11
Storage devices

---

 

 

11.1 Introduction

Digital technology depends heavily on reliable storage devices. A number of storage technologies co-exist because there is as yet no one ideal solution. Given the use of error correction to allow arbitrary reliability, the main ways of comparing storage devices are cost per bit and access time. These attributes are usually incompatible. The hard disk drive has evolved to provide rapid access whereas the rotary head tape has evolved to provide low-cost bulk storage. Technology will continue to advance, and as a result recording densities of all media will continue to increase, along with improvements in transfer rate. The relative merits of different media will stay the same; disks will always be faster than tape, tape will always be cheaper. Clearly it makes sense to combine these in some way.

In a network-based system it is useful to have tape and disk drives which can transfer faster than real time in order to speed up the process. The density of tape recording has a great bearing on the cost per bit. One limiting factor is the track width. In current DVTR formats, the track width is much greater than theoretically necessary because of the difficulty of mechanically tracking with the heads. This is compounded by the current insistence on editing to picture accuracy on the tape itself. This is a convention inherited from analog VTRs, and it is unnecessary in digital machines. Digital VTRs employ read–modify–write, and this makes the edit precision in the data independent of the block size on tape. Future DVTRs may only be able to edit once per second, by employing edit gaps. This allows the tracks to be much narrower and the recording density can rise. Picture accurate editing requires the block to read intact, edited elsewhere, and written back whole. The obvious way to do this is on a disk. Thus a future for hybrid storage systems is to integrate the DVTR with the disk system and give it a format that it could not use as a standalone unit. This brings the DVTR very close in functionality to the streaming tape recorder which has evolved in the computer industry.

11.2 Disk storage

Disk drives came into being as random-access file-storage devices for digital computers. The explosion in personal computers has fuelled demand for low-cost high-density disk drives and the rapid access offered is increasingly finding applications in digital video. After lengthy development, optical disks are also emerging in digital video applications.

Figure 11.1 shows that, in a disk drive, the data are recorded on a circular track. In hard-disk drives, the disk rotates at several thousand rev/min so that the head-to-disk speed is of the order of 100 miles per hour. At this speed no contact can be tolerated, and the head flies on a boundary layer of air turning with the disk at a height measured in microinches. The longest time it is necessary to wait to access a given data block is a few milliseconds. To increase the storage capacity of the drive without a proportional increase in cost, many concentric tracks are recorded on the disk surface, and the head is mounted on a positioner which can rapidly bring the head to any desired track. Such a machine is termed a moving-head disk drive. An increase in capacity could be obtained by assembling many disks on a common spindle to make a disk pack. The small size of magnetic heads allows the disks to placed close together. If the positioner is designed so that it can remove the heads away from the disk completely, it can be exchanged. The exchangeablepack moving-head disk drive became the standard for mainframe and minicomputers for a long time.

Later came the so-called Winchester technology disks, where the disk and positioner formed a compact sealed unit which allowed increased storage capacity but precluded exchange of the disk pack alone.

Disk drive development has been phenomenally rapid. The first flying head disks were about 3 feet across. Subsequently disk sizes of 14, 8, 5 1⁄4, 3 1⁄2 and 1 7⁄8 inches were developed. Despite the reduction in size, the storage capacity is not compromised because the recording density has increased and continues to increase. In fact there is an advantage in making a drive smaller because the moving parts are then lighter and travel a shorter distance, improving access time.

There are numerous types of optical disk, which have different characteristics. The basic principles of optical disk readout were introduced in section 10.7. Optical disks fall into three broad groups which can usefully be compared:

1 The Compact Disc, its data derivative CD-ROM and the later DVD are examples of a read-only laser disk, which is designed for mass duplication by stamping. They cannot be recorded.

2 Some laser disks can be recorded, but once recorded they cannot be edited or erased because some permanent mechanical or chemical change has been made. These are usually referred to as write-onceread- many (WORM) disks.

3 Erasable optical disks have essentially the same characteristic as magnetic disks, in that new and different recordings can be made in the same track indefinitely, but there is usually a separate erase cycle needed before a new recording can be made since overwrite is not always possible.

Figure 11.2 introduces the essential subsystems of a disk drive which will be discussed here. Magnetic drives and optical drives are similar in that both have a spindle drive mechanism to revolve the disk, and a positioner to give radial access across the disk surface. In the optical drive, the positioner has to carry a collection of lasers, lenses, prisms, gratings and so on, and will be rather larger than a magnetic head. The heavier pickup cannot be accelerated as fast as a magnetic-drive positioner, and access time is slower. A large number of pickups on one positioner makes matters worse. For this reason and because of the larger spacing needed between the disks, multi-platter optical disks are uncommon. Instead ‘juke box’ mechanisms have been developed to allow a large library of optical disks to be mechanically accessed by one or more drives. Access time is sometimes reduced by having more than one positioner per disk; a technique adopted rarely in magnetic drives. A penalty of the very small track pitch possible in laser disks, which gives the enormous storage capacity, is that very accurate track following is needed, and it takes some time to lock onto a track. For this reason tracks on laser disks are usually made as a continuous spiral, rather than the concentric rings of magnetic disks. In this way, a continuous data transfer involves no more than track following once the beginning of the file is located.

Rigid disks are made from aluminium alloy. Magnetic-oxide types use an aluminium oxide substrate, or undercoat, giving a flat surface to which the oxide binder can adhere. Later metallic disks having higher coercivity are electroplated with the magnetic medium. In both cases the surface finish must be extremely good owing to the very small flying height of the head. As the head-to-disk speed and recording density are functions of track radius, the data are confined to the outer areas of the disks to minimize the change in these parameters. As a result, the centre of the pack is often an empty well. In fixed (i.e. non-interchangeable) disks the drive motor is often installed in the centre well.

The information layer of optical disks may be made of a variety of substances, depending on the working principle. This layer is invariably protected beneath a thick transparent layer of glass or polycarbonate.

Exchangeable optical and magnetic disks are usually fitted in protective cartridges. These have various shutters which retract on insertion in the drive to allow access by the drive spindle and heads. Removable packs usually seat on a taper to ensure concentricity and are held to the spindle by a permanent magnet. A lever mechanism may be incorporated into the cartridge to assist their removal.

11.3 Magnetic disks

In all technologies there are specialist terms, and those relating to magnetic disks will be explained here. Figure 11.3 shows a typical multiplatter magnetic disk pack in conceptual form. Given a particular set of coordinates (cylinder, head, sector), known as a disk physical address, one unique data block is defined. A common block capacity is 512 bytes. The subdivision into sectors is sometimes omitted for special applications. A disk drive can be randomly accessed, because any block address can follow any other, but unlike a RAM, at each address a large block of data is stored, rather than a single word.

Magnetic disk drives permanently sacrifice storage density in order to offer rapid access. The use of a flying head with a deliberate air gap between it and the medium is necessary because of the high medium speed, but this causes a severe separation loss which restricts the linear density available. The air gap must be accurately maintained, and consequently the head is of low mass and is mounted flexibly.

The aerohydrodynamic part of the head is known as the slipper; it is designed to provide lift from the boundary layer which changes rapidly with changes in flying height. It is not initially obvious that the difficulty with disk heads is not making them fly, but making them fly close enough to the disk surface. The boundary layer travelling at the disk surface has the same speed as the disk, but as height increases, it slows down due to drag from the surrounding air. As the lift is a function of relative air speed, the closer the slipper comes to the disk, the greater the lift will be. The slipper is therefore mounted at the end of a rigid cantilever sprung towards the medium. The force with which the head is pressed towards the disk by the spring is equal to the lift at the designed flying height. Because of the spring, the head may rise and fall over small warps in the disk. It would be virtually impossible to manufacture disks flat enough to dispense with this feature. As the slipper negotiates a warp it will pitch and roll in addition to rising and falling, but it must be prevented from yawing, as this would cause an azimuth error. Downthrust is applied to the aerodynamic centre by a spherical thrust button, and the required degrees of freedom are supplied by a thin flexible gimbal. The slipper has to bleed away surplus air in order to approach close enough to the disk, and holes or grooves are usually provided for this purpose in the same way that pinch rollers on some tape decks have grooves to prevent tape slip.

