JPS5999563A - Magnetic disk device - Google Patents

Abstract

PURPOSE:To convert the speed of a signal so that the signal meets connection standard requirements and output the signal by providing a signal speed converting circuit to the circuit of a signal system among disk control circuits which connect an external computer and disk devices together. CONSTITUTION:Data generated by a CPU 1 is inputted to the controller 6 of the signal converting circuit 5 and stored in the 1st memory 7 and the 2nd memory 8; and speed conversion and modulation by a modulating circuit 9 are performed in order and the resulting signal is applied to a head 10 and recorded on a rotating disk 11. When reproduction is performed, a signal read out by the head 10 is demodulated by a modulating and demodulating circuit 9 and then the signal converting circuit 5 converts the signal speed to send the signal to the CPU 1. When a track on the disk 11 is selected, a carriage 12 is moved, but the carriage is moved to which track is commanded by the CPU 1 and the command is applied to a carriage controller 3 to perform actual seeking operation.

Description

[Detailed description of the invention] [Technical field of invention] The present invention relates to a magnetic disk device.

``1. Parallel External Storage Devices 4. Personal computer converters, which have become old, and now flexible disk devices and fixed disk devices d are often used. As the machine becomes smaller, there is a demand for a smaller storage device d, and at the same time there is a demand for an increase in the storage capacity. To implement the disk group d, which is small and has a large storage capacity, it is sufficient to increase the linear density and track density on the disk surface. Flexible disk devices d and u61)1 disk devices have standardized and unified close-reading regulations regarding signal transfer rules, procedures, seek timing, and the like. The track density is −ρ
If you increase it, the number of tracks will only increase by 7)S, so
(There is no significant deviation from each.On the other hand, if the linearity, that is, the recording density in the circumferential direction, is increased and the rotational speed of the disk is kept as before, the signal transfer speed 1ψ will increase, and If it doesn't fit, in other words, the distortion of the disk is made extremely large, resulting in a device that does not meet the standards.1.
One possible way to bring the No. 3 transfer speed into compliance with the standard is to reduce the number of rotations of the disk. However, if one mistake is made, the following inconvenience will occur.

First of all, for both flexible disks and fixed disks, the number of times and sheets (if it falls, the waiting time and door after the henodo seek will be longer). Specifically, reduce the number of rotations to 1/2,
Most flexible disks have a rotation speed of 35Qrpm, but when using a fixed disk, the rotation speed is 3600rpm (2t800).
When lowered to rpm, the average waiting time is 8.3m5eC, and υS I becomes 6.6m5ec. A longer time before rotation means that it takes more time to read out the signal after seeking without fixing the old one. against the party's request.

[Object of the Invention] The object of the present invention is to increase the gt of the disk without changing the disk rotation problem or increasing the access time (conventional disk model 1). 1. Provide a disk device d that can output signals suitable for Kyoto.

[Summary of the Invention 4] The present invention provides a signal speed change circuit in the circuit of the No. system of the disk control wholesale circuit that connects the external 1iB computer and the disk device d, and changes the speed of one No. 1 to one vote. The output is changed to match the profanity inkstone.

[Effect of the invention] From this jb description, notes 1 per disk device 4
[Embodiments of the Invention] No. 11'4 Obtains a disk device 4θ1 which maintains the same access time or is shorter than the conventional one even when the storage capacity is increased, and which complies with the conventional disk connection standards. The figure shows a block diagram of the disk VAld according to the present invention. 75o external CPU<INK disk control circuit (2), disk machine groove (side) magnetic disk group! (4) are connected. The circuit of the magnetic disk drive and the structure of the shallow groove part 1 are different from the flexible 7 pull disk k tt (!: L-D fixed disk drive 4, including the presence or absence of a carriage servo circuit/) 1
, which are not related to the essence of the present invention, will be omitted. A newly added part according to the present invention is a signal conversion circuit (5
), the data generated from the CPU II) is input to the controller (6) of the word/position conversion circuit (5), stored in the first memory (7) and second memo (January 8), and then speed converted. , modulation circuit (9), the head 110
) The K signal is input 11 times and the signal is recorded on the disk θD which rotates -C. Next, in the case of 14 heads, the head (1
The signal read from 0) is demodulated by the modem circuit (9), and then the signal speed is converted by the signal conversion circuit (5).
0 is sent to the CPU (1) after the signal is executed.
When selecting a track on the disk plate, the command to move the carriage a to the fourth track is generated from CPU (1) and sent to the carriage controller (3). In fact, more than 71 additional operations are performed on the zero-order ((For a more detailed explanation of the operation of signal speed conversion,
I can talk.・Figure 2 shows a conventional disk loading device and a 1d signal lamp with increased linear density. As shown in (a), to write and read signals to the disk, there is a seek period in which the head seeks at the target track, and a period in which signals are written or read on the other side is repeated, and the seek time is currently selected. The distance between the currently selected truck and the next selected truck varies by one city. Also, the inside of the truck is divided into /JX small sectors, but it is not always possible to extract one word from every sector. In the sector, the data viewed by &f with jlM and l□# is recorded by 1j, but linesman 1. ! When t is increased by J, the linear density in Fig. 2(b) is increased by 9, and as shown in the signal -1, data from 741f flights are collected and recorded by 1j. (Reli indicates double the linear density.) It is clear that if the linear density is doubled and read at the same rotation speed, the signal transfer per car will be doubled. . The present invention aims to adapt the increase in transfer speed due to the increase in linear density to the connection standard of 1 degree to 2 degrees using a signal conversion circuit.

