JPS6238574A - System for detecting head position of disc memory device - Google Patents

Abstract

PURPOSE:To reduce an area in a disc required for reference information by using a ramp signal as a threshold to count the number of repetitive pattern signals crossing the threshold in reference information from each of both areas and detecting the position of a head in accordance with the difference of the counted value between both areas. CONSTITUTION:A reference signal IS corresponds to the state where a head 4 is placed in a normal position PO, and the peak value of the reference signal IS corresponding to the (i-1)th reference information is equal to that corresponding to the i-th reference information. Counting pulses CP whose number is equal to the number of times of crossing of the peak of the reference signal IS to a ramp signal RS are issued from a comparing circuit 40 as the result of comparison between the reference signal IS and the ramp signal RS. When peak values of the reference signal IS from the (i-1)th and i-th reference information are equal to each other, the number of counting pulses CP is 2 for both reference information. When the head position 4 is placed in a position P1, peak values of the reference signal from the (i-1)th and i-th reference information are different from each other, and the number of counting pulses for the (i-1)th reference information is 2 and that for the i-th reference information is 1. These counting pulses CP are counted for reference information adjacent to each other and the difference is obtained.

Description

[Detailed description of the invention] [Technical field to which the invention pertains]

The present invention relates to a method for detecting the current position of a head in a disk storage device, particularly a hard disk storage device, in which the position of a transducer head, such as a magnetic head, for reading and writing information is controlled in a so-called closed loop.

[Prior art and its problems]

