JPS6326893B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to, for example, a head position control system of a magnetic disk device or a position control system of an NC machine tool, and particularly relates to a malfunction detection method for detecting a malfunction of a movable part in the position control system itself.

Conventionally, in this type of position control system, a movable part (for example, a magnetic head in a magnetic disk device)
As a method for detecting malfunctions in the position control system itself, a method has been used to detect the winding of the coil. With this method, it is difficult to prevent malfunctions of the detection circuit itself, and the detection accuracy cannot be set strictly. For example, in a magnetic disk drive, a method has been adopted in which information regarding track addresses is recorded in advance on each track of the disk, and the recorded information is read each time to confirm the track address. However, since this method cannot distinguish between read errors and positioning errors,
Both were used together in disks.

In view of these circumstances, the present invention provides a new malfunction detection method that can easily detect malfunctions of movable parts in the position control system itself. A reference position detector is provided at a predetermined position other than both ends of the operating range of the section, the output signal value being different depending on when the operating position detector section is on one side and on the other side of the predetermined position, When a movement command is given to the movable part, it is determined in advance whether or not the operating position detector crosses the reference position detector when executing the movement command, and then the movable part is moved to the The present invention is characterized in that the position control system is actually moved in accordance with the movement command, and if the result does not match the above-mentioned determination, it is determined that there has been a malfunction in the position control system.

Hereinafter, one embodiment of the method according to the present invention will be described in detail with reference to the drawings.

FIG. 1 is a diagram showing a head position control system of a magnetic disk device to which the method according to the present invention is applied.
1A shows the case where the heads 1,1 are in the outermost track position, and FIG. 1B shows the case where the magnetic heads 1,1 are in the innermost track position.

In these figures, reference numeral 2 is a disk, and concentric circles are formed on both sides of the disk 2.
There are 200 magnetic tracks (0 to 199 tracks). In this case, the outermost track is the 0 track, the innermost track is the 199 track, and 3 is a spindle motor for rotating the disk 2.

The magnetic heads 1, 1 are mounted on head arms 4, 4, respectively, and these head arms 4, 4 are both mounted on a carriage 5. The carriage 5 is used to move the magnetic heads 1, 1 to a designated track (seek operation), and this seek operation actuator 6 performs this operation. The actuator 6 is composed of a voice coil 7 attached to the carriage 5, magnets 8, 8, etc. that create a magnetic field in the moving range of the voice coil 7, and when a direct current is applied to the voice coil 7, the applied direct current Depending on the direction of the current, the voice coil 7 receives electromagnetic force in the left or right direction, and this electromagnetic force moves the carriage 5 in the left or right direction.

An inductosyn moving scale 10 is attached to the side surface of the carriage 5, and an inductosyn fixed scale 11 is fixed to a chassis or the like at a position facing the inductosyn moving scale 10. Inductosyn fixed scale 1
As is well known, a comb-shaped copper foil pattern is formed on the surface facing the inductosin moving scale 10 of No. 1, and a high frequency current is applied to this copper foil pattern. Further, a one-turn copper foil pattern is formed on the surface of the inductosyn moving scale 10. Then, when the carriage 5 moves and the inductosin moving scale 10 moves along the inductosin fixed scale 11, an alternating current is induced in the copper foil pattern of the inductosin moving scale 10. This alternating current is converted by the waveform conversion circuit 9 (see FIG. 3) and output as a sawtooth position detection signal as shown in FIG. 2A. Based on this sawtooth position detection signal, the electric circuit shown in FIG. 3 detects the position of the carriage 5. Further, in this case, zero-crossing points C, C, . . . of the output of the waveform conversion circuit 9 correspond to respective track positions on the disk 2, respectively.

At the bottom of the carriage 5 is a shutter 1 whose long side is slightly longer than the moving distance of the carriage 5.
2 is attached, and the left end 1 of this shutter 12
2a, that is, the distance L shown in FIG.
A photo sensor 13 is fixedly arranged on a chassis or the like in the middle. Then, the carriage 5 moves to the left in FIG.
When the magnetic heads 1 and 1 are located exactly between the 99th track and the 100th track, the light from the photo sensor 13 is blocked by the shutter 12, and the light from the photo sensor 13 is subsequently blocked by the shutter 12 until the magnetic heads 1 and 1 reach the 199th track.
The light is blocked by the shutter 12. As a result, the photo sensor 13 detects that the magnetic heads 1, 1 are 0.
When located on tracks 100 to 199, it outputs a "low" level signal, and when located on tracks 100 to 199, it outputs a "high" level signal.

