JPH02266710A - Semiconductor integrated circuit filter - Google Patents

Abstract

PURPOSE:To keep an amplitude level constant even when a pole frequency is changed by providing a Q control circuit which controls the feedback quantity of a feedback circuit network. CONSTITUTION:The feedback circuit network 7 is inserted between an output terminal 6 and the (1) input terminal of a transconductance amplifier 2a, and the Q control circuit 8 changes the Q value of a low-pass filter by controlling the on and off of the switch 71 of the feedback circuit network 7. In other words, the feedback quantity of the feedback circuit network 7 is controlled so that the Q valves in a calculation value at a low pole frequency can be set higher than that at a high pole frequency by the Q control circuit 8 when the pole frequency is decreased, and adversely, the Q valves can be suppressed when it is increased from the low pole frequency to the high pole frequency. In such a way, the amplitude level can be controlled at a decided amplitude level when the pole frequency of a filter is decreased or increased.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention is applicable to, for example, consumer equipment (color television,
VTR (video tape recorder).

When an analog signal such as R-DAT (low head digital audio tape recorder) passes through a filter, the pole frequency of the filter (
This invention relates to a semiconductor integrated circuit filter that can maintain the amplitude level of the filter constant at a certain ratio by varying the amplitude (hereinafter referred to as fo) at the same amplitude level or by raising and lowering it at a certain ratio.

[Conventional technology]

FIG. 6 is a circuit diagram showing a conventional second-order low-pass filter circuit using a transconductance amplifier. In FIG. 6, 1 is a signal input terminal, and 2 is a differential input voltage signal between the input voltage signal Vt from the signal input terminal 1 and the output voltage signal v output from the output terminal 6, which is converted into a current iI. It is a transconductance amplifier.

This signal 11 is integrated by a capacitor 4, passed through a buffer circuit 3, and is input to a transconductance amplifier 2a.

This transconductance amplifier 2a has the same function as the transconductance amplifier 2 described above, and receives an input voltage signal v1 input from the buffer circuit 3.
The difference between the output voltage signal V and the output voltage signal V is converted into one current and output.

This current i is integrated by a capacitor 4a, and an output signal voltage V is outputted from a signal output terminal 6 via a buffer circuit 3a.

Further, current source circuits 5a and 5b are connected between the transconductance amplifiers 2 and 2a and the power source, respectively, and a current source circuit 5 is also connected between the power source and the ground.
C is connected.

These current source circuits 5a to 5C are controlled by a control circuit 5, and as a result,
The transconductance gll+ and g2 value of the transconductance amplifier 2°2a are controlled to control f6. Note that 9 is an r-fist control signal terminal.

Next, the operation will be explained. When a human input signal voltage Vi is input to the signal input terminal 1, it is input to the (g) input terminal of the transconductance amplifier 2.

Since the (c) input terminal of the transconductance amplifier 2 is connected to the output terminal 6, the output voltage signal V is applied thereto.

Therefore, the transconductance amplifier 2 generates a differential voltage (vt
A t current 11 proportional to V@) appears at the output of the transconductance amplifier 2.

The current 11 appearing at the output end of the transconductance amplifier 2 is integrated by the capacitor 4, and when the integrated current passes through the buffer circuit 3, it appears at the output end as a voltage signal .

This voltage signal vl is applied to the (e) input terminal of the transconductance amplifier 2a in the subsequent stage.

In this case, the buffer circuit 3 is designed to prevent interference in coupling the transconductance amplifiers 2 and 28.

The signal processing operation after the transconductance amplifier 2a is performed by the transconductance amplifier 2 and the buffer circuit 3.
, the operation is similar to that of capacitor 4.

The main relational expression representing the main operations of the previous operations and the transfer function of the low-pass filter of FIG. 6 obtained from the result of this main relational expression are shown below.

Main relational expression % formula % (1) In these formulas (1) to (4), 1+ is the output current of transconductance amplifier 2, 11 is the output current of transconductance amplifier 2a, ■1 is the integrated voltage by capacitor 4 , gllll is the transconductance of the transconductance amplifier 2, and g2 is the transconductance of the transconductance amplifier 2a.

+B) Result formula A resultant formula is calculated using the above formulas (1)2 to (4).

Transfer function □(S) = H(S) V 盈S'''C:'' 10Cr, Cz However, in this equation (5), jω is set to S, and Ct and Ct are the capacitances of the capacitor 4.4a, respectively. It is.

The general expression for the transfer function of this kind of filter of second order is:

In this equation (6), ω. is the pole angular frequency, ro is the pole frequency, Q is Paul Q1 It is.

Comparing equations (5) and (6) above, from equation (7), becomes.