In exchangeable-pack drives, there will be a ramp on the side of the cantilever which engages a fixed block when the heads are retracted in order to lift them away from the disk surface.

Figure 11.4 shows how disk heads are made. The magnetic circuit of disk heads was originally assembled from discrete magnetic elements. As the gap and flying height became smaller to increase linear recording density, the slipper was made from ferrite, and became part of the magnetic circuit. This was completed by a small C-shaped ferrite piece which carried the coil. Ferrite heads were restricted in the coercivity of disk they could write without saturating. In thin-film heads, the magnetic circuit and coil are both formed by deposition on a substrate which becomes the rear of the slipper.

In a moving-head device it is not practicable to position separate erase, record and playback heads accurately. Erase is by overwriting, and reading and writing are carried out by the same head. The presence of the air film causes severe separation loss, and peak shift distortion is a major problem. The flying height of the head varies with the radius of the disk track, and it is difficult to provide accurate equalization of the replay channel because of this. The write current is often controlled as a function of track radius so that the changing reluctance of the air gap does not change the resulting record flux. Automatic gain control (AGC) is used on replay to compensate for changes in signal amplitude from the head.

Equalization may be used on recording in the form of precompensation, which moves recorded transitions in such a way as to oppose the effects of peak shift in addition to any replay equalization used.

Early disks used FM coding, which was easy to decode, but had a poor density ratio. The invention of MFM revolutionized hard disks, and further progress led to run-length-limited codes such as 2/3 and 2/7 which had a high density ratio without sacrificing the large jitter window necessary to reject peak shift distortion. Partial response is also suited to disks.

Typical drives have several heads, but with the exception of specialpurpose parallel-transfer machines, only one head will be active at any one time, which means that the read and write circuitry can be shared between the heads. The read channel usually incorporates AGC, which will be overridden by the control logic between data blocks in order to search for address marks, which are short unmodulated areas of track. As a block preamble is entered, the AGC will be enabled to allow a rapid gain adjustment.

11.4 Accessing the blocks

The servo system required to move the heads rapidly between tracks, and yet hold them in place accurately for data transfer, is a fascinating and complex piece of engineering. In exchangeable pack drives, the disk positioner moves on a straight axis which passes through the spindle. Motive power is generally by moving-coil drive, because of the small moving mass which this technique permits.

When a drive is track-following, it is said to be detented, in fine mode or in linear mode depending on the manufacturer. When a drive is seeking from one track to another, it can be described as being in coarse mode or velocity mode. These are the two major operating modes of the servo.

Moving-coil actuators do not naturally detent and require power to stay on-track. The servo system needs positional feedback of some kind. The purpose of the feedback will be one or more of the following:

1To count the number of cylinders crossed during a seek

2To generate a signal proportional to carriage velocity

3To generate a position error proportional to the distance from the centre of the desired track

Magnetic and optical drives obtain these feedback signals in different ways. Many positioners incorporate a tacho which may be a magnetic moving-coil type or its complementary equivalent the moving-magnet type. Both generate a voltage proportional to velocity, and can give no positional information.

A seek is a process where the positioner moves from one cylinder to another. The speed with which a seek can be completed is a major factor in determining the access time of the drive. The main parameter controlling the carriage during a seek is the cylinder difference, which is obtained by subtracting the current cylinder address from the desired cylinder address. The cylinder difference will be a signed binary number representing the number of cylinders to be crossed to reach the target, direction being indicated by the sign. The cylinder difference is loaded into a counter which is decremented each time a cylinder is crossed. The counter drives a DAC which generates an analog voltage proportional to the cylinder difference. As Figure 11.5 shows, this voltage, known as the scheduled velocity, is compared with the output of the carriage-velocity tacho. Any difference between the two results in a velocity error which drives the carriage to cancel the error. As the carriage approaches the target cylinder, the cylinder difference becomes smaller, with the result that the run-in to the target is critically damped to eliminate overshoot.

Figure 11.6(a) shows graphs of scheduled velocity, actual velocity and motor current with respect to cylinder difference during a seek. In the first half of the seek, the actual velocity is less than the scheduled velocity, causing a large velocity error which saturates the amplifier and provides maximum carriage acceleration. In the second half of the graphs, the scheduled velocity is falling below the actual velocity, generating a negative velocity error which drives a reverse current through the motor to slow the carriage down. The scheduled deceleration slope can clearly not be steeper than the saturated acceleration slope. Areas A and B on the graph will be about equal, as the kinetic energy put into the carriage has to be taken out. The current through the motor is continuous, and would result in a heating problem, so to counter this, the DAC is made nonlinear so that above a certain cylinder difference no increase in scheduled velocity will occur. This results in the graph of Figure 11.6(b). The actual velocity graph is called a velocity profile. It consists of three regions: acceleration, where the system is saturated; a constant velocity plateau, where the only power needed is to overcome friction; and the scheduled run-in to the desired cylinder. Dissipation is only significant in the first and last regions.

The track-following accuracy of a drive positioner will be impaired if there is bearing runout, and so the spindle bearings are made to a high degree of precision.

In order to control reading and writing, the drive control circuitry needs to know which cylinder the heads are on, and which sector is currently under the head. Sector information used to be obtained from a sensor which detects holes or slots cut in the hub of the disk. Modern drives will obtain this information from the disk surface as will be seen. The result is that a sector counter in the control logic remains in step with the physical rotation of the disk. The desired sector address is loaded into a register, which is compared with the sector counter. When the two match, the desired sector has been found. This process is referred to as a search, and usually takes place after a seek. Having found the correct physical place on the disk, the next step is to read the header associated with the data block to confirm that the disk address contained there is the same as the desired address.

11.5 Servo-surface disks

One of the major problems to be overcome in the development of highdensity disk drives was that of keeping the heads on-track despite changes of temperature. The very narrow tracks used in digital recording have similar dimensions to the amount a disk will expand as it warms up. The cantilevers and the drive base all expand and contract, conspiring with thermal drift in the cylinder transducer to limit track pitch. The breakthrough in disk density came with the introduction of the servosurface drive. The position error in a servo-surface drive is derived from a head reading the disk itself. This virtually eliminates thermal effects on head positioning and allows great increases in storage density.

In a multiplatter drive, one surface of the pack holds servo information which is read by the servo head. In a ten-platter pack this means that 5 per cent of the medium area is lost, but this is unimportant since the increase in density allowed is enormous. Using one side of a singleplatter cartridge for servo information would be unacceptable as it represents 50 per cent of the medium area, so in this case the servo information can be interleaved with sectors on the data surfaces. This is known as an embedded-servo technique. These two approaches are contrasted in Figure 11.7.

The servo surface is written at the time of disk pack manufacture, and the disk drive can only read it. Writing the servo surface has nothing to do with disk formatting, which affects the data storage areas only. As there are exactly the same number of pulses on every track on the servo surface, it is possible to describe the rotational position of the disk simply by counting them. All that is needed is an unique pattern of missing pulses once per revolution to act as an index point, and the sector transducer can also be eliminated.

The advantage of deriving the sector count from the servo surface is that the number of sectors on the disk can be varied. Any number of sectors can be accommodated by feeding the pulse signal through a programmable divider, so the same disk and drive can be used in numerous different applications.

11.6 Winchester technology

In order to offer extremely high capacity per spindle, which reduces the cost per bit, a disk drive must have very narrow tracks placed close together, and must use very short recorded wavelengths, which implies that the flying height of the heads must be small. The so-called Winchester technology is one approach to high storage density. The technology was developed by IBM, and the name came about because the model number of the development drive was the same as that of the famous rifle.

Reduction in flying height magnifies the problem of providing a contaminant-free environment. A conventional disk is well protected whilst inside the drive, but outside the drive the effects of contamination become intolerable.

In exchangeable-pack drives, there is a real limit to the track pitch that can be achieved because of the difficulty or cost of engineering headalignment mechanisms to make the necessary minute adjustments to give interchange compatibility.

The essence of Winchester technology is that each disk pack has its own set of read/write and servo heads, with an integral positioner. The whole is protected by a dust-free enclosure, and the unit is referred to as a head disk assembly, or HDA.

