JPH046129B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an amplification device that amplifies audio signals and the like with a low distortion factor.

In audio signal amplifiers that use vacuum tubes, transistors, FETs, etc., the nonlinearity of these elements appears in the output of the amplifier as nonlinear distortion. This effect is particularly large in power amplifiers that require high output. Further, in a power amplifier using a class B push-pull circuit, crossover distortion and switching distortion exist, which are considered to have an adverse effect on sound quality. Conventionally, gain control means called feedforward and feedback have been devised to reduce distortion in such amplifiers.
The latter is generally put into practical use. As shown in Figure 1, the principle of feed forward is that an attenuator 2 extracts a 1/A signal from the output of an amplifier 1 with a certain distortion and a gain of A, and compares this with the input signal to determine the difference. A method in which the differential amplifier 3 etc. amplifies the distortion component by a factor of A to create a component with the opposite phase to the distortion appearing in the output of the amplifier 1, and this is added to the output of the amplifier 1 using the synthesis circuit 4 to cancel the distortion component. It is.
However, this feed forward method has the following drawbacks. It is difficult to match the gains of the amplifier 1 and the differential amplifier 3 and the attenuation of the attenuator 2. In the combining circuit 4, it is difficult to reduce power loss. For these reasons, the feed forward method has not been put into practical use much.

On the other hand, the principle of the feedback method is that, as shown in FIG. By subtracting the input signal from the input signal, an input signal is created in which a distortion multiplied by β is added in the opposite phase, thereby reducing the output distortion. When feedback is applied using this method, the gain is A/1+βA, and the distortion is approximately 1/1+βA times. Therefore, in an amplifier with a large open gain, distortion (static distortion) can be significantly reduced by applying a large amount of feedback. However, the above feedback method has the following drawbacks. Even if the amount of feedback is increased, distortion cannot be completely reduced to zero. If the feedback loop is large, applying feedback may make the amplifier unstable or cause TIM
(Transient Inter Modulation)
Dynamic distortion such as IIM (interface intermodulation) may occur.

In addition, TIM is an amplifier circuit that has multiple amplification stages, and even if negative feedback is applied from the output stage to the input stage, the negative feedback may not operate properly due to the time difference between the input and output. , when inputting a pulse waveform with large transients, etc., dynamic distortion may occur due to the time delay of the negative feedback operation. It refers to this dynamic distortion. IIM also means that when a load that generates a back electromotive force, such as a speaker, is connected to an amplifier, the back electromotive force generated by the load returns to the input stage via a feedback circuit, and the phase relationship with the input signal is changed. In some cases, a negative feedback circuit may operate like a positive feedback circuit. The dynamic distortion generated by connecting a load in this way is called IIM.

This invention was made in view of the above circumstances, and it incorporates the advantages of the feedback method and the feedback method and minimizes the drawbacks. Feedforward method operation can be realized, and static distortion and dynamic distortion can be extremely reduced.Especially in power amplification circuits, power loss and output can be reduced when combining the feedforward output. It is an object of the present invention to provide an amplifying device that suppresses an increase in impedance.

Embodiments of the present invention will be described below with reference to the drawings. In FIG. 3, 10 is an input terminal and 16 is an output terminal. The signal at input terminal 10 is applied to one input terminal of synthesis circuit 11 and the positive input terminal of differential amplifier 14 . The output of combiner 11 is applied to amplifier 12 and amplified. The output of the amplifier 12 is applied to one input terminal of a combiner 15 and also to a feedback circuit 13. Feedback circuit 1
The output of No. 3 is connected to the other input terminal of the synthesizer 11,
It is applied to the negative input terminal of the differential amplifier 14. The output of the differential amplifier 14 is then applied to the other input terminal of the combiner 15, and the output of the combiner 15 is led out to the output terminal.

