JPH087627B2 - Feedforward drive signal generation method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial field of application] The present invention relates to a magnetic disk or optical disk recording / reproducing apparatus or the like, and when an actuator equipped with a recording / reproducing head is driven by the motor, a drive applied to the motor. The present invention relates to a method of generating a signal.

[Conventional technology]

In a large-capacity magnetic disk or optical disk device used for online processing, a multi-actuator configuration is generally adopted for the purpose of accessing stored contents at high speed. In this case, a plurality of actuators are used. Since it operates independently, the driving force for the actuator moving at high speed is transmitted as a disturbance to other actuators performing position tracking control via the actuator support mechanism, which may cause deviation in the head position. There is.

As a countermeasure against this, a method has been proposed in which the drive signal applied to the drive motor of the actuator is trapezoidal and the drive force component near the mechanical resonance frequency is removed.
Further details are disclosed in 263699.

That is, as shown in FIG. 8, the access period II and III inserted in the middle of the position follow-up control period I of the actuator are divided into the left acceleration and the right deceleration, respectively.
In each of them, trapezoidal drive signals S ₁ and S ₂ having equal amplitudes and opposite polarities to each other and substantially equipodal are used, whereby a suitable actuator is driven.

Actually, the access period is divided into the feedback speed control period III, which is the remaining 20 to 30% of the feedforward speed control period II, and the actuator is decelerated to the target position while removing the control error. In the period III, the reference speed profile V is set as V = f (x) according to the residual distance x to the target position, and the value of the drive signal is controlled following this. .

[Problems to be solved by the invention]

However, the magnetic material used for the drive motor of the actuator has a large characteristic deviation at the time of manufacture and a large characteristic change due to temperature.
The deviation from the target position due to the feed forward control becomes large due to the change in the driving force resulting from this, and it is not possible to completely compensate for the deviation in the feed back control near the target position. It is possible, and the end side of the drive signal waveform does not have the desired isosceles trapezoidal shape, causing vibration interference between actuators or causing a seek error due to deviation from the target position. .

[Means for solving problems]

In order to solve the above problems, the present invention is configured by the following means.

That is, in the method of generating the drive signal described above,
The first invention detects that the actuator has traveled 1/2 distance of the travel distance, and when the actual time until the actuator travels 1/2 distance is shorter than the scheduled reference time, the acceleration direction is determined according to the detection. The generation of the trapezoidal drive signal is stopped, and the deceleration direction trapezoidal drive signal is generated after the difference between the reference time and the actual time is subtracted from the generation period of the deceleration direction trapezoidal drive signal. When it is long, the deceleration direction trapezoidal drive signal is generated from the time when the difference between the real time and the reference time has elapsed from the time when the detection was performed.

Further, in the second invention, on the same premise as in the first invention, until the actuator detects that the actuator has moved a predetermined distance in the moving distance in the upper bottom section of the acceleration direction trapezoidal drive signal, and moves the predetermined distance. When the actual time of is shorter than the scheduled reference time, the generation of the upper base section is stopped in accordance with the detection, and the subsequent leg sections are generated, and then the upper base section has the same upper period as the upper base section. When generating a deceleration direction trapezoidal drive signal, when the actual time is longer than the reference time, the upper base section is extended and generated from the time when the detection is performed until the time difference between the actual time and the reference time elapses, Next, it is assumed that a leg section is generated and then a deceleration direction trapezoidal drive signal having an upper base section having the same period as the extended upper base section is generated.

[Action]

Therefore, in each of the inventions, the acceleration direction trapezoidal drive signal generation period is corrected according to the short or long real time according to the torque constant of the drive motor, and the deceleration direction symmetrical to this is also corrected. A trapezoidal drive signal will be generated and changes in the torque constant will be compensated for by adjusting the drive energy.

〔Example〕

Hereinafter, the details of the present invention will be described with reference to the drawings illustrating examples.