As the HDA contains its own heads, compatibility problems do not exist, and no head alignment is necessary or provided for. It is thus possible to reduce track pitch considerably compared with exchangeable pack drives. The sealed environment ensures complete cleanliness which permits a reduction in flying height without loss of reliability, and hence leads to an increased linear density. If the rotational speed is maintained, this can also result in an increase in data transfer rate. The HDA is completely sealed, but some have a small filtered port to equalize pressure.

An exchangeable-pack drive must retract the heads to facilitate pack removal. With Winchester technology this is not necessary. An area of the disk surface is reserved as a landing strip for the heads. The disk surface is lubricated, and the heads are designed to withstand landing and takeoff without damage. Winchester heads have very large air-bleed grooves to allow low flying height with a much smaller downthrust from the cantilever, and so they exert less force on the disk surface during contact. When the term ‘parking’ is used in the context of Winchester technology, it refers to the positioning of the heads over the landing area.

Disk rotation must be started and stopped quickly to minimize the length of time the heads slide over the medium. This is conveniently achieved with a servo-controlled brushless motor which has dynamic braking ability. A major advantage of contact start/stop is that more than one head can be used on each surface if retraction is not needed. This leads to two gains: first, the travel of the positioner is reduced in proportion to the number of heads per surface, reducing access time; and, second, more data can be transferred at a given detented carriage position before a seek to the next cylinder becomes necessary. This increases the speed of long transfers. Figure 11.8 illustrates the relationships of the heads in such a system.

Figure 11.9 shows that rotary positioners are feasible in Winchester drives; they cannot be used in exchangeable-pack drives because of interchange problems. There are some advantages to a rotary positioner. It can be placed in the corner of a compact HDA allowing smaller overall size. The manufacturing cost will be less than a linear positioner because fewer bearings and precision bars are needed. Significantly, a rotary positioner can be made faster since its inertia is smaller. With a linear positioner all parts move at the same speed. In a rotary positioner, only the heads move at full speed, as the parts closer to the shaft must move more slowly. The principle of many rotary positioners is exactly that of a moving-coil ammeter, where current is converted directly into torque.

One characteristic of rotary positioners is that there is a component of windage on the heads which tends to pull the positioner in towards the spindle. Windage can be overcome in rotary positioners by feeding the current cylinder address to a ROM which sends a code to a DAC. This produces an offset voltage which is fed to the positioner driver to generate a torque which balances the windage whatever the position of the heads.

When extremely small track spacing is contemplated, it cannot be assumed that all the heads will track the servo head due to temperature gradients. In this case the embedded-servo approach must be used, where each head has its own alignment patterns. The servo surface is often retained in such drives to allow coarse positioning, velocity feedback and index and write-clock generation, in addition to locating the guard bands for landing the heads.

Winchester drives have been made with massive capacity, but the problem of backup is then magnified, and the general trend has been for the physical size of the drive to come down as the storage density increases in order to improve access time and to facilitate the construction of storage arrays (see section 11.9). Very small Winchester disk drives are now available which plug into standard integrated circuit sockets. These are competing with RAM for memory applications where non-volatility is important.

11.7 The disk controller

A disk controller is a unit which is interposed between the drives and the rest of the system. It consists of two main parts; that which issues control signals to and obtains status from the drives, and that which handles the data to be stored and retrieved. Both parts are synchronized by the control sequencer. The essentials of a disk controller are determined by the characteristics of drives and the functions needed, and so they do not vary greatly. It is desirable for economic reasons to use a commercially available disk controller intended for computers. Such controllers are adequate for still store applications, but cannot support the data rate required for real-time moving video unless data reduction is employed. Disk drives are generally built to interface to a standard controller interface, such as the SCSI bus. The disk controller will then be a unit which interfaces the drive bus to the host computer system.

The execution of a function by a disk subsystem requires a complex series of steps, and decisions must be made between the steps to decide what the next will be. There is a parallel with computation, where the function is the equivalent of an instruction, and the sequencer steps needed are the equivalent of the microinstructions needed to execute the instruction. The major failing in this analogy is that the sequence in a disk drive must be accurately synchronized to the rotation of the disk.

Most disk controllers use direct memory access, which means that they have the ability to transfer disk data in and out of the associated memory without the assistance of the processor. In order to cause a file transfer, the disk controller must be told the physical disk address (cylinder, sector, track), the physical memory address where the file begins, the size of the file and the direction of transfer (read or write). The controller will then position the disk heads, address the memory, and transfer the samples. One disk transfer may consist of many contiguous disk blocks, and the controller will automatically increment the disk-address registers as each block is completed. As the disk turns, the sector address increases until the end of the track is reached. The track or head address will then be incremented and the sector address reset so that transfer continues at the beginning of the next track. This process continues until all the heads have been used in turn. In this case both the head address and sector address will be reset, and the cylinder address will be incremented, which causes a seek. A seek which takes place because of a data transfer is called an implied seek, because it is not necessary formally to instruct the system to perform it. As disk drives are block-structured devices, and the error correction is codeword-based, the controller will always complete a block even if the size of the file is less than a whole number of blocks. This is done by packing the last block with zeros.

The status system allows the controller to find out about the operation of the drive, both as a feedback mechanism for the control process and to handle any errors. Upon completion of a function, it is the status system which interrupts the control processor to tell it that another function can be undertaken.

In a system where there are several drives connected to the controller via a common bus, it is possible for non data-transfer functions such as seeks to take place in some drives simultaneously with a data transfer in another.

Before a data transfer can take place, the selected drive must physically access the desired block, and confirm this by reading the block header. Following a seek to the required cylinder, the positioner will confirm that the heads are on-track and settled. The desired head will be selected, and then a search for the correct sector begins. This is done by comparing the desired sector with the current sector register, which is typically incremented by dividing down servo-surface pulses. When the two counts are equal, the head is about to enter the desired block. Figure 11.10 shows the structure of a typical magnetic disk track. In between blocks are placed address marks, which are areas without transitions which the read circuits can detect. Following detection of the address mark, the sequencer is roughly synchronized to begin handling the block. As the block is entered, the data separator locks to the preamble, and in due course the sync pattern will be found. This sets to zero a counter which divides the data-bit rate by eight, allowing the serial recording to be correctly assembled into bytes, and also allowing the sequencer to count the position of the head through the block in order to perform all the necessary steps at the right time.

The first header word is usually the cylinder address, and this is compared with the contents of the desired cylinder register. The second header word will contain the sector and track address of the block, and these will also be compared with the desired addresses. There may also be bad-block flags and/or defect-skipping information. At the end of the header is a CRCC which will be used to ensure that the header was read correctly. Figure 11.11 shows a flowchart of the position verification, after which a data transfer can proceed. The header reading is completely automatic. The only time it is necessary formally to command a header to be read is when checking that a disk has been formatted correctly.

During the read of a data block, the sequencer is employed again. The sync pattern at the beginning of the data is detected as before, following which the actual data arrive. These bits are converted to byte or sample parallel, and sent to the memory by DMA. When the sequencer has counted the last data-byte off the track, the redundancy for the error-correction system will be following.

During a write function, the header-check function will also take place as it is perhaps even more important not to write in the wrong place on a disk. Once the header has been checked and found to be correct, the write process for the associated data block can begin. The preambles, sync pattern, data block, redundancy and postamble have all to be written contiguously. This is taken care of by the sequencer, which is obtaining timing information from the servo surface to lock the block structure to the angular position of the disk. This should be contrasted with the read function, where the timing comes directly from the data.

When video samples are fed into a disk-based system, from a digital interface or from an A/D converter, they will be placed in a buffer memory, from which the disk controller will read them by DMA. The continuous-input sample stream will be split up into disk blocks for disk storage.

The disk transfers must by definition be intermittent, because there are headers between contiguous sectors. Once all the sectors on a particular cylinder have been used, it will be necessary to seek to the next cylinder, which will cause a further interruption to the data transfer. If a bad block is encountered, the sequence will be interrupted until it has passed. The instantaneous data rate of a parallel transfer drive is made higher than the continuous video data rate, so that there is time for the positioner to move whilst the video output is supplied from the FIFO memory. In replay, the drive controller attempts to keep the FIFO as full as possible by issuing a read command as soon as one block space appears in the FIFO. This allows the maximum time for a seek to take place before reading must resume. Figure 11.12 shows the action of the FIFO. Whilst recording, the drive controller attempts to keep the FIFO as empty as possible by issuing write commands as soon as a block of data is present. In this way the amount of time available to seek is maximized in the presence of a continuous video sample input.

