JPH07192209A - Method and device for reading data - Google Patents

Abstract

PURPOSE: To provide a further higher track density without complicating a servo device by performing read so as to provide a data block read by a read element related to a read channel in all the read channels. CONSTITUTION: The respective read elements 1 are connected to data read channels 2 through amplifiers 1A and data are alternately supplied from the channels 2 to digital channel butters 3. The respective read channels 2 try to read all the blocks. Every time a new block is correctly read to the buffer 3, the buffer 3 reports a main controller 4 that the new block can be used through a signal line 5. Then, the main controller 4 reads the address of the block from the butter 3. When the address corresponds to the next desired block in a data sequence, the controller 4 reads all the blocks from the butter to the main memory 7 through a switch S controlled by the data line 8 and the controller 4.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is a method for reading data contained in a plurality of blocks, the blocks being arranged in a plurality of tracks on a storage medium in a sequence, the storage medium being along a nominal path. Moving, each track has a width substantially perpendicular to the nominal path, and movement of the storage medium causes the tracks to deflect approximately over a displacement area about the nominal path, the displacement area Have respective maximum eccentrics on opposite sides of the nominal path, and each block in the direction of the nominal path requires the time required to read the block for the track to move between the maximum eccentrics. The present invention relates to a data reading method having a length that is shorter than a predetermined time.

The present invention also includes a storage medium moving along a nominal path, the storage medium having a plurality of data blocks arranged in a plurality of tracks in an identifiable sequence, each track being With a track width substantially perpendicular to the nominal path, movement of the storage medium causes the track to deflect approximately over a displacement area about the nominal path, the displacement area including the nominal path. On each opposite side of each of the blocks, each block in the direction of the nominal path requires more time to read the block than a track needs to travel between the maximum eccentrics. A data read-out device for a data storage and correction device having a length such as short.

[0003]

2. Description of the Related Art One means for increasing storage capacity in magnetic data storage devices such as disk drives and tape drives is to increase the number of recording tracks.
During the last 20 years, both disk drives and tape drives have seen a marked increase in the number of recording tracks (often represented by the number of recording tracks per inch of recording media). On the other hand, however, storage capacity has also been increased by increasing linear bit density and / or data coding. Drive developers often prefer increased track density. I mean,
For example, a decrease in S / N ratio is less severe than a decrease in S / N ratio of the same, for example, than when linear bit density is increased.

However, increasing the track density is difficult when the tolerance of the magnetic device and the driving device itself is so large that the read head cannot follow the recording track during the reproducing operation. For example, in a disc drive, the recording track does not depict a precise circle like the disc rotational movement. The disk may oscillate due to defects in the mechanical construction of the spindle motor or spindle mount, for example. Therefore, the recording track does not follow the exact circle, but follows a path that oscillates around the ideal circular track (very often this movement resembles a sine wave).

FIG. 1 shows a part of an ideal track path (shown by a solid line) and an actual path of a recording track during a read operation by a dotted line.

The same problems as track position tolerances occur in tape drives. For example, in the case of longitudinal recording tracks, where each track is recorded in the tape movement direction, the track does not follow exactly the same path during a subsequent read operation as during a write operation. Again, the track draws a curve that changes around the original recording track. One of the reasons for this property
First, the tape position is typically controlled by two or more tape guides, and there is often a small amount of play between the tape and those guides. Additionally, as the tape passes over the head, there is usually no guide and the tape floats slightly above its nominal exact position.

Therefore, the track density cannot be increased beyond a certain limit. This limit is dictated at least in part by the recording medium itself, and in part by mechanical configuration and drive limitations, unless some improvement is introduced to reduce this position change.
Embedded servo devices are commonly used in disk drives. The actual principle is different, but the basic principle is that the disc has pre-recorded information (in combination with a separate disc plate or data area). This servo information is read by the disk drive during disk rotation and is used to correct any small rapid variations from the correct position. Often solenoids
Used to quickly move the head to the correct position.

By this means, even if the read data track from the disk draws a curve that changes around the original recording track, the deviation from the original curve is reduced, so that the read head reproduces the data. can do.

Helical scan tape drives (such as video drives) often use alternative servo schemes. In this way the deviation from the correct position is read by the drive. The drive alternately increases and decreases the tape speed slightly until the read head returns exactly to the track.