The above-mentioned shutter 12 and photo sensor 13
is for detecting the reference position of the magnetic heads 1,1. That is, the inductosyn possible scale 10 and the inductosin fixed scale 11
In the head position detection by
It is not possible to detect absolute position. In this embodiment, inductosin is used to detect the head position, but the head position can also be detected using an optical scale or based on servo information recorded on the disk surface. Various mechanisms have been adopted.
However, with either of these mechanisms it is usually not possible to detect absolute position. Therefore, a reference position detector or a track on which the reference position is written (in the case of a servo) is required to detect the absolute position of the magnetic head. In this embodiment, as described above, the magnetic heads 1, 1
When moving from the 99th track to the 100th track, the output of the photo sensor 13 changes from "low" level to "high" level, and when moving from the 100th track to the 99th track, the output of the photo sensor 13 changes from "high" level to " changes to low level. Therefore, this photo sensor 1
The absolute position of the magnetic head can be detected based on the output change of 3.

The above is the mechanical configuration of the embodiment shown in FIG. Incidentally, this embodiment differs from the head position control system in a conventional magnetic disk device in the position of the photo sensor 13. That is, in the conventional head position control system, the photo sensor 13 was provided at a position A shown by a dashed line in FIG. Then, magnetic heads 1 and 1 are 1 to 199
When the photo sensor is in the track position, the light from the photo sensor is blocked by the shutter 12, and the magnetic heads 1, 1 are
Only when the magnetic head reaches the 0 track position does the light from the photo sensor stop being blocked by the shutter 12, whereby the absolute position of the magnetic heads 1, 1 is detected.

On the other hand, in this embodiment, the photo sensor 13 is located exactly in the middle of the moving distance L of the left end 12a of the shutter 12, as described above. and,
More importantly, in addition to the function of detecting the absolute position of the magnetic heads 1, 1, the photo sensor 13 also has the following functions:
It also has the function of a position detector for detecting malfunctions of the magnetic heads 1, 1 (ie, malfunctions of the carriage 5).

Next, the electric circuit of the position control system described above will be explained with reference to FIG. In this figure, the same parts as those in FIG. 1 are given the same reference numerals. In addition, in the following explanation, "1" signal,
The "0" signals each indicate a signal at a binary logic level.

In Fig. 3, inductosin movable scale 1
The output of 0 is converted by the waveform conversion circuit 9 and then supplied to the waveform shaping circuit 20 . Waveform shaping circuit 2
0 converts the output of the sawtooth waveform conversion circuit 9 into a pulse signal, and when the output of the waveform conversion circuit 9 shown in FIG. 2A is supplied, its zero cross point C,
A thin pulse signal shown in FIG. 2B is generated at C. The track address counter 21 stores the current address of the magnetic heads 1, 1, and is a presetable up/down counter, and is connected to the up/down designation terminal UD.
and a preset terminal PS. and,
When a "1" signal is supplied to the up/down designation terminal UD, the output of the waveform shaping circuit 20 is counted up, and when a "0" signal is supplied, the output of the waveform shaping circuit 20 is counted up.
Counts down the output of the
When a "1" signal is supplied to PS, a preset value in the track address counter 21 is preset.

On the other hand, the output of the photo sensor 13 is supplied to a waveform shaping circuit 22. The waveform shaping circuit 22 shapes the output of the photo sensor 13, and outputs a "1" signal when the output of the photo sensor 13 is greater than a predetermined threshold level TH, and outputs a "0" signal when it is smaller. (See Figure 4). The output of the waveform shaping circuit 22 is supplied to a malfunction detection circuit 23 and a restore circuit 24, respectively.The malfunction detection circuit 23 is a circuit for detecting malfunction of the magnetic heads 1,1. In other words, this malfunction detection circuit 2
3 detects a malfunction of the magnetic heads 1, 1 based on the output of the track address counter 21, the output of the track address register 25, and the output of the waveform shaping circuit 22, and sends the detected output to an alarm (lamp, buzzer, etc.) 26. supply Note that this malfunction detection method will be described in detail later. The track address register 25 stores a target address (target track address to which the magnetic heads 1, 1 should seek) supplied from an external device, and its output is supplied to a malfunction detection circuit 23 and a comparison circuit 27. Further, when a "1" signal is supplied to the reset terminal R, it is reset. The comparison circuit 27 compares the output (current address) of the track address counter 21 and the output (target address) of the track address register 25. If the current address > the target address, a "1" signal is output from the terminal _T2 . If current address < target address, a "0" signal is output from terminal _T2 , and if current address = target address, a "0" signal is output from terminal _T1 . Please note that the current address≠
In the case of the target address, a "1" signal is output from the terminal _T1 .