11.8 Defect handling

The protection of data recorded on disks differs considerably from the approach used on other media in digital video. This has much to do with the intolerance of data processors to errors when compared with video data. In particular, it is not possible to interpolate to conceal errors in a computer program or a data file.

In the same way that magnetic tape is subject to dropouts, magnetic disks suffer from surface defects whose effect is to corrupt data. The shorter wavelengths employed as disk densities increase are affected more by a given size of defect. Attempting to make a perfect disk is subject to a law of diminishing returns, and eventually a state is reached where it becomes more cost-effective to invest in a defect-handling system.

In the construction of bad-block files, a brand new disk is tested by the operating system. Known patterns are written everywhere on the disk, and these are read back and verified. Following this the system gives the disk a volume name, and creates on it a directory structure which keeps records of the position and size of every file subsequently written. The physical disk address of every block which fails to verify is allocated to a file which has an entry in the disk directory. In this way, when genuine data files come to be written, the bad blocks appear to the system to be in use storing a fictitious file, and no attempt will be made to write there. Some disks have dedicated tracks where defect information can be written during manufacture or by subsequent verification programs, and these permit a speedy construction of the system bad-block file.

11.9 RAID arrays

Whilst the MTBF of a disk drive is very high, it is a simple matter of statistics that when a large number of drives is assembled in a system the time between failures becomes shorter. Disk drives are sealed units and the disks cannot be removed if there is an electronic failure. Even if this were possible the system cannot usually afford down time whilst such a data recovery takes place.

Consequently any system in which the data are valuable must take steps to ensure data integrity. This is commonly done using RAID (redundant array of inexpensive disks) technolgy. Figure 11.13 shows that in a RAID array data blocks are spread across a number of drives.

An-error correcting check symbol (typically Reed–Solomon) is stored on a redundant drive. The error correction is powerful enough to fully correct any error in the block due to a single failed drive. In RAID arrays the drives are designed to be hot-plugged (replaced without removing power) so if a drive fails it is simply physically replaced with a new one. The error-correction system will rewrite the drive with the data which were lost with the failed unit.

When a large number of disk drives are arrayed together, it is necessary and desirable to spread files across all the drives in a RAID array. Whilst this ensures data integrity, it also means that the data transfer rate is multiplied by the number of drives sharing the data. This means that the data transfer rate can be extremely high and new approaches are necessary to move the data in and out of the disk system.

11.10 Disk servers

The disk controller will automatically divide files up into blocks of the appropriate size for recording. If any partial blocks are left over, these will be zero stuffed. Consequently disk stores are not constrained to files of a particular size. Unlike a DVTR which always stores the same amount of data per field, a disk system can store a different amount of data for each field if needs be.

This means that disks are not standards dependent. A disk system can mix 4:4:4, 4:2:2 and 4:2:0 files and it doesn’t care whether the video is interlaced or not or compressed or not. It can mix 525- and 625-line files and it can mix 4:3 and 16:9 aspect ratios. This an advantage in news systems where compression is used. If a given compression scheme is used at the time of recording e.g. DVCPRO, the video can remain in the compressed data domain when it is loaded onto the disk system for editing. This avoids concatenation of codecs which is generally bad news in compressed systems.

One of the happy consequences of the move to disk drives in production is that the actual picture format used need no longer be fixed. With computer graphics and broadcast video visibly merging, interlace may well be doomed. In the near future it will be possible to use noninterlaced HD cameras, and downconvert to a non-interlaced intermediate resolution production format.

As production units such as mixers, character generators, paint systems and DVEs become increasingly software driven, such a format is much easier to adopt than in the days of analog where the functionality was frozen into the circuitry. Following production the intermediate format can be converted to any present or future emission standard.

11.11 Optical disk principles

In order to record MO disks or replay any optical disk, a source of monochromatic light is required. The light source must have low noise otherwise the variations in intensity due to the noise of the source will mask the variations due to reading the disk. The requirement for a lownoise monochromatic light source is economically met using a semiconductor laser.

In the LED, the light produced is incoherent or noisy. In the laser, the ends of the semiconductor are optically flat mirrors, which produce an optically resonant cavity. One photon can bounce to and fro, exciting others in synchronism, to produce coherent light. This is known as Light Amplification by Stimulated Emission of Radiation, mercifully abbreviated to LASER, and can result in a runaway condition, where all available energy is used up in one flash. In injection lasers, an equilibrium is reached between energy input and light output, allowing continuous operation with a clean output. The equilibrium is delicate, and such devices are usually fed from a current source. To avoid runaway when temperature change disturbs the equilibrium, a photosensor is often fed back to the current source. Such lasers have a finite life, and become steadily less efficient. The feedback will maintain output, and it is possible to anticipate the failure of the laser by monitoring the drive voltage needed to give the correct output.

Many rerecordable or eraseable optical disks rely on magneto-optics as considered in section 10.7. The storage medium is magnetic, but the writing mechanism is the heat produced by light from a laser; hence the term ‘thermomagneto-optics’. The advantage of this writing mechanism is that there is no physical contact between the writing head and the medium. The distance can be several millimetres, some of which is taken up with a protective layer to prevent corrosion. Originally, this layer was glass, but engineering plastics have now taken over.

The laser beam will supply a relatively high power for writing, since it is supplying heat energy. For reading, the laser power is reduced, such that it cannot heat the medium past the Curie temperature, and it is left on continuously.

Whatever the type of disk being read, it must be illuminated by the laser beam. Some of the light reflected back from the disk re-enters the aperture of objective lens. The pickup must be capable of separating the reflected light from the incident light. When playing prerecorded disks such as CDs or DVDs, the phase contrast readout process results in a variation of intensity of the light returning to the pickup. When playing MO disks, the intensity does not change, but the magnetic recording on the disk rotates the plane of polarization one way or the other depending on the direction of the vertical magnetization. Figure 11.14(a) shows that a polarizing prism is required to linearly polarize the light from the laser on its way to the disk. Light returning from the disk has had its plane of polarization rotated by approximately ±1°. This is an extremely small rotation. Figure 11.14(b) shows that the returning rotated light can be considered to be composed of two orthogonal components. Rx is the component which is in the same plane as the illumination and is called the ordinary component and Ry is the component due to the Kerr effect rotation and is known as the magnetooptic component. A polarizing beam splitter mounted squarely would reflect the magneto-optic component Ry very well because it is at right angles to the transmission plane of the prism, but the ordinary component would pass straight on in the direction of the laser. By rotating the prism slightly a small amount of the ordinary component is also reflected. Figure 11.14(c) shows that when combined with the magneto-optic component, the angle of rotation has increased. Detecting this rotation requires a further polarizing prism or analyser as shown. The prism is twisted such that the transmission plane is at 45° to the planes of R_x and R_y. Thus with an unmagnetized disk, half of the light is transmitted by the prism and half is reflected. If the magnetic field of the disk turns the plane of polarization towards the transmission plane of the prism, more light is transmitted and less is reflected. Conversely if the plane of polarization is rotated away from the transmission plane, less light is transmitted and more is reflected. If two sensors are used, one for transmitted light and one for reflected light, the difference between the two sensor outputs will be a waveform representing the angle of polarization and thus the recording on the disk. This differential analyser eliminates common-mode noise in the reflected beam.

High-density recording implies short wavelengths. Using a laser focused on the disk from a distance allows short wavelength recordings to be played back without physical contact, whereas conventional magnetic recording requires intimate contact and implies a wear mechanism, the need for periodic cleaning, and susceptibility to contamination.

The information layer is read through the thickness of the disk; this approach causes the readout beam to enter and leave the disk surface through the largest possible area. Despite the minute spot size of about 1 micrometre diameter, light enters and leaves through a 1 mm diameter circle. As a result, surface debris has to be three orders of magnitude larger than the readout spot before the beam is obscured. This approach has the further advantage in MO drives that the magnetic head, on the opposite side to the laser pickup, is then closer to the magnetic layer in the disk.

