JPS6138641B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to low frequency amplifiers, particularly push-pull amplifiers, and more particularly to bias circuits for these amplifiers.

In recent years, as audio amplifiers have improved in performance and output, a pure complementary class B push-pull circuit is usually used in the output stage of the audio amplifier. FIG. 1 shows a pure complementary source follower B-class push-pull circuit with a conventional bias circuit. In Figure 1, 1 is an N-channel enhancement type MOSFFH, 2 is a P-channel enhancement type MOSFFT, 3 is a load, 4, 5
is a power supply, 6 is a constant current source, 7 is a diode, 8 is a
NPN transistors, 9 and 10 are resistors, 11 is a coupling capacitor, and 12 is a signal source. The operation shown in FIG. 1 will be briefly explained. The gate potential of the MOSFETs 1 and 2, that is, the collector potential of the NPN transistor 8, is biased by bias resistors 9 and 10 so that the source connection point of the output MOSFETs 1 and 2 becomes zero when there is no signal. Now, when a signal is applied from the signal source 12, the signal is amplified to the opposite phase by the NPN transistor 8, and in the positive half cycle of the signal,
MOSFET2 is in a conductive state, MOSFET1 is in a non-conductive state, and current flows from the load 3 to the power supply 5 through the MOSFET2. In the negative half cycle of the signal
MOSFET1 becomes conductive, MOSFET2 becomes non-conductive, and current flows from power supply 4 to load 3 through MOSFET1. In this way, the load 3 is supplied with a current proportional to the signal. The diode 7 is a bias circuit that allows a no-signal current to flow through the MOSFETs 1 and 2 when there is no signal, and removes crossover distortion.

Generally, the causes of distortion in class B push-pull circuits include:
Distortion can be broadly divided into two types: distortion due to nonlinearity of the input/output characteristics of the amplifier element and distortion due to switching.
Distortion due to nonlinearity of input/output characteristics can be further divided into two types. This is due to the extreme nonlinearity in the small current region near the origin of the input/output characteristics of the general semiconductor amplifying element shown in FIG. 2, and the gradual nonlinearity in the larger current region. The distortion caused by the former is generally called crossover distortion, and its amount is much larger than the distortion caused by the latter. The amount of the latter distortion is small. Moreover, it can be reduced to a practically negligible level by negative feedback in a general negative feedback amplifier.

Here, the crossover distortion will be further explained, and the function of the diode 7 in FIG. 1 will be supplementarily explained. The input/output characteristics of a pure complementary push-pull circuit as shown in FIG. 3 can be obtained by combining the input/output characteristics of FIG. 2. That is, the combined input/output characteristics are as shown in FIG.
When a sine wave is applied as the input voltage, the output current becomes extremely distorted near the crossover where the current is zero, as shown in FIG. This is crossover distortion.
In order to remove this crossover distortion, a constant DC bias is applied to the input in advance as shown in Figure 6, and the combined input/output characteristics pass through the origin of the graph as shown in Figure 7. It should be a straight line. In this way, when a sine wave is added to the input voltage, the output current becomes a sine wave without distortion as shown in FIG. Since a constant current is supplied to diode 7 in FIG. 1 by constant current source 6, a constant voltage is generated between the anode and cathode of diode 7. Therefore diode 7
performs the same function as the DC bias battery in FIG. 6, and serves to eliminate crossover distortion. In this way, crossover distortion is eliminated in the circuit of FIG.