The restore circuit 24 is a circuit used when restoring heads 1, 1, that is, returning heads 1, 1 to 0 track. The voice coil drive circuit 28 is a circuit that drives the voice coil 7. That is, when the voice coil drive circuit 28 receives a "1" signal to both its input terminals I ₁ and I ₂ ,
The output terminal T ₁ supplies a positive DC voltage and the output terminal T ₂ supplies a negative DC voltage to the voice coil 7, thereby driving the voice coil 7 to the left in FIG.
When a “1” signal and a “0” signal are supplied to I ₁ and I ₂ respectively, the output terminal T ₁ supplies a negative DC voltage and the output terminal T ₂ supplies a positive DC voltage to the voice coil 7. Drive to the right in FIG. Further, when a "0" signal is supplied to the input terminal _I1 , no DC voltage is output from the output terminals _T1 and _T2 , and therefore the voice coil 7 is in a stopped state.

Next, the operation of the head position control system shown in FIGS. 1 and 3 will be explained.

First, the case where heads 1, 1 perform a normal seek operation will be described.

When a target address is supplied from an external device to the track address register 25, the target address is compared in a comparison circuit 27 with the output of the track address counter 21, that is, with the current address of the magnetic heads 1,1. If the target address > the current address, the output terminal of the comparison circuit 27
A “1” signal is output from each of T ₁ and T ² , and the signal is sent to the voice coil drive circuit 28 via OR circuits 29 and 30.
is supplied to the input terminals I ₁ and I _{2 of} . As a result, from the output terminals T ₁ and T ₂ of the voice coil drive circuit 28,
A DC voltage with T ¹ =positive and T ₂ =negative is output, thereby driving the voice coil 7 to the left in FIG. The voice coil 7 is driven to the left,
When the carriage 5 moves to the left, the inductosin fixed scale 10 also moves to the left, and as a result, the sawtooth position detection signal shown in FIG. 2A is output from the waveform conversion circuit 9. . This sawtooth current is converted into a pulse signal shown in FIG.
At this time, the "1" signal is supplied to the up/down designation terminal UD of the track address counter 21 from the terminal _T2 of the comparator circuit 27, and therefore the track address counter 21 receives the output of the waveform shaping circuit 20. Count up. When the magnetic heads 1, 1 reach the track position corresponding to the target address, the output of the track address counter 21 and the output of the track address register 25 match, and as a result, the terminal of the comparator circuit 27
A “0” signal is output from T ₁ , and this “0” signal is supplied to the input terminal I ₁ of the voice coil drive circuit 28 via the OR gate 29 . As a result, no DC voltage is output from the output terminals T ₁ and T ₂ of the voice coil drive circuit 28, and the voice coil 7 stops.

In addition, if the target address < the current address, a "1" signal and a "0" signal are output from the output terminals T ₁ and T ₂ of the comparator circuit 27, respectively, which causes the voice coil 7 to move to the right in FIG. While being driven, the track address counter 21 counts down the output of the waveform shaping circuit 20. and,
When the magnetic heads 1, 1 reach the track position corresponding to the target address, the output of the track address counter 21 and the output of the track address register 25 match, and as a result, a "0" signal is output from the output terminal _T1 of the comparator circuit 27. is output, and the voice coil 7 stops.

Next, the restore operation of heads 1, 1 will be explained. Note that this restore operation is performed when the power is turned on or when the heads 1, 1 make a seek error for some reason. When a restore signal (signal for commanding restoration) is supplied from a control circuit (not shown) to the restore circuit 24, the restore circuit 24 first disables the comparator circuit 27 and reshapes the waveform being supplied to the input terminal _I1 . Examine the output of circuit 22. When the output is a "0" signal (when heads 1 and 1 are on tracks 0 to 99), a "1" signal is output from the output terminals _T1 and _T2 , respectively. These "1" signals are supplied to input terminals I ₁ and I ₂ of the voice coil drive circuit 28 via OR gates 29 and 30, respectively, thereby driving the voice coil 7 to the left in FIG. When the voice coil 7 moves to the left, the shutter 12 moves to the left, and the left end 12a of the shutter 12 crosses the photo sensor 13, the output of the waveform shaping circuit 22 changes from "0" to "1". This change is detected by the restore circuit 24. Restore circuit 24
is the first track after detecting the above change (i.e.,
When head 1,1 reaches track 100,
Outputs a “0” signal from output terminal _T1 , and
A "1" signal is output from the output terminal _T2 , and the comparator circuit 27 is further enabled. The “0” signal output from the above output terminal _T1 is the OR gate 2
9 to the input terminal of the voice coil drive circuit 28
I ₁ , which causes the voice coil 7 to stop. The "1" signal output from the output terminal _T3 is supplied to the reset terminal R of the track address register 25 and the preset terminal PS of the track address counter 21, whereby the track address register is reset and the track address counter 21 is reset. The value “100” is preset. When the comparator circuit 27 is enabled, the normal seek operation described above is performed, and the magnetic heads 1, 1 are returned to the track specified by the track address register 25, that is, the 0 track.

