JPH0566701U - Digital servo circuit - Google Patents

Abstract

(57) [Summary] [Object] To provide a digital servo circuit capable of suppressing the influence of a quantization error without increasing the number of quantization bits. An A / D converter 7 for converting an error signal indicating a difference from a target value into a digital signal, and means for providing a predetermined compensation characteristic, which are controlled according to an output of the A / D converter 7. In a digital servo circuit having a PID controller 91 for generating a control signal for converging an object and a D / A converter 8a for converting the control signal into an analog signal and outputting the analog signal, a dither signal Sd1 corresponding to a minute signal is generated. At the same time, the dither signal generator 9 generates the dither signal Sd2 by adding the compensation characteristic to the dither signal Sd1.
3 and PID controller 92, an adder 6 for adding and outputting the dither signal Sd1 to the error signal, and a subtractor 10 for subtracting and outputting the dither signal Sd2 from the control signal.

Description

[Detailed description of the device]

[0001]

[Industrial applications]

 The present invention relates to a digital servo circuit suitable for use in, for example, a magnetic disk device that positions a magnetic head with high accuracy.

[0002]

[Prior Art]

 As is well known, in recent years, digital servo control has been widely used as a technique for automatically bringing displacement, speed, etc. of an object to be controlled close to a target value. There is a digital servo circuit that utilizes this digital servo control, and this is used, for example, for control that requires high accuracy such as head positioning in a magnetic disk device. Generally, in this type of circuit, the analog signal supplied as an error signal is once converted into a digital signal and subjected to predetermined digital processing, and then this is converted back into an analog signal as a control signal to be controlled. It is supposed to be output. Therefore, in the digital servo circuit, an A / D (analog / digital) converter and a D / A (digital / analog) converter for mutually converting an analog signal and a digital signal are essential.

By the way, when an analog signal is converted into a digital signal by the A / D converter, the signal is quantized as shown in FIG. 5, but at this time, the quantization is performed due to the limitation of the number of quantization bits. An error occurs. Here, the quantization error will be described with reference to FIG. First, (a) of the figure shows the characteristics of the analog signal before being quantized, and (b) of the figure shows the characteristics of the quantized digital signal. In other words, the quantized digital signal Q (x) (output) shows a gradual change in units of the quantization decomposition width d with respect to the analog signal (input) which shows a continuous change. It is a discrete value, which is caused by the well-known quantization error shown in FIG.

Therefore, in the conventional digital servo circuit, in order to suppress the influence of the quantization error, for example, the number of quantization bits is increased to improve the resolution, or the apparent quantization resolution is increased. Some measures were taken to improve it.

[0005]

[Problems to be solved by the device]

 However, in the above-mentioned conventional digital servo circuit, if the number of quantization bits is increased in order to suppress the influence of such quantization error, the A / D converter and the D / A converter become expensive, resulting in In addition, there is a problem that the cost of the servo circuit itself is increased. Further, in the method of improving the apparent quantization resolution, the range in which a good dynamic range can be obtained is limited, and there is a problem that high accuracy can be obtained only in a narrow range.

The present invention has been made under such a background, and an object thereof is to provide a digital servo circuit capable of suppressing the influence of a quantization error without increasing the number of quantization bits. I am trying.

[0007]

[Means for Solving the Problems]

 In order to solve the above-mentioned problems, the present invention is a means for converting an error signal indicating a difference from a target value into a digital signal and outputting the digital signal, and means for providing a predetermined compensation characteristic. And a compensating means for generating a control signal for converging the controlled object according to the output of the first converting means, and a second converting means for converting the control signal into an analog signal and outputting the analog signal. In the servo circuit, a superimposition signal generating means for generating a first superimposition signal corresponding to a minute signal and generating a second superimposition signal obtained by adding the compensation characteristic to the first superimposition signal, and the first superimposition signal generating means. Adding means for supplying the addition result of the superimposition signal and the error signal to the first converting means, and subtracting means for subtracting and outputting the second superimposition signal from the control signal. It is a feature.

[0008]

[Action]

 According to the above configuration, the superimposition signal generating means generates the first superimposition signal corresponding to the minute signal, and also generates the second superimposition signal having the compensation characteristic added to the first superimposition signal. Then, the adding means adds the first superimposed signal and the error signal and supplies the error signal to the first converting means, and the subtracting means subtracts the second superimposed signal from the control signal and outputs it. As a result, the finally obtained control signal is hardly affected by the quantization error associated with the A / D conversion and the D / A conversion.