FIG. 3 is a diagram for explaining a first embodiment of the magnetic disk 'Ii:tt' according to the present invention, and is explained in terms of signal timing. Also, in this figure, it is assumed that the linear density is twice that of the conventional one.

First, FIG. 3(a) shows input data to the device. After the head of the gL seeks to the target trunk, the CPU manually inputs data to which a track number and sector number are added according to the standard. This data is sequentially read into the memory as shown in ())) and read out from the memory at the timing shown in (c). At the same time as reading from the memory, data is recorded in a certain sector of a fixed track on the disk surface, but since the recording density in the circumferential direction of the device according to the present invention has increased compared to the conventional device, data of 1
Can be recorded on a track. In other words, conventionally, the track number assigned to the data corresponded to the track number on the actual disk surface, but with the device of the present invention, this correspondence cannot be established, so the data has a new track number on the actual disk surface. Three types of codes are added to the data and recorded on the disk surface: a conventional track number attached to a number and a word, and a conventional trunk number and sector number attached to a signal. To explain the situation between 1 and 1 using g314, T and s in ta) are track numbers and sector numbers assigned according to the conventional format.

T; 'r1s1 is a new number assigned within the device, and T1 indicates the actual track number of the disc.

The task of assigning these numbers is performed by the controller (6) in FIG. 41. The read signal from the memory is read out faster than the input data (a) as shown in Figure 3 (C), but this increases the linear density, so the disk plate is rotated at the same rate as before. If (( is the line ′
! It is necessary to position the signal at a higher speed by increasing the B degree θS
This is because there is. While signals are being output from the first memory M1 and recorded on the disk, the second memory M2 continues to read missing data (T181) that is continuously sent, and after reading is completed, it is read and recorded on the disk. . By using two memories, the fJ1 memory and the second memory, it is possible to ignore the period during which signals are read from the memory and recorded on the disk, and it is possible to receive data continuously. If the trunk of the device is full (
Until it becomes ζ! When the machine sells and is sold, it is run, and then the next data is manually recorded.

No. 41J shows the recording pattern on the track on the ki□d disk surface, and it can be seen that 7j was not recorded in the 111-hundredth order after the data was sent. The reason for this is ・'ii As you can see from the timing in Figure 3 (C), there is a difference between the memory read time and the read time, so the signals are sent consecutively to I, αI memory M1, and second memory IJ M2. I can't get it out. (C) When the data is thinly indented into the rotating disk surface at the same time as the selling timing, as shown in PJ4 figure, the data is not sent in the order in which it was sent, but at random. Side 1
If you try to record it in speed reading, for example, T1. T1. Sl
and record the next T;, 'r1. When I try to record s2, M2 does not contain all the data, and it is not the timing to take out one line from memory, so the fl"l-product I want to record passes, and as soon as I try to record it, I start crying. It is necessary to wait for a period of time outside the rotation, and the data to be sent will also have to wait for that amount of time, resulting in a lot of time loss.

No. 51A shows the state of data reading of data i according to the first embodiment of the present invention. First, Figure 5(a)
As shown in Figure 2, the data recorded at intervals is synchronized to create a continuous memo.
l. The second memo IJ M2 is read alternately as shown in (b). It is also possible to share the memory for recording with the first memory M1 and the second memory M2. And...
If the data of each sector recorded in the memory M1 and the second memory M2 is output in accordance with the transfer rate of the conventional disk, it is possible to output the data as continuous data in sector order.

In particular, the magnetic disk drive +6 operates the PLL circuit in synchronization with the recorded signal, and determines the read signal J@lQ+ based on the oscillation clock M of the PLL circuit. Also, a synchronization signal for pulling in the PLL is added to the beginning of each sector. That is, as in the first embodiment of the present invention, signals are transferred to the magnetic disk in units of sectors. If you play, the recorded signal will be continuous, and the knob number will have the same frequency ripple as the synchronization signal added to the beginning of the sector, and the signal will be divided in the middle of the sector and recorded/written in memory and read out. When recording is performed, there is a possibility that the discontinuous sudden frequency of the signal in the divided portions may not match the synchronizing signal added to the beginning of the sector due to rotational jitter or the like. If data is grouped in one order like this, errors are likely to occur in discontinuous portions of signals or portions where frequencies do not match. That is, in the first embodiment of the present invention, one sector is 1r17. There is great significance in doing so.

