JPS63120507A - Gain control circuit - Google Patents

Abstract

PURPOSE:To obtain a decibel-linear characteristic with simple circuit constitution by forming a part of an ideal transfer function showing linear change by a linear circuit system approximately so as to omit an input resistor of the amplifier. CONSTITUTION:A transfer function between input and output terminals of an amplifier is selected to be (RB/(RA+RB))(1+Rf/Rs), where RB is a resistor of a variable resistance circuit network obtained through the changeover by a digital code signal. Moreover, each value in the transfer function is selected so that the transfer function change region between the input and output obtained by varying the resistor RB is in the linear region where a difference between the linear characteristic in the ideal transfer dunction and the decibel- linear characteristic by logarithmic representation is small. In varying a value Rf/Rs by using a high order digital code signal, the transfer function change region is expanded. Thus, the decibel-linear characteristic is obtained while the simple circuit constitution is applied without the need of a resistor at the input side of the amplifier.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a gain control circuit that digitally controls gain, and particularly relates to a gain control circuit that can obtain linear decibel characteristics with a simple circuit configuration, such as a facsimile receiving circuit. If applied to the automatic gain control circuit inside, it is possible to effectively prevent output fluctuations due to different attenuation amounts each time a line is connected.

[Prior art] For example, dynamic range 51 dB, resolution 0.2 d
B's digitally controlled variable gain control circuit (hereinafter referred to as a programmable gain amplifier) can be configured with a single stage amplifier circuit equipped with 256 switches and resistors, but the number of components is extremely large. ,In general,
The number of parts is reduced by dividing the amplifier circuit, which has 16 switches and resistors per stage, into two stages: a front stage and a rear stage.

However, even with this configuration, the number of parts is still large, so it is necessary to devise a circuit.

Therefore, as disclosed in Japanese Patent Application Laid-Open No. 55-25209, there has been a method in which the above-mentioned problems have been solved by adopting a circuit system in which the entire transfer function is a logarithmic approximation formula. That is, as shown in FIG. 4, the transfer function V=YU shown by the solid line...For (1), using the transfer function shown by the dotted line that approximates this, G
A circuit with decibel linear characteristics due to the variation of the oscilloscope is realized as shown in FIG.

Incidentally, assuming that the combined resistance in the dashed line of this circuit is R8, the transfer function V becomes RA R1 (R8 + R,) R, (1-1-G) Therefore, it can be seen that equation (2) is equal. .

The circuit shown in Figure 5 is designed to vary the predetermined gain control range in 24 steps, so if it is used in the subsequent amplifier circuit, it will eliminate the 16 switches and resistors that were previously required. The switch can be reduced by one-fourth, and the resistance by one-half.

By the way, in the above circuit, as mentioned above, the transmission amount -A
- Since the circuit system uses a circuit system in which the entire number is an approximate expression of logarithm, in addition to resistors R, R8, R8, R, a resistor R1 was required to be inserted between the 8 input terminals and the inverting input terminal. .

[Problems to be solved by the invention] As mentioned above, in the conventional circuit, a resistor R is placed on the input side of the amplifier.
1, there was a problem in that the circuit was complicated.

Therefore, it is an object of the present invention to solve the problems of the prior art described above and provide a gain control circuit that can obtain linear decibel characteristics while having a simple circuit configuration that does not require the resistor R1. There is a particular thing.

[Means for Solving the Problems] The gain control circuit of the present invention is a variable resistance circuit consisting of an amplifier having a constant gain between input and output terminals, a plurality of resistors, and a plurality of switches for selectively switching the resistors. It is composed of a net.

The gain control circuit configured in this manner sets the transfer function between its input and output terminals, that is, between the input and output terminals of the amplifier, as shown in the following equation.

Here, R8 is the resistance of the variable resistance network obtained by switching the switch with a digital code signal.

In addition, the change area of the transfer function between the input and output obtained by varying the resistance RB is a linear area in which the difference between the linear characteristic of the ideal transfer function and the decibel linear characteristic expressed by its logarithm is small. The six values in the formula are set.