FIG. 3 is a block diagram of the first invention, which is a memory (hereinafter, MM) 11 in a microcomputer (hereinafter, μCP) 1.
And the processor (hereafter CPU) omitted in the figure
According to the program stored in the memory M11, an adder (hereinafter referred to as ADD) 12 and a timer (hereinafter referred to as TIM) 13 are configured to generate a drive signal, and the contents of ADD12 are converted into a digital-to-analog converter ( Hereinafter, the signal is sent to the DAC 2), which then gives it to the power amplifier (hereinafter, PA) 3 as an analog signal, amplifies it to a predetermined power, and then drives it to the drive motor of the actuator (not shown). It is supposed to be sent as S _D.

In addition, the position of the head mounted on the actuator is detected as a two-phase position signal S _P by general means not shown in the figure, the signal S _P is counted by a counter (hereinafter, CUT) 4, and the actuator is moved. Therefore, the number of tracks on the recording medium that the head has traversed is calculated, and the number of tracks and the data showing the 1/2 distance with respect to the required moving distance of the actuator given by the contents of MM11 are used to detect the moving distance (hereinafter, MDD) 5 compares and the detection signal is given to μCP 1 when the two match.

In addition, a feedback control system is provided separately, and μCP1 compares the drive signal value V _F resulting from this with the drive signal value V _B generated by itself, and the condition of V _F > V _B is satisfied. If this is done, a control signal is sent to a switching circuit (not shown) on the PA3 input side to switch to feedback control.

FIG. 4 is a principle diagram showing a situation in which a drive signal is generated by μCP1, in which the drive signal waveform is divided into unit intervals T _{0 to} T ₇ at intervals of each change point, and the change state of each unit. Determine the increment values Δ _{0 to} Δ ₇ according to the above, store them in the MM11 in μCP1 as a table, and use this to store the CPU.
Is to execute the processing of FIG.

That is, FIG. 5 is a flow chart of the signal generation state by the CPU in the μCP1, and the access command ACC to the μCP1.
Is given as residual track data on how many tracks on the recording medium should be moved, the “access start?” 101 becomes Y (TES) accordingly, and the movement distance of the actuator indicated by the residual track data. According to
This value is stored in the MM11 by the “1/2 distance data set” 102, and the “parameter load (Tn, Δn, n = 1
・ 1/2 distance) ”103, and Tn and Δn are read out when n = 0, and this is calculated by“ TIM = Tn · Δ = Δn ”111.
It is set to M13 and is prepared as an added value by ADD12, and 1/2 distance is sent to MDD5.

Then, “OUT = OUT + Δn · TIM = TIM−1” 112 is used to add Δn by ADD12 and subtract the contents of TIM13, and check “1/2 distance move?” 121 by the detection output of MDD5, which is N ( during NO) is repeated step 112 after through the N of "TIM = 0?" 122, if step 122 is Y, a "interval T ₃ End?" 123 in FIG. 4 and a checking, which is N When the counter configured in the CPU is "n =
The signal from the section T ₀ to the section T ₃ is generated by incrementing by “n + 1” 124 and repeating Step 111 and subsequent steps until Step 123 becomes Y.

If step 123 becomes Y, a "detection flag set?" 131, which will be described later, is judged, and Y of this is taken as a premise, and the above "V
_F ＞ V _B ? Check 132 and, depending on its N, step 12
3 in the same manner as in "interval T ₇ end?" 133 was a checking, this is N
Between steps 111 and subsequent steps through step 124,
Generates signals up to interval T ₇ .

As a result, the drive signal shown in FIG. 4 is transmitted as the output OUT of the ADD 12, and the acceleration direction trapezoidal drive signal S ₁ and the deceleration direction trapezoidal drive signal S ₂ are generated.

