JPS5951005B2 - servo circuit - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a servo circuit including a magnetostrictive potentiometer, and in particular to an improved dead zone compensating means.

The servo circuit including the magnetostrictive potentiometer was newly developed by the applicant.

Hereinafter, before explaining the present invention, the principle of a servo circuit including a magnetostrictive potentiometer will first be explained. FIG. 1 is a block diagram of a servo circuit including a magnetostrictive potentiometer.

In the figure, 1 is a magnetostrictive wire made of linear magnetostrictive material such as Ni-SPANC or Ni, 2 is a first receiving coil placed at one end of the magnetostrictive wire 1, and 3 is the other end of the magnetostrictive wire 1. A second receiving coil, 4, arranged at the end of the magnetostrictive wire 1 is movably arranged between the receiving coils 2 and 3 via the magnetostrictive wire 1.
This is an excitation coil that generates ultrasonic signals inside the device. 05 is an excitation pulse generator, and its output pulse PE is applied to the excitation coil 4, as well as flip-flops FF, FF, which will be described later.

is applied to the set terminal S and the monomulti M. OP
,,OP. are both amplifiers, and OP is the receiving coil 2.
The detection signal e, is amplified and applied to the reset terminal R of the flip-flop FF. Also, OP. is the detection signal e from the receiving coil 3. Amplify and flip-flop F
F. is applied to the reset terminal R of. CK is a flip-flop FF,,FF. Two types of time width signals P obtained from
W,,PW. This is an arithmetic circuit that takes as input. In the arithmetic circuit CK, Es, -Es are standard power supplies, R, , R. is a resistor, 5, 50 is a flip-flop FF, FF.
A switch driven by a pulse width signal from IG
is an integrator and an operational amplifier OP. and a capacitor Cl. SH is a circuit that samples and holds the output signal of the integrator IG, and is composed of a switch 50, a capacitor Co, and an operational amplifier OP. INT
are input terminals, R, , R are respective input resistors, SA is a servo amplifier, and SM is a servo motor, and the rotating shaft of this servo motor is mechanically connected to the excitation coil 4. The output terminal of the sample-and-hold circuit SH is connected to the voltage dividing point of the resistor R2, and is connected to the servo amplifier S so as to be differentially connected to the input signal Ei applied from the input terminal 1NT.
Connected to the input terminal of A. The operation of the servo circuit having such a configuration will be explained as follows.

First, an excitation pulse PE is applied from the excitation pulse generator OS to the excitation coil 4 as shown in FIG. 2A. When a pulse is applied to the excitation coil 4, the so-called J
An ultrasonic signal is generated inside the magnetostrictive wire 1 due to the Oule effect, and this signal propagates toward both ends of the magnetostrictive wire 1. This ultrasonic signal eventually reaches the first receiving coil 2 and the second receiving coil 3 provided at both ends of the magnetostrictive wire 1. When an ultrasonic signal passes through the magnetostrictive wire 1, pulse-like voltage signals El and e2 are generated in the receiving coils 2 and 3 due to the so-called Villari effect, as shown in the opening of FIG. Now, if we assume that an ultrasonic signal is generated within the magnetostrictive wire 1 at the same time as the excitation pulse PE is applied to the excitation coil 4, this ultrasonic signal propagates within the magnetostrictive wire 1 from the position of the excitation coil 4, and then reaches the receiving coil. The times t1 and T2 until reaching 2 and 3 are calculated by formula (1) and (
2) It can be expressed by the following equation. However, VS: Speed at which the ultrasonic signal propagates within the magnetostrictive wire 1 11: Distance between the excitation coil 4 and the first receiving coil 2 12: Distance between the excitation coil 4 and the second receiving coil 3 Excitation pulse PE is the excitation coil 4 and at the same time flip-flop FF
The voltage is applied to the set terminals S of flip-flops FF1 and FF2, and the flip-flops FF1 and FF2 are set to the set state as shown in FIG.

Further, the pulsed voltage signal El obtained by the receiving coils 2 and 3 after Tl, t2 after applying the excitation pulse PE
, e2 are applied to the reset terminals R of the lower flip-flops Fl and FF2 through amplifiers 0P1 and 0P2, respectively, to bring the flip-flops FF1 and FF2 into the reset state as shown in FIG. 2C and 2. Therefore, flip-flop FFl
, 11 from the output end of FF2, as shown in Fig. 2C, 2.
, 12, time width signals PWl, PW2 having time widths Tl, t2 are obtained. In the arithmetic circuit CK, switches Sl and S2 are connected to the lower flip-flops Fl and FF.
It is made conductive by time width signals PWl and PW2 from 2.