11.12 Focus and tracking systems

The frequency response of the laser pickup and the amount of crosstalk are both a function of the spot size and care must be taken to keep the beam focused on the information layer. If the spot on the disk becomes too large, it will be unable to discern the smaller features of the track, and can also be affected by the adjacent track. Disk warp and thickness irregularities will cause focal-plane movement beyond the depth of focus of the optical system, and a focus servo system will be needed. The depth of field is related to the numerical aperture, which is defined, and the accuracy of the servo must be sufficient to keep the focal plane within that depth, which is typically ±1μm.

The track pitch of a typical optical disk is of the order of a micrometre, and this is much smaller than the accuracy to which the player chuck or the disk centre hole can be made; on a typical player, runout will swing several tracks past a fixed pickup. The non-contact readout means that there is no inherent mechanical guidance of the pickup and a suitable servo system must be provided.

The focus servo moves a lens along the optical axis in order to keep the spot in focus. Since dynamic focus-changes are largely due to warps, the focus system must have a frequency response in excess of the rotational speed. A moving-coil actuator is often used owing to the small moving mass which this permits. Figure 11.15 shows that a cylindrical magnet assembly almost identical to that of a loudspeaker can be used, coaxial with the light beam. Alternatively a moving magnet design can be used. A rare-earth magnet allows a sufficiently strong magnetic field without excessive weight.

A focus-error system is necessary to drive the lens. There are a number of ways in which this can be derived, the most common of which will be described here.

In Figure 11.16 a cylindrical lens is installed between the beam splitter and the photosensor. The effect of this lens is that the beam has no focal point on the sensor. In one plane, the cylindrical lens appears parallelsided, and has negligible effect on the focal length of the main system, whereas in the other plane, the lens shortens the focal length. The image will be an ellipse whose aspect ratio changes as a function of the state of focus. Between the two foci, the image will be circular. The aspect ratio of the ellipse, and hence the focus error, can be found by dividing the sensor into quadrants. When these are connected as shown, the focus-error signal is generated. The data readout signal is the sum of the quadrant outputs.

Figure 11.17 shows the knife edge-method of determining focus. A split sensor is also required. At (a) the focal point is coincident with the knife edge, so it has little effect on the beam. At (b) the focal point is to the right of the knife edge, and rising rays are interrupted, reducing the output of the upper sensor. At (c) the focal point is to the left of the knife edge, and descending rays are interrupted, reducing the output of the lower sensor. The focus error is derived by comparing the outputs of the two halves of the sensor. A drawback of the knife-edge system is that the lateral position of the knife edge is critical, and adjustment is necessary. To overcome this problem, the knife edge can be replaced by a pair of prisms, as shown in Figure 11.17(d)–(f). Mechanical tolerances then only affect the sensitivity, without causing a focus offset.

The cylindrical lens method is compared with the knife-edge/prism method in Figure 11.18, which shows that the cylindrical lens method has a much smaller capture range. A focus-search mechanism will be required, which moves the focus servo over its entire travel, looking for a zero crossing. At this time the feedback loop will be completed, and the sensor will remain on the linear part of its characteristic. The spiral track of CD and MiniDisc starts at the inside and works outwards. This was deliberately arranged because there is less vertical runout near the hub, and initial focusing will be easier.

In addition to the track runout mentioned above, there are further mechanisms which cause tracking error. A warped disk will not present its surface at 90° to the beam, but will constantly change the angle of incidence during two whole cycles per revolution. Owing to the change of refractive index at the disk surface, the tilt will change the apparent position of the track to the pickup, and Figure 11.19 shows that this makes it appear wavy. Warp also results in coma of the readout spot. The disk format specifies a maximum warp amplitude to keep these effects under control. Finally, vibrations induced in the player from outside, particularly in portable and automotive players, will tend to disturb tracking. A track-following servo is necessary to keep the spot centralized on the track in the presence of these difficulties. There are several ways in which a tracking error can be derived.

In the three-spot method, two additional light beams are focused on the disk track, one offset to each side of the track centre-line. Figure 11.20 shows that, as one side spot moves away from the track into the mirror area, there is less destructive interference and more reflection. This causes the average amplitude of the side spots to change differentially with tracking error. The laser head contains a diffraction grating which produces the side spots, and two extra photosensors onto which the reflections of the side spots will fall. The side spots feed a differential amplifier, which has a low-pass filter to reject the channel-code information and retain the average brightness difference. Some players use a delay line in one of the side-spot signals whose period is equal to the time taken for the disk to travel between the side spots. This helps the differential amplifier to cancel the channel code.

The side spots are generated as follows. When a wavefront reaches an aperture which is small compared to the wavelength, the aperture acts as a point source, and the process of diffraction can be observed as a spherical wavefront leaving the aperture as in Figure 11.21. Where the wavefront passes through a regular structure, known as a diffraction grating, light on the far side will form new wavefronts wherever radiation is in phase, and Figure 11.22 shows that these will be at an angle to the normal depending on the spacing of the structure and the wavelength of the light. A diffraction grating illuminated by white light will produce a dispersed spectrum at each side of the normal. To obtain a fixed angle of diffraction, monochromatic light is necessary.

The alternative approach to tracking-error detection is to analyse the diffraction pattern of the reflected beam. The effect of an off-centre spot is to rotate the radial diffraction pattern about an axis along the track.

Figure 11.23 shows that, if a split sensor is used, one half will see greater modulation than the other when off-track. Such a system may be prone to develop an offset due either to drift or to contamination of the optics, although the capture range is large. A further tracking mechanism is often added to obviate the need for periodic adjustment. Figure 11.24 shows that in this dither-based system, a sinusoidal drive is fed to the tracking servo, causing a radial oscillation of spot position of about ±50 nm. This results in modulation of the envelope of the readout signal, which can be synchronously detected to obtain the sense of the error. The dither can be produced by vibrating a mirror in the light path, which enables a high frequency to be used, or by oscillating the whole pickup at a lower frequency.

In prerecorded disks there is obviously a track to follow, but in recordable disks provision has to be made for track following during the first recording of a blank disk. This is typically done by pressing the tracks in the form of continuous grooves. The grooves may be produced with a lateral wobble so that the wobble frequency can be used to measure the speed of the track during recording.

11.13 Structure of a DVD player

Figure 11.25 shows the block diagram of a typical DVD player, and illustrates the essential components. The most natural division within the block diagram is into the control/servo system and the data path. The control system provides the interface between the user and the servo mechanisms, and performs the logical interlocking required for safety and the correct sequence of operation.

The servo systems include any power-operated loading drawer and chucking mechanism, the spindle-drive servo, and the focus and tracking servos already described.

Power loading is usually implemented on players where the disk is placed in a drawer. Once the drawer has been pulled into the machine, the disk is lowered onto the drive spindle, and clamped at the centre, a process known as chucking. In the simpler top-loading machines, the disk is placed on the spindle by hand, and the clamp is attached to the lid so that it operates as the lid is closed.

The lid or drawer mechanisms have a safety switch which prevents the laser operating if the machine is open. This is to ensure that there can be no conceivable hazard to the user. In actuality there is very little hazard in a DVD pickup. This is because the beam is focused a few millimetres away from the objective lens, and beyond the focal point the beam diverges and the intensity falls rapidly. It is almost impossible to position the eye at the focal point when the pickup is mounted in the player, but it would be foolhardy to attempt to disprove this.

The data path consists of the data separator, the de-interleaving and error-correction process followed by a RAM buffer which supplies the MPEG decoder.

The data separator converts the EFMplus readout waveform into data. Following data separation the error-correction and de-interleave processes take place.

Because of the interleave system, there are two opportunities for correction, first, using the inner code prior to de-interleaving, and second, using the outer code after de-interleaving. In Chapter 9 it was shown that interleaving is designed to spread the effects of burst errors among many different codewords, so that the errors in each are reduced. However, the process can be impaired if a small random error, due perhaps to an imperfection in manufacture, occurs close to a burst error caused by surface contamination. The function of the inner redundancy is to correct single-symbol errors, so that the power of interleaving to handle bursts is undiminished, and to generate error flags for the outer system when a gross error is encountered.

