WO1998022940A1

WO1998022940A1 - Apparatus and method for servo system calibration of a removable diskette medium

Info

Publication number: WO1998022940A1
Application number: PCT/US1997/007388
Authority: WO
Inventors: Lawrence K. C. Ko; Fei Hong Zhu
Original assignee: Swan Instruments
Priority date: 1996-11-19
Filing date: 1997-05-02
Publication date: 1998-05-28
Also published as: AU2823397A; JPH10247365A

Abstract

An ultra-high capacity removable media diskette data storage system includes a floppy-type disk drive (10) having a flexible media disk (40) rotatable at a controlled angular velocity which includes a pre-recorded pattern of a plurality of repeating servo bursts (A, B, C, D), each burst having a predetermined integral fraction radial offset relative to the other bursts. A read/write head (34) is positionable relative to data tracks (42) in response to information derived from the servo bursts (A, B, C, D). The data storage system further includes a plurality of calibration tracks (-1, 0, +1), each track including a first series of head-positioning servo sectors interspersed with a second series of calibration servo sectors. The calibration servo sectors include full amplitude servo bursts (A and B) which are read and processed by an electronic circuit (72) which defines a magnitude and polarity signal representing the electronic offset introduced by the circuitry.

Description

APPARATUS AND METHOD FOR SERVO SYSTEM CALIBRAΗON OF A REMOVABLE DISKETTE MEDIUM

CROSS-REFERENCE TO RELATED APPLICATION

This application is related to co-pending application Serial No. 08/752,478 filed November 19, 1996, entitled APPARATUS AND METHOD FOR SERVO SYSTEM CALIBRATION FOR IMPROVED DISK TRACK FORMATTING AND HEAD

SERVOLNG FOR ULTRA-HIGH CAPACITY REMOVABLE MEDIA DISKETTE, the entire disclosure of which is expressly incorporated herein by reference.

FIELD OF THE INVENTION This invention relates generally to a system and method for head position servoing for ultra-high capacity removable diskette-type disk drives, and relates more particularly to a system and method for servo calibration for use in such disk drives employing servo calibration patterns.

BACKGROUND OF THE INVENTION

Over the past several years, there has been an increase in the number and variety of disk drives which are able to store and retrieve information from various forms of removable media. In particular, the 3.5" diskette cartridge, commonly associated with floppy disk drives, has seen a near 100 fold in storage capacity. These ultra-high capacity diskettes have been introduced in response to the ever increasing demand for greater data storage capacity in removable media, as well as addressing the increase in the size of programs and files which must be ported from computer to computer. Indeed, many commercially available application programs are provided to a consumer on a multiplicity of conventional 3.5" diskettes, with the number of these diskettes often exceeding 10 or 15 for the complete program.

The reason for this becomes evident when it is realized that the formatted data storage capacity of a conventional 3.5" removable diskette is approximately 720 kilobytes to 1.44 megabytes with a recording density of approximately 17,500 bpi (bits per inch). Track densities of conventional flexible magnetic storage disks for conventional floppy disk drives approximately 135 tracks per inch (TPI).

In contrast, ultra-high capacity diskettes have realized formatted data storage capacities in excess of 100 megabytes. These higher capacities are achieved by an at least order of magnitude increase in the disk rotational speed (from about 300 RPM to approximately 3,600 RPM), the use of advanced read/write head design such as thin film or double MIG, and advanced closed loop servoing techniques allowing track densities in excess of 2,500 TPI, including the use of embedded servo sectors which are pre-written at spaced-apart intervals along the data tracks.

Extensive research efforts in the field of ultra-high capacity removable flexible media diskettes have been directed to developing practical techniques for increasing the areal recording density on the media surface and thus increasing the formatted data storage capacity for such media. Improved techniques for increasing areal recording density are an important enabling factor in the development of ultra-high capacity diskettes and drives.

Areal recording density is the product of the track density (i.e. the number of concentric tracks per inch or TPI) on the surface of a disk, and the bit density (i.e., the number of bits per inch or BPI) that can be recorded along a given track. As track density increases, thereby reducing the spacing between tracks, it is necessary to provide more precise radial positioning of the recording or read/write head during read and/or write operations. Also, as bit density increases, more accurate positioning of the recording head over the centerline of the desired track becomes necessary to ensure accurate writing or reading back of the desired information. Thus, some form of head-positioning servo system is required to ensure that the read/write head is maintained over the centerline of a desired data track with the necessary precision.