The amplifier 12 generates a distortion component N in its output, has a bare gain of _A1 , and is fed back by a feedback circuit 13 with a feedback rate of _β1 . Therefore, the apparent distortion of the amplifier 12 is reduced to approximately N/1+β ₁ A ₁ . The signal returned through the feedback circuit 13 contains a distortion β ₁ times that of the output, so the differential amplifier 14 takes the difference between the input signal and the feedback signal and amplifies the opposite-phase distortion component. Then, the synthesizer 15 adds it to the output of the amplifier 12 to cancel the distortion component.

The gain of the differential amplifier 14 is _A2 , and the combiner 15
Expressing _{the distortion} at the output using _{the formula , the distortion level is :} Therefore, when A ₂ ≈1/β ₁ , the distortion is significantly reduced. However, since A ₂ fluctuates under the influence of operating temperature and power supply voltage, it is difficult to reduce distortion to completely zero.

On the other hand, when the input signal is S, the signal level at each point is as follows.

Output of amplifier 12 = A ₁ /1 + β ₁ A ₁ S Output of feedback circuit 13 = β ₁ A ₁ /1 + β ₁ A ₁ S Output of differential amplifier 14 = (S - β ₁ A ₁ /1 + β ₁ A ₁ S) A ₂ = A ₂ /1 + β ₁ A ₁ S Output of the synthesis circuit 15 = A ₁ /1 + β ₁ A ₁ S +A ₂ /1 + β ₁ A ₁ S = A ₁ +A ₂ /1 + β ₁ A ₁ S Note that if β1 is 1 In this case, an emitter follower transistor or a source follower FET may be used as the amplifier 12 with the feedback circuit 13.

FIG. 4 shows an example of a specific circuit using the present invention, and the same parts as those in FIG. 3 will be described with the same reference numerals.
The amplifier 12 for signal amplification has a gain of _A1 ,
The amplification factor of the feedback circuit 13 is β ₁ =R ₁ /R ₁ +R ₂ . Further, the differential amplifier 14 for amplifying distortion components has a feedback circuit section in order to obtain stability of gain and improvement of phase characteristics. This differential amplifier 14 further includes differential amplifiers 17 and 18, resistors R ₃ , R ₄ , R ₅ ,
Consists of R ₆ and R ₇ . Resistor _R7 is a variable resistor for gain adjustment. The positive input terminal of the differential amplifier 17 is
This is an inversion input terminal. The gain in this differential amplifier 14 is determined as follows. First, the open loop gains, feedback factors, and voltages at the non-inverting input terminal, inverting input terminal, and output terminal of the differential amplifiers 17 and 18 are A ₃ , β ₃ , V ₁ , V ₂ , V ₃ , and A , respectively. ₄ , _β4 , _V4 , _V5 , and _V6 . Differential amplifier 17
Regarding, V ₃ =A ₃ ×V ₁ /(1+β ₃ ×A ₃ ) ≒V ₁ /β ₃ V ₂ =β ₃ ×A ₃ ×V ₁ /(1+β ₃ ×A ₃ ) ≒V _1where , β ₃ = R ₃ R ₇ / (R ₄ + (R ₃ R ₇ )) (indicates the resistance value in parallel connection) holds true. Next, considering the path from the differential amplifier 17 to the differential amplifier 18, V ₆ = -R ₆ ×V ₃ /R ₅ -R ₆ ×V ₂ /R _7In this formula, the V ₃ obtained above is , substituting V ₂ ,
When calculating V ₆ /V ₁ (gain in the differential amplifier 14), the gain in the differential amplifier 14 is −[R ₄ +(R ₃ R ₇ )/R ₃ R ₇ ·R ₆ /R ₅ +R ₆ / R ₇ ] = - [R ₄ R ₆ /R ₃ R ₅ +R ₄ R ₆ /R ₅ R ₇ +R ₆ /R ₅ +R ₆ /R ₇ ]. On the other hand, the positive input terminal of the differential amplifier 18 is a non-inverting input, and the gain of the differential amplifier 18 is determined as follows. First, regarding only the amplifier 18, V ₆ =A ₄ ×V ₄ /(1+β ₄ ×A ₄ ) ≒V ₄ /β ₄ V ₅ =β ₄ ×A ₄ ×V ₄ /(1+β ₄ ×A ₄ ) ≒V ₄ holds true. In addition, in the above calculation, the output of the amplifier 18 was obtained assuming that the output of the amplifier 17 is zero, but in reality, the output of the amplifier 17 has the output of the amplifier 1
Contains a component due to input V5 of 8. Therefore, the output V ₆ can be found by adding these components based on the principle of superposition. That is, it is input to the amplifier 17 via the resistor R ₇ , and the output of this amplifier 17 is input to the amplifier 18 again via the resistor R ₅ .
The output of the amplifier 18 corresponding to this component is V ₅ ×(−R ₄ /R ₇ )×(−R ₆ ×R ₅ ). Based on the principle of superposition, the above value is
When added to equation 6, it becomes V ₆ ≈V ₄ /β ₄ +V ₅ ×(−R ₄ /R ₇ )×(−R ₆ ×R ₅ ). Therefore, by substituting the previously obtained V ₅ into this equation to find V ₆ /V ₄ (gain in the differential amplifier 18), the gain in the differential amplifier 18 is R ₆ + (R ₅ R ₇ ) /R ₅ R ₇ +R ₄ /R ₇・R ₆ /R ₅ =R ₆ /R ₅ +R ₆ /R ₇ +R ₄ R ₆ /R ₅ R ₇ +1. Here, if R ₃ =R ₆ and R ₄ =R ₅ , the absolute values of the differential amplifier 14 and the differential amplifier 18 will be 1+R ₃ /R ₄ +2R ₃ /R ₇ and will be equal. Distortion is most reduced when the gain of the differential amplifier 14 is made equal to the reciprocal of the feedback factor β1 of the amplifier 12 (=1/β1=(R1+R2)/R1). However, since the above calculation uses an approximation formula, it is necessary to make fine adjustments to be exact. This fine adjustment is performed by variable resistor R7. In this way, if a differential amplifier is implemented using two differential amplifiers with feedback elements, negative feedback can be achieved even if the open loop gain of these two differential amplifiers changes slightly due to temperature changes, etc. The gain can be improved by