In addition, the output pulse SP of the monomulti M (see Fig. 2 E) triggered by the output pulse of the excitation pulse generator 0S turns on the switch S3, and the output of the integrator 1G changes to The hold circuit SH allows the sun full
will be held. Now, assuming that the absolute values of the standard power supply voltages Es and Es are equal, the switches Sl and S2
During the period when both are conducting, the standard voltages Es and -Es are applied to the input terminal of the integrator 1G through the resistor R1 and the switch Sl.
, S2, and the output EO of the sample-and-hold circuit SH is applied via a resistor R2. Therefore, the output voltage EI of the integrator 1G becomes a voltage that decreases as shown by the section f in FIG. In a state where only one of the switches SL and S2 is conducting, the switch S
2), the standard power supply voltage Es of either polarity and the output EO of the sample-and-hold circuit SH are applied to the integrator G. Therefore, integrator 1
The output voltage EI of G becomes a rising voltage as shown in the section g in FIG. In this way, the output E of the integrator 1G
is sampled and held by the sample-and-hold circuit SH every time the sampling pulse SP shown in FIG. A voltage EO as shown in the figure is output. This output voltage E
O is differentially matched with input signal E1 applied from input terminal 1NT and applied to servo amplifier SA. The deviation signal output by the servo amplifier SA is applied to the servo motor SM, and causes the excitation coil 4 connected to the rotating shaft of this servo motor to be displaced on the magnetostrictive wire 1. When the deviation signal output from the servo amplifier SA becomes zero, this servo circuit is balanced like a normal servo system. In such a servo circuit, the output EO of the sample-and-hold circuit SH after the MT time (m is an integer) has elapsed is expressed by the following equation (3). If so, equation (3) becomes equation (5) when m=ω.

In formula (5), 1, +1.

is the distance between the first receiving coil 2 and the second receiving coil 3, and is a constant value regardless of the movement of the excitation coil 4. In addition, the constant K is also a value determined by the resistances R, ~R,
Therefore, the position (1, -1) of the excitation coil 2 corresponds to the input signal E. Therefore, if a pointer or a recording pen is provided in conjunction with the excitation coil 4, the servo circuit shown in FIG. 1 including the magnetostrictive potentiometer can be used as a servo recorder. By the way, FIG. 3 shows a block diagram of an example of a dead zone compensation means used in a conventional servo circuit.

In FIG. 3, NF is a dead zone compensation circuit, which applies the output Asinft of the oscillator to the input of the servo amplifier to vibrate the load at a frequency f to cancel out the frictional force F. This compensation means is called the dither method, and the dither signal Asinft obtained from the oscillator has an amplitude A large enough to cancel out the frictional force F, and a frequency f of a high frequency (usually 50 to 500). Such dead zone compensation means using the dither method has been used in the past, but since it requires an oscillator to generate the dither signal, the number of parts increases, resulting in increased costs and decreased reliability. The present invention has been made in view of this point.

That is, as explained in FIG. 1, in a servo circuit using a magnetostrictive potentiometer as a feedback element, switches S,
, S2 are conducting, the arithmetic circuit CK performs the (
3) The calculation shown in the formula must be performed, and during the calculation period, the resistance R. The voltage applied to the voltage dividing point must not vary. Therefore, in a servo circuit using such a magnetostrictive potentiometer as a feedback element, the sample-and-hold circuit SH is indispensable as explained in FIG. By the way, as shown in FIG. 2, the output EO of the sample-and-hold circuit SH always has ripples. The present invention utilizes this ripple to compensate for the dead zone. FIG. 4 is a connection diagram showing an embodiment of the servo circuit of the present invention using a magnetostrictive potentiometer. In FIG. 4, HA indicated by a dashed line is an adder circuit added according to the present invention, and the circuit other than this adder circuit HA is completely the same as the circuit in FIG. Re-explanation of the same parts will be omitted. The adder circuit HA includes a high frequency amplification function and is an operational amplifier 0P.

and its feedback resistance R,, addition resistance R,,R. and a resistor R for the high frequency amplifier circuit. and a capacitor C2. One end of the addition resistor R3 is connected to the human power terminal INT, and the other end is connected to the operational amplifier 0P. is connected to the (-) input terminal of the One end of the other summing resistor R is connected to the output terminal of the sample-and-hold circuit SH, and the other end is connected to the (-) input terminal of the operational amplifier OP. Also, the resistance R. and capacitor C. are connected in series, and this series circuit is connected in parallel to the resistor R,. Operational amplifier 0P. The output terminal of is connected to the input terminal of servo amplifier SA. In the apparatus of the present invention having such a configuration, the basic operation including the magnetostrictive potentiometer is exactly the same as that shown in FIG. 1, so a re-explanation thereof will be omitted.