As is well known, the aforementioned hard disk or fixed disk device is equipped with a single or multiple magnetic disks as information storage media, and this disk is rotated at a high speed of usually several thousand r, p, s by a so-called spindle motor. are doing. A plurality of concentric tracks are set on the disk for recording information. The head for writing information to or reading information from this track is one for each disk, or more precisely, one for one side of the disk, and is positioned on one of the plurality of tracks, In principle, it is positioned at the center of the trunk. As technology progresses, this disk device is required to be smaller and capable of storing a larger amount of mistresses, and as the diameter of the disk becomes smaller, the recording density is increased. When the recording density is relatively low, open loop control has conventionally been used to control the position of the information read/write head. In this control method, a predetermined number of pulse commands are only issued to the head actuator, such as a pulse motor, so that the head is positioned at a specified track number, but the current position of the head is not disclosed.
Therefore, no further fine-tuning of the head position is performed, but such an oven loop control scheme is
Taking a 5-inch fixed disk as an example, 5007PT (
track/inch). As the recording density becomes higher and exceeds 700↑Pl, it becomes necessary to adopt a closed-loop method for head position control. The basis of this closed-loop control is, of course, accurate detection of the head position, and various methods have been devised for this purpose. In principle, it is accurate to write reference information for position detection in the head, read this reference information by the head itself whose position is to be detected, and learn the head position from the read signal of this information. And it's easy. Conventional inventions made from this point of view include, for example, Japanese Patent Application Laid-Open No. 50-9
It is described in No. 9709 and JP-A-50-99709. In these conventional technologies, the pattern of reference information recorded on the disk and the processing method of the readout signal thereof are devised, but in general, the signal processing circuit for position detection tends to become quite complex, and the reference information is For a pattern to have meaning, writing to the disk must be reasonably accurate (and, including these examples, at the beginning of closed-loop control implementation, it was assumed that the original information on the disk was always located during reading and writing. Since we were based on the concept of control, it was necessary to provide reference information for head position detection in parallel with the original information recording area, so at least one side of the disk was used for recording reference information. However, for disks that rotate at high speed, such parallel control is not necessarily necessary, and it is possible to maintain accurate positioning by controlling the position once per rotation. It has become clear that the information can be maintained sufficiently, and it has become possible to reduce the area of the disc area eaten up by reference information.An example of such a conventional technique will be explained with reference to FIGS. 7 to 10. FIG. A side view of the main parts of a fixed disk device, in which a plurality of disks 1 as magnetic storage media are attached to a spindle shaft 1.
a, and is driven at high speed by a spindle motor lb. As shown in FIG. 8, a circumferential area of a plurality of tracks 2 is set on each surface, and a sector area is provided in which reference information 3 is written by cutting into a part of the circumferential direction of the ring. area has been set. disk l
A transducer head 4 for reading information is provided on each side of the transducer, and these heads 4 are carried by a moving carrier 5 via a head arm.
The carrier 5 moves forward and backward in the direction indicated by arrow P in the figure. The carrier 5 is directed eastward by guides 5a, 5a, and is driven or position controlled by an actuator 6 via a thin metal/Nl lead 5b. The conventional pattern of the reference information 3 is as shown in FIG. 9, for example, and the three tracks 28 to 2G shown in the figure
The reference information for the center track 2b is 3a, 3.
Both reference information 3a and 3b indicated by b are recorded by repeating S and H from left to right in the figure, but when the disc moves in the direction of arrow Q in the figure, the starting point a and the ending point are In the center of the , it moves from N to S simultaneously,
On both sides, recording patterns are created such that the points from S to N are shifted from each other. Furthermore, such a
The pattern "-c" is usually repeated multiple times. The signal waveform when the reference information 3a, 3b is read out by the head 4 is shown in FIG. 10. Considering the case where the position is at Pl which is shifted upward in the diagram by 2 inter-track binches from the normal position PO for the track 2b, the position of the head 4 at this time is the same as the position of the reference information 3a. Pl in Figure 10
As shown in the case of , the readout signal has a peak va due to the reference information 3a and a peak vm corresponding to the center m of the reference information.
appears, and both have different polarities as shown in the figure.
Next, when the head 4 is at a position P2 shifted downward in the drawing from the normal position PO by two inter-track pitches, only the peak b and center beak V- based on the reference information 3b appear in the read signal. From now on, the head 4 will be at the normal position P. , it can be seen that the peak shears a and vb and the central peak V- due to both reference information 3a and 3b all appear. Also, this regular position il! Peak v when at PO
The heights of a and vb are equal to each other, and as can be easily imagined, they are shifted by two pitches at the position P1. The heights of the peaks va and vb that appear independently in P2 are approximately equal to A, while the height of the inversely radial central peak va does not change depending on the position of the head 4. Furthermore, both reference information 3a,
The timings at which the peaks va and vb appear according to 3b are always shifted by a certain amount of time Δt corresponding to the pattern shift described above. From the waveform of the readout signal as described above, the central beat vm
It is possible to distinguish two types of peaks va and vb with a timing deviation Δt from each other and extract them from each other, and also to determine the amount of deviation of the head 4 from the normal position PO from the difference in height of the two types of peak shears a and vb. can be detected. Also,
When the peak Va is higher than the peak Si b, the head 4 is shifted toward the upper Pl, and in the opposite case, the head 4 is shifted toward the lower P2.
It can be seen that it is shifted towards . All of these conventional techniques, including the first two conventional examples, detect the position of the head or the amount of deviation from the normal position by measuring the height of the peak appearing in the readout signal of the reference information Il or the difference therebetween. It is necessary to strictly evaluate the analog quantities represented. In this case, the height of each individual peak usually needs to be stored in analog form using a peak hold circuit before evaluation, and a fairly sophisticated circuit is required to reduce errors at this stage. In this way, the circuits for separating the peaks that appear in the reference information readout signal, holding the peak height values individually, and comparing and evaluating the peak heights become more sophisticated or complicated. Since it is necessary to improve the accuracy of the reference information itself, the means for writing the reference information becomes difficult, and a certain amount of space must be secured for the area in which the reference information is written.

[Object of the Invention] The main object of the present invention is to provide a head position detection method for a disk storage device that can simplify the processing circuit for reference information signals read out through the head and can determine the head position with high accuracy. shall be. A further object of the present invention is to use such a detection method to facilitate the writing of reference information onto a disc, and to reduce the size of the area within the disc required for the reference information.