[0009]

【Example】

 An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention. The embodiment shown in this figure is applied to the head positioning of a magnetic disk device. In this figure, the magnetic head 1 reads the servo information recorded between the sectors on the track of the magnetic disk 2 and outputs a signal corresponding to this. This servo information includes information indicating the displacement from the center of the track.

Reference numeral 3 denotes a head amplifier that amplifies a signal corresponding to servo information output from the magnetic head 1. A position signal demodulation circuit 4 demodulates a signal corresponding to the servo information amplified by the head amplifier 3 to generate a position signal corresponding to the position of the magnetic head 1. This position signal is supplied to the subtractor 5 as a feedback signal Sf in the servo control system.

The subtractor 5 subtracts the feedback signal Sf from the head position target value signal St supplied from the outside, and generates a position error signal Se indicating a difference from the head position target value. The adder 6 adds the position error signal Se and a dither signal Sd1 supplied via a D / A converter 8a described later, and outputs the addition result to the A / D converter 7. The A / D converter 7 converts the addition result supplied from the adder 6 into a digital signal and outputs it.

Reference numeral 9 denotes a digital controller, which performs a process of compensating for an error in the head position according to the difference between the target value of the head position and the actual head position. This digital controller 9 includes two PID controllers 91 and 92 that perform compensation processing by a digital PID (proportional / integral / derivative) control method, and a dither signal generator 93 that generates a dither signal that is a slight vibration of white noise. Have

The PID controller 91 takes in the digital signal supplied from the A / D converter 7 described above, performs a predetermined compensation process on the digital signal, and outputs the digital signal to the D / A converter 8b. The D / A converter 8b converts the output signal of the PID controller 91 into an analog signal and supplies it to the subtractor 10 as the manipulated variable signal Sc. On the other hand, the PID controller 92 takes in the dither signal output from the dither signal generator 93, performs a process equivalent to that of the PID controller 91, and then outputs it to the D / A converter 8c. The D / A converter 8c converts the output signal of the PID controller 92 into an analog signal and supplies it to the subtractor 10 as a dither signal Sd2. Further, the dither signal generator 93 outputs the same dither signal as that output to the PID controller 92 to the D / A converter 8a. The D / A converter 8a converts the dither signal into an analog signal, and then supplies it as the dither signal Sd1 to the adder 6 described above.

The subtractor 10 subtracts the compensated dither signal Sd2 supplied via the D / A converter 8c from the manipulated variable signal Sc supplied from the D / A converter 8b. The subtraction result is output as the final manipulated variable signal. A VCM (voice coil mortor) driver 11 drives the VCM 12 based on the final manipulated variable signal supplied from the subtractor 10. The VCM 12 moves the magnetic head 1 so as to converge on the target position by the drive control of the VCM driver 11, and positions the magnetic head 1.

With such a configuration, first, the feedback signal Sf output from the position signal demodulation circuit 4 is subtracted from the head position target value signal St 1 supplied from the outside, and this becomes the position error signal Se. Then, the dither signal Sd1 output from the D / A converter 8a is added to the position error signal Se and input to the A / D converter 7. The input characteristics at this time are shown in FIG. In this figure, the noise width of the added dither signal Sd is ± d / 2 when the 1-bit decomposition width of the A / D converter 7 is d.

The output of the A / D converter 7 is subjected to a predetermined compensation process by the PID controller 91, and then output from the D / A converter 8b as a manipulated variable signal Sc. The output characteristic of the D / A converter 8b is shown in FIG. In this figure, (a) shows the output characteristics after quantization, and (b) shows the quantization error that occurs corresponding to the output characteristics after quantization shown in (a). In this case, the output Q (x) (vertical axis) is a discrete value with the 1-bit decomposition width d including the influence of noise due to the dither signal as a unit, and the width of the quantization error generated at this time is the maximum. ± d.