Longitudinal recording tape drives, such as the 1/4 "cartridge tape drive (often referred to as the QIC drive), rely solely on the mechanical improvement of the tape cartridge or the drive itself to increase the number of tracks. However, devices have now been introduced that use tapes with unique pre-recorded servo information: the drive has a uniquely configured head with multiple tracks, each channel being defined from the other channels. During a write or read operation, one of the channels follows one set of pre-recorded servo information tracks and the other channel writes or reads data.Used to sense servo track information. The channel uses this information to detect whether the head is accurately following the servo track. Otherwise, information from the servo head is used by the drive to reposition the head in the correct position, including the use of small solenoids or fast stepper motors for this repositioning operation. , Various means are used.

FIG. 2 is a front view of a typical QIC driver head for performing the servo operation as described above. This head has three channels, each with two read elements RA, RB, RC and one write element WA,
It has WB and WC. Here, alternately one of the channels is simultaneously used for tracking the servo track, while the other one or two channels are used for recording or reading. Servo information is usually pre-recorded on the tape during manufacture.

Therefore, as mentioned above, the number of recording tracks can be increased by using a servo device to correct for small variations such as would occur in any device. However, increasing the number of tracks further often complicates the servo system. Also, in many applications, cost and / or
Or electricity is not acceptable.

[0013]

SUMMARY OF THE INVENTION An object of the present invention is to provide a method and a device which can increase the track density without using a servo device. The method and apparatus also enables higher track densities with the servo system without further complicating the servo system.

[0014]

According to the present invention, a plurality of read-out elements are arranged on the storage medium so that each of the read-out elements covers at least a part of the track width of the track to be read out. And each of the read elements has a unique read channel, and the data is read simultaneously using all of the read elements as follows: all of the read channels are associated with the read channel. Read to have a block of data read by a device, monitor the read channel, select the first read block that has been correctly read from the read channel, and the block is the previously selected block. Is the next block following in the sequence and the selected block is related to its position in the sequence. It is solved by storing in the memory location.

The invention will now be described based on its use with a longitudinal recording tape. However, the present invention is also suitable for use with other magnetic recording devices such as spiral scan (video) tape recorders and magnetic disks.

The basic idea of the invention is to position the read heads exactly on the recording tracks by having the read headsets (each read head reads a track) at relatively slightly different positions. Overcoming the problem of not being able to often. The information read from each individual read head is for determining whether the read information is the desired information (ie, the data block having the next address following the most recent read address in succession). To the logical unit. If desired, this information is used in the usual manner by electronic circuits. Inaccurate information is simply ignored.

Before describing the details, it should be mentioned that the invention is based on the premise supported by the fact that the actual track position cannot change so rapidly. There is always a certain tape or disk transit time for moving the track to be read from one eccentric track to another (opposite) eccentric position. Often, including tape or disk drives, the varying frequency depends on the particular recording medium and drive associated with conditions such as reel size and tape speed (or disk size and disk speed). For example, in a 1/4 "cartridge tape drive, experiments have shown that the dynamic change in instantaneous track position can be represented almost exactly by the sum of sinusoids of different amplitude and frequency. And for cartridge formats, the minimum sinusoidal frequency is typically 10-
It is on the order of 50 Hz, and the maximum frequency is about 200-40
It is 0 Hz. However, the main frequency is typically 100
The range is up to 120 Hz.

The frequency and amplitude patterns will vary for other recording media and drives. However, the present invention
Most of the patterns are based on the insight that they are predictable and reproducible.

The invention is also based on the fact that in almost all forms of data recording the data is divided into smaller groups or blocks than before the recording.
Typically, each group can be identified by the type of address or block number. Also, each block of data typically has a unique character used to determine (CRC) whether the data block was read correctly. Additionally, in many applications the data blocks are grouped into frames. This is shown in FIG. 3, which shows the layout of the QIC-2GB tape format.

In the QIC-2GB standard, data is sequentially recorded in blocks along each track (one track at that time). As shown in FIG. 3, in addition to the data bytes, each block has unique information such as a block address byte, a control byte (CRC), and so on.

The block groups are mutually recorded in frames. This is also shown in FIG. Each frame has 16 blocks consisting of 14 data blocks and 2 eigenerror correction blocks. The error correction block can be used by the tape drive to recover one or two of the 14 data blocks of a frame if the drive cannot read it accurately. .

Accordingly, a small group of data blocks is recorded along each track, which data blocks are grouped into frames. Other tape and disk devices can use slightly different layouts. The common point is that group data is divided into blocks in all devices, and each block is uniformly defined by the type of numbering system. Also, because each block has its own control or check byte (CRC), the driver can determine whether the read operation was successful when reading the block.