[Key points of the invention]

According to the present invention, the above object is to write repetitive pattern information in each area as reference information that is shifted with respect to the longitudinal position of the track from the reference information in the adjacent area, and to generation of a time-varying ramp-like signal synchronized with the read-out of the reference information of the reference information, and using the ramp-like signal as a threshold value to cut the threshold value in the reference information from each of both regions; This is achieved by counting the number of pattern signals and detecting the head position from the difference between the counted values for both areas. Each reference information in the above configuration may be a repetition of a very simple N, S or on/off pattern, and in this case, the repetition period can be shortened, so about 100 pieces of information can be stored in a narrow area. It is possible to improve the detection accuracy of the head position by writing reference information having the number of pattern repetitions. In addition, the reference information written in two adjacent slots are offset from each other in the longitudinal direction of the track, and therefore the read signals from both areas come out of the head clearly separated in time, so the mutual separation is extremely high. Easy and accurate. The waveform of the ramp-shaped threshold signal that should be generated in synchronization with the readout of the reference information requires high accuracy, but after comparing this and the reference information using, for example, a comparison circuit, the reference information Since the threshold of the repeating pattern signal in the filter is digitized, the detection result is not affected by the details of the waveform, and its evaluation is simply a matter of counting the number of repeating patterns. Because it is good, there is no error,
Of course, as a means for counting the number of 0 repetition patterns, which greatly simplifies the circuit configuration, it is also possible to provide a counter for each signal read out from both apex regions, but as mentioned above, it is possible to provide a counter for each signal read out from both apex regions, but as mentioned above, if both signals are read out at different timings, Therefore, if one up-down counter is used to amplify one signal and then down-count the other signal, it is possible to use only one counter including the counting difference calculation circuit. Note that two advantageous modes for generating the ramp signal are:
It is possible to generate one triangular waveform signal for the readout signal from each reference information area, or to generate the same sawtooth waveform signal for the signal from each area. Furthermore, the temporal gradient of the rise or fall of these waveforms does not necessarily require strict linearity, and it is sufficient that both peak region signals have the same or symmetrical waveforms. Other more detailed embodiments are described in the next section.

[Embodiments of the invention]