On the other hand, if the result of checking the output of the waveform shaping circuit 22 in the restore circuit 24 is a "1" signal (when heads 1 and 1 are on tracks 100 to 199),
The restore circuit 24 outputs a "1" signal and a "0" signal from its output terminals T ₁ and T ₂ , respectively, thereby driving the voice coil 7 to the right in FIG. Then, when the left end 12a of the shutter 12 passes the photo sensor 13 and the output of the waveform shaping circuit 22 changes from "1" to "0", the restore circuit detects this change and returns the first track after detecting the change. When the heads 1 and 1 reach the 99th track, a "0" signal and a "1" signal are output from the output terminals T ₁ and T ₃ , respectively, and the comparison circuit 27 is enabled. When a "0" signal is output from the output terminal _T1 , the voice coil 7 stops, and when a "1" signal is output from the output terminal _T3 , the track address register 25 is reset, and the track address register 25 is reset. counter 21
The number “99” is preset. When the comparison circuit 27 is enabled, a normal seek operation is performed and heads 1, 1 are returned to 0 track.

In this manner, the circuit shown in FIG. 3 detects the reference track position (ie, 99th track or 100th track) based on the output change of the waveform shaping circuit 22, and adjusts the heads 1 and 1 to 0 according to this detection result.
I'm back on track.

Next, a malfunction detection operation to which the malfunction detection method according to the present invention is applied will be explained.

For example, if the current position of the magnetic heads 1 and 1 is on track 50, and the target track is track 150, then the track address counter 21
The address "50" is supplied from the track address register 25 to the malfunction detection circuit 23, and the address "150" is supplied from the track address register 25 to the malfunction detection circuit 23. Then, from these addresses "50" and "150", the malfunction detection circuit 23 detects heads 1 and 1.
It is possible to detect in advance that the output of the waveform shaping circuit 22 changes during the seek operation.
Next, a seek operation is started and completed, but if the output of the waveform shaping circuit 22 does not change during this time, the malfunction detection circuit 23 detects malfunction of the heads 1, 1 and drives the alarm 26. Similarly, if the current position of heads 1, 1 is on the 120th track and the target track is on the 30th track, the malfunction detection circuit 23 detects whether or not the output of the waveform shaping circuit 22 changes. 1 malfunction can be detected. Also, for example, the current position of heads 1 and 1 is on track 20, and the target track is
In the case of 30 tracks, the malfunction detection circuit 23 activates the waveform shaping circuit 2 during the seek operation of heads 1 and 1.
It can be known in advance that the output of No. 2 will not change. If the output of the waveform shaping circuit 22 changes between the start and end of the seek operation, the malfunction detection circuit 23 detects malfunction of the heads 1, 1,
The alarm device 26 is activated.

As described above, in the malfunction detection method according to the present invention, the malfunction detection circuit determines in advance whether or not the output of the waveform shaping circuit 22 changes based on the current address and the target address (that is, if the left end 12a of the shutter 12 is 13) is detected. The waveform shaping circuit 2 uses this pre-detected result and the waveform shaping circuit 2 during the actual seek operation.
A malfunction of the heads 1, 1 is detected based on the output change of the head 2. That is, the method according to the present invention uses the current address, the target address, and the photo sensor 1.
A malfunction of the heads 1, 1 can be detected only by the output of the waveform shaping circuit 22 (the output of the waveform shaping circuit 22). Therefore, it is particularly effective when driving a movable part (carriage 5 in the above-mentioned embodiment) with a pulse motor. However, in this malfunction detection method, if the left end 12a of the shutter 12 does not cross the photo sensor 13 during the seek operation (for example, if the current address is "20",
(If the target address is "30", or if the current address is "150" and the target address is "170", etc.)
Furthermore, even in this malfunction state, if the left end 12a of the shutter 12 does not cross the photo sensor 13 (for example, if it does not move at all), malfunction of the heads 1, 1 cannot be detected. However, if an abnormality occurs during the process of successively performing various seek operations as in the normal use of a magnetic disk device, it will eventually be detected as a malfunction.