Combiner 15 of amplifier 12 and differential amplifier 18
Therefore, it is preferable to use a matrix with resistors R ₈ R ₉ so that R ₈ =R ₉ so as not to be affected by load fluctuations. Note that in order to avoid mutual influence between the amplifier 12 and the differential amplifier 18, resistors R ₈ and R ₉ are connected.
The value of can not be made very small, and when used as a power amplification, it is necessary to use an output transformer in order not to reduce the damping factor, so the above-mentioned circuit is suitable for small signal amplification. When multiple amplifiers with different output voltages are connected in parallel in this way, power loss and impedance increase occur at the output, which has been a major obstacle to constructing a feed forward circuit, but according to the circuit below. Solvable.

In other words, the output terminals of amplifiers X and Y, which have internal output resistances Rx and Ry, are connected to impedance elements.
When connecting to a common load via Zx and Zy (here, the output of amplifier X is the signal component and the output of amplifier Y is the distortion canceling component), the output current Ix from amplifier The current also flows through the amplifier Y (the current flowing through the amplifier Y is assumed to be Ixy), and has a power loss of Ixy×Ry due to the internal output resistance Ry of the amplifier Y. Furthermore, the output impedance of the amplifier as seen from the load increases due to the impedance elements Zx and Zy interposed between the load and the amplifier. The power loss is
As (Vx−Vy) increases, Zx, Zy
cannot be made very large. On the other hand, according to this circuit, since the input signals are operated in the same phase and at the same level, the difference between the outputs of the two amplifiers is only the distortion component and the correction for canceling the distortion, so the power loss during synthesis is Less is enough. Conversely, the output impedance of the entire amplifier can be lowered by reducing the values of the impedance elements Zx and Zy required for output synthesis.