On the other hand, the torque constant of the drive motor is large,
If the actuator moves to 1/2 distance before step 123 becomes Y and step 121 becomes Y, "OUT = -OUT
・ MDD reset / detection flag set ”141
Inverting the output polarity, and is thereby reset the MDD4, and excisional detection flag corresponding to step 131, and a checking similarly "interval T ₂ running?" 142 and step 123, when it is Y ' T ₅ = T ₂ ′ · n = n + 2 ”151, the time T ₂ ′ at which the section T ₂ has been executed up to now is obtained from the contents of TIM 13 as the execution time of the section T ₅ , and“ 2 ”is added to the counter. Then, the execution of the sections T ₃ and T ₄ is skipped, and if step 142 is N, the section T ₃ is already being executed, and this execution time T ₃ ′ is obtained in the same manner as described above, and “T ₄
= T ₃ ′ ”152, the execution time of section T ₄ is determined, and step 1
As in 24, the counter is incremented by "n = n + 1" 153, and then steps 111 and subsequent steps are repeated.

At this time, since MDD5 has been reset in step 141, step 121 becomes N and step 1
It will be via 22 or later.

On the other hand, if the torque constant of the drive motor is small and the actuator does not move to 1/2 distance even if step 123 becomes Y, and step 131 is N, step 123 becomes Y until the time. To show the planned reference time t _s required to move to 1/2 distance shown in Fig. 4, this and the actual time t _RD required for the actuator to move to 1/2 distance
For the purpose of obtaining the time difference _ΔtD from
After TIM13 is incremented by, the same "jump 1/2 distance?" 162 as in step 121 is judged, and while this is N, step 161 and subsequent steps are repeated, while step 162 becomes Y. , "MDD reset / detection flag set" 171 is performed in the same manner as in step 141, and this time, the content of TIM13 is subtracted by "TIM = TIM-1" 181 and step 181 is executed until "TIM = 0?" 182 becomes Y. The subsequent steps are repeated, and as step 182 becomes Y, the time difference ^ΔtD has elapsed, so step 12
Repeat steps 111 through 4 through 4.

At this time as well, since step 121 becomes N and step 131 becomes Y, step 111 and subsequent steps are repeated until step 133 becomes Y.

When step 132 becomes Y, "switching circuit control signal transmission" 191 is immediately performed, switching to feed back control is performed as described above, and then the processing returns to step 101 via "RET".

Therefore, as shown in the waveform of the drive signal in FIG.
The waveform is corrected according to the magnitude of the torque constant, and the drive status of the actuator becomes normal due to the decrease or increase of the drive energy.

However, in the figure, it is clear that T ₃ = T ₄ = 0 and the correction status of the waveform is clear. The thick solid line shows the reference waveform, the dotted line shows the large torque constant, and the thin solid line shows the small torque constant. , If the torque constant is large, the real time t
_{When RF} is shorter than the reference time t _{S and the} MDD4 detects it at the time t _DF when the actuator moves 1/2 distance, the generation of the drive signal S ₁ is stopped in response to this detection and the drive signal S ₂ generates the drive signal S _2F after the time difference Δt _F is subtracted from the occurrence time period T _W , and when the torque constant is small and the real time t _RD is longer than the reference time t _S , the time difference from the detection time point t _DD Drive signal S _2D from the point when Δt _D
Is generated.

Incidentally, CUT4 is for performing counting by 2-phase position signals S _P in units of 1/2 of 1 Toratsukupitsuchi, detection by MDD5 is made by precision of 1/2 Toratsukupitsuchi, it can be carried out sufficiently accurately detected.

Further, in the case shown by the dotted line in FIG. 1, the detection time point t _DF
, The polarity of the drive signal is immediately inverted, which increases the high-frequency component and increases mutual vibration interference, while the thin solid line shows that the waveform of the drive signal hardly changes and the high-frequency component Therefore, it is preferable to select the torque constant in such a direction that the real time t _R becomes the reference time t _S or more.