Next, the position of the photo sensor 13 will be considered. First, in FIG. 5, the moving range of the heads 1, 1 is L, the position corresponding to the photo sensor 13 is H, the area between this position H and the innermost track is _L1 , and the area between it and the outermost track is L1. Let the area of L ₂ be L 2 . Furthermore, if the ratio of the number of tracks included in area L ₁ to the total number of tracks is P, then area L ₂
The ratio of the number of tracks included in the number of tracks to the total number of tracks is (1-P). However, head 1,1
The probability E that the head 1,1 in the area _L1 moves to the area L2, or _{the head 1,1 in the area L2 crosses} the position H corresponding to the photo sensor 13.
The probability E that 1 moves to area _L1 is E=2P(1-P)...(1). When this equation (1) is expressed in orthogonal coordinates with P on the horizontal axis and E on the vertical axis, as shown in Figure 6, the horizontal axis is crossed at P = 0 and 1, and P = 1/ 2
It becomes a quadratic curve that takes the maximum value at . That is, the probability that heads 1,1 will cross position H (probability that malfunction of heads 1,1 can be detected) is maximum when P=1/2. Therefore, in the method according to the present invention, when the photo sensor 13 is set exactly in the middle of the moving range L, the probability of detecting a malfunction can be maximized.

Incidentally, the probability E' of not being able to detect a malfunction of the heads 1 and 1 can be determined by the following equation: E'=P ² +(1-P) ² =1-2P+2P ² (2).

As can be easily seen from the above description, in this embodiment, the time required to return the heads 1, 1 to their reference positions can be shortened. That is,
When heads 1 and 1 are on tracks 0 to 99, move heads 1 and 1 in the forward direction (toward the middle of disk 2), and move heads 1 and 1 on tracks 100 and 209.
When in the track, the heads 1, 1 only have to be moved in the reverse direction, so the return time is shortened compared to the conventional case where the heads 1, 1 are always moved in the reverse direction (0 track side).

Further, in the above embodiment, the photo sensor 1
3 abnormalities can also be detected.

As explained above, according to the present invention, in the position control system that controls the position of the movable part, the operating position detection part that moves in accordance with the position of the movable part is placed at a predetermined position other than both ends of the operating range. A reference position detector is provided that outputs different signal values depending on when the operating position detector section is on one side and on the other side of the predetermined position, and when a movement command is given to the movable section, , when executing the movement command, it is determined in advance whether or not the operating position detector crosses the reference position detector, and then the movable part is actually moved according to the movement command, and as a result, Since it is determined that a malfunction has occurred in the position control system if a match with the above judgment cannot be obtained, it is possible to easily detect a malfunction of a movable part in the position control system itself, and also to be able to return to the reference position. time can be shortened. Further, the present invention has the advantage that it can be easily implemented by simply shifting the position of the reference position detector in the conventional device.

[Brief explanation of the drawing]

Figures 1A and 1B are diagrams showing a head position control system of a magnetic disk device, which is an embodiment of the present invention;
2A is a waveform diagram showing the output of the waveform conversion circuit 9 shown in FIG. 3, FIG. 2B is a diagram showing the output of the waveform shaping circuit 20 shown in FIG. 3, and FIG. 3 is shown in FIG. A block diagram showing the electric circuit of the head position control system, Fig. 4A shows the photo sensor 13 shown in Figs.
FIG. 4 (b) is a diagram showing the output of the waveform shaping circuit 22 shown in FIG. 3, and FIG. 6 is a graph showing the same probability E. 1... Magnetic head, 5... Carriage, 12...
...Shutter, 13...Photo sensor.

Claims (1)

[Claims] 1. In a position control system that controls the position of a movable part, at a predetermined position other than both ends of the operating range of an operation position detection part that moves in accordance with the position of the movable part,
A reference position detector is provided that outputs different signal values depending on when the operating position detector section is on one side and on the other side of the predetermined position, and when a movement command is given to the movable section, , when executing the movement command, it is determined in advance whether or not the operating position detector crosses the reference position detector, and then the movable part is actually moved according to the movement command, and as a result, A method for detecting a malfunction in a position control system, characterized in that if a coincidence with the determination is not obtained, it is determined that a malfunction has occurred in the position control system.