Furthermore, according to the first practical example of the present invention, although the recording density is 1 or more, the number of Sikhs is decreasing, and 21
When reading data of δ71, it is possible to read data that is more difficult to read than the conventional one.

The ratio of two sectors to one part! The signal speed is determined by a certain amount of memory (in the case of a flexible/pull disk, it is therefore a byte).
+C conversion and conformance to standard connection standards, so the effect is great.

Figure g6 shows a row of a second embodiment of the device according to the invention. This example shows a method of saving a large amount of seek time by reducing the memory capacity by 1 inch according to the present invention (only the data recorded on one track of the road can be skipped). In the case of this example,
It is assumed that the linear density is twice that of the conventional method. First <a>
The input data to d is 1 according to the standard.
Six tracks are sent in sequence. The first memory has (1))
After storing data equivalent to one revolution of the device according to the present invention, as shown in (C), track 1 is stored.
One 1t K of laps is recorded on the track in advance.

During this time, the data of the next track is input to the second memory M2, but since there is a time to move to the next track on the device surface, a seek operation is performed. The average seek time for both flexible and fixed disks is about 1/2 revolution, so on average, as shown in the figure, the seek can be completed before the next write. , the seek time can be drastically reduced. In the case of reading, reverse this and input the reproduced data into the first and second memories alternately as shown in (d), and first read (
The data read into M1 as in e) is read out at the conventional data transfer rate as in (f). The speed of transferring data is 1/2 of the speed of sending out one page of data, so you can use the extra time to do the next transfer,
1+i when seeking]! (Data can be read out continuously without taking any time, and signals can be read out at a speed that ignores the seek time of conventional disk devices.

According to the embodiment of the present invention, the memory capacity is relatively large, but since the /- time can be almost ignored, data output is faster by the /- time than in the past, and data can be taken in as quickly as possible. It has a very large effect on the demand.

[Other implementations of the invention] In 41 implementations of the invention, the memory capacity; k is 1.
Although the data is for one sector, one sector or more may be stored. If the amount of memory is increased and the signal is read while reading the signal from the memory, the seek time can be saved f)3.

In addition, in the first and second embodiments of the present invention, two memories, the first memory and the second memory, are used and the memories are operated alternately, so that when reading from the memory, 1... appears to be 1 Hi and 0. But memory g! 11: For simple IT systems that cannot use a lot of data, even if you use one memory and provide one memory readout IFj'1, the transfer archive +B' which is the gist of the present invention can still be converted. . Nobility will never be lost.

[Brief explanation of drawings]

Figure 1 is a block diagram of the magnetic disk fA iii according to the present invention, Figure 2 is a block diagram of the magnetic disk fA iii according to the present invention, and Figure 2 is a diagram showing the conventional device and the device according to the present invention (Figure 3 showing the recording state of the disk). FIG. 4 is a diagram showing the timing of signals according to the first embodiment of the disk device cliff, and FIG.
FIG. 45 is a diagram showing the timing of signal reproduction according to the embodiment of l and δ1 of the magnetic disk device according to the present invention, and FIG. 61 is a diagram showing the third embodiment of the magnetic disk PLif according to the present invention.
FIG. 3 is a diagram illustrating the timing of signals according to an embodiment of the present invention. 1...CPU, 2...Disk control circuit, etc.
...Disk mechanism section, ±...Magnetic disk device d, 5
...Signal conversion circuit, 6..con)o-ra, 7..JX
1 memory, 8...second memory, 11...disk board. Agent: Patent attorney Kensuke Chika (and 1 other person)

Claims (3)

[Claims] (1) In a magnetic disk device that performs magnetic recording on a disc-shaped magnetic recording medium, the de-recording subcircuit external packaging (
A magnetic disk drive that has a signal conversion circuit for converting the signal transfer rate in the disk control circuit that connects the direct drive. (2) The signal conversion circuit waits for one sector of the signal to be completely recorded on the disk, and one or more memo IJs.
It is composed of a control circuit that controls the transfer rate, and the transfer rate conversion signal is designed to treat at least 17 units as one unit. (!: 'f Claims・'A''t Magnetic disk group described in item 1+t
. (3) The signal conversion circuit converts one of the signals recorded on the disk.
One or more memories with a capacity for tracks and a control circuit that controls them,
2. The magnetic disk device according to claim 1, wherein the magnetic disk device is controlled to perform an open (C seek operation) while reading data stored in the memory.