FIG. 5 shows another embodiment of the invention, in which the performance of the circuit shown in FIG. 4 is further improved. In FIG. 5, 20 is an input terminal, and 21 is an output terminal. The basic principle of operation is to operate the two amplifiers 25 and 35 with the input signals in phase and at the same level, while canceling out each other's distortion components. Since there is almost no power loss and impedance increase, it is particularly suitable for power amplification.

That is, the input signal is applied to one input terminal of each of the combiners 22 and 32. This synthesizer 22, 3
Distortion components from the paired amplifiers 25 and 35 are mutually input to the other input terminals of the amplifiers 2 and 2. The output of combiner 22 is applied to one input terminal of each of combiners 23 and 24, and the output of combiner 32 is applied to one input terminal of each of combiners 33 and . Feedback components from feedback circuits 26 and 36 are input to the other input terminals of the combiners 23 and 33, respectively. Similarly, combiners 24, 34
The feedback components of the feedback circuits 26 and 36 are also input to the other input terminals of the feedback circuits 26 and 36, respectively. The outputs of the combiners 23 and 33 are then added to the other input terminal of the combiners 32 and 22 as distortion components.
The outputs of the combiners 24 and 34 are sent to amplifiers 25 and 34, respectively.
35, and the output of each amplifier 25, 35 is
They are applied to feedback circuits 26 and 36, respectively, and to one input terminal and the other input terminal of combiner 27.

According to the circuit shown in FIG. 5 above, the amplifiers 25, 35
has some degree of distortion. Now each amplifier 25, 35
The gains of A ₁ and A ₂ are respectively assumed to operate as a power amplifier. Considering the amplifier 25 side, its output includes distortion components reduced to some extent by negative feedback, but after passing through the feedback circuit 26,
The distortion component is _β1 times. This feedback circuit 2
The output of 6 is input to the combiner 23. In the synthesizer 23, signal components are largely removed and a distortion component with an opposite phase is output, which is sent to the synthesizer 32.
is input. Therefore, this distortion component is transmitted to the amplifier 3.
5 to the final stage synthesizer 27,
At this time, they cancel each other out together with their own distortion from the amplifier 25 side. Distortion generated on the amplifier 35 side is also canceled out.

In the above system, the signal input to the combiner 22 from the output of the combiner 33 is x,
If the signal input to the synthesizer 32 from the output of is y, the output of the entire system (output of the synthesizer 27) is A ₁ /1+β ₁ A ₁ (S+x)+A ₂ /1+β ₂ A ₂ (S+y)... ...(1) x=S+y-β ₂ A ₂ /1+β ₂ A ₂ (S+y) = 1/1+β ₂ A ₂ (S+y) ...(2) y=S+x-β ₁ A ₁ /1+β ₁ A ₁ (S+y ) = 1/1 + β ₁ A ₁ (S + x) ... (3) Calculating x and y from equations (2) and (3), x = 2 + β ₁ A ₁ / (1 + β ₁ A ₁ ) (1 + β ₂ A ₂ ) -1
Sy=2+ _β2A2 /(1+ _β1A1 )(1 _{+ β2A2 ) −1}
It becomes S. Substituting these into equation (1), the output of the entire system is 2(A ₁ +A ₂ )+(β ₁ +β ₂ )A ₁ A ₂ /β ₁ A ₁ +β ₂ A ₂ +β
₁ β ₂ A ₁ A ₂ S. Here, if we set β ₁ = β ₂ = β, the output of the entire system is 2 (A ₁ + A ₂ ) + 2 βA ₁ A ₂ / β (A ₁ + A ₂ ) + β ² A ₁ A ₂ S
=2/βS That is, if β ₁ and β ₂ are equal, the output of the system is uniquely determined without being influenced by other parameters. Even if the gains A ₁ and A ₂ of the amplifiers 25 and 35 change, the output remains constant, and the distortion of these amplifiers is completely eliminated.