Various prior art-type rigid disk drives, with non-removable media, have been implemented including various known types of head-positioning servo systems. Pertinent such servos include servo burst fields that are pre-recorded on the surface of a, for example, rigid magnetic disk, during manufacture of the disk drive. Such servo burst fields typically comprise a sequence of magnetic flux reversals whereby write current flowing through the read/write head alternates in polarity thereby sequentially inducing magnetic flux reversals which define an oscillating, substantially constant frequency read back signal referred to herein as a servo signal. The frequency of a servo signal is typically in excess of 1 MHz (megahertz).

In a pertinent prior art-type servo system, often referred to as an "embedded servo", the prerecorded servo burst fields occupy portions (servo sectors) of each recording surface, with the servo sectors being angularly spaced and interspersed among the data sectors of the track. Each servo sector is pre-recorded on the disk's recording surface with each having a discrete angular position such that as the disk's recording surface is rotated beneath a read/write head, servo sectors pass beneath the head in time quantifiable phases. Each servo phase represents the angular position of that servo sector on the recording surface and defines a time period for servo processing circuitry in which servo information is valid. During operation of such a conventional servo system, a read/write head "flies" above the disk media and reads servo burst fields during a sequence of time windows to produce a servo signal. Signal processing circuitry responds to the servo signal to define sequential analog signals that represent the amplitude of the servo signal during successive time windows. Additional circuitry operates on the sequential analog signals to define a position error signal (PES) which defines a magnitude and direction of an error between the actual position of a read/write head and the desired position of the head over the centerline of a data track. The position error signal, in turn, is used to command a head-positioning actuator assembly to move the read/write head such that it is positioned most precisely over the centerline of the desired track.

One common prior approach to servo burst fields is referred to as an "A-B" 2-burst servo pattern. Such an "A-B" pattern may comprise a radially spaced outer "A" servo burst field and a radially spaced inner "B" servo burst field, located in each servo sector along each data track. Each "A" burst field occupies an arcuate space having burst-field centerline that is one half-track distance radially to one side of the centerline of an associated data track, and each "B" burst field occupies an arcuate space having a burst-field centerline that is one halftrack distance radially to the opposite side of the centerline of the data track. When the head flies exactly over the centerline, the sequential reading of the "A" and "B" burst fields should cause the servo system to produce a null position error signal, and would do so under the following idealized circumstances. The "A" burst signal (i.e., the signal produced by the head during reading of the "A" burst field) would have the same demodulated analog signal magnitude as the "B" burst (i.e., the signal produced by the head during reading of the "B" burst field. The ideal circumstances involve the two demodulated signals having the same value, such that the difference between them, as represented by the position error signal, is null.

When the head is not exactly on the track centerline, the sequential reading of the "A" and "B" burst fields, and the demodulation and comparison thereof, should cause the servo system electronics to provide a position error signal with appropriate magnitude and direction to use in driving the head-positioning actuator assembly to move the head to the desired track centerline radial position on the track. For example, if the head is offset towards the "A" side of the track centerline, the analog magnitude value of the "A" burst demodulation signal should be larger than that of the "B" burst demodulation signal. Conversely, if the head is offset towards the "B" side, the analog magnitude value of the "A" burst signal should have a lower amplitude than that of the "B". The farther the head is located away from the track centerline, the higher such difference in value will be. The polarity of the difference indicates the direction that the head is offset relative to the track centerline. By comparing the values of the "A" and "B" demodulated signals, a servo can determine the head's transducer radial position relative to a data track centerline, and thereby determine the correction required of the head-positioning actuator assembly.

While playing a particularly important role in the implementation and improvement of high-capacity rigid disk drives, such prior art servo systems are not easily adaptable for use with flexible disks such as comprise today's ultra-high capacity removable diskettes. The main drawback of prior art servo approaches is that they have been designed to accommodate the eccentricities or wobble of hard disk drive systems, where the mechanical design of the rotating media system is such that the total resulting uncorrected radial position error of the mechanical system (i.e., motor bearing noise, spindle eccentricity, disk misalignment, recurring and non-recurring runout, and the like) are typically on the order of less than Vi of a data track width. Were the mechanical system of any hard disk drive such that a read/write head could be mis-positioned, or be off-track by an entire track width, a re-seek operation would result. Such a mechanical design would be totally unacceptable in the marketplace.