The EFMplus coding is a group code which means that a small defect which changes one channel pattern into another could have corrupted up to eight data bits. In the worst case, if the small defect is on the boundary between two channel patterns, two successive bytes could be corrupted. However, the final odd/even interleave on encoding ensures that the two bytes damaged will be in different inner codewords; thus a random error can never corrupt two bytes in one inner codeword, and random errors are therefore always correctable.

The de-interleave process is achieved by writing sequentially into a memory and reading out using a sequencer. The outer decoder will then correct any burst errors in the data. As MPEG data are very sensitive to error the error-correction performance has to be extremely good.

Following the de-interleave and outer error-correction process an MPEG program stream (see Chapter 9) emerges. Some of the program stream data will be video, some will be audio and this will be routed to the appropriate decoder. It is a fundamental concept of DVD that the bit rate of this program stream is not fixed, but can vary with the difficulty of the program material in order to maintain consistent image quality. The bit rate is changed by changing the speed of the disk. However, there is a complication because the disk uses constant linear velocity rather than constant angular velocity. It is not possible to obtain a particular bit rate with a fixed spindle speed.

The solution is to use a RAM buffer between the transport and the MPEG decoders. The RAM is addressed by counters which are arranged to overflow, giving the memory a ring structure as described in Chapter 1. Writing into the memory is done using clocks derived from the disk whose frequency rises and falls with runout, whereas reading is done by the decoder which, for each picture, will take as much data as are required from the buffer.

The buffer will only function properly if the two addresses are kept apart. This implies that the amount of data read from the disk over the long term must equal the amount of data used by the MPEG decoders. This is done by analysing the address relationship of the buffer. If the disk is turning too fast, the write address will move towards the read address; if the disk is turning too slowly, the write address moves away from the read address. Subtraction of the two addresses produces an error signal which can be fed to the spindle motor.

The speed of the motor is unimportant. The important factor is that the data rate needed by the decoder is correct, and the system will drive the spindle at whatever speed is necessary so that the buffer neither underflows nor overflows.

The MPEG decoder will convert the compressed elementary streams into PCM video and audio and place the pictures and audio blocks into RAM. These will be read out of RAM whenever the time stamps recorded with each picture or audio block match the state of a time stamp counter. If bidirectional coding is used, the RAM readout sequence will convert the recorded picture sequence back to the real-time sequence. The time stamp counter is derived from a crystal oscillator in the player which is divided down to provide the 90 kHz time stamp clock.

As a result the frame rate at which the disk was mastered will be replicated as the pictures are read from RAM. Once a picture buffer is read out, this will trigger the decoder to decode another picture. It will read data from the buffer until this has been completed and thus indirectly influence the disk speed.

Owing to the use of constant linear velocity, the disk speed will be wrong if the pickup is suddenly made to jump to a different radius using manual search controls. This may force the data separator out of lock, or cause a buffer overflow and the decoder may freeze briefly until this has been remedied.

The control system of a CD player is inevitably microprocessor-based, and as such does not differ greatly in hardware terms from any other microprocessor-controlled device. Operator controls will simply interface to processor input ports and the various servo systems will be enabled or overridden by output ports. Software, or more correctly firmware, connects the two. The necessary controls are Play and Eject, with the addition in most players of at least Pause and some buttons which allow rapid skipping through the program material.

Although machines vary in detail, the flowchart of Figure 11.26 shows the logic flow of a simple player, from start being pressed to pictures and sound emerging. At the beginning, the emphasis is on bringing the various servos into operation. Towards the end, the disc subcode is read in order to locate the beginning of the first section of the program material.

When track-following, the tracking-error feedback loop is closed, but for track crossing, in order to locate a piece of music, the loop is opened, and a microprocessor signal forces the laser head to move. The tracking error becomes an approximate sinusoid as tracks are crossed. The cycles of tracking error can be counted as feedback to determine when the correct number of tracks have been crossed. The ‘mirror’ signal obtained when the readout spot is half a track away from target is used to brake pickup motion and re-enable the track-following feedback.

11.14 Digital video tape

Whilst numerous experimental machines were built previously, the first production DVTR, launched in 1987, used the D-1 format which recorded colour difference data according to CCIR-601 on 3 4-inch tape. Whilst it represented a tremendous achievement, the D-1 format was too early to take advantage of high-coercivity tapes and its recording density was quite low, leading to large cassettes and high running costs. The majority of broadcasters then used composite signals, and a component recorder could not easily be used in such an environment. Where component applications existed, the D-1 format could not compete economically with Betacam SP and M-II analog formats. As a result D-1 found application only in high-end post-production suites.

The D-2 format came next, but this was a composite digital format, handling conventional PAL and NTSC signals in digital form, and derived from a format developed by Ampex for a robotic cart. machine. The choice of composite recording was intended to allow broadcasters directly to replace analog recorders with a digital machine. D-2 retained the cassette shell of D-1 but employed higher-coercivity tape and azimuth recording (see Chapter 10) to improve recording density and playing time. Early D-2 machines had no flying erase heads, and difficulties arose with audio edits. D-2 was also hampered by the imminent arrival of the D-3 format.

D-3 was designed by NHK, and put into production by Panasonic. This had twice the recording density of D-2; three times that of D-1. This permitted the use of 1/2-inch tape, making a digital camcorder a possibility. D-3 used the same sampling structure as D-2 for its composite recordings. Coming later, D-3 had learned from earlier formats and had a more powerful error-correction strategy than earlier formats, particularly in audio recording.

By this time the economics of VLSI chips had made made data reduction in VTRs viable, and the first application was the Ampex DCT format which used approximately 2:1 data reduction so that component video could be recorded on an updated version of the 3 4-inch cassettes and transports designed for D-2.

When Sony were developing the Digital Betacam format, compatibility with the existing analog Betacam format was a priority. Digital Betacam uses the same cassette shells as the analog format, and certain models of the digital recorder can play existing analog tapes. Sony also adopted data reduction, but this was in order to allow the construction of a digital component VTR which offered sufficient playing time within the existing cassette dimensions.

The D-5 component format is backward compatible with D-3. The same cassettes are used and D-5 machines can play D-3 tapes. However, in standard definition, compression is not used; the tape speed is doubled in the component format in order to increase the bit rate. With mild compression D-5 recorders can handle high-definition video.

During the development of the DVTR, hard disk storage was developing rapidly and as costs fell, the advantages of disk-based video storage began to erode the DVTR market. In this environment the most successful tape-based solution recently has been the DV format and its production relative DVCPRO. DV has used compression to allow a highly miniaturized mechanism which is ideal for portable use and which outperforms disk-based technology in that application.

In the future recording technology will continue to advance and further formats are inevitable as manufacturers perceive an advantage over their competition. This does not mean that the user need slavishly change to every new format, as the cost of format change is high. Astute users retain their current format for long enough to allow a number of new formats to be introduced. They will then make a quantum leap to a format which is much better than the present one, missing out those between and minimizing the changeover costs.

11.15 The rotary head tape transport

The high bit rate of digital video could be accommodated by a conventional tape deck having many parallel tracks, but each would need its own read/write electronics and the cost would be high. However, the main problem with such an approach is that the data rate is proportional to the tape speed. The provision of stunt modes such as still frame or picture in shuttle are difficult or impossible. The rotary head recorder has the advantage that the spinning heads create a high head-to-tape speed offering a high bit rate recording with a small number of heads and without high tape speed. The head-to-tape speed is dominated by the rotational speed, and the linear tape speed can vary enormously without changing the frequencies produced by the head by very much. Whilst mechanically complex, the rotary head transport has been raised to a high degree of refinement and offers the highest recording density and thus lowest cost per bit of all digital recorders. Figure 11.27 shows that the tape is led around a rotating drum in a helix such that the entrance and exit heights are different. As a result the rotating heads cross the tape at an angle and record a series of slanting tracks. The rotating heads turn at a speed which is locked to the video field rate so that a whole number of tracks results in each input field. Time compression can be used so that the switch from one track to the next falls within a gap between data blocks. Clearly the slant tracks can only be played back properly if linear tape motion is controlled in some way. This is the job of the linear control track which carries a pulse corresponding to every slant track. The control track is played back in order to control the capstan. The breaking up of fields into several tracks is called segmentation and it is used to keep the tracks reasonably short. The segments are invisibly reassembled in memory on replay to restore the original fields.