A more specific example of the circuit shown in Fig. 5 above is shown in Fig. 6.
As shown in the figure. Components corresponding to the same parts as those in the circuit of FIG. 5 will be described with the same reference numerals. In this circuit, the feedback circuit 26 is composed of resistors R ₁₁ and R ₁₂ , and the feedback circuit 36 is composed of resistors R ₁₃ and R ₁₄ . Further, the synthesizer 23 includes a differential amplifier 4
1, 42 and resistors R ₁₇ to R ₂₀ , and the synthesizer 33 consists of differential amplifiers 43 and 44 and resistors R ₂₃ to _{R 26}
It is composed of etc. Furthermore, combiners 24, 3
4 uses differential amplifiers as the amplifiers 25 and 35, and achieves the synthesis function by inputting a feedback signal to their inverting input terminals.

The feedback amplification factor on the amplifier 25 side is β ₁ =R ₁₁ /R ₁₁ +R ₁₂ and the feedback amplification factor on the amplifier 35 side is β ₂ =R ₁₃ /R ₁₃ +R ₁₄ . In order to eliminate distortion, β ₁ = β ₂ , that is,
The following conditions are required: R ₁₁ /R ₁₁ +R ₁₂ =R ₁₃ /R ₁₃ +R ₁₄ . Q ₁ is a FET that constitutes a source follower for impedance conversion, and together with resistors R ₁₅ and R ₁₆ ,
Low output impedance prevents voltage drop. An emitter follower may be used instead of a source follower. To explain the differential amplifiers 41 and 42 side, if we now set the resistances R ₁₇ = R ₁₈ = R ₁₉ = R ₂₀ , the amplitude of the signal input to the positive input terminal of the differential amplifier 42 will be doubled. , the phase is inverted and appears at the output of the differential amplifier 41. On the other hand, the signal input to the positive input terminal of the differential amplifier 41 has twice the amplitude and appears at the output of the differential amplifier 41 with the phase unchanged. That is, the differential amplifiers 41 and 42 perform differential amplification with twice the gain. Therefore, if the values of resistors R ₂₁ and R ₂₂ are made equal, this differentially amplified output is added to the output from the source follower by 1/2. If the gains do not match between the differential amplifiers 41 and 42 due to errors in resistance values, etc.
A variable resistor as shown in FIG. 4 may also be used.
A similar operation can be obtained on the differential amplifiers 43 and 44 side. Protection circuits P ₁ and P ₂ are provided at the outputs of the power amplifiers 25 and 35 to ensure safety. Resistance R ₂₀ ,
A low value (for example, 0.5Ω) is used for R ₃₀ to lower the output impedance of the entire system. In this system, there are two resistor combiners on one side, so the voltage amplification factor is 2/β x 1/2 x 1/2 = 1/2
Becomes β.