Embodiments of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a diagram showing the basic structure of the system of the present invention, in which the same parts as in FIGS. 7 to 10 regarding the prior art are given the same reference numerals. The upper part of Fig. 1 shows a disk 1 as a storage medium.
is shown in expanded form, and above this t-t+i+
++1st three tracks are shown, reference information 10 is written so as to interrupt a part of them, and from now on head 4
A reference signal Is for detecting the position is read out through the reference signal Is. This reference signal is amplified by the amplifier circuit 20. As in the past, this disk 1 is driven at high speed by a spindle motor 1a such as an electronic motor, and the head 4 is driven or position controlled by an actuator 6. The actuator 6 is, for example, a multi-polar pulse motor equipped with two-phase stator coils 6a and 6b and a rotor 6c as shown in the figure, and the rotor 6C is driven to rotate with an accuracy of 1 degree or less per pulse. In addition, both coils 6
By relatively increasing or decreasing the current in one or both of a and 6b, the rotor 6 can be used as a kind of torque motor.
The angular position of c can be finely adjusted. Driver 7
generates such driving pulses and adjusts the two-phase current. In the following description, it is assumed that the driver 7 moves the head 4 to the i-th track 2 in response to a numerical command indicated by 1 from the right side thereof. A detailed example of the reference information IO is shown in FIG. As shown in the figure, the area of each reference information 10 is radially shifted from the track 2 by an inter-track pinch A, and the i-th reference information 10 is located between the i-th and i+1-th tracks 2. The first reference information 10 is written between the i-1th and i-th tracks 2. The reference information 1° in this example, indicated by the striped pattern in the figure, is
The reference information 10 is alternately shifted in the longitudinal direction of the track with the center m as the border, and adjacent reference information does not overlap.As mentioned above, each reference information 10 is
The disk has an on-off pattern, and even if the pattern between the starting edge l and the ending edge n shown in the figure is repeated several times to ensure detection accuracy, it only occupies about 1 to 2% of the total circumference of the disk. do not have. The left side of FIG. 2 shows positions PO and P1 of head 4 at 311.
゜P2 is shown, and as before the position PO is the normal position for the i-th track 2, and the position! ! P1 and P2 are positions shifted upward and downward in the figure, respectively, from this normal position PO by an amount smaller than the z inter-track pitch. The waveform of the reference signal tS from the head 4 for these positions PO to P2 is as shown in FIG. As can be easily seen, when the adjacent pieces of reference information 10 are shifted from each other, the readout signal of the t-1th reference information is between the starting edge l and the center m in the diagram, and the readout signal of the t-1th reference information is between the center m and the ending edge n. In between, read signals of the i-th reference information appear temporally distinct from each other. Further, when the head 4 is at the normal position PO, the peak values of the reference signal ■S for the i-1th and 1st reference information are equal to each other, but when the head 4 is at the position P1 shifted upward, the peak value is higher for the former. is larger than the latter, and at a position lP2 shifted downward, the relationship is the opposite. Returning to FIG. 1, the ramp signal generating circuit 30 generates a ramp-shaped threshold value for this reference signal rs, and generates a medium-high triangular wave at the center m as shown in the upper part of the frame. or the center m as shown below.
generates a sawtooth wave that repeats on both sides of the Use this kind of ramp signal R3 as a reference signal! A synchronization signal SS for synchronizing with S is given from a synchronization circuit 60 on the left side thereof,
Furthermore, the trigger for generating this synchronization signal SS includes a first
In the illustrated example, a so-called index pulse IP is used. As is well known, the index pulse IP is a synchronizing pulse for reading and writing information to the disk, and is generated once per rotation of the disk by, for example, an angular position pulse generator 1c attached to the spindle motor 1a. The synchronization circuit 60 has, for example, a counter 61 as shown in the figure, and the counter 61 receives an index pulse IP.
and counts clock pulses from clock 62. Therefore, if the synchronization signal SS is taken from the final stage of this counter 61, the index pulse IP
is reset to "0" at the start end l, and "1" at the center m.
It is possible to obtain a signal with a waveform that rises to . Figure 4(a). (b) shows this index pulse IP and the counter 61.
The waveform of the synchronizing signal SS, which is obtained by inverting the output signal of the inverter 35, is shown. However, the synchronization circuit 60
The synchronization with the reference signal Is does not necessarily need to be performed using the index pulse IP, and it is also possible to use a method such as writing information that provides a reference pulse in the reference information. The reference signal Is described above is evaluated by the comparator circuit 40 using the thus generated ramp signal R5 as a threshold value. This pattern is shown in FIG. 4, tel-+81. The reference signal Is shown in FIG. 4fcl corresponds to when the head 4 is at the normal position PO, and therefore
The peak value of the reference signal Is corresponding to the first reference information is the same as the value corresponding to the i-th p-reference information. As a result of comparing this reference signal IS with the ramp signal RS shown in the figure (d+), the comparator circuit 40 outputs the same number of counting pulses CP as the number of times the peak in the reference signal IS crosses the ramp signal 11S. In the case of Fig. 4, the peak values of the reference signals ■S from the f-1st and i-th reference information are the same, so the number of counting pulses CP is the same for both reference information. In the case shown in FIG. 5, the head 4 is at the above-mentioned position P1, and i
-Since the peak values of the first and i-th reference signals are different,
The numbers of counting pulses for the i-1th and i-th reference information are 2 and 1, respectively. The counting circuit 50 shown in FIG. 1 counts such counting pulses CP for adjacent reference information and determines the difference between them. The count value is reset in synchronization with the index pulse IP described above. Therefore, in the JIJI converting circuit 60, the trigger pulse TP is extracted from the stage next to the counting stage from which the synchronizing pulse SS was taken before the counter 61, and the index pulse IP is extracted via the inverter 63 and the one-shot circuit 64. A reset pulse synchronized with is generated and applied to the up/down counter 51. Also, the 1.0 counter 51 is increased. The previous synchronization pulse SS or its complementary pulse signal as shown is given to the down operation designation input υ/D,
