifier units receive power from two conductors presently represented by Vdd and Ground.

(10) The amplifier units Amp1/Amp2 have, for this example, inverted characteristics so that the amplifier unit Amp2 is current limited when the output signal requires current to be received by the output, such as when a voltage of the output signal is negative. Thus, the amplifier unit Amp2 is configured to receive, at a given voltage, when a maximum voltage causes a maximum current (such as a maximum obtainable output voltage), a first current, whereas it is able to deliver a second, higher current, when the output voltage is positive.

(11) Conversely to that, the amplifier unit Amp1 is current limited when the output signal requires current to be delivered from the output out, such as when a voltage of the output signal is positive. Thus, the amplifier unit Amp1 is configured to deliver, at a given voltage (such as a maximum obtainable output voltage) a third current, whereas it is able to receive a fourth, higher current, when the output voltage is negative.

(12) The basic idea is to combine an amplifier unit (Amp1), which is configured to source a large current to a load connected to the output out, but sinking a limited current from the load, with another amplifier unit (Amp2) which is configured to sink a large current from the load. The two complementary buffer amplifiers or amplifier units are connected in parallel at their inputs and at their outputs.

(13) When combining the output of the two amplifier units, the combined current will be the sum of the first and fourth currents and the sum of the second and third currents. Thus, the amplifier units may be selected or paired so that these sums give the sought after symmetry or behaviour. Clearly, the maximum currents of the amplifier units may be different, as long as the combination gives the desired overall current behaviour. It may be easier to design a circuit where the first and third currents are at least substantially the same and where the second and fourth currents are at least substantially the same. On the other hand, there may be a large difference between the first/third currents and the second/fourth currents.

(14) The outputs of the amplifier units are connected to the output of the overall amplifier. Naturally, the outputs may be directly connected, such as galvanically, to the output. Alternatively, the first and second outputs may be connected to a combiner connected to the output. The combiner may be as simple as a resistor allowing the output voltages of the first and second amplifier units to vary. The combiner may additionally perform a smooth transition from when the output from only one amplifier unit is connected to the output to when the output of only the other amplifier unit is connected to the output. The switching/transition may simply be controlled by the sign of the voltage output from the output, as this identifies the direction of the current on the output. Again, the amplifier units or their characteristics may be selected to arrive at the sought after symmetry.

(15) FIG. 1 illustrates the characteristics of two amplifier units with complementary, actually inverted, characteristics. The graph shows the output current drive capabilities of the two amplifier unit types. Amplifier unit 2 (marked with A), is e.g. a source follower, biased with 10 A, which is configured to source up to 10 A and to sink more current. Amplifier unit 1 (marked with B) is a complementary type, configured to sink up to 10 A, and of sourcing more current.

(16) This means that either amplifier unit type has a region where the maximum output current is smaller than the required output current, so that slewing may occur, such as when a capacitive load is connected, where the output voltage then is non-linearly distorted from each individual amplifier unit.

(17) FIG. 2 illustrates a graph showing the actual output current of an amplifier that is composed of parallel connected Amp1 type and Amp2 type amplifier units.

(18) Both amplifier units deliver an actual output current that is smaller than the amplifier's maximum output current capability, while the sum of both actual output currents equals the required output current.

(19) This means the combined amplifier can deliver any required output current, and the output voltage will not be distorted.

(20) FIG. 3 illustrates a graph showing the current drawn from the power supply (Vdd; cf FIG. 4). It may be clear that the supply current of Amp2 always equals its bias current (10 A in this example), while the supply current of Amp1 can vary between a value that is lower than its bias current (approx. 5 A this example) and the current available from the external power source.

(21) For small signals, the required output current is close to zero, and the supply current will equal the quiescent current level (marked with C), which is the sum of the bias currents of both buffers.

(22) Clearly, the amplifier units Amp1 and Amp2 need not be of the same type or same technology,

(23) In FIG. 5, an amplifier unit is illustrated with characteristics similar to those of Amp2 in FIG. 1, i.e. an amplifier unit which is limited in the current delivered to the output, due to the limitation of the current source Idd, but not in the current received from the output, which current may be fed directly to ground.

(24) In FIG. 6, an example is illustrated of an amplifier unit with characteristics similar to those of Amp1 in FIG. 1, i.e. which is configured to source a large current. This is in the form of folded source follower. In this circuit, the drain current of the P-type MOSFET is determined by a current source or current limiter Idd, and a feedback path provides the supply current. In order to maximize the output voltage swing, a level shifter is provided in the feedback path.

(25) For the implementation of the level shifter, there are several possibilities, which will not be elaborated here.

(26) A current limiter may be implemented in a number of manners. A suitable current limiter type often used in amplifier units is a constant current source.

(27) The constant current source is a well known element for MOSFET integrated circuits. The advantage is that it can be used to provide a current that is to a high degree independent of the supply voltage level, and the operating temperature, and production variance. Also the feedthrough of any noise from the supply connection can be greatly reduced. Such independency may be obtained by generating the required bias voltage for a MOSFET current source by means of a reference circuit that can be integrated in the same integrated circuit die as the other elements of the amplifier. Alternatively, a self-biased configuration can be used for a constant current source.

(28) Thus, due to the current limiter Idd, the current which may be received on the output is limited, whereas the current output from the output may be derived directly from the Vdd supply.

(29) The circuits of FIGS. 5 and 6 may thus be connected in parallel. This is seen in FIG. 7. Here, both amplifier units are biased at the same Idd, the DC-voltage at their outputs is the same, so that no DC-current will flow through the filter resistors.

(30) The proposed circuit can also be used without a capacitive load, when a low impedance resistive load must be driven, and still a low quiescent current is required.