[0023]

DETAILED DESCRIPTION OF THE INVENTION The present invention provides that the time to read one block is less than the typical time required for the tape to move from one maximum eccentric to the opposite maximum eccentric (shown in FIG. 4). It is based on the recognition that The areas between the lines that are parallel to the nominal path and pass through this maximum eccentric are displacement areas. That is, a predetermined track is found in this area.

In FIG. 4, it is assumed that the track to be read follows a line from A to B. A and B represent the maximum opposing eccentricity from the ideal track line. As described above, the tape requires a predetermined time to move from the displacement position A to the displacement position B. This time is assumed to be longer than the time it takes to read the block as shown in FIG.

The actual time from A to B will vary depending on the recording medium used, speed, drive configuration, etc. However, for a given configuration, the block length should be chosen as follows: That is, the block read time should always be chosen to be shorter than the time required to move from one maximum eccentricity of the track to the opposite maximum eccentricity.

The present invention is based on the above principles and the use of read headsets mounted in close proximity to each other. This is shown in FIG.

The head configuration shown in FIG. 5 is one of the embodiments. This illustrates the basic principle of the invention. The write gap (Wa or Wb depending on the direction of tape movement) is used for writing tracks. A read gap, which typically consists of a single gap, is replaced by two or more gaps located close to each other. In FIG. 5, there are three gaps, one central gap Rc, one upper gap Ro and one lower gap Ru. In conventional recording without a servo device, the device tolerances would be chosen so that the central read channel Rc can always read the recording track. However, if the tracks are extremely narrow and located close to each other, the probability that the central read gap will actually deviate completely or partially from the recording track and hit adjacent tracks is non-negligible. is there.

According to the invention, three heads will cover such a large area and at least one of the heads will always read the data accurately (but not necessarily the same read head). . Sometimes two of the heads can read the data correctly, and sometimes all three read the data correctly (depending on configuration and tolerances). What is important in the present invention is to have a device that constantly monitors the reading of all (3) read gaps and selects the desired data from the data provided by the read element and the read channel.

The actual configuration for the read element can vary. 6B to 6F show various embodiments. Note that the final head configuration for most applications also has a write element. 6B-F show only different variants of the read element used in the construction of the invention, for the sake of simplicity.

The read elements are arranged such that they are as wide as the normal track width, or in the same area as the normal track width, or in areas smaller than the normal track width (measured perpendicular to the nominal path). Can be covered. This depends on the actual configuration of the embodiment, in particular on the distance between one track and an adjacent track.

FIG. 6A shows a recording track and a conventional single read gap A.

FIG. 6B shows two embodiments of the invention using two adjacent read elements. Two elements B1 can be used to cover an area wider than the track width, and two read elements can be used to cover an area narrower than the track width. In two embodiments, two read elements can read the contents of the track. However, since the relative position of the recording track changes in relation to the head gap, only one read gap can accurately read the track content at any given time.

FIG. 6C shows an embodiment using two read elements C spaced apart from each other and partially overlapping. The readout electronics must be configured to handle the difference in read timing between the two elements.

FIG. 6D shows an embodiment using three closely mounted read elements. Two examples are shown. Three elements D1 can be used to cover an area wider than the track width, and three elements D2
Can be used to cover approximately the same area as the track width.

FIG. 6E also shows an embodiment of the invention using three read elements E. However, in this example, the central element is slightly offset with respect to the other two elements, and these elements also partially overlap.

FIG. 6F shows an embodiment of the invention having five read elements F.

The embodiments shown in FIGS. 6B-F are not all read head configurations suitable for use with the present invention. A person skilled in the art can appropriately configure the read element separately. What is important is that at least two components are used and the corresponding electronic circuit is constructed as follows. That is, the read data from all (at least two) read channels is monitored and the correct data is continuously provided to the rest of the driver electronics.

The read configuration shown in FIGS. 6B-F is usually combined with the write section shown in FIG. 5 to form a complete read / write head.

As shown in FIGS. 5 and 6B-F,
The basic ideal of the present invention is to use two common single read heads.
One or more read elements, at least one of which always covers the recording track regardless of track position variations, and at the same time distinguishes the read data from each element and selects the correct data for the read operation. It is to have an electronic circuit in the driving device.