The up/down counting operation is switched at a time point corresponding to the center m of the reference information. As a result, the up/down counter synchronizes with the index pulse IP, and therefore starts counting up the count pulse CP from the starting edge l of the reference information, and downcounting the count pulse CP from the center m of the reference information, so that At time n, the difference in the number of count pulses CP corresponding to two adjacent pieces of reference information is held in the counter 51 as a count value. This count value of the difference is taken into the buffer 71 of the control circuit 70 in synchronization with the rise of the next trigger pulse TP. The control circuit 70 is, for example, one microcomputer, and sends a position correction command CS to the head based on the count value read from the counting circuit 50 representing the positional deviation of the head 4 from the normal position PO. 2 to the driver 7. The driver 7 receives this correction command CP.
Based on, for example, the stator coil 6 of the actuator 6
The current distribution to a and 6b is controlled, and the position of the head 4 is adjusted to its normal position PO. If the actuator 6 is a pulse motor as in this example,
The above-mentioned ml adjustment operation is performed by using the actuator 6 as a kind of torque motor. In the above operation, each of the three adjacent pieces of reference information stores a repetitive pattern of about 100 times, so when the position of the head 4 deviates considerably from its normal position, the difference of about several tens of times can be counted. The value is outputted as the amount of positional deviation and given from the counting circuit 50 to the control circuit 70, and even if the positional deviation is small, the count value of several differences is obtained as the amount of positional deviation. When the number is small, correction or fine adjustment of the head position is usually done by 1
・Even if the positional deviation is relatively large, it usually takes 2 times.
It is possible to make slight adjustments to the number of times. Next, a specific circuit example of the main part of the configuration shown in FIG. 1 will be explained with reference to FIG. 6. Amplification circuit 2 for reference signal Is
0 includes a differential input/differential output high-precision operational amplifier 21 and an attached circuit on its output side. Capacitors 22a and 22b receiving the two differential outputs cant off the DC component in the signal and convert it to AC. Only that amount is given to the next stage resistor circuit. The pair of resistors 24a and 24b are load resistors for two differential outputs, and the amplified output is actually taken from the interconnection point of the resistors 24a and 24b, but the differential amplification effect of the operational amplifier 21 is A pair of resistors 23 to maintain symmetry
a, 23b are added. The center of the ramp signal generation circuit 30 is an operational amplifier 31, and the amplifier 31 is equipped with a known feedback capacitor 32 between its input and output to carry out an integrating operation. This is the potential of the connection point V, to which the synchronization signal SS from the inverter 35 described above is applied via the resistor 35a. Therefore, the potential of the connection point ■ from the starting edge l to the center m of the reference information is "
1'', and from the center m to the terminal n, the integral speed, and hence the slope of the ramp signal R3, is 0. Selected by time constant. Connection point V
According to the change in , the integrator circuit integrates the rlJ value in the positive direction at this set integration speed, and then integrates it in the negative direction at the same integration speed. It becomes a triangular wave. Note that the potential of the input on the upper side of the diagram of the operational amplifier 31 on the time constant circuit side is connected to a pair of resistors 36a.
, 36b, and the potential of the lower input thereof is set by the other resistors 38a, 38b and is applied via the resistor 37. Further, the two diodes 31a and 31b and the resistor 31c attached to the operational amplifier 31 are a kind of glue circuit for ensuring the above-mentioned operation of the amplifier. The comparison circuit 40 uses the ramp signal R5 as a variable threshold value and compares the AC component of the amplified reference signal Is with the ramp signal R5, but includes a low-pass filter for removing unnecessary high frequency components that may be included in the reference signal Is. A circuit is provided. This filter circuit has an axle 42 and a capacitor 4 on the ground side.
3 and a capacitor 44 on the B power supply side. In addition, in order to set the potential of the input of the comparison amplifier 41 on the side receiving the reference signal IS,
A pair of resistors 45a. 45b is provided, whereby a set DC potential is applied to the human power via the resistor 46. Comparison amplifier 41 compares the reference signal applied to this input with ramp signal R5 applied to its lower input, and generates the aforementioned count pulse CP. As mentioned above, the ramp signal R5 may be a twice-repeated sawtooth wave, in which case the ramp signal generation circuit 30
is rather simple, and most simply, C is reset or discharged at the time corresponding to the starting edge l and the center m of the reference signal.
It can be configured with a charging/discharging circuit of R. Ramp signal R in this case
Although the linearity of the slope of No. 3 is not necessarily guaranteed, there is no problem in principle as long as the repeated waveforms are the same. In addition, the head position detection method of a disk storage device according to the present invention can be implemented in various ways. For example, it is not necessary to set the area for reference information 10 by intervening in track 2; It can also be provided on the disk surface, and the entire dedicated disk surface can be referenced. 8
10, the present invention can be implemented in applications where fine adjustment of the position of the head 4 is continued in parallel with the reading and writing of information on the track 2. When the control circuit 70 is a microcomputer, the function of the counting circuit 50 can also be shared by the microcomputer. actuator 7,
The synchronization circuit 60 or the control circuit 70 is not limited to the specific configuration and operation mode shown in the embodiments, but may be implemented with appropriate modifications or with some parts omitted within the scope of the present invention. can.