On the other hand, the dither signal output from the dither signal generator 93 is subjected to processing equivalent to that of the PID controller 91 by the PID controller 92, and then input to the D / A converter 8c. Then, after this, it is supplied to the subtractor 10 as the dither signal Sd2. Here, the output characteristic of the subtractor 10 is shown in FIG. In this figure, (a) shows the output characteristic, and (b) shows the change in the quantization error that occurs corresponding to the output characteristic shown in (a). As is clear from this figure, although the output is noisy, the step-like characteristic unlike the output Q (x) shown in FIG. 5B is not seen. Although the quantization error at this time is also noisy, the sawtooth characteristic shown in FIG. 5C is not seen. That is, the characteristics are exactly the same as the input characteristics shown in FIG.

Further, FIG. 6 is a diagram showing a comparison between the characteristics after quantization according to the conventional technique shown in FIG. 5B and the characteristics after quantization according to the present embodiment shown in FIG. 4A. is there. In this figure, (a) shows the spectrum that appears due to the characteristics of the prior art, and not only the fundamental input frequency f but also the high frequency components such as 2f, 3f, ... Equivalent to) has appeared. On the other hand, (b) shows the spectrum that appears due to the characteristics of this embodiment, and although it is somewhat noisy overall, the spectrum appears only at the basic input frequency f.

Based on the signal after the dither signal subtraction having such an output characteristic, the VCM driver 12 drives the VCM 11 and the magnetic head 1 is positioned so as to converge to the target position. Therefore, in the positioning control of the magnetic head 1, the influence of the quantization error due to the A / D conversion and the quantization accompanying the D / A conversion hardly appears. Further, since the error generated in this embodiment is like white noise, it has almost no adverse effect on the positioning control. In this embodiment, the digital controller 9 is shown to be composed of the PID controllers 91 and 92 and the dither signal generator 93, but the present invention is not limited to this, and for example, a CPU realizes these functions. It may be done.

[0020]

[Effect of the device]

 As described above, according to the present invention, the superimposition signal generating means generates the first superimposition signal corresponding to the minute signal, and the second superimposition in which the compensation characteristic is added to the first superimposition signal. Generate a signal. Then, the adding means adds the first superposed signal and the error signal and supplies them to the first converting means, and the subtracting means subtracts the second superposed signal from the control signal and outputs it. As a result, the finally obtained control signal is hardly affected by the quantization error associated with the A / D conversion and the D / A conversion, so that the quantization error can be obtained at a low cost without increasing the number of quantization bits. The effect of being able to suppress the influence of is obtained. Further, there is an effect that the range in which a good dynamic range is obtained is not limited.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is a diagram showing an input characteristic to an A / D converter 7 in the embodiment.

FIG. 3 is a diagram showing an output characteristic of a D / A converter 8a in the embodiment, (a) showing an output characteristic,
(B) shows a change in the quantization error that occurs corresponding to the output characteristic shown in (a).

FIG. 4 is a diagram showing the output characteristic of the subtractor 10 in the same embodiment, where (a) shows the output characteristic and (b) shows the quantum generated corresponding to the output characteristic shown in (a). The error is shown.

FIG. 5 is a diagram showing characteristic changes before and after quantization, (a)
Shows the characteristics of the analog signal before quantization, and (b)
Indicates the characteristics of the quantized digital signal. Further, (c) shows a quantization error generated by the quantization.

FIG. 6 is a diagram showing a comparison between the characteristics after quantization according to the conventional technique shown in FIG. 5B and the characteristics after quantization according to the present embodiment shown in FIG. 4A; Shows the spectrum that appears according to the conventional technique, and (b) shows the spectrum that appears according to this embodiment.

[Explanation of symbols]

 1 magnetic head 2 magnetic disk 3 head amplifier 4 position signal demodulation circuit 5,10 subtracter 6 adder 7 A / D converter 8a-8c D / A converter 9 digital controller 11 VCM 12 VCM controller 91, 92 PID controller 93 Dither signal generator

Claims (1)

[Scope of utility model registration request] 1. A first conversion means for converting an error signal indicating a difference from a target value into a digital signal and outputting the digital signal, and a means having a predetermined compensation characteristic, the output of the first conversion means. In a digital servo circuit having a compensating means for generating a control signal for converging the controlled object accordingly and a second converting means for converting the control signal into an analog signal and outputting the analog signal, a first signal corresponding to a minute signal is provided. Generate a superimposed signal,
A superimposition signal generating means for generating a second superimposition signal in which the compensation characteristic is added to the first superimposition signal, and an addition result of the first superimposition signal and the error signal are supplied to the first conversion means. And a subtraction unit for subtracting and outputting the second superimposed signal from the control signal.