Figure 11.28 shows the important components of a rotary head helical scan tape transport. There are four servo systems which must correctly interact to obtain all modes of operation: two reel servos, the drum servo and the capstan servo. The capstan and reel servos together move and tension the tape, and the drum servo moves the heads. For variable-speed operation a further servo system will be necessary to deflect the heads.

There are two approaches to capstan drive, those which use a pinch roller and those which do not. In a pinch roller drive, the tape is held against the capstan by pressure from a resilient roller which is normally pulled towards the capstan by a solenoid. The capstan only drives the tape over a narrow speed range, generally the range in which broadcastable pictures are required. Outside this range, the pinch roller retracts, the tape will be driven by reel motors alone, and the reel motors will need to change their operating mode; one becomes a velocity servo whilst the other remains a tension servo.

In a pinch-roller-less transport, the tape is wrapped some way around a relatively large capstan, to give a good area of contact. The tape is always in contact with the capstan, irrespective of operating mode, and so the reel servos never need to change mode. A large capstan has to be used to give sufficient contact area, and to permit high shuttle speed without excessive motor rpm. This means that at play speed it will be turning slowly, and must be accurately controlled and free from cogging. A multipole ironless rotor pancake type brush motor is often used, or a sinusoidal drive brushless motor.

The simplest operating mode to consider is the first recording on a blank tape. In this mode, the capstan will rotate at constant speed, and drive the tape at the linear speed specified for the format. The drum must rotate at a precisely determined speed, so that the correct number of tracks per unit distance will be laid down on the tape. Since in a segmented recording each track will be a constant fraction of a television field, the drum speed must ultimately be determined by the incoming video signal to be recorded. The phase of the drum rotation with respect to input video timing depends upon the time delay necessary to shuffle and interleave the video samples. This time will vary from a minimum of about one segment to more than a field depending on the format.

In order to obtain accurate tracking on replay, a phase comparison will be made between offtape control track pulses and pulses generated by the rotation of the drum. If the phase error between these is used to modify the capstan drive, the error can be eliminated, since the capstan drives the tape which produces the control track segment pulses. Eliminating this timing error results in the rotating heads following the tape tracks properly. Artificially delaying or advancing the reference pulses from the drum will result in a tracking adjustment. Alternatively, the capstan phase can be controlled by analysing tracking signals embedded in the slant tracks. This approach is more accurate and allows a finer track pitch leading to higher recording density. The fixed head is also eliminated.

11.16 Digital video cassettes

The main advantages of a cassette are that the medium is better protected from contamination whilst out of the transport, and that an unskilled operator or a mechanical elevator can load the tape. The digital cassette contains two fully flanged reels side by side. The centre of each hub is fitted with a thrust pad and when the cassette is not in the drive a spring acts on this pad and presses the lower flange of each reel firmly against the body of the cassette to exclude dust. When the cassette is in the machine the relative heights of the reel turntables and the cassette supports are such that the reels seat on the turntables before the cassette comes to rest. This opens a clearance space between the reel flanges and the cassette body by compressing the springs.

The use of a cassette means that it is not as easy to provide a range of sizes as it is with open reels. Simply putting smaller reels in a cassette with the same hub spacing does not produce a significantly smaller cassette. The only solution is to specify different hub spacings for different sizes of cassette. This gives the best volumetric efficiency for storage, but it does mean that the transport must be able to reposition the reel drive motors if it is to play more than one size of cassette.

Cassettes typically have hinged doors to protect the tape when not in a transport, and a reel-locking mechanism to prevent the tape forming a slack loop in storage. There is also typically a write protect tab. Most cassettes have provision for a barcode for use in automated handling systems. Some contain the equivalent of a smart card which carries metadata describing the recording. This can be read quickly without lacing the tape and independently of how far the tape has been wound.

11.17 DVTR block diagram

Figure 11.29(a) shows a representative block diagram of a PCM (i.e. uncompressed) DVTR. Following the convertors will be the distribution of odd and even samples and a shuffle process for concealment purposes. An interleaved product code will be formed prior to the channel coding stage which produces the recorded waveform. On replay the data separator decodes the channel code and the inner and outer codes perform correction as in section 10.24. Following the de-shuffle the data channels are recombined and any necessary concealment will take place. Figure 11.29(b) shows the block diagram of a DVTR using compression. Data from the convertors are rearranged from the normal raster scan to the DCT blocks upon which the compression system works. A common size is eight pixels horizontally by four or eight vertically. The blocks are then shuffled spatially. This has two functions; first, it aids concealment purposes and second, it makes the entropy of the picture more uniform. The shuffled blocks are passed through the compression process. The output of this is distributed and then assembled into product codes and channel coded as for a conventional recorder. On replay data separation and error correction takes place as before, but there is now a matching decoder which outputs DCT blocks. These are then de-shuffled prior to the error-concealment stage. As concealment is more difficult with pixel blocks, data from another field may be employed for concealment as well as data within the field.

The various DVTR formats largely employ the same processing stages, but there are considerable differences in the order in which these are applied. Distribution is shown in Figure 11.30(a). This is a process of sharing the input bit rate over two or more signal paths so that the bit rate recorded in each is reduced. The data are subsequently recombined on playback. Each signal path requires its own tape track and head. The parallel tracks which result form a segment.

Segmentation is shown in Figure 11.30(b). This is the process of sharing the data resulting from one video field over several segments. The replay system must have some means to ensure that associated segments are reassembled into the original field. This is generally a function of the control track.

Figure 11.30(c) shows a product code. Data to be recorded are protected by two error-correcting codeword systems at right angles; the inner code and the outer code (see Chapter 10). When it is working within its capacity the error-correction system returns corrupt data to their original value and its operation is undetectable.

If errors are too great for the correction system, concealment will be employed. Concealment is the estimation of missing data values from surviving data nearby. Nearby means data on vertical, horizontal or time axes as shown in Figure 11.30(d). Concealment relies upon distribution, as all tracks of a segment are unlikely to be simultaneously lost, and upon the shuffle shown in Figure 11.30(e). Shuffling reorders the pixels prior to recording and is reversed on replay. The result is that uncorrectable errors due to dropouts are not concentrated, but are spread out by the deshuffle, making concealment easier. A different approach is required where compression is used because the data recorded are not pixels representing a point, but coefficients representing a DCT blocks and it is these which must be shuffled.

There are two approaches to error correction in segmented recordings. In D-1 and D-2 the approach shown in Figure 11.31(a) is used. Here, following distribution the input field is segmented first, then each segment becomes an independent shuffled product code. This requires less RAM to implement, but it means that from an error correction standpoint each tape track is self-contained and must deal alone with any errors encountered.

Later formats, beginning with D-3, use the approach shown in Figure 11.31(b). Here following distribution the entire field is used to produce one large shuffled product code in each channel. The product code is then segmented for recording on tape. Although more RAM is required to assemble the large product code, the result is that outer codewords on tape spread across several tracks and redundancy in one track can compensate for errors in another. The result is that size of a single burst error which can be fully corrected is increased. As RAM is now cheaper than when the first formats were designed, this approach is becoming more common.

11.18 Picture in shuttle

A rotary head recorder cannot follow the tape tracks properly when the tape is shuttled. Instead the heads cross the tracks at an angle and intermittently pick up short data blocks. Each of these blocks is an inner error-correcting codeword and this can be checked to see if the block was properly recovered. If this is the case, the data can be used to update a frame store which displays the shuttle picture. Clearly the shuttle picture is a mosaic of parts of many fields. In addition to helping the concealment of errors, the shuffle process is beneficial to obtaining picture-in-shuttle. Owing to shuffle, a block recovered from the tape contains data from many places in the picture, and this gives a better result than if many pixels were available from one place in the picture. The twinkling effect seen in shuttle is due to the updating of individual pixels following deshuffle.

When compression is used, the picture is processed in blocks, and these will be visible as mosaicing in the shuttle picture as the frame store is updated by the blocks.

