JPS5912206B2 - Direct connection type B class amplifier - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is a low-level, low-power direct-coupled class 8 IC amplifier, and more specifically, a low-level IC that is ideal for use in hearing aids with extremely low idle link current levels and low distortion levels. This relates to a low-power direct-coupled class 8 amplifier.

Class 8 amplifiers used in hearing aids and the like typically include a three-stage direct-coupled preamplifier connected to a transistor output stage.

Distortion has long been a problem with this amplifier.

The main cause of distortion is that when current flows through a transistor, the gain of that transistor increases.

When a suitable input signal enters the amplifier and the current flowing through the output transistor increases, the gain of that output transistor also increases.

As the current increases, the gain increase can reach as much as 500 times.

Unfortunately, the decrease in gain of the pre-AE transistor, the three-stage preamplifier transistor, is less than the increase in gain of the output transistor (due to the difference in the resistive loads connected to the transistors in the direct-coupled amplifier species). Therefore, the overall gain of the amplifier increases with the magnitude of the current.

This causes waveform distortion on the output side.

In order to solve this distortion problem, conventionally the output transistor is biased with a relatively bi-level idle link current to reduce the change in the output transistor's current, that is, the change in the gain.

In conventional class 8 amplifiers such as those shown in Canadian Patent Nos. 811,844 and 844,156, an idling current of 200 μA is used for each of the two output amplifiers.

When used in a hearing aid or the like that uses one extremely small battery as a power source, it is desirable that the idle link current be small in order to extend the life of the battery.

Another problem with low-level direct-coupled Class 8 amplifiers for hearing aids is that the load of the amplifier is a hearing aid receiver whose impedance changes with frequency.

Generally, the impedance of a hearing aid receiver increases as the frequency increases.

The gain of the output transistor of a conventional class 8 amplifier increases approximately in proportion to the load impedance.

Therefore, the harmonics of the amplified signal are amplified more than the fundamental wave, resulting in even greater distortion.

In view of the above circumstances, an object of the present invention is to provide a low-level, low-power class 8 amplifier using an IC that has a low level of idle link current and a low level of distortion.

According to the present invention, there is provided a two-channel amplifier in which the collector of each output transistor is connected to a first-stage preamplifier transistor via a feedback loop made up of only hollow resistors.

This provides AC negative feedback that can produce extremely low distortion, allowing the amplifier to operate with an idle current of approximately 50 μA; The distortion is smaller than that of conventional class 8 amplifiers.

The resistive portion of each AC feedback loop connected to the collector of its output transistor is electrically isolated from the power supply voltage such that the collector of that output transistor can swing more than 0.6 volts beyond the power supply voltage.

Additionally, common mode distortion forces are eliminated by connecting the emitters of at least one pair (preferably at least two pairs) of preamplifier transistors together through a resistor to ground.

Embodiments of the present invention will be described in detail below with reference to the drawings.

In FIG. 1, a low-level, low-power IC direct-coupled class 8 amplifier 2 according to an embodiment of the present invention has an input signal (Vin) source 4. In FIG.

This input signal source 4 is typically a double-ended hearing aid microphone, but it can also be a microphone with a single-ended output connected to a phase splitter that converts the single-ended microphone signal into a double-ended signal. good.

The amplifier 2 of this embodiment has exactly the same two channels 8 and 1.
It has a two-channel preamplifier 6 with 0.

The channels 8 and 10 each amplify one half of the input signal Vin.

The signals from the preamplifier 6 are further amplified at the output stage 12, combined at the transformer 14, and supplied to the load 16.

Next, the channel 8 of the preamplifier and the upper half of the output stage 12 (7) will be explained in more detail.

In addition, another channel 10 of the preamplifier and output stage 1
The lower half of the channel 2 has exactly the same structure as the channel 8 and the upper half, respectively, so a detailed explanation will be omitted.

Channel 8 of the preamplifier each has a transistor Q1
, Q2, Q3 (all NPN in the figure) are shown as transistors.

).

(This preamplifier always has an odd number of stages.

) One end of the signal source 4 is connected to the base of the first stage transistor Q1 via a capacitor C2, and the emitter of this transistor Q1 is connected to ground via a resistor R2.

The collector of this transistor Q1 is connected to the positive electrode vb of the power supply via resistors R3, R, and further connected to the base of the second stage transistor Q2.