The signal to be measured E is a resistance R. The output of the sample-and-hold circuit SH (shown in FIG. 2) is applied to the operational amplifier 0P through the resistor R. and capacitor C. and an operational amplifier 0P via a series circuit of and a resistor R. added to. The DC component of the output EO of the sample-and-hold circuit SH is connected to a resistor R.
, and are added with opposite polarity to the input to be measured E, and the deviation signal is applied to the servo amplifier SA. On the other hand, the magnitude of the ripple in the θ output EO of the sample-and-hold circuit SH is usually about 5 mVp-P, and the frequency is about 500 ratio. Such ripples occur in the resistor R that constitutes the high frequency amplifier circuit. and capacitor C. The signal is applied to the operational amplifier 0P5 via the amplifier 0P5 and amplified to a predetermined magnitude. In this case, if you try to simply amplify this ripple with a circuit that can amplify from DC to high frequencies, the feedback gain will change like DC, so when this servo circuit is applied to a pen recorder, etc. If you try to change the degree of ripple amplification, the pen recording span will change, which is inconvenient, but with the present invention, which uses a high-frequency amplification circuit, it is possible to arbitrarily amplify only the ripple amount. . The ripple thus amplified is applied to the servo motor SM as a dither signal via the servo amplifier SA, thereby canceling the dead zone. A block diagram of such a servo circuit according to the present invention is shown in FIG.

In FIG. 5, HAS represents the transfer function of the adder circuit HA shown in FIG. 4, TD represents the differential time constant, and n represents the differential gain. The values of TD and n are determined by the gain KA of the servo amplifier SA, the torque gain KT of the servo motor SM, and the friction force F, but the maximum value of the vibration force due to the dither signal is about twice the friction force F. In order to do this, n and TD given by the following equations may be used. In addition, since the high frequency amplification function in the adder circuit HA acts as a forward compensation circuit for the servo system as a whole, we actually considered n and TD from the viewpoint of the phase compensation circuit of the servo system, and as a result, (6 ) and (7
) It is necessary to take care to ensure that the conditions given by the formula are compatible.

As explained above, according to the present invention, in a servo circuit including a magnetostrictive potentiometer, the ripple signal output from the sample hold circuit SH, which is an essential requirement, is amplified by a high-frequency amplifier circuit and input to the servo amplifier SA. Since it is configured to be used as a dither signal by adding it to , the dead zone can be canceled without using a dedicated oscillator.

Moreover, since the high-frequency amplifier circuit is simply a circuit in which a capacitor and a resistor are connected in series to the summing resistor of an summing circuit using a normal operational amplifier, there is no increase in cost and decrease in reliability compared to an oscillator. It is possible to obtain a servo circuit that requires less.

[Brief explanation of the drawing]

Fig.] is a connection diagram of a servo circuit including a magnetostrictive potentiometer, Fig. 2 is an operating waveform diagram of the circuit in Fig.], Fig. 3 is a schematic diagram of a conventional servo circuit including a dead zone compensation means, and Fig. 4 is a diagram of the present invention. A connection diagram showing an example of the servo circuit of FIG.
2 is a block diagram of the circuit shown in FIG. 1... Magnetostrictive wire, 2J3... Receiving coil, 4... Excitation coil, CK... Arithmetic circuit, SH... Sample hold circuit , SA...
... Servo amplifier, HA ... High frequency amplifier circuit.

Claims (1)

[Claims] 1. A magnetostrictive wire made of a magnetostrictive material, a first receiving coil, a second receiving coil, and an excitation coil coupled to the magnetostrictive wire, the distance between the excitation coil and the first receiving coil, and the distance between the excitation coil and the second receiving coil. a servo motor that changes the distance in response to mechanical displacement; Circuit means for obtaining two types of dynamically changing time width signals, two with different polarities
two standard power supplies, two switches each connected to these standard power supplies and driven by the two types of time width signals, an integrator to which the standard power supply voltage is applied via these switches; a sample-and-hold circuit for sampling and holding the output of the integrator; means for applying the output of the sample-and-hold circuit to the integrator together with the standard power supply voltage; A high voltage amplifier is constructed of an operational amplifier in which a circuit in which a capacitor and a resistor are connected in series is connected to a summing resistor in parallel, and the output of the sample-and-hold circuit is applied to this parallel circuit with a polarity opposite to that of the input signal. A servo circuit comprising an adder circuit including a range amplification function and a servo amplifier to which the output of the adder circuit is applied, and the output of the servo amplifier is applied to the servo motor.