FIG. 6 is a block diagram of the second invention. An analog digital converter (hereinafter referred to as ADC) 6 is added to FIG. 3, and each value of the two-phase position signal S _P is converted into a digital signal by the ADC 6. In addition, it is given to MDD5, and MDD5 detects the movement of a predetermined distance according to the sum of the upper bit by CUT4 and the lower bit by ADC6.

If the ratio of the upper bottom section length to the lower bottom section length of the drive signal is 0.47, when the torque constant is the reference value corresponding to the reference time t _S , the actuator is terminated at the time when the generation of the section T ₁ is finished. Since it is moved by 1/4 of the moving distance, the resolution of the ADC6 will be sufficient if it is defined as 1/4 track pitch.

However, if a 4-phase position signal is used as signal S _P , CUT4
By this, the count is performed for each 1/4 track pitch, so that the ADC 6 can be omitted.

Moreover, in FIG. 6, the upper bottom section T is used instead of the half distance.
_The predetermined distance by the movement in ₁ is used, and MDD5 detects that this matches the actual movement distance.

FIG. 7 is a flow chart of the situation in which the drive signal is generated by the CPU in the μCP1, and it is assumed that it is performed based on each data of FIG. 4 as in FIG. 5, and is almost the same as in FIG. However, the contents of the following steps are different.

That is, the data of a predetermined distance is stored in the MM 11 by the “predetermined distance data set” 202, and the predetermined distance is M by the “parameter load (Tn, Δn, n = 0, predetermined distance)” 203.
Given to DD5, "predetermined distance?" Determines a predetermined distance movement of the actuator by 221, "interval T ₁ End?" 223
By then a checking generation end of the upper base section T _1, "MDD reset and detection Furagusetsuto" 241 without polarity reversal by OUT = -OUT in, _"1 running interval T?" 242 if Y, the TIM13 The execution time T ₁ ′ so far indicated by the contents is used as the execution time of the section T ₆ by “T ₆ = T ₁ ” 251, and when the step 242 is N, the execution of the section T ₀ is still in progress. , “T ₅ = T ₀ ′ · T ₆ = 0 · n = n + 1” 252,
The execution time T ₀ ′ so far is set as the execution time of the section T ₅ , and
The execution time of the section T ₆ is set to zero and the execution of the section T ₁ is skipped.

Moreover, "predetermined distance?" 262 if the Y, the "MDD reset and detection Furagusetsuto _{· T 6 = T 6 +2 ·} Δt D " 271, adds 2 times the value of the difference time Delta] t _D to T ₆ The execution time of section T ₆ .

Therefore, the waveform of the drive signal is as shown in FIG.
Similar to FIG. 1, the waveform is modified according to the magnitude of the torque constant, and the drive status of the actuator becomes normal.

Also in FIG. 2, T ₃ = T ₀ = 0, the thick solid line shows the reference waveform, the dotted line shows the large torque constant, and the thin solid line shows the small torque constant. If so, the real time t required to travel a given distance t
_RF is shorter than the reference time t _S , the actuator moves a predetermined distance, at the time t _DF detected by MDD5, the generation of the upper base section T ₁ is stopped, and the subsequent leg section T ₂ is generated,
Then, the drive signal S _2F having the upper bottom section T ₆ of the same period as the real time T ₁ ′ of the section T ₁ is generated.

Also, the torque constant is small and the real time t _RD is the reference time t
_If it is longer than _S, it occurs by extending the upper base section T ₁ from the detection time t _DD to the time when the time difference Δt _D has elapsed, and then the leg section
The T ₂ occurs, then the driving signals S _2D having upper base section T ₆ +2 · Δt _D of extended upper base section T ₁ +2 · Δt _D same period
Will be generated.

Therefore, the drive signal waveform changes in accordance with the torque constant, but the trapezoid itself does not change, the high-frequency component hardly increases, and the increase in vibration mutual interference becomes extremely small.

However, it is more difficult to completely compensate for the change in the torque constant than in the case of FIG. 1, and it is suitable when high-precision position control is not required, but when extremely high-precision position control is required, By using the method of FIG. 1 and the method of FIG. 2 together, a 20% torque constant change can be elevated to a position deviation of about 5%.