Here, the ideal transfer function is a transfer function of an amplifier circuit having a resistance network consisting of 16 switches and resistors, for example, in the case of a 4-bit configuration.

Note that it is possible to make R and /R8 variable depending on the higher-order digital code signal to expand the transfer function change area.

[Function] In Fig. 4, when trying to cover a large range (0 to 10 times) of change (gain = 6- change) in the ideal transfer function V with a single stage amplifier circuit, the entire transfer function becomes a logarithmic approximation. It is necessary to adopt a circuit system that satisfies the equation, but it is 1 to 1.4
In the linear range of gain change with small double culmability, the difference between decibel linear characteristics and linear characteristics is small, so instead of using a logarithmic approximation circuit system for the entire ideal transfer function,
By approximating a part of it using a linear circuit system, a gain in the above range can be obtained by decibel linear approximation.

Therefore, by suppressing the maximum gain to be obtained from a single-stage circuit within the above range, it is possible to adopt a circuit system that uses a linear approximation formula using a combination of resistors, and from that circuit, the necessary If you remove the resistor R1 (connected between the input terminal and the inverting input terminal of the amplifier) and set the transfer function as However, the transfer function expressed in logarithm can also be varied linearly.

[Example] An example of the present invention will be described below based on FIGS. 1 to 3.

FIG. 1 shows an overall configuration diagram of a two-stage programmable gain amplifier in which the gain control circuit of the present invention is applied in the latter stage.

The reason why the circuit configurations of the front stage and the rear stage are different is to reduce the dynamic range of the latter stage, approximate decibel linearity using a combination of resistors, and reduce the number of parts.

FIG. 2 is a specific example of the circuit shown in FIG. 1, and shows a programmable gain amplifier with a dynamic range of 51 dB and a resolution of 0.2 dB. It is characterized by a wide dynamic range and high precision resolution for each set value.

The input signal is guided to the A/D converter 1, A/D converted, and the CPU
Calculate the error in step 2, set the required gain, and set the CP
AGC as 8-bit digital code signal from U2
The analog switches 5WO-3W5 of the programmable gain amplifier 10 are controlled by outputting control signals bo to b7. This ensures that the output level remains constant even if the input level fluctuates.

Table 1 shows the relationship between the 8-bit control signal from the CPU 2 and the gain.

Although it is conceivable that the rear stage 3 of the programmable gain amplifier 10 may have a 6-bit configuration, in order to reduce the dynamic range and error, it is configured as a 5-bit configuration, and the front stage 4 is configured as a 3-bit configuration.

The latter stage 3 changes the gain between input and output from 0 to 6.2 at 0.2 dB intervals.
[dB]. That is, it is variable in 25-32 steps. The gain of the latter stage 3 mainly composed of the amplifier 5 and the variable resistance network 6 can be expressed as follows.

-GB -G4... (1) Here, 2
If we set 0LOg10G = L a [dB], (1
) can be expressed as LG = LG8 + LG4 [dB].

The gain change G8 corresponding to the lower 4 bits (b3 to bo) of the latter stage is OdB to 3. <1 dB=1 to 1.43 times, and this is done by making the resistance RB variable. The resistance R is the resistance value indicated by the variable resistance network 6 consisting of four resistors RB□"RB3 which are connected in parallel with 88g and four switches SWO to SW3 that select and switch these resistances. be.

In addition, the gain change G4 corresponding to the remaining 1 bit (b4) in the latter stage is achieved by varying R and /R8.For this purpose, Rf+R8=R1+R+R, and resistor R2 is incorporated on the R side by switching SW4. Dari (Rf-R2
+R3) or incorporated into the R8 side (R8-R1+R2)
It is now possible to do so.

In addition, the second part 3. It goes without saying that if both the front stage 4 has a 4-bit configuration, the resistor R2 and switch SW4 become unnecessary.

The method of calculating the circuit constants for performing a decibel linear gain change (Od8 to 3.0 dB) with the combination of the four resistors and switches described above is as follows.

First, Table 2 shows the relationship between the states of bits 3 to 0 necessary for constant calculation and the gain.

Table 2 Relationship between states of bits 3 to 0 and gain However, the switch is turned on when it is 0. Since, the following is obtained. Here, when RA=2 and Ω and RB3 to RBo are calculated for the above state, Table 3 shows the results.

Table 3 Circuit constants Enter the RB3 to RBo and RBg obtained here as initial values into a personal computer, and set them in all states (00H-F
R83 to R6o close to the ideal values are determined by calculating FH) and comparing it with the ideal value obtained by a 16-resistance switching system consisting of 16 resistors and switches.

The gain characteristics obtained by comparing the ideal value and the calculated value are shown in FIG. The solid line is the ideal value corresponding to a decibel linear line from 0 to 3.0 dB (1 to 1.43 times), and is 0.2 (IB
Although it is realized by 16 resistors at intervals, since it is a nearly straight line, it can be sufficiently approximated by a combination of 4 resistors as shown by the calculated value indicated by the O symbol. Also, the error that inherently occurs due to approximation is 0.01
[times], a very small 0.1 [dB] step.

The front stage 4 has a ratio of expanding the dynamic range of 0 to 6.2 [dB] obtained in the rear stage 3 to θ to 51 [dB]. That is, the gain setting is changed in steps 23-8.

The gain can be expressed as follows.

The gain change G corresponding to the upper three bits (b7 to b5) of the first stage 4 is achieved by making the resistor R9 variable. Resistor R8 consists of seven series resistors RD1 to Ro7 (R9o
= OkΩ) is determined by the selection of each connection point,
The selection is made by switching the switch SW5, which is controlled by a switch control signal obtained by converting the 3-bit control signal by the decoder 7.

As mentioned above, according to the programmable gain amplifier according to this embodiment, a linear circuit system is adopted in which the dynamic range of the subsequent stage is reduced and errors are minimized in the latter stage, so that the input resistance R1 required for the logarithmic circuit system is reduced. By eliminating the need for the gain of the latter stage and supplementing the gain of the latter stage with the former stage, good decibel linear characteristics with a dynamic range of 51 dB and a resolution of 0.2 dB can be obtained despite the circuit configuration having a small number of parts.

[Effects of the Invention] According to the present invention, by approximating a part of the ideal transfer function exhibiting a linear change using a linear circuit system, the entire ideal transfer function can be approximated by a logarithmic approximation formula. Unlike the conventional circuit configured using the following circuit system, the input resistor R1 of the amplifier is omitted, and decibel linear characteristics can be obtained with a simple circuit configuration.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram of a programmable gain amplifier incorporating a gain control circuit in a subsequent stage according to an embodiment of the present invention;
Figure 2 is a detailed diagram of the P[1-grammar pull gain amplifier shown in Figure 1, Figure 3 is a gain characteristic diagram showing the error of the subsequent gain control circuit shown in Figure 2, and Figure 4 is a diagram of the logarithmic approximation method. FIG. 5 is a gain characteristic diagram showing the error of the conventional circuit employed, and FIG. 5 is a diagram showing the configuration of a conventional circuit employing a logarithmic circuit system. In the figure, 5 is an amplifier, 6 is a variable resistance network, and RB. ~RB3 and RBg are multiple resistors, and swo~SW3 are switches.

Claims (1)

[Claims] Consisting of an amplifier and a variable resistance network consisting of a plurality of resistors and a plurality of switches for selectively switching the resistors,
When the resistance of the variable resistance network obtained by switching the above switch using a digital code signal is R_B, the transfer function between the input and output terminals of the amplifier is expressed by the following formula, [R_B/(R_A+R_B)] [1+(R_f/
R_s)] and the change area of the transfer function between the input and output by varying the resistance R_B is a linear area where the difference between the linear characteristic of the ideal transfer function and the decibel linear characteristic represented by its logarithm is small. A gain control circuit characterized in that each value in the above formula is set.