In contrast to the precision required by hard disk drives, in the manufacture and implementation of their rotating disk and associated servo systems, flexible media disks used by floppy disk drives, exhibit additional substantial and inherent mechanical instability and nonuniformity. In particular, along with technical problems of hard disk drives, flexible media disks are susceptible to a significant degree of warping and/or lateral swelling caused by temperature and/or humidity changes in the surrounding ambient atmosphere. And, because flexible media disks are removable, they are exposed not only to the surrounding environment, but also to those mechanical systems whose dimensional tolerances vary from drive to drive. Given the prevalence of floppy disk drives in virtually every computer sold today, as well as the vast array of floppy disk drive manufacturers, a typical removable media diskette cartridge will be subjected to environmental and mechanical stresses that would cause a hard disk drive to cease to function.

An additional drawback of prior art embedded sector servo systems, as exemplified by hard disk drives, is that there is no provision for quickly calibrating a servo system "on-the- fly". While prior art servo systems do, indeed, perform a calibration function, such calibration is performed during the initial drive spin-up, and involves the creation of an offset table which defines an offset value for the position error signal, based on observed repeatable perturbations, for each servo sector on the surface of a disk. Since hard disk drives comprise a multiplicity of servo sectors in each track, and thousands of tracks on each major surface of the disk, such an offset table of correction values would comprise several tens of thousands of values and occupy a significant portion of the memory available to the disk drive control electronics. In addition, such a calibration scheme would necessarily be very computationally intensive, requiring a high-level firmware algorithm and an arithmetic processor operating under firmware program control. As can be seen, a hitherto unsolved need has arisen for a low cost, high performance servo system for an ultra-high capacity removable media diskette cartridge operating in combination with a floppy disk drive which overcomes the limitations and drawbacks of prior art hard disk drive servo systems. Such a servo system should be able to accommodate the larger eccentricities or wobble and instabilities of flexible removable media, and provide an efficient, low cost method of performing "on-the-fly" self calibration.

SUMMARY OF THE INVENTION

In general, the present invention provides a low cost, high performance servo system for an ultra-high capacity removable media diskette cartridge operating in combination with a floppy disk drive which increases the amount of useful data which may be stored thereon and which improves overall servo loop performance without significantly reducing the available user data megabyte storage space. In particular, the present invention provides for an ultra- high capacity removable media diskette cartridge data storage system of the type including a UHD floppy-type disk drive. The disk drive includes a head-positioning actuator assembly having a read/write transducer head for reading/recording data on a flexible media surface of a removable diskette cartridge and is adapted for linear motion radially across the surface of the flexible disk. The disk drive includes a spindle motor assembly for rotating a flexible media disk at a constant, controlled angular velocity. An ultra-high capacity removable media diskette cartridge comprises a flexible media storage disk that is rotatable at a controlled angular velocity by the spindle motor assembly, and defines thereon a plurality of concentric, pre-formatted data tracks each containing a plurality of sectors. Each sector comprises a servo portion at the beginning of each sector followed by a data portion. Each servo portion comprises a pattern of circumferentially, spaced-apart servo bursts with each burst having a pre-determined interval fraction radial offset relative to the others. The read/write transducer head is radially positionable relative to the tracks in response to servo information derived from the spaced-apart servo bursts. A microprocessor is provided for controlling the position of the read/write transducer head during track following operations in response to servo burst information read by the read/write head's transducer from each servo portion. Servo information is provided to the microprocessor by circuit means which receive servo burst information from the read/write head's transducer and which senses the relative amplitudes of the servo bursts in order to allow electronic demodulation circuitry to determine thereby a position error signal. The position error signal is processed by the microprocessor to define head-positioning actuator commands to maintain the actuator and read/write head in substantial alignment with the centerline of a select data track during track following operations of the device. Calibration servo burst patterns are pre-formatted on selected ones of the concentric data tracks and provide calibration information to the circuit means, thus allowing the circuit means to define a magnitude and sign of the offset error introduced by the circuit means to the position error signal.