The collector of this second-stage transistor Q2 is connected to the collector of the first-stage transistor Q1 via a capacitor C3 and a resistor R12, and is connected to the positive electrode V of the power supply via a resistor R7.
b, and further connected to the base of the third stage transistor Q3.

Further, the emitter of the transistor Q2 is connected to ground via a resistor R5 and a collector emitter circuit of the transistor Q5, which are parallel to each other.

The emitter of the third stage transistor Q3 is connected to ground via a resistor R6.

The collector of this transistor Q3 is a resistor RB t R
It is connected to the positive electrode of the power supply via IQ, and further connected to the base of transistor Q4 forming the upper half of the output stage.

The emitter of this transistor Q4 is connected to ground, and the collector is connected to one end of a primary coil 18 having a center tap of the transformer 14.

(The center tap of this primary coil 18 is connected to the positive electrode of the power source.

) The secondary coil 20 of the transformer 14 is connected to the load 16.

Similar to conventional class 8 amplifiers, when the signal Vin is differentially input to the inputs of the preamplifier (the bases of transistors Q1 and Q1'), channels 8 and 10 are 18
[d] increases the out-of-phase single-ended signal.

The amplified signal is then applied to the input of output stage 12, ie the base of output transistor Q4 tQ4'.

As a result, the output transistor Q4 tQ+' is turned ON and OFF' alternately.

In other words, either output transistor Q+ or Qa' is ON during either the upper or lower half of the input signal cycle.
However, during that time, the other transistor is OFF'.
Conversely, during the other half of the input signal cycle, the other transistor is ON; during that time, said one transistor is OFF.
I'm on FF.

The half-wave signals appearing at the collectors of the output transistors Q4 tQ*' are differentially combined by the transformer 14 to re-form a complete sine wave across the load 16.

An important characteristic of the output stage of a class 8 amplifier is that it can output high-power signals, yet remains idle at extremely low quiescent currents, an important characteristic when used in hearing aids. .

In order to control the idling current of the output stage transistor Q4 yQa', D is used to keep the base voltage of the output stage transistor ChsQ4' constant as in the conventional case.
A C negative feedback circuit is provided.

The DC negative feedback circuit consists of a connection including a resistor from the output of the preamplifier back to the base of the first stage transistor of the preamplifier.

That is, the connection is from the connection between resistor R8 and RIO to the base of transistor Q1 via resistor R9°R1.

AC by decoupling by capacitor 01
Signals are prevented from returning through this connection.

This capacitor C1 shorts all AC signals to ground.

The operation of this DC negative feedback circuit is well known in the art and is explained in detail in the aforementioned Canadian Patent No. 84.4156.

As mentioned above, a major drawback of conventional class B amplifiers, such as those shown in Canadian Patent No. 844,156, is distortion due to the extremely large amplitude on the output side.

This is due to the following reasons.

That is, when the transistor Q4 is turned on, the current flowing through the transistor Q4 is a DC idle link current (this may be about 50 μA).

) to a peak current as high as 25 mA, an increase of 500 times.

Since the gain of transistor Q4 is proportional to the current flowing through it, its gain also increases by a factor of 500.

The load seen by transistor Q3 is transistor Q4 and resistor R8 connected in parallel with each other.

It is RIQ.

(Resistance R0 is extremely large and can be ignored in a first-order approximation.

) The input impedance presented by transistor Q, is inversely proportional to the current flowing through transistor Q, and decreases by a factor of 500 on the same order of magnitude as that current increases.

However, due to the effect of resistor Rgt R10, the load on transistor Q3 is reduced not by a factor of 500, but by a much smaller percentage.

That is, transistor Q4 causes transistor Q3 to
The load RL3 of the transistor Q3 is expressed as Rin and Q4.

As is clear from this equation, due to the presence of the resistor RgyRtg, the load RL3 of the transistor Q3 is not proportional to the input impedance Rl n of the transistor Q4, Q4.

When the input impedance of transistor Q4 decreases (
In other words, when the current flowing through transistor Q4 increases)
The load on transistor Q3 decreases at a rate less than the decrease in the input impedance of transistor Q4.

Since the gain of transistor Q is proportional to the load of transistor Q3 up to a first order approximation, the gain of transistor Q3 does not decrease by the root of the increase in the gain of transistor Q4.