It should be noted that in FIGS. 3 and 6, μCP1 may not be used and a combination of various logic circuits and arithmetic circuits may be used. When the drive motor responds to a digital signal,
DAC2 can be omitted.

Further, in FIG. 5 and FIG. 7, depending on the conditions,
Swap steps, or replace with other equivalents,
Alternatively, various modifications are possible, such as unnecessary elements may be omitted.

〔The invention's effect〕

As is apparent from the above description, according to the present invention, when the trapezoidal drive signal is used, the waveform is corrected according to the change of the torque constant of the drive motor, and the error occurrence of the position control situation based on the change of the torque constant is prevented. At the same time, the vibration interference between the actuators is also prevented, so that a remarkable effect can be obtained in the generation of the drive signal to be given to the actuator drive motors of various recording / reproducing devices.

[Brief description of drawings]

1 to 7 show an embodiment of the present invention, FIGS. 1 and 2 are waveform diagrams of generated drive signals, FIG. 3 is a block diagram of the first invention, and FIG. FIG. 5 is a flow chart of a situation in which a drive signal is generated by the configuration of FIG. 3, FIG. 6 is a block diagram of the second invention,
FIG. 7 is a flow chart of the situation in which a drive signal is generated by the configuration of FIG. 6, and FIG. 8 is a waveform diagram showing a conventional example. 1 ... μCP (microcomputer), 4 ... CUP (counter), 5 ... MDD (moving distance detection circuit), 11 ... MM
(Memory), 12 ...... ADD (adder), 13 ...... TIM (Timer), S _D ...... drive signal, S _P ...... 2-phase position signals, t _S ...... reference time, t _RF, t _RD ... … Real time, Δt _F , Δt _D …… Difference time, T
₁ , T ₆ ...... Upper base section, T ₂ ...... Leg section.

Continuation of the front page (56) References JP-A-56-60910 (JP, A) JP-A-55-102009 (JP, A) JP-A-49-37342 (JP, A) JP-A-49-71377 (JP , A)

Claims (2)

[Claims] 1. A method for generating a trapezoidal drive signal of which the amplitude is equal during acceleration and deceleration applied to a drive motor of an actuator and which have mutually opposite polarities and which are substantially isosceles.
When it is detected that the actuator has moved 1/2 distance of the moving distance, and when the actual time until moving the 1/2 distance is shorter than the scheduled reference time, the acceleration direction trapezoidal drive signal according to the detection. Of the deceleration direction trapezoidal drive signal is generated, and the deceleration direction trapezoidal drive signal is generated after the difference between the reference time and the actual time is subtracted from the generation period of the deceleration direction trapezoidal drive signal, and the actual time is longer than the reference time. In some cases, the feedforward drive signal generation method is characterized in that the deceleration direction trapezoidal drive signal is generated from a time point after a difference time between the real time and the reference time from a time point when the detection is performed. 2. A method for generating a trapezoidal drive signal which has equal amplitudes and opposite polarities to each other during acceleration and deceleration applied to a drive motor of an actuator, and which is substantially equipodal.
Detecting that the actuator has moved a predetermined distance in the moving distance in the upper bottom section of the acceleration direction trapezoidal drive signal, and when the actual time until moving the predetermined distance is shorter than a predetermined reference time, the detection is performed. According to the above, the generation of the upper base section is stopped, and the subsequent leg sections are generated, and then the deceleration direction trapezoidal drive signal having the upper base section having the same period as the upper base section is generated, and the real time is used as a reference. When it is longer than the time, it occurs by extending the upper base section from the time when the detection is performed to the time when the difference time between the real time and the reference time elapses, then the leg section is generated, and then the extension is performed. A method of generating a feedforward drive signal, comprising generating a deceleration direction trapezoidal drive signal having an upper base section having the same period as an upper base section.