From this equation (9), the pole frequency (r in Figures 5 and 7)
, or fx) is the square root of the product of t→− and 4 divided by 2π.

Also, from equation (8), the amplitude level at the pole frequency is the ratio of both transconductances g"+ + g'lz" and the capacitance CI+ CzO of both capacitors 4 and 4a.
It can be seen that it is determined by the square root of the product multiplied by the ratio J C2 (2).

A or b in FIG. 7 shows the characteristics of a general low-pass filter (LPF) when the value of Q is larger.

[Problem to be solved by the invention]

Since the conventional low-pass filter is configured as described above, as explained in equation (8) above, for example, a high signal is input to f, control signal terminal 9 in FIG. 6, and the f0 control signal is input. The circuit 5 controls the current source circuits 5a to 5c that control the transconductances gra+ and gllg, for example, in a negative manner,
If it is possible to keep the ratio of −(1− gllg constant), capacitors in semiconductor integrated circuits are originally supposed to have an extremely accurate ratio, and f
From equation (9), o is proportional to the square root of the multiplied value of gll·gllz.

Therefore, for example, it is quite possible from the calculated values to realize both gllll and gllg at a constant level without reducing the amplitude level, and there is nothing wrong with it theoretically.

That is, as shown in FIG. 5, even if the pole frequency is lowered from the pole frequency f2 to a lower pole frequency f, it is calculated that the amplitude level remains constant.

However, this does not necessarily work in actual circuits; for example, as shown in Figure 7, the pole frequency is changed from f2 to fl.
If it is lowered to 1, the problem is that the Q is lowered and the amplitude level is lowered as shown in FIG. 7b.

There are various main causes of this conventional problem, but the main causes are f, the tracking accuracy of the current ratio of the current source circuits 5a to 5c by the control circuit 5, and the change in transconductance g due to the current adjustment of the current source circuit 5a. Rate (=9g
ml, current source circuit 5bal.

It is difficult to make the rate of change in the transconductance gll due to current adjustment < = a g wh z > constant.

al.

Note that I,,1. are current source circuits 5a + 5, respectively.
Transconductance g1m++ gll by b
This is the control current that determines z.

The reason why it is difficult to make the above rate of change the same is, for example, as shown in FIG.
The control current T+, in which the offset voltage generation status of H determines the transconductance g1, and the control current 1, which determines the transconductance get.
The transconductance of the first stage differential amplifier in the transconductance amplifier does not change in a similar manner at a constant ratio, and as a result, the ratio of -(1-) determines the Q in equation (8). This is because ID does not change linearly.

This invention was made to solve the above problems, and provides a semiconductor integrated circuit filter in which the amplitude level remains almost the same even if the pole frequency is changed from a predetermined pole frequency, and which can be made using simple means. The purpose is to obtain.

[Means to solve the problem]

The semiconductor integrated circuit filter according to the present invention includes a feedback circuit network inserted between the (c) input terminal of a subsequent transconductance amplifier and the output terminal of the filter and feeding back an output voltage at a predetermined ratio, and this feedback circuit network. A Q control circuit is provided to control the amount of feedback.

[For production]

In this invention, the feedback circuit network converts the output voltage of the filter into the transconductance amplifier (
C) In addition to suppressing and feeding back at a predetermined rate to the input terminal, when the pole frequency is lowered to a predetermined pole frequency, the Q control circuit makes the calculated Q at the low pole frequency higher than the Q at the high pole frequency. However, when increasing from a low pole frequency to a high pole frequency, the feedback amount of the feedback circuit network is controlled so as to suppress the Q.

〔Example〕

Embodiments of the semiconductor integrated circuit filter of the present invention will be described below with reference to the drawings. FIG. 1 is a circuit diagram showing the configuration of one embodiment.

In Fig. 1, when explaining the configuration, the same parts as in Fig. 6 are given the same reference numerals to avoid duplicate explanation.
I will mainly describe the parts that are different from the diagram.

As is clear from comparing Figure 1 with Figure 6,
In the figure, a feedback circuit network 7 and a Q control circuit 8 are newly added to the configuration of FIG. 6.

That is, a feedback circuit m7 is inserted between the output terminal 6 and the (c) input terminal of the transconductance amplifier 2a, and the resistor RB is connected between this (c) input terminal and the output terminal 6. .

Further, the (c) input terminal of the transconductance amplifier 2a is connected to the positive terminal of a DC power source E via a resistor RA, and the negative terminal of this DC power 'aE is grounded.

Positive electrode of DC power supply E and transconductance amplifier 2a
A switch 71 is connected between the input terminals of (c), and this switch 71 is controlled to open and close by the output of the Q control circuit 8.

The input terminals of the Q control circuit 8 and f control circuit 5 are connected to the f0 control signal terminal 9.