This causes distortion because large signals are amplified more than small signals.

Therefore, the waveform of the signal supplied to the load 16 will be distorted.

This problem of distortion becomes even more serious when the load consisting of the transformer 14 and resistor 16 is a transducer (receiver) for a hearing aid.

The gains of the output stage transistors Q4 and Q4' are approximately proportional to the load impedance, and the impedance of the receiver increases as the frequency increases, so harmonics resulting from distortion are amplified to a greater extent than in the basic board.

Therefore, distortion will be emphasized.

Conventional class B amplifiers usually attempt to solve this problem by increasing the idle link current of the output stage (to reduce the rate of change of current in the output transistor).

This idle link current is typically 200 μA or more per output transistor.

However, even with such relatively large idle link currents, acoustic distortion in hearing aids using conventional class B amplifiers was typically greater than 10%.

By using a novel circuit, the present invention is capable of reducing distortion even when the idle link current of the output stage transistors is reduced (up to about 50 μA for each output transistor) compared to a conventional class B amplifier with an idling current of 200 μA. This is what I did.

In fact, the distortion of a prototype hearing aid class 8 amplifier assembled in accordance with the present invention was approximately 2%.

This is comparable to the performance of a conventional class 8 amplifier for hearing aids that uses an idling current of 500 μA.

One of the important features of the present invention is the provision of a novel AC negative feedback circuit.

That is, the collector of each output transistor Q4, Q+' is connected by a resistor, that is, resistor R1□, R3(R1
□′, first stage transistor Q1 tQ via R3
It is connected to the collector circuit of t'.

It has been discovered that this AC negative feedback circuit linearizes the gain of the amplifier and reduces distortion due to output-dependent nonlinearity.

This feedback reduces the influence of load impedance on the gain of the amplifier, and reduces emphasis on distortion when a load whose impedance changes depending on frequency, such as a receiver, is used.

When the open loop gain of transistors Q2 and Q3.94 stages multiplied by the ratio R11/R of resistors R1 and R4 is greater than 1, the dependence of the gain of the circuit consisting of the preamplifier and output stage on the output stage current is extremely low. Become.

As mentioned above, in the class 8 amplifier of the present invention, even if the idle link current for each output transistor is approximately 50 μA, the distortion is only 2.
%, which is an acceptable value.

The AC feedback loop through resistor all is not connected to the base of transistor Q1.

This is because if the loop is connected to the base of the transistor Q1, it will interfere with the aforementioned DC feedback loop and lower the input impedance of the trough 291 stage.

If the input impedance of the trough 291 stage decreases, the gain decreases and tone control becomes difficult, which is not desirable for use in a hearing aid.

In addition, it must be connected to the first stage of the preamplifier via a resistor R1□ and connected to the base of the transistor Q1' to form a negative feedback circuit (this is known as cross-coupling).

). In this case, if the transducer does not function as a good transformer, its AC feedback loop may be incompletely formed, resulting in unexpected gain and distortion in the circuit. .

A compensation circuit consisting of a resistor R1□ and a capacitor 03 is provided to stabilize the amplifier having the two feedback circuits.

The illustrated amplifier is constructed as an integrated circuit (IC) on a P-type substrate (depending on the type of transistors illustrated).

In such integrated circuits, the resistor is typically formed as a diffused P-type layer in an epitaxial N-shaded region.

The epitaxial region is typically biased at a DC power supply voltage so that the diffusion-insulator diode does not conduct.

However, in the circuit of the present invention, the collector voltages of the output transistors Qa and Q4' may swing above and below the power supply voltage due to the action of the transformer.

If the voltage of the diffusion layer becomes 0.6 V higher than the DC power supply voltage, the diffusion layer-insulating layer diode of the resistors R□I and R11 becomes conductive, and the output is short-circuited.

In order to prevent this, an unprecedented arrangement as shown in FIG. 2 is used when forming the resistor R11°R1□'.

In FIG. 2, the P-type substrate 30 on which the circuit is formed is conventionally grounded, and resistor R11 is shown as P-type material 32 in the epitaxial N-shaded region 34.

Further, both contacts of the resistor all are indicated by 36.38.

Furthermore, no contact is provided in the region 34, and the region 34 is kept in a floating state.