FIG. 7 shows an example in which the present invention is applied to a BTL connection, and since the operating principle is similar to the circuit shown in FIG. 5, corresponding parts will be described with the same reference numerals. That is, in this system, a phase inversion circuit 45 is provided between the input terminal 20 and the synthesizer 32. Input signal S, output of amplifier 25 is V ₁ output of amplifier 35 is V ₂ , voltage applied to load 46 is
V ₀ , the signal input from the combiner 33 to 22 is x, and the signal input from the combiner 23 to 32 is y.
The amplification characteristics of this system are expressed as follows: V ₀ =V ₁ −V ₂ =β ₁ A ₁ /1+β ₁ A ₁ (S+x) −A ₂ /1+β ₂ A ₂ (−S+y) ……() x=-(-S+y)+β ₂ A ₂ /1+β ₂ A ₂ (-S+x) ...() y=-(S+x)+β ₁ A ₁ /1+β ₁ A ₁ (S+x) ...() ()() Therefore, solving for x and y, x=2+β ₁ A ₁ /(1+β ₁ A ₁ )(1+β ₂ A ₂ )−1
Sy=2+ _β2A2 /(1+ _β1A1 )( _{1 + β2A2} ) ₋₁
S Substitute these into () to find V ₀ : V ₀ = 2 (A ₁ + A ₂ ) + (β ₁ + β ₂ ) A ₁ A ₂ / β ₁ A ₁ + β ₂ A ₂
+β ₁ β ₂ A ₁ A ₂ S This is similar to the case of the circuit explained in FIG. 5, and if β ₁ =β ₂ =β, then V ₀ =2/βS. FIG. 8 shows an example of a circuit that further embodies this system. This circuit basically corresponds to the circuit configuration shown in FIG. 6, so corresponding parts will be described with the same reference numerals. That is, the phase inversion circuit 45 is
It is composed of FETQ ₂ , Q ₃ , resistors R ₃₁ , R ₃₂ , R ₃₃ and the like. Note that C ₀ and C ₁ are coupling capacitors. This phase inversion circuit 45 sets the gain of FETQ ₂ to -1 with the resistance R ₃₁ =R ₃₂ . Further, a speaker is connected as the load 46. The differential amplifiers constituting the combiners 23 and 33 are the same as the circuit shown in FIG. 6, and have a gain of 2. Resistors R ₂₁ , R ₂₂ , R ₂₇ , and R ₂₈ for signal synthesis all have the same value,
Add 1/2 of each input signal. amplifier 2
If the feedback factor of 5,35 is R ₁₁ /R ₁₁ +R ₁₂ =R ₁₃ /R ₁₃ +R ₁₄ =β, then the voltage amplification factor of this system is 2/β×1/2=1/β.

As described above, according to the present invention, by combining the feedback method and the feedback method without relying completely on the feedback method, which is said to have problems with stability and dynamic characteristics if used extensively, Output distortion components can be significantly reduced,
Further, in the synthesis section at the latter stage of the power amplifier using the feed-forward method, it is possible to provide an amplifier device that can be applied without causing power loss or increasing output impedance.

[Brief explanation of drawings]

1 and 2 are basic configuration diagrams of a conventional amplifier device, FIGS. 3 and 4 are basic configuration diagrams and specific circuit diagrams of an embodiment of the present invention, and FIG.
6 are basic configuration diagrams and specific circuit diagrams of other embodiments of the present invention, and FIGS. 7 and 8 are basic configuration diagrams and specific circuit diagrams showing applied examples of the present invention. 12...Amplifier, 13...Feedback circuit, 14...
Differential amplifier, 15...Synthesizer, 22, 23, 2
4, 32, 33, 34... combiner, 25, 35...
...Amplifier, 26, 36...Feedback circuit.

Claims (1)

[Claims] 1. First and second amplifiers each equipped with negative feedback circuits having substantially equal feedback rates, and a signal that supplies the same signal in phase or in reverse phase to the input terminals of the first and second amplifiers. a first supply means for extracting the distortion component by comparing the input signal of the first amplifier with the signal containing the distortion component fed back to the input side of the amplifier;
and a second comparing means for extracting the distortion component by comparing the input signal of the second amplifier and the signal containing the distortion component fed back to the input side of the amplifier.
means for inputting the output of the first comparing means to a second amplifier, means for inputting the output of the second comparing means to the first amplifier, and a means for inputting the output of the first comparing means to the first amplifier; 1. An amplifier comprising: distortion canceling means for canceling distortion components by supplying the distortion components to the amplifier. 2. The signal supply means supplies the same in-phase signal to the input terminals of the first and second amplifiers, and the distortion cancellation means:
2. The amplifying device according to claim 1, wherein the amplifying device is configured to combine the outputs of the first and second amplifiers and supply the combined output to a load. 3. The signal supply means supplies the same signal of opposite phase to the input terminals of the first and second amplifiers, and the distortion cancellation means:
2. The amplifier device according to claim 1, wherein the amplifier device is configured to supply the outputs of the first and second amplifiers to both ends of the load, respectively.