On the other hand, switching distortion is the distortion that occurs when a semiconductor amplification element transitions from a conductive state to a non-conductive state, or vice versa, from a non-conductive state to a conductive state.
This distortion is caused by a time delay when the semiconductor amplifying element changes from a non-conducting state to a conducting state or from a conducting state to a non-conducting state. The frequency components of this distortion consist of higher-order components. For example, at a frequency of 1KHz, the negative feedback amount of a normal negative feedback amplifier is
It is about 40 dB, but at a frequency of 20 KHz it becomes about 15 dB, and as the frequency increases further, the amount of negative feedback decreases. Generally, the switching distortion components for a signal with a frequency of 20KHz are 40KHz, 60KHz, and 80KHz.
Hz, 100KHz, etc., but for the above-mentioned reason, the amount of negative feedback for these is equal to zero, and the distortion reduction effect of negative feedback is greatly impaired. In the bias circuit of the circuit in Figure 1,
It is clear that switching distortion occurs because MOSFETs 1 and 2 alternately repeat conductive and non-conductive states. As is clear from the above, it is extremely difficult to reduce this distortion by negative feedback for high frequency signals. In this way, in conventional class B push-pull circuits, the effect of reducing switching distortion due to negative feedback deteriorates at high frequencies, leading to an increase in the distortion rate. It has the disadvantage of poor sound quality. A class-A push-pull circuit in which the amplifying element is not switched is considered as a solution to the above-mentioned drawbacks of the push-pull circuit, but this has the drawback that the efficiency is significantly lower than that of the class-B push-pull circuit.

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the above-mentioned drawbacks of the prior art and to provide a bias circuit that prevents switching distortion in a class B push-pull circuit without impairing the efficiency of the prior art.

In order to achieve the above-mentioned object, the present invention always prevents the occurrence of switching distortion caused by the transition between conduction and non-conduction in the semiconductor amplification element constituting the B-class push-pull circuit. The feature is that a field effect transistor with constant current characteristics is inserted as a bias circuit between the input terminal of the semiconductor amplification element and the ground terminal in order to keep the semiconductor amplification element in a conductive state.

FIG. 9 shows an embodiment of the present invention. In the figure, the same reference numerals as in FIG. 1 indicate the same parts, 13 is a P-channel enhancement type MOSFET, and 14 is an N-channel enhancement type MOSFET. P
Channel and N-channel enhancement type
MOSFETs 13 and 14 constitute a bias circuit according to the present invention. The operation of this bias circuit will be explained below. In FIG. 9, the source electrode and gate electrode of P-channel enhancement type MOSFET 13 are connected to the gate electrode and source electrode of N-channel enhancement type MOSFET 1 constituting the output stage, and the source electrode and gate electrode of N-channel enhancement type MOSFET 14 are P channel enhancement type that constitutes the output stage
Connected to the gate electrode and source electrode of MOSFET2, respectively. In addition, P channel and N channel enhancement types that make up the bias circuit
The drain electrodes of MOSFETs 13 and 14 are connected to each other. A bias circuit like this
The P-channel and N-channel enhancement type MOSFETs 1 and 2, which are connected between the gate electrodes of the MOSFETs 1 and 2 and constitute an output stage, are driven by an NPN transistor 8 with a constant current source 6 as a load. In this way, the P-channel and N-channel enhancement type MOSFETs 13 and 14 are biased to a constant current by the constant current source 6. As is well known, the general enhancement type MOSFET
As shown in Fig. 10, the output characteristics are divided into the triode characteristic region (on the left side of the broken line in the figure) and the saturation characteristic region (on the right side of the broken line) by the boundary indicated by the broken line.
It can be divided into These areas generally have the following relationship. In the triode _{characteristic} region, _equation (1 ₎ is established, _and in the saturation characteristic region, equation ( ² ₎ is established. Also, the broken line boundary I _D =-β/2 (V _GS -V _th ) ² (2) is expressed by equation (3). Here, I _D is the drain current V _DS = V _GS − V _th (3) current, V _GS is the gate-source voltage, V _DS is the drain-source voltage, V _th is the threshold voltage, and β is a constant. be. In addition, in the above equation, these values are V _GS −V _th >0 for an N-channel enhancement type device:
V _th >0; V _GS >0; β <0; V _D , I _D > 0, while for P-channel enhancement type devices V _GS −V _th <0; V _th <0; V _GS <0;>0;
V _D , _ID <0. Consider the relationship between V _GS and V _DS when an enhancement MOSFET with such characteristics is biased with a constant current I _Bias . For simplicity, let V _GS ≫V _th and V _th ≒0, as shown in FIG. 11. The characteristics in Figure 11 are the first
This can be easily inferred from Figure 0. In Figure 10, assuming a constant current as I _(Bias) , V _D
When _S > V _GS , V _GS is V _{GS (Bias)} regardless of V _DS
becomes a constant value. This is shown on the right from point A in Figure 11. Furthermore, in FIG. 10, if V _DS <V _GS , then V _GS must become large in order to allow I _(Bias) to flow. This shows the left side of point A in FIG. These relationships can also be determined from equations (1) and (2).