In this way, whether the voltage at the contacts of resistor all is above or below the supply voltage, the diode formed by the P-type material 32 and the N-shaded region 34 will not conduct and short-circuit the output. There is.

The resistors R1□ and R11' may also be insulated by other methods, such as using thick or thin film deposition techniques or using separate resistors.

The resistors R11tR11' are insulated both from the power supply voltage and from the substrate 30, and further insulated from each other (both are located on different and independent floating regions 34).

), the other resistors of the amplifier 2 can be placed in one region biased with the supply voltage, if desired.

This is shown in FIG.

In another area 50 in FIG. 2, a resistor 52 with contacts 54 is provided.

This resistor 52 may be any resistor shown in FIG. 1 other than the resistor R1□ and the resistor R1
, R1□' may be included.

Region 50 is connected to a power supply via contact 56 and is biased at power supply voltage vb.

Generally, the region in which the resistor is located must be biased with a power supply voltage to provide a reverse bias leakage current between region 50 and substrate 300, thereby providing the reverse bias leakage current from the resistor in that region. It is desirable to avoid this.

If the leakage current is supplied from a resistor, a PNP transistor action will occur, causing excessive current to flow from the resistor to the substrate.

However, if the voltage across the resistor becomes greater than 0°6 above the supply voltage (as in the case of resistors R11, R1□'), its undesirable transistor action becomes unavoidable.

To this end, the area for the resistor R1□JR11' is isolated from the power supply, and by making the area 34 as small as possible, the current gain of its undesired transistor action is minimized as much as possible.

As shown in FIG. 2A, the boundaries of region 34 are brought into close proximity to material 32 forming resistor R1□.

In other words, the contact area between the region 34 and the substrate 30 is made as small as possible.

Another feature of the amplifier of the present invention is the elimination of common mode distortion.

Eliminating common mode distortion is particularly important in hearing aids since hearing aid power supplies typically have an output impedance of 2 to 15 ohms.

The output impedance of this power supply may generate a signal on the power supply line that is comparable in magnitude to the signal to be amplified.

Signals on this power supply line easily enter the normal signal path through the preamplifier collector circuit, causing distortion.

The signal of this power line should be inputted.
) It is important that the amplifier distinguish between the common mode signal and the difference signal so that the gain of the common mode signal is less than the gain of the difference signal.

Common mode distortion rejection (CMR) can be measured by the ratio of the gain of the difference signal and the gain of the common mode signal.

Common mode distortion is eliminated by transistor Qa tQ3'
This is done in the third stage of the preamplifier by connecting the emitters of 2 to 40 at a node 40, and connecting that node 40 to ground via a resistor R6.

According to such a configuration, when the same signal in the positive direction appears at the bases of both transistors Q3, Q3', both transistors conduct better and bring their emitters to a positive potential.

Therefore, in the end, the base-emitter voltages of transistors Qa, Qs' do not change at all or very little (to a first order approximation) and the common mode signal is not amplified at all or very little.

However, if the differential mode signal is
, Q3', for example, a positive-going signal appears at the base of transistor Q3, and when a corresponding negative-going signal appears at the base of transistor Q31, the positive-going voltage of transistor Q3 increases. Since the change in is canceled by the change in the voltage of transistor Qa' in the negative direction, the change in the voltage difference at the connection point 40 is extremely small.

Therefore, the change in the base-emitter voltage of transistor Q3 increases, and the positive-going signal appearing at the base of transistor Q3 is amplified.

Thus, in this stage, the common mode signal is not amplified as much as the differential mode signal, ie, the common mode signal is eliminated.

On the other hand, in the transistor Q1.911 stage, common mode distortion is not eliminated.

This is because although the resistor R2t R2' reduces the gain given to the common mode signal, it also reduces the gain given to the differential mode signal.

In the second stage of the preamplifier consisting of transistors Q2 and Q2', common mode distortion is eliminated by transistor Q5 and resistor R5.

The collector emitter circuit of transistor Q5 is resistor R5.
are connected in parallel.

Transistor Q5 is biased to act as a power supply by transistor Q6, which is used in diode fashion.

The transistor Q6 is the transistor Q5 as shown.
, and its collector is further connected to the positive electrode vb of the power supply via a resistor R13, so that the current flowing through transistor Q6 sets the current flowing through transistor Q5.