【Effect of the invention】

As explained above, in the method of the present invention, the current position of a position-controllable transducer head for selectively reading and writing information from multiple recording tracks set on a storage medium disk corresponds to each track on the disk. When detecting reference information written in an area by reading it out through the head, each area contains repetitive pattern information that is shifted from the reference information in the adjacent area with respect to the longitudinal position of the track. is written, and when reading reference information from two adjacent areas, a time-varying ramp signal synchronized with the reading is generated, and the reference information from each of the two head areas is generated using the ramp signal as a threshold value. The number of repetitive pattern signals that cross the threshold in the information is counted, and the head position is detected from the difference between the counted values for both head regions.
By appropriately selecting the number of repeating patterns in the reference information, the position of the transducer head can be accurately detected with the required accuracy.The repeated pattern can be very simple, so it is possible to accurately detect the position of the transducer head with the required accuracy. Even if the number of repetitions for position detection is increased considerably, there is no risk of the area for reference information becoming too large, and disk storage using the so-called data surface servo method! It can also be advantageously applied to devices. In implementing the method of the present invention, it is desirable to use a high-precision electronic circuit to generate a ramp signal as a threshold value that changes over time and to amplify the reference information readout signal. After comparing the signal to the digital signal, the signal to be evaluated is digitized.
There is no need to worry about errors occurring, and the evaluation circuit can be made much simpler than before. In many cases, the functions after counting can be performed by existing computer circuits, so in practice there is no need to provide a special evaluation circuit. Since the signals can be separated easily and accurately, there is no need to provide a special signal separation circuit. In this way, the means for reading reference information becomes accurate and simple, and the means for writing reference information becomes correspondingly easy, and repeating patterns in reference information can be simplified to make it easier to use for reference information. By reducing the area on the disk surface, the storage capacity of the disk storage device can be increased.

[Brief explanation of the drawing]

Fig. 1 is a diagram showing the principle configuration of the head position detection method of a disk storage device according to the present invention, Fig. 2 is a diagram showing an example of writing reference information on the disk surface, and Fig. 3 is a diagram showing the layout of reference information according to the head position. FIGS. 4 and 5 are waveform diagrams showing the waveform of the information readout signal. FIGS. 4 and 5 are waveform diagrams for explaining the process of cutting the reference signal at a threshold value by a ramp signal to generate a count pulse. FIG. A circuit diagram showing a specific configuration example of the main circuit for implementing the present invention, FIG. 7 is a side view of the main part of the disk storage device, FIG. 8 is a front view of the main part, and FIG. 9 is a conventional 10 is a waveform diagram showing a waveform of a reference information readout signal according to the head position in the conventional technology. , 4: head,
6: Actuator, 7: Actuator driver, 1
0: Reference information, 20: Reference signal amplifier circuit, 30: Ramp signal generation circuit, 40: Comparison circuit, 50: Counting circuit, 6
0: Synchronization circuit, 70: Control circuit or microcomputer, CP: Count pulse, C8: Hand position correction command signal, ■P: Index pulse, ■s: Reference signal, i4+i+i+1: Track number or reference (
# fi1 number, l: start end of reference information, m: center of reference information, n: end of reference information, R5: ramp signal as L threshold, SS: synchronization (No. 8, TP:) trigger pulse,
It is. Figure 1 Figure 2 Figure 3 Figure 6

Claims (1)

[Claims] 1) Reference information written in an area corresponding to each track on the disk, indicating the current position of a position-controllable transducer head for selectively reading and writing information on multiple recording tracks set on the storage medium disk. In this method, repetitive pattern information is written in each area as reference information, which is shifted with respect to the longitudinal position of the track from the reference information in the adjacent area. When reading the reference information from each region, a time-varying ramp signal synchronized with the reading is generated, and the ramp signal is used as a threshold value to set the threshold value in the reference information from each of the two regions. 1. A head position detection method for a disk storage device, characterized in that the number of repetitive pattern signals that cut the value is counted, and the head position is detected from the difference between the counted values for both areas.