FIG. 7 is a block circuit diagram of the basic system of the present invention. Each read element 1 is connected to a data read channel 2 via an amplifier 1. The data read channel 2 alternately supplies data to the digital channel buffer 3. Each read channel attempts to read all blocks. Whenever a new block is read into the buffer 3 correctly, the buffer 3 sends the signal line 5 to the main controller 4.
To notify that a new block can be used.
Next, the main controller 4 reads the address of the block from the buffer 3. If this address corresponds to the next desired block in the data sequence, the controller 4 transfers all blocks from this buffer to its main memory 7 on the data line 8.
And read out via the switch S controlled by the controller 4. The controller then waits for a signal from the same data buffer 3 or one of the data buffers in sequence for the next block.

The main controller 4 is arranged so that it can continuously monitor each read channel (via the signal line). This allows the controller 4 to determine whether a new block has been read into the channel buffer 3, whether it has been read correctly, and whether this block has the correct address. If the answers to all these questions are “yes”, the main controller 4 can transfer data from the channel data buffer 3 to the main memory 7 via the switch 5.

For the embodiment of the present invention using more than one read element, each additional read element would have the same read channel configuration as shown in FIG.

The read channel shown in FIG. 7 is somewhat simplified. The actual embodiment may vary according to the needs of the device. For most devices, each data buffer 3 should be constructed large enough to store data from two blocks. In this way, the main controller 4 has one block (for example, block N).
Can be transferred to the main memory 7, and at the same time, the contents of the block N + 1 are transferred to the channel buffer 3 in the channel under the reading process. The dual operation of all such channel data buffers is simply achieved by configuring each data channel buffer as two parallel buffers as shown in FIGS. 8 and 9. Block N is section 2 of channel buffer 3
Block N + 1 is loaded into Section 1 at the same time as it is read from. After that, the block N + 1 is read from the section 1 to the main memory 7. On the other hand, the data of the block N + 2 is loaded into the section 2 of the channel buffer 3.

The operation of the apparatus will be described in more detail by the following examples. The device is assumed to read 10 data blocks numbered 100 to 109. It is also assumed that the device uses two read elements. Two
The two read elements are a read element A and a read element B.
Further, the read track position shall be changed such that at some times both channels can read data and at one time only one channel can read. This operation will be described using the table below. Reading is done from the top of the table to the bottom.

Block number Channel A Channel B Output from operation of main controller ─────────────────────────────── ────── None None None None, only standby 100 100 100 Detects whether both channels read the desired block. Transfer the data from channel A to the main buffer. Both channels are then instructed to read the next block.

101 101 Bad block Channel B failed to read a block. However, the data block from channel A can be used. Therefore, this data is transferred from buffer A to the main buffer. Both channels are instructed to read the next block.

102 102 Bad block Same as the case of block 101.

103 103 103 Both channels have the desired block. The main controller transfers the data block from channel A to the main buffer and instructs both channels to read the next block.

104 Bad Block 104 Channel B has the desired block, but Channel A has failed. The main controller transfers the block from channel B to the main buffer and instructs both channels to read the next block.

105 bad block 105 Same as the case of block 104.

106 bad block same as 106 block 105.

107 Bad Block Same as 107 block 106.

108 108 Bad block Same processing as for block 102.

109 109 109 Both channels have the desired block. The main controller transfers the data from the channel A buffer to the main buffer and instructs both channels to read the next block.

The operation is essentially the same even if the reading device uses more than two elements. As with the device with two read elements shown in FIG. 7, each read element of a device with three or more elements is connected to its own independent read channel and block buffer. The master controller continuously monitors the status from all channels and selects and uses each data block from the first channel.

According to the present invention, data can be accurately read from a recording track even when the track position is very different from the nominal position at the time of recording. The details of the head configuration and other parts of the device depend on the specific device requirements. As previously mentioned, this multiple readhead redundancy system can be used with conventional servo track followers.
In this case, the requirements imposed on the servo operation can be reduced.

Although the use of the invention has been described herein as applied to magnetic recording, the invention can also be combined with an optical recording device. In this case the magnetic read-out element is replaced by an optical read-out element, which is shown in FIG.
.. to F. The optical recording track can be followed.

If the read elements are arranged at a predetermined interval (see E in FIG. 6), one read element may read the block earlier than the other read element. In fact, one read element can read block N + 1 at the same time or before another read element can read block N. The main controller 4 is then arranged to store each block in its main memory in its correct position for the sequence of block reads.

The advantageous configurations of the present invention are summarized below.

The read channel has a buffer memory,
Each buffer memory has a first and a second memory section, in succession with the data blocks read by the read element, in relation to a read channel, the first and the second of the buffer memories in the read channel. Inputting to a memory section, reading a data block from one of said memory sections of said buffer memory in a read channel and inputting a data block into the other of said memory sections in said read channel.