In one aspect of the invention, each servo portion includes four servo bursts, all written at a width of one full track pitch, in quadrature relationship and arranged such that a first burst in each servo portion is offset radially from a track centerline by one half of a track pitch or width increment, a second burst offset radially from the track centerline in the opposite direction by one half of a track pitch increment, and a third burst which spans the track centerline. A read back null portion alternates between the third burst position and the fourth burst position on respective odd and even numbered tracks. The qualification servo burst pattern comprises four servo bursts in which the first and second bursts span each respective track centerline and are written at a width of one full track pitch. A third servo burst spans the centerline of each even numbered calibration track and is also written at a width of a full track pitch, and a read back null is provided in alternating third and fourth burst locations for the tracks comprising the calibration means.

In an additional aspect of the invention, the electronic circuitry comprises a preamplifier section comprising analog demodulation circuitry, which evaluates the relative magnitudes of the first and second servo burst signals and develops a position error signal representative of the difference in magnitude between the first and second signals. The sign of the position error signal indicates the direction of the radial offset of the read/write transducer head off the track centerline.

The preamplifier and demodulator circuitry, in combination with the first and second full track width burst patterns of the calibration sectors define means for determining a magnitude and sign of an offset error signal. The offset error signal is representative of an offset error introduced into the position error signal by the analog components comprising the demodulator and preamplifier circuitry.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features, aspects and advantages of the present invention will be more fully understood when considered with regard to the following detailed description, appended claims and accompanying drawings wherein:

FIG. 1 is a semi-schematic, exploded perspective view of an ultra-high capacity disk drive detailing a head-positioning actuator assembly including a voice-coil motor and head- positioning carriage assembly and a diskette positioning carriage assembly useful in practice of principles of the invention;

FIG. 2 is a semi-schematic, diagrammatic view of a removable media disk drive subsystem, including a head and disk in accordance with practice of principles of the invention;

FIG. 3 is a diagrammatic, signal pattern-level representation of the layout of a servo data sector in accordance with practice of principles of the invention;

FIG. 4 is a diagrammatic, block-level representation of five contiguous concentric data tracks showing the arrangement of servo burst patterns in accordance with practice of principles of the invention;

FIG. 5 is a diagrammatic representation of the read-back amplitude of odd and even track servo bursts in accordance with the four burst servo pattern of FIG. 4;

FIG. 6 is a diagrammatic, block-level representation of servo circuitry in accordance with the invention, including a preamplifier and demodulator section depicting the electrical input model for the four burst servo pattern of FIG. 4;

FIG. 7 is a diagrammatic, block-level representation of the layout of calibration tracks in accordance with practice of principles of the invention; and

FIG. 8 is a diagrammatic flow chart depicting the calibration sequence in accordance with practice of principles of the invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT(S)

Turning now to FIG. 1, a removable media disk drive 10 suitable for operation in combination with a host personal computer for data storage and retrieval purposes includes a base 12 formed of cast aluminum alloy through which a spindle hub/motor assembly 14 extends upwardly and which defines a central hub configured to receive and rotatably transport a flexible media data storage disk including an ultra-high capacity diskette cartridge. A cover assembly (not shown) fits tightly over the disk drive base 12 to form an interior space into which diskette cartridge may be inserted. The base and cover are sized and configured to locate and secure all of the movable components of the drive 10 and includes side walls which define the form factor of a standard three and one half inch floppy diskette drive. A cover comprising a bezel 16 which includes an access door 18 is provided at one end of the drive 10 and completes the enclosure of the drive interior space. The access door 18 is pivotally mounted to cover an access slot, formed in the bezel, which has horizontal and vertical dimensions suitable for receiving therethrough a diskette cartridge, for example, a three and one half inch removable diskette. An electronically commutated three-phase direct drive brushless DC spindle motor 15 is mounted to the base 12 and includes a rotating spindle or hub 14 onto which the mating hub of a flexible media disk is mounted. A flexible media disk is rotated counter clockwise at a controlled angular velocity of about 3,600 RPM by the spindle motor 15.

A linear voice coil motor-type actuator assembly 20, including an actuator carriage assembly 21 , is mounted in the interior closure of the drive 10 and includes a pair of torroidal, generally rectangular molded coils 22A and B, each of which are linearly displaced between a pair of generally flat magnetically polar aligned permanent magnets 24A and B (the lower magnets of each pair are partially obscured in FIG. 1). The magnets 24A and B are positioned such that they lie above and below each of the corresponding coils 22 A and B. Ferromagnetic flux return plates 25 A and B are mounted so as to pass through the central opening of the coils such that the magnetic flux developed by the magnets is essentially directed to the region above and below the ferromagnetic flux return plates 25 A and B in which the coil travels. The winding coils 22A and B are freely movable between the permanent magnets 24 A and B within a limited range of in-line displacement or stroke of the actuator's carriage assembly 21, in this instance, the stroke is approximately 1-1/8 inches.