11.19 DV and DVCPRO

This component format uses quarter-inch wide metal evaporated (ME) tape which is only 7 micrometres thick in conjunction with compression to allow realistic playing times in miniaturized equipment. The format has jointly been developed by all the leading VCR manufacturers. Whilst DV was originally intended as a consumer format it was clear that such a format is ideal for professional applications such as news gathering and simple production because of the low cost and small size. This led to the development of the DVCPRO format.

In addition to component video there are also two channels of sixteenbit uniformly quantized digital audio at 32, 44.1 or 48 kHz, with an option of four audio channels using twelve-bit non-uniform quantizing at 32 kHz.

Figure 11.32 shows that two cassette sizes are supported. The standard size cassette offers 41⁄2 hours of recording time and yet is only a little larger than an audio Compact Cassette. The small cassette is even smaller than a DAT cassette yet plays for one hour. Machines designed to play both tape sizes will be equipped with moving-reel motors. Both cassettes are equipped with fixed identification tabs and a moveable write-protect tab. These tabs are sensed by switches in the transport.

DV has adopted many of the features first seen in small formats such as the DAT digital audio recorder and the 8 mm analog video tape format. Of these the most significant is the elimination of the control track permitted by recording tracking signals in the slant tracks themselves. The adoption of metal evaporated tape and embedded tracking allows extremely high recording density. Tracks recorded with slant azimuth are only 10μm wide and the minimum wavelength is only 0.49μm resulting in a superficial density of over 0.4 Megabits per square millimetre.

Segmentation is used in DVC in such a way that as much commonality as possible exists between 50 and 60 Hz versions. The transport runs at 300 tape tracks per second; Figure 11.33 shows that 50 Hz frames contain 12 tracks and 60 Hz frames contain 10 tracks.

The tracking mechanism relies upon buried tones in the slant tracks. From a tracking standpoint there are three types of track shown in Figure 11.34; F₀, F₁ and F₂. F₁ contains a low-frequency pilot and F₂ a highfrequency pilot. F₀ contains no pilot tone, but the recorded data spectrum contains notches at the frequencies of the two tones. Figure 11.34 also shows that every other track will contain F₀ following a four-track sequence.

The embedded tracking tones are recorded throughout the track by inserting a low frequency into the channel-coded data. Every 24 data bits an extra bit is added whose value has no data meaning but whose polarity affects the average voltage of the waveform. By controlling the average voltage with this bit, low frequencies can be introduced into the channel-coded spectrum to act as tracking tones. The tracking tones have sufficiently long wavelength that they are not affected by head azimuth and can be picked up by the ‘wrong’ head. When a head is following an F₀ type track, one edge of the head will detect F¹ and the other edge will detect F₂. If the head is centralized on the track, the amplitudes of the two tones will be identical. Any tracking error will result in the relative amplitudes of the F₁ F₂ tones changing. This can be used to modify the capstan phase in order to correct the tracking error. As azimuth recording is used requiring a minimum of two heads, one head of the pair will always be able to play a type F₀ track.

In simple machines only one set of heads will be fitted and these will record or play as required. In more advanced machines, separate record and replay heads will be fitted. In this case the replay head will read the tracking tones during normal replay, but in editing modes, the record head would read the tracking tones during the pre-roll in order to align itself with the existing track structure.

Figure 11.35 shows the track dimensions. The tracks are approximately 33 mm long and lie at approximately 9° to the tape edge. A transport with a 180° wrap would need a drum of only 21 mm diameter. For camcorder applications with the small cassette this would allow a transport no larger than an audio ‘Walkman’. With the larger cassette it would be advantageous to use time compression to allow a larger drum with partial wrap to be used. This would simplify threading and make room for additional heads in the drum for editing functions.

The audio, video and subcode data are recorded in three separate sectors with edit gaps between so that they can be independently edited in insert mode. In the case where all three data areas are being recorded in insert mode, there must be some mechanism to keep the new tracks synchronous with those which are being overwritten. In a conventional VTR this would be the job of the control track.

In DV there is no control track and the job of tracking during insert is undertaken by part of each slant track. Figure 11.35 shows that the track begins with the insert and track information (ITI) block. During an insert edit the ITI block in each track is always read by the record head. This identifies the position of the track in the segmentation sequence and in the tracking tone sequence and allows the head to identify its physical position both along and across the track prior to an insert edit. The remainder of the track can then be recorded as required.

As there are no linear tracks, the subcode is designed to be read in shuttle for access control purposes. It will contain timecodes and flags.

^{Figure 11.36} shows a block diagram of the DV signal system. The input video is eight-bit component digital according to CCIR-601, but compression of about 5:1 is used. The colour difference signals are subsampled prior to compression. In 60 Hz machines, 4:1:1 sampling isused, allowing a colour difference bandwidth in excess of that possible with NTSC. In 50 Hz machines, 4:2:0 sampling is used. The colour difference sampling rate is still 6.75 MHz, but the two colour difference signals are sent on sequential lines instead of simultaneously. The result is that the vertical colour difference resolution matches the horizontal resolution.

A 16:9 aspect ratio can be supported in standard definition by increasing the horizontal pixel spacing as is done in Digital Betacam. High-definition signals can be supported using a higher compression factor.

As in other DVTRs, the error-correction strategy relies upon a combination of shuffle and product codes. Frames are assembled in RAM, and partitioned into blocks of 8 × 8 pixels. In the luminance channel, four of these blocks cover the same screen area as one block in each colour difference signal as Figure 11.37 shows. The four luminance blocks and the two colour difference blocks are together known as a macroblock. The shuffle is based upon reordering of macroblocks. Following the shuffle compression takes place. The compression system is DCT based and uses techniques described in Chapter 9. Compression acts within frame boundaries so as to permit frame accurate editing.

Intra-frame compression uses 8 × 8 pixel DCT blocks and allows a higher compression factor because advantage can be taken of redundancy between the two fields when there is no motion. If motion is detected, then moving areas of the two fields will be independently coded in 8 × 4 pixel blocks to prevent motion blur. Following the motion compensation the DCT coefficients are weighted, zig-zag scanned and requantized prior to variable-length coding. As in other compressed VTR formats the requantizing is adaptive so that the same amount of data is output irrespective of the input picture content. The entropy block occupies one sync block and contains data compressed from five macroblocks.

The DV product codes are shown in Figure 11.38. The video product block is shown at (a). This block is common to both 525- and 625-line formats. Ten such blocks record one 525-line frame whereas 12 blocks are required for a 625-line frame.

The audio channels are shuffled over a frame period and assembled into the product codes shown in Figure 11.38(b). Video and audio sync blocks are identical except for the ID numbering. The subcode structure is different. Figure 11.38(c) shows the structure of the subcode block. The subcode is not a product block because these can only be used for error correction when the entire block is recovered. The subcode is intended to be read in shuttle where only parts of the track are recovered. Accordingly only inner codes are used and these are much shorter than the video/audio codes, containing only 5 data bytes, known as a pack. The structure of a pack is shown In Figure 11.39. The subcode block in each track can accommodate 12 packs. Packs are repeated throughout the frame so that they have a high probability of recovery in shuttle. The pack header identifies the type of pack, leaving four bytes for pack data, e.g. timecode.

Following the assembly of product codes, the data are then channel coded for recording on tape. A scrambled NRZI channel code is used which is similar to the system used in D-1 except that the tracking tones are also inserted by the modulation process.

In the DVCPRO format the extremely high recording density and long playing time of the consumer DVC was not a requirement. Instead ruggedness and reliable editing were needed. In developing DVCPRO, Panasonic chose to revert to metal particle tape as used in most other DVTRs. This requires wider tracks, and this was achieved by increasing the tape linear speed. The wider tracks also reduce the mechanical precision needed for interchange and editing. However, the DVCPRO transport can still play regular DVC tapes.

The DVCPRO format has proved to be extremely popular and a number of hard disk-based editors are now designed to import native DVCPRO data to cut down on generation loss. With a suitable tape drive this can be done at 4 × normal speed. The SDTI interface (see Chapter 12) can also carry native DVC data. As DVCPRO has developed, later format variants increase the bit rate of the transport so that the compression factor can be reduced or high-definition pictures can be recorded.