The feedback circuit network 7 gives a certain proportion of the output signal voltage V appearing at the output terminal 6, that is, in the case of FIG. 1, the resistance value of the resistors RA-RC, respectively, to give a feedback amount to the inductance 2a.

Further, the Q control circuit 8 controls the on/off state of the switch 71 (analog switch, etc.) of the feedback circuit 1i17, and changes the Q value of the low-pass filter.

Next, the operation will be explained. FIG. 2 simplifies the internal configuration of the feedback circuit M47 of the embodiment shown in FIG. 1, and omits the Q control circuit 8. The operation of the embodiment shown in FIG. 1 will be explained using FIG. 2. Then, calculate the pole frequency and Q value.

The resistance RAO value of the resistance RA of the feedback circuit network 7 is adaptively changed with respect to the desired value of fo.

The main relational expressions and resultant expressions (=transfer functions) similar to those of the conventional circuit configuration shown in FIG. 6 will be described below.

In this invention, the difference from the above-mentioned conventional example is that the feedback amount of the transconductance amplifier 2aRAI after the filter is 100%, and a formula is created by paying attention to the fact that only = -RA+RBK.

(1) Main relational expression %formula% (a Result expression A resultant formula is calculated using the above formulas 0θ to 05).

Transfer function □(S) −H(S) i Here, the general formula for the transfer function of this type of filter of second order (
Compared to (6) 2) above, ω. is the same as the above equation (7), and therefore it is clear that r is also the same as the above equation (9).

However, when Q is determined by comparing the first-order coefficients of equation (7), we get: Q=-'-oax K-C* mt knee 1 x g m +・gllggl C
+, (g x Kx C.

becomes.

That is, compared to the conventional circuit shown in FIG. 6, the Q of the circuit shown in FIG. 2 is expressed as twice as high. other ω. , fo do not change at all from the above equation.

The circuit of the embodiment shown in FIG. 1 is based on the circuit configuration shown in FIG. 2, and the value of Q expressed as the above equation 01 is set in parallel to the resistor RA, and the resistor RC is turned on by a switch 71.

When the resistor RC is in parallel with the resistor RA, K increases.

That is, by doing this, for example, the pole frequency is set at an amplitude level of 18, and then the transconductance g Tsuru 1 + g is set in the current power supply circuits 5a and 5b.
+*fi controltfE I It l z'
c By decreasing the transconductance g1
1++ When g1 is decreased and the pole frequency is set to fl, the nonlinear characteristics of the circuit, especially 95 m
1 -111-, as shown in Figure 8, 91. 〉 a+.

Control current 1. ．． 1. When the transconductance is varied, it is not constant, and gllz
G-kaku! decreases by a certain factor.

As a result of this, the Q expressed by equation (8) or equation 07 decreases, and as shown in FIG. 8, the amplitude level decreases at the pole frequency f. It becomes possible to compensate for the decrease in the amplitude level by increasing K by sitting at .

That is, the amplitude level of al+b+ can be made constant as shown in FIG.

Next, a description will be given of FIG. 3, in which the blocks of FIG. 1 are expanded into an actual tC conversion circuit. The reference numerals of each block in FIG. 3 are the same as those of the blocks in FIG. 1, and in FIG. 3 only the parts newly added to the conventional structure of FIG. 6 will be explained.

First, the bias circuit 11 is a circuit that sets the bias current of the circuit, and when the Q control signal current flows from the Q control signal terminal 9 in the Q control circuit 8, the transistor 8a is turned on.

As a result, the transistor 8b whose base and collector are connected from the emitter of the transistor 8b is connected to the resistor 8c1.
A route of the on-state a (Vex=0) of the transistor 8a is formed, and a current flows.

As a result, transistors 8d and 8e forming a current mirror with transistor 8b are turned on. The collector current of the transistor 8d is connected to the feedback network 7.
The resistor RC is connected in parallel to the resistor RA by entering the base of the switch 71 (analog switch) and turning on the two transistors of the switch 71.

As a result, as explained in FIGS. 1 and 2, the value of K is increased.

Here, the collector current of the transistor 8d is the switch 7
1 and flows to the collector of transistor 8f. This is because the collector current from the transistor 8e is used to activate the transistors 8f and 8g forming the current mirror circuit, and the collector of the transistor 8f operates to absorb the current flowing from the transistor 8d to the switch 71. It is.

The DC power supply E for bias application shown in FIG. 1 includes a bias power supply 8h in the Q control circuit 8 shown in FIG. 3, an emitter follower transistor 81 activated by the bias power supply 8h, and a current source transistor 8j. The emitter electrode of the transistor 81 corresponds to the positive electrode of the DC power supply E in FIG. The other parts perform the same operations as in FIG.