The magnets 24A and B are premagnetized in a manner such that current passing through the winding coils 22A and B in one direction causes the actuator assembly 20 to move linearly toward the spinner motor hub 14, while current passing through the winding coils in the other direction causes the actuator assembly 20 to move linearly in the other direction, away from the hub 14. The actuator assembly's range of linear motion is limited, at both ends, by crash stops (not shown). Indeed, the winding coils 22A and B would come into undesirable physical contact with the folded-over front and rear ends of the permanent magnets if the actuator assembly's range of linear motion were not so limited.

In addition, upper recording head loading means are provided to rotatably or pivotally move the upper head's load beam 30 away from the lower head support frame 36 which, in turn, separates the upper and lower read/write heads 32 and 34 a sufficient distance to allow a removable diskette cartridge to be inserted into the drive 10, with the recording disk positioned between the upper and lower read/write heads (32 and 34 respectively). Such means suitably comprises a disk cartridge carriage assembly 60 which includes a support frame 64 movably mounted on the base 12 of the drive, in combination with a support housing 66 mounted for vertical displacement at the urging of support frame 64. A pair of lifting fingers 86 extend rearwardly from the top surface of the support housing 66 and are separated by a slot-like opening having width sufficient for unobstructed motion of the head- positioning actuator's carriage assembly therebetween. The slot between the fingers 86 extends into the material of the body portion of the support housing 66 a distance sufficient for the terminal edge of the slot to be positioned substantially over the drive hub 14 of the disk drive spinner motor 15. Accordingly, the slot allows the head-positioning actuator's carriage assembly to move linearly along the entire radial length of the recording band(s) of a flexible disk contained within a diskette cartridge. Once the support housing 66 is mounted on the support frame 64, the lifting fingers 86 engage the underside of a lateral lifting bar 88 coupled to the upper load beam 30 on the voice coil actuator's carriage assembly 21. As the support housing 66 is caused to move vertically, the lifting fingers 86 engage the lateral lifting bar 88 and forces the upper load beam 30 away from the lower support frame 36, thus separating the upper and lower read/write heads.

Additional details of a three and one half inch floppy disk drive suitable for practice of principles of the present invention may be had with reference to U.S. Patent Application entitled Ultra-High Capacity Removable-Diskette Drive, filed on instant date herewith, and commonly owned by the assignee of the present invention, the disclosure of which is expressly incorporated herein by reference.

A diagrammatic view of a removable media disk drive subsystem, including a head and disk in accordance with practice of the principles of the invention is shown in FIG. 2.

For purposes of convenience, features that are shared with the exemplary embodiment of FIG. 1 will be identified by common reference numerals.

In addition to the linear voice coil motor, the head-positioning actuator assembly 20 (FIG.l) further comprises a head stack assembly, generally indicated at 28, which further includes an upper load beam 30 which supports a first magnetic read/write head 34 which is formed with a 70% slider of e.g. calcium titanate ceramic material. In addition, the head stack assembly 28 further comprises a lower head support frame 36 which, in turn, supports a lower read/write head 38 (seen in FIG. 1) which is also formed with a 70% slider of e.g. calcium titanate ceramic material. Referring to FIG. 2, there is shown a flexible recording media disk 40, of the type typically contained within the casing of a diskette cartridge (not shown). The disk is rotated by a spindle motor assembly 15 which is coupled to the recording medium by a spindle hub 14. Each of the two major surfaces of the disk 40 are coated with a magnetic recording medium which comprises a thin (approximately 0.1 to 0.5 microns) layer of metal particles coated over a non-magnetic titanium substrate layer, bonded in turn to a Mylar™ disk body. Each surface of the disks 40 comprises a multiplicity of radially spaced-apart concentric data tracks 42 and are further divided into a plurality of sectors, represented by sector boundary lines 43, each of which, as is well known in the art, comprises on each track a servo portion and a data portion. The track density achieved by the system is approximately 2700 tracks per radial inch of the disk, and recording densities of approximately 71 ,000 bits per linear inch along each track (bit density being about 71 ,218 bits per inch as measured along the circumferential direction at the innermost concentric data track).