It would be ideal if the resistor R5 was not used and the connection point 42 was grounded only by the transistor Q to eliminate common mode distortion (because the power supply by the transistor Q5 theoretically has infinite impedance). However, in this case, the voltage at the connection point 42 is freely determined by the power supply, and can be any voltage.

If this happens, the DC gain provided by the second stage transistors Q2 and Q2' will be destroyed, and the resistors R9 and
The regulation effect provided by the DC feedback loop consisting of R1 is destroyed.

For this purpose, it is necessary to connect the connection point 42 to ground through a finite resistor to control the emitter voltage of the transistors Q2 and Q2', and therefore it is necessary to connect the resistor R5 in parallel with the transistor Q. be.

Furthermore, even if a small resistor is inserted between the emitters of transistors Qa and Qs' and connection point 40, and between the emitter of transistors Q21Q2' and connection point 420, the common mode distortion elimination function of those stages will hardly decrease. .

Next, FIG. 3, which shows a circuit very similar to the circuit of the embodiment shown in FIG. 1, will be described.

In FIG. 3, the same parts as in FIG. 1 are given the same numbers.

The circuit of FIG. 3 is designed to have a smaller gain than the circuit of FIG. 1, and therefore the transistors Q2 and Q2' of FIG. 3 are PNP transistors.

The current flowing through the emitter-collector circuit of transistors Q2 and Q2' is actually transistor Q31Q3'.
Since the base current is , the gain and power consumption of the circuit shown in FIG. 3 are smaller than those of the circuit shown in FIG.

In this circuit, the emitter of transistor Q3 and the emitter of transistor Q3' are connected and grounded via resistor R4, and similarly transistor Q1
Common mode distortion is eliminated by connecting the emitter of transistor Q 1/ to the emitter of transistor Q 1/ and also grounding them through resistor R14.

In the circuit of Figure 3, capacitor C3.

03' requires a fairly large charging current to follow the signal.

In some cases, the collector current of transistor Q2 is small and cannot cover the appropriate charging current.
This will cause distortion.

Therefore, as shown by the dashed line, transistor Q2
, Q2' may be connected to ground through resistors R209R20' to increase the current in the Q2' stage through transistors Q2 and Q2'.

Further, the collector of the transistor Q2 and the collector of the transistor Q2' are connected by the resistor R21, thereby compensating for the increase in gain caused by the increase in the collector current of the transistor Q2jQ2' due to the addition of the resistors R29 and R2. You may also do so.

In addition, if necessary, between the collector of transistor Q3 and the base of transistor Q4 and between the collector of transistor Q3' and the base of transistor Q4, as shown in dashed lines.
Resistors R2□ and R22' between the bases of '
may be provided to make the operation of transistor Q4 more linear by reducing the change in the input impedance of transistor Q4 tQ4' due to current.

Except for the changes mentioned above, the circuit of FIG. 3 is substantially the same as the circuit of FIG. 1.

The circuit shown in FIG. 4 is also substantially the same as the circuit shown in FIG. 1, and in FIG. 4, parts corresponding to those in FIG. 1 are given the same numbers.

The difference between the circuit of FIG. 1 and the circuit of FIG. 4 is that transistor Q5 and resistor R5 are eliminated in the circuit of FIG.

Therefore, the second stage of the preamplifier does not have a function to eliminate common mode distortion.

Instead, a common mode distortion elimination function is provided to the first stage by connecting the emitter of transistor Q1 and the emitter of transistor Q1' and connecting to ground via resistor R6.

Furthermore, compensation resistance R? 2 has also been deleted.

The typical values of each element in the first circuit are shown in Table 3 below.

Table ■ Element Value R, R1' 27 KΩ R2, R2' Ω at 2.3 R, R3' 52 KΩ R4, R4' 120Ω Ω at R510 R6, R6' Ω at 1°5 Ω R7, R7' at 24 R3, R3' 400Ω Rg, Rg' 47 KΩ RIOt, O' 9. IKΩR1□t
Ω to R1□'15 Ω to R1□tR1□' 8゜2 Ω to R1336 Load resistance 16 1 KΩ C,01' 10 μf c2 o6o3μfC3, C3
'100 p fVb
1゜55VQ1, Qt', Q2)
Q2" Heavy industry mitter's Qa, Qs,', Q5, Q
a NPN) transistor Q4 a Q4'
Triple emitter NPN) transistor transformer 14 With center tap and 1:1 winding ratio.As shown in Table ■, the resistance value of resistor R1 is smaller than that of resistors R3 and R1+.