In order to understand the operation of the bias circuit shown in FIG.
have the same electrical characteristics (β and V in equations (1) and (2)
It is assumed that _th is the same value and V _DS and _ID have different signs. Furthermore, the N-channel and P-channel enhancement type MOSFETs 1 and 2 that constitute the source follower B-class push-pull circuit have the same electrical characteristics, and the P-channel and N-channel enhancement type MOSFETs 13 and 2 for the bias circuit, respectively, have the same electrical characteristics.
It is assumed that the area ratio of the semiconductor substrate is different from 14. In other words, β in equations (1) and (2) differs only in area ratio, and under certain same bias conditions—for example, gate
Source voltage V _GS = V _{GS (Bias)} for bias circuit
If the drain current of the MOSFET is 1mA, and the area of the output stage MOSFET is 100 times larger, a drain current of 100mA will flow through the output stage MOSFET. Further, in FIG. 9, if the bias current I _BIAS from the constant current source is 1 mA, then when there is no signal, between the drain and source of the bias circuit MOSFETs 13 and 14, as shown in FIG. 11, V _DS = V _{GS ( Bias)} voltage is high. Therefore, there is a voltage of 2.|V _G between the sources of MOSFETs 13 and 14.
_{S (Bias)} | voltage rises, each MOSFET 1, 2
A voltage of V _{GS (Bias)} is applied between the gate and source of. Therefore, for this bias circuit
Gate-source voltage V _G of MOSFETs 13 and 14
_{S (Bias)} biases the gates and sources of the output stage MOSFETs 1 and 2 with a voltage of V _{GS (Bias)} . And since the output stage is based on the previous assumption that the area ratio is 1:100 and the characteristics are equal, I _BIA
If _S is 1 mA, a no-signal current of 100 mA flows through the output stage MOSFETs 1 and 2. However, this current does not flow to the load 3. And this V _GS(Bias
) has the same function as the diode 7 in FIG. 1, and functions to eliminate crossover distortion of the class B push-pull circuit.

Next, consider the sine wave shown in FIG. 12a as a signal. The collector potential of the NPN transistor 8 changes in proportion to the signal and in opposite phase as shown in FIG. 12b. When the collector potential of the NPN transistor 8 rises, the source potentials of the output stage MOSFETs 1 and 2 also rise from zero by approximately the same amount. This voltage increase is due to an increase in the gate-source voltage of the output stage MOSFET 1, in other words, an increase in the gate-source voltage of the bias MOSFET 13, and an increase in the current flowing from the power supply 4 to the load 3. In other words, at this time, the gate-source voltage of the bias MOSFET 13 is V _{GS (
Bias)} . Therefore, as can be seen from FIG. 11, the drain-source voltage of the bias MOSFET 13 decreases. for screwdriver
The voltage between the source electrodes of MOSFETs 13 and 14 is
Since it is equal to the voltage between the gate electrodes of output stage MOSFETs 1 and 2, the voltage decrease between the drain and source of MOSFET 13 for bias described above is for bias.
This is achieved by increasing the drain-source voltage of MOSFET 14. At this time, for the driver
The drain-source voltage of MOSFET14 is V _GS(
However, as shown in _Figure 11, the voltage between the gate and source remains fixed at V _{GS (Bias)} , and this voltage causes the output stage to
The gate and source of MOSFET2 are biased. Therefore, a constant current of 100 mA, which is the same as when there is no signal, flows from the power source 5 through the load 3 and the MOSFET 2.