Electronic circuit elements for electronically controlling the operation of the head disk assembly are preferably formed on a printed circuit board having approximately the same length and width dimensions as the base portion (12 of FIG. 1) of a floppy disk drive (10 of

FIG. 1) and which is attached typically to the underside portion of the base. In FIG. 2, the major functional circuit elements of the electronic circuit elements include a preamplifier and demodulator 44, connected to an analog-to-digital converter 45, which provides digital signal information to a monolithic microprocessor 46 which operates under software or firmware program control to command and manage the various operations of the disk drive system.

A digital-to-analog converter 47 receives digital control words from the microprocessor 46 and converts those words to analog signal values which are operatively directed to a power driver circuit 48. The power driver 48 generates an appropriate current which is directed through the voice coil motor windings of the actuator assembly 20 in order to radially move the head assembly from its current location to a destination track location during track seeking operations, and to maintain the head's magnetic recording gap in centerline alignment with a data track during track following operations.

As will be described in greater detail below, the servo portions of each sector have recorded therein information for identifying the track number, synchronization signals and servo burst pattern signals which, when decoded by means of the preamplifier and demodulator 44 and analog-to-digital converter 45, provide a signal which may be processed by the microprocessor 46 to generate an indication of the position of a read/write transducer head 34 relative to the centerline of a particular data track. Each data portion of a sector following the servo portion contains user data which may be read from and recorded to under software or firmware program control of a host computer (not shown).

The intelligent portion of the servo system is commonly implemented in the microprocessor 46 which calculates the head-positioning actuator motion control information, at least once per sample of information read from the tracks on the disk 40 during a seek operation by using one of a variety of conventional mathematical algorithms well known to those having skill in the art. Calculated values of the control motion are applied to control the head-positioning actuator motor through the digital-to-analog converter 47 and the power driver circuit 48.

In accordance with practice of the present invention, the exemplary embodiment of the disk drive 10 has been depicted in FIG. 1 and described with a linear voice coil motor-type head-positioning actuator assembly 20. It should be noted that such a head-positioning actuator assembly is provided in accordance with the invention to accommodate positioning a read/write head within the media access openings commonly provided on a 3.5" flexible media diskette cartridge. These conventional media access openings are configured to support linear motion of a head-positioning actuator, in well known fashion, in the radial direction along a flexible media disk. However, practice of the present invention is not limited to a head-positioning actuator of the type depicted, with a linear voice coil motor-type actuator assembly. It is well within the skill of a routineer to substitute a rotary voice coil motor-type head-positioning actuator, such as are commonly employed in contemporary rigid disk or hard disk drives. Such a rotary actuator may be easily positioned within the confines of the disk drive 10 in accordance with the invention, and configured to move the read/write head arcuately over the surface of a flexible media disk. Such arcuate motion could easily be accommodated by a simple reconfiguration of the media access openings of a diskette cartridge, or by a variety of methods by which the two major surfaces of a flexible media disk could be exposed to contact by a pair of read/write heads. Such a configuration, i.e., a rotary voice coil motor-type head-positioning actuator, would allow the resulting floppy disk drive to be constructed with a smaller footprint, and allow for some degree of commonality between floppy disk drive components and those of contemporary hard disk drives. This would further result in an increase in production efficiency along with a consequent reduction in cost.

SECTOR LAYOUT

FIG. 3 illustrates the layout of an exemplary embodiment of the read back signals of a servo data sector, e.g. that portion of each data track lying between the sector boundaries (43 of FIG. 2). The servo sector pattern depicted in FIG. 3 is replicated, in sequential fashion, along each servo data track in each servo data sector portion, and also replicated concentrically on each servo data track of the disk (40 of FIG. 2).

In FIG. 3, the exemplary data sector is depicted in two different formats. The upper portion of FIG. 3 depicts the exemplary data sector graphically, wherein the transitions are expressed by exemplary read back waveforms. Beneath this graphical representation, there is depicted a digital signal pattern-level representation, wherein the presence or absence of a magnetic flux transition is represented by a " 1 " or "0", respectively, in the "transition" row.

In