The plurality of read elements are arranged along one line and combined to cover an area wider than the track width.

There are two read elements.

There are three read elements.

There are five read elements.

The plurality of read elements are arranged along one line and combined to cover an area narrower than the track width.

In this case, the number of read elements is two or three.

The read element is arranged along two lines extending perpendicularly to the nominal path, at least 1
One read element overlaps one of the lines and at least one read element overlaps the other of the lines.

The plurality of read elements are two, and one read element is arranged on one of the lines.

The plurality of read elements are three, and the two read elements are arranged in one of the lines,
One read element is arranged on the other line.

There is a buffer memory arranged in each of the read elements, each buffer memory has a first and a second memory section, and each channel has a contiguous block of data in said first and second memory. Means for alternating input to the sections and means for reading data blocks from one of the memory sections and writing data blocks to the other of the memory sections.

The plurality of read elements are arranged in a read head, and the read head also has at least one write element.

The storage medium is a magnetic storage medium and the read element is an inductive air gap read element.

[0073]

The present invention provides a method and apparatus that can increase track density without the use of servo devices. The method and apparatus of the present invention also enables higher track densities with servo systems without further complicating the servo system.

[Brief description of drawings]

FIG. 1 is a diagram showing an ideal (nominal) track path by a solid line and an actual recording track path generated by a conventional drive device by a broken line.

FIG. 2 is a schematic diagram showing a conventional read head configuration for a 1/4 "tape cartridge driving device using a servo device.

FIG. 3 is a schematic diagram showing a frame and block format for the QIC-2GB standard.

FIG. 4 is a diagram for explaining the basic principle of the present invention that the time required for the storage medium to move between the maximum eccentric portions is longer than the time required to read one data block.

FIG. 5 is a schematic diagram showing a head configuration of the present invention having both a read element and a write element.

FIG. 6 is a schematic view showing a conventional configuration having a single read element for a recording track in A and a head configuration of the present invention in B to F.

FIG. 7 is a block circuit diagram of an electronic circuit for processing a signal obtained using the read element according to the present invention.

FIG. 8 is a schematic diagram for explaining an operation of a dual section of a data channel buffer.

FIG. 9 is a schematic diagram illustrating an operation of a dual section of a data channel buffer.

[Explanation of symbols]

 A, B maximum eccentricity part Wa, Wb write gap Ro, Rc, Ru read gap

Claims (2)

[Claims] 1. A method of reading data contained in a plurality of blocks, the blocks being arranged in a plurality of tracks on a storage medium in a sequence, the storage medium moving along a nominal path, and each track Has a width substantially perpendicular to the nominal path, and the movement of the storage medium causes the track to deflect approximately over a displacement area about the nominal path, the displacement areas each of the nominal path. Each block has a maximum eccentricity on the opposite side, and each block in the direction of the nominal path is such that the time required to read the block is less than the time required for the track to move between the maximum eccentrics In a data read method having different lengths, a plurality of read elements are tracked to the storage medium, and the tracks to be read by each of the read elements. Arranged to cover at least a portion of the width, each of said read elements having a unique read channel, said data being read simultaneously using all of said read elements as follows: Read such that all of the channels have a data block read by a read element associated with the read channel, monitor said read channel, select a correctly read first-arriving block from said read channel, The block is a next block following the preceding selected block in the sequence, and the selected block is stored in a memory location associated with the position in the sequence. 2. A storage medium moving along a nominal path, the storage medium having a plurality of data blocks arranged in a plurality of tracks in an identifiable sequence, each track being in the nominal path. Having a track width substantially perpendicular thereto, movement of the storage medium causes the tracks to deflect approximately over a displacement region about the nominal path, the displacement regions being respectively opposite of the nominal path. Each block has a maximum eccentricity, and each block in the direction of the nominal path is such that the time required to read the block is less than the time required for the track to move between the maximum eccentricities. A data read-out device for a data storage and correction device having a length, wherein the storage medium covers at least a part of a track width of a track to be read out. A plurality of read elements arranged in parallel, a plurality of read channels each having a unique read element, and all of the read elements having data read by the read elements associated with the read channels. Used to read the block to have a block, means for monitoring the read channel, means for selecting a correctly read first incoming block from the read channel, the block Is a next block following the previously selected block in the sequence, and means for storing the selected block in the memory location associated with that position in the sequence. apparatus.