On the other hand, when the collector potential of NPN transistor 8 falls, the output stage MOSFET is
The source potentials of 1 and 2 fall below zero. This voltage drop is caused by an increase in the gate-source voltage of the output stage MOSFET 2, causing current to flow from the power source 5 through the load 3. At this time, the bias MOSFET
The gate-source voltage of No. 14 is greater than V _{GS (Bias)} , and the drain-source voltage is V _{GS (Bias).
)} . This decrease is for the same bias
The drain-source voltage of MOSFET13 is V _GS(
Bias) . At this time, the voltage between the drain and source of the bias MOSFET 13 is higher than V _{GS (Bias)} , the voltage between the gate and source remains fixed at V _{GS (Bias)} , and this voltage causes the gate and source of the output stage MOSFET 1 to A bias is applied between the sources, and a constant current (100 mA) is applied to the load 3 from the power source 4.

Figure 12c shows each MOSFET 1, 2, 1.
3, 14 gate-source voltage (curve V _GS-1,
13 ; V _GS-2,14 ) and the drain-source voltage (curve V _DS-13 ; V _DS-14 ). Also the first
FIG. 2d shows the relationship between the drain currents _ID-1 and _ID-2 of the output stage MOSFETs 1 and 2 and the current I _R flowing through the load. As is clear from FIG. 12d, according to the present invention, the output stage MOSFETs 1 and 2 are always in a conductive state, and switching distortion due to repeated conduction and non-conduction does not occur, and the Sound quality will be improved. Also for output stage
If the current value due to V _{GS (Bias)} of MOSFETs 1 and 2 is selected to be the same as the non-signal current of a normal class B push-pull circuit, the efficiency of the circuit is the same as that of a normal class B push-pull circuit.

FIG. 13 shows another embodiment of the present invention. In the operation explanation of FIG. 9, V _th is 0, but if V _th ≠ 0, a voltage of 2 {|V _GS(Bias) |+|V _th |} is applied between the source electrodes of the bias MOSFETs 13 and 14. It's rickety. Therefore, when there is no signal, the output stage
Between the gate electrodes of MOSFETs 1 and 2, there is a voltage of 2｜V _GS(
Bias) |+|V _th |} voltage rises, and a voltage of V _GS(Bias) rises between each gate and source electrode, 2{|V _GS(Bias) |+|V _th |}=2 ｜VGS _(Bi
as) | An inconvenient event occurs. FIG. 13 solves this problem. In this figure, the same symbols as in FIG. 9 indicate the same parts, and 15 and 16 are resistors. The voltage of I p R, which is the product of the drain current I _p flowing due to V _{GS (Bias)} between the gate and source of output stage MOSFETs 1 and 2 and the resistance value _R of resistors 15 and 16, becomes equal to V _th . It is clear that if the value of R is set as such, the above-mentioned disadvantage will not occur. Since the operation in FIG. 13 is the same as that in FIG. 9, the explanation will be omitted.

In Figure 9, the means for adjusting the no-signal current of output stage MOSFETs 1 and 2 is for bias.
What is necessary is to change the current value of the constant current source 6 that drives the MOSFETs 13 and 14. That is, the value of I _BIAS in FIG. 10 may be changed to change the value of V _{GS (Bias)} , and the gate-source bias voltage V _{GS (Bias)} of the output stage MOSFETs 1 and 2 may be changed.
FIG. 14 shows another embodiment of the present invention in which the non-signal current of the output stage MOSFETs 1 and 2 can be adjusted. In this figure, the same reference numerals as in FIGS. 9 and 13 indicate the same components, and 17 and 18 are resistors. Resistors 17 and 18 are bias MOSFET 1
The voltage drop caused by the current flowing through MOSFETs 3 and 14 applies a bias to the substrates of bias MOSFETs 13 and 14, equivalently changing the static characteristics. For this reason, when the same MOSFET as in FIG. 9 is used, the voltage appearing between the gate and source will be different from V _{GS (Bias)} even if the constant current source current I _BIAS is the same. Therefore the output stage MOSFET
The bias voltages between the gates 1 and 2 are also different, and a non-signal current different from that shown in FIG. 9 can flow. In order to continuously change this no-signal current, a resistor of 1
All you have to do is change 7 and 18 to variable resistors. The operation in FIG. 14 is also the same as that in FIG. 9, so the explanation will be omitted.