This reduces the influence of the AC feedback loop containing the resistor all on the prevailing C state in the amplifier, and helps maintain the stability of the amplifier.

Also, compensation circuits other than connecting capacitor C3 between the base of transistor Q2 and the collector of transistor Q3 could be used.

For example, a series-connected compensation capacitor and resistor may be inserted between the base of transistor Q3 and ground.

[Brief explanation of the drawing]

Fig. 1 is a circuit diagram of an amplifier according to an embodiment of the present invention, Fig. 2 is a diagram for explaining a method of insulating an AC feedback resistor from a power supply voltage, and Fig. 2A is a circuit diagram of an amplifier used in the embodiment of Fig. 1. 3 and 4 are circuit diagrams of other embodiments, respectively. 4...Input signal source, 6...Channel preamplifier,
8.10... Channel, 12... Output stage, 14...
...Transformer, 16...Load resistance.

Claims (1)

[Claims] 1. A low-level, low-power, direct-coupled class 8 amplifier having the following characteristics, having a power supply terminal for connection to a power supply, and formed as an integrated circuit on a substrate. (a) It has two amplification channels with a similar configuration. (b) the two channels are each characterized by: 1) Having a preamplifier channel having three preamplifier gain transistors directly connected to each other from the first stage to the third stage. 11) It has an output transistor directly connected to the third stage preamplifier transistor. 111) A DC that connects the collector of the third-stage preamplifier transistor to the base of the first-stage preamplifier transistor and controls the DC current of the output transistor.
Having a negative feedback loop. 1v) The collector of the first stage preamplifier transistor is connected to one end of a first resistor, the other end of the first resistor is connected to one end of a second resistor at a connection point, and The other end of the second resistor is connected to the power supply terminal, and the resistance value of the first resistor is considerably higher than the resistance value of the second resistor. (2) It has an AC negative feedback circuit consisting essentially of only resistance in order to reduce the dependence of the gain of each channel on the current flowing through the output transistor;
The C negative feedback circuit consists of a main body and a feedback resistor having a pair of terminals, one terminal of the feedback resistor is connected to the collector of the output transistor, and the other terminal is connected to the collector of the output transistor. The feedback resistor is connected to the connection point of the second resistor, and the resistance value of the feedback resistor is considerably higher than the resistance value of the second resistor. vl) The body of the feedback resistor is insulated from the power supply terminal, so that the voltage across the feedback resistor can swing beyond the power supply voltage. (c) the emitters of at least two of said preamplifier transistors are connected to each other and to ground via resistive means, so as to eliminate common mode distortion of said preamplifier; To be there. (d) having a load for the output transistor including a pair of terminals and a center tap; (e) The output transistors of both channels have the same polarity, the emitters of both output transistors are connected, and the collector of one output transistor is connected to one terminal of the load. and the collector of the other output transistor is connected to the other terminal of the load. 2. The emitters of both preamplifier transistors in the second stage are connected, and both emitters are further connected to ground through resistive means, thereby eliminating common mode distortion in the second stage of the preamplifier. The amplifier according to claim 1, characterized in that the amplifier is adapted to be 3. The emitters of both preamplifier transistors in the third stage are connected, and both emitters are further connected to ground via a third resistor, and
The emitters of both preamplifier transistors in the second stage are connected to each other, and both emitters are connected to ground via a fourth resistor.
3. The amplifier according to claim 2, further comprising a power supply means connected in parallel with the resistor, the power supply means being a collector emitter circuit of a transistor. 4. 1. above. 3. The amplifier according to claim 2, wherein all emitters of the third preamplifier transistors are connected to each other and further connected to ground via a common resistor. 5. Each of said feedback resistors is formed of a material of the same shape as said substrate, and further said body is electrically connected to said substrate by a region of material of an opposite shape that is not directly connected to said power supply terminal. 3. The amplifier according to claim 1 or 2, wherein the amplifier is insulated. 6. Amplifier according to claim 5, characterized in that the edges of said region of oppositely shaped material are in close proximity to the edges of said feedback resistor.