In the above embodiment, simply the pole frequency f.

We have described an example in which the amplitude level ratio of only two points, pole frequency f,, f, is constant, but it is clear that this can be done by setting multiple frequency points, such as three or four points. .

Furthermore, it is clear that the present invention can be applied not only to low-pass filters but also to band-pass filters (BPF), phase circuits, and the like.

Further, as another example of reusing the embodiment of the present invention shown in FIG. 1, in place of the resistor RB of the feedback circuit M47 in FIG. 1, as shown in FIG. The third transconductance amplifier 72 (transconductance is g113) is connected to RB =-! -, it is clear that even if the value of the transconductance gll13 is controlled, the same effect as in the embodiment of FIG. 1 can be achieved.

〔Effect of the invention〕

As described above, according to the present invention, a feedback circuit network that feeds back the output signal voltage at a predetermined ratio is inserted into the subsequent transconductance amplifier, and the feedback amount of this feedback circuit network is Q-controlled based on the pole control signal. Since it is integrated into a semiconductor integrated circuit so that the amplitude level remains constant even if the pole frequency is controlled by the circuit, it is possible to control the nonlinear parameters of the semiconductor integrated circuit, especially the current transformer that controls the transconductance of the transconductance amplifier. Amplitude level fluctuations caused by the Q of the filter due to different conductance change coefficients (differential coefficients) can be inversely corrected to a constant ratio.

Along with this, VTR, D-VTR (digital VTR)
When the pole frequency of the filter is decreased or increased by a certain ratio in standard mode, long-time mode, etc. during recording and playback of consumer equipment such as , it becomes possible to omit external parts, and it becomes possible to compactly organize the electrical circuits of equipment, and this effect is particularly significant when converting waveform equivalent circuits used in magnetic recording/reproducing circuits into LSIs.

[Brief explanation of drawings]

FIG. 1 is a circuit diagram of a semiconductor integrated circuit filter according to an embodiment of the present invention, FIG. 2 is a circuit diagram of a prototype of the semiconductor integrated circuit filter of FIG. 1, and FIG. 3 is a circuit diagram of the semiconductor integrated circuit filter of FIG. 1. 4 is a circuit diagram showing a detailed circuit configuration of a semiconductor integrated circuit filter according to another embodiment of the present invention. FIG. 5 is a circuit diagram showing a semiconductor integrated circuit filter according to another embodiment of the present invention. FIG. Fig. 6 is a circuit diagram of a conventional low-pass filter, and Fig. 7 shows the pole frequency due to a decrease in Q when changing the pole frequency of a conventional low-pass filter. Figure 8 is a characteristic diagram showing changes in amplitude level for conventional low-pass filters, and shows that the factor that causes a decrease in amplitude level due to a change in Q in a conventional low-pass filter is that the differential coefficient of transconductance with respect to the control current of this transconductance amplifier differs. FIG. 3 is an explanatory diagram of the factors by which the value of the transconductance ratio changes depending on the value of the control current value. 2.2a, 72...transconductance amplifier,
3.3a...Buffer circuit, 4,4a...Capacitor, 5a-5C...Current source circuit, 5...F0 control circuit, 7...Feedback circuit network, 8...
Q control circuit. In addition, in the figures, the same reference numerals indicate the same or equivalent parts. Agent: Masuo Oiwa Diagram: Diagram: Funeral diagram Procedures for amendment (self-motivated) Name of the invention: Semiconductor integrated circuit filter 3, Person making the amendment Representative: Moriya Kiki Within Mitsubishi Electric Corporation: Diagram: 5゜Details subject to amendment Corrected "It is possible. For this reason" to "Because it is possible" in line 18 of the Detailed Description of the Invention column and Drawing 6° Amendment Contents Specification, page 7 (8). do. Same line ■ Correct Figure 8 of the 5th drawing as shown in the attached sheet. 7゜ Attached document catalog correction drawings

Claims (1)

[Claims] a pre-stage transconductance amplifier that converts the differential input voltage signal between the input voltage signal and the output voltage signal into a current;
A pre-stage capacitor that integrates the current output from the pre-stage transconductance amplifier; a feedback circuit that feeds back the output voltage signal at a predetermined ratio; a transconductance amplifier in the subsequent stage that converts the voltage difference between the signal and the signal corresponding to the current into a current; a capacitor in the subsequent stage that integrates the current output from the transconductance amplifier in the subsequent stage to obtain the output voltage signal; and a pole. a pole control circuit that controls a pole frequency by controlling the transconductance of the preceding and subsequent transconductance amplifiers using a control signal;
and a Q control circuit that controls the amount of feedback of the feedback circuit network using the pole control signal to change the Q value of the filter when changing the pole frequency to maintain the amplitude level at the same level.