In each of the above embodiments, the case where an enhancement type MOSFET is used in the output stage has been explained, but if an enhancement type MOSFET with saturation characteristics is used as the bias MOSFET, it is possible to use the same enhancement type bipolar transistor in the output stage. It is clear that the same effect can be obtained. In this case, the bipolar transistor in the output stage must be connected to Darlington to increase the input impedance. Also, the output stage is of depreciation type.
If MOSFET is used, bias MOSFET
depletion type with the same saturation characteristics as
It is also clear that the same effect can be achieved using MOSFETs. Furthermore, when an element that does not have saturation characteristics is used in the output stage, a bias element that has saturation characteristics may be used. In other words, an element that has the same input characteristics as the output stage element and has saturation characteristics in its output characteristics may be used as the biasing element.

As described above, according to the present invention, switching distortion can be removed without impairing the efficiency of the conventional B-class push-pull circuit, and the sound quality at high frequencies can be significantly improved.

[Brief explanation of the drawing]

Figure 1 shows an example of a conventional class B push-pull circuit, Figure 2 shows an input/output characteristic diagram of a MOSFET, Figure 3 shows an example of pure complementary connection using MOSFETs, and Figure 4 shows an example of a pure complementary connection using MOSFETs. Figure 5 is the output current waveform diagram of Figure 3, Figure 6 is an example of a circuit with a conventional pure complementary connection in which a bias power supply is inserted to remove crossover distortion, Figure 7 is the input/output characteristic diagram. is an input/output characteristic diagram of FIG. 6, FIG. 8 is an output current waveform diagram of FIG. 6, FIGS. 9, 13, and 14 are circuit diagrams of the embodiment of the present invention, and FIG.
MOSFET output characteristic diagram, Figure 11 is the same as Figure 10.
Gate voltage when MOSFET is driven by constant current source
FIG. 12 is a diagram showing the relationship between source voltage and drain-source voltage, and is a diagram explaining the operation of the circuit of FIG. 9, which is an embodiment of the present invention. 1, 2... Output stage MOSFET, 3... Load, 6... Constant current source, 12... Input signal source, 13, 14... Bias MOSFET.

Claims (1)

[Claims] 1. A first field effect transistor having a gate electrode, a drain electrode, and a source electrode; and a second field effect transistor having complementary characteristics to the first field effect transistor; 1st and 2nd
It has an output stage circuit in which the source electrodes of field effect transistors are connected to each other, a gate electrode, a drain electrode, and a source electrode, and the source-grounded transfer characteristic is only the first field effect transistor and drain current value of the output stage circuit. and a fourth field effect transistor having complementary characteristics to the third field effect transistor; A bias circuit connecting the drain electrodes of the third and fourth field effect transistors, a constant current source connected to the bias circuit, and a connection between the first and second field effect transistors of the output stage circuit. the gate electrodes of the first and second field effect transistors of the output stage circuit and the source electrodes of the third and fourth field effect transistors of the biasing circuit, respectively; the source electrodes of the first and second field effect transistors of the output stage circuit and the third field effect transistor of the bias circuit;
and the gate electrode of the fourth field effect transistor, respectively, and a signal source is connected to the source electrode of one of the field effect transistors of the bias circuit, so that even when there is no signal, a predetermined amount of signal is sent to the output stage circuit. A push-pull amplifier characterized in that it is configured to allow a non-signal current to flow. 2 By inserting a voltage drop element between each source electrode of the first and second field effect transistors of the output stage circuit and the load, the threshold values of the third and fourth field effect transistors of the bias circuit can be adjusted. A push-pull amplifier according to claim 1, characterized in that the voltage is compensated. 3. Each field effect transistor is of an insulated gate type, and the substrate electrodes of the third and fourth field effect transistors of the bias circuit are connected to the gate electrodes of the first and second field effect transistors of the output stage circuit, respectively. By inserting voltage drop elements between the source electrodes of the third and fourth field effect transistors of the bias circuit and the gate electrodes of the first and second field effect transistors of the output stage circuit, respectively, the output 3. The push-pull amplifier according to claim 1, wherein the push-pull amplifier controls the value of the non-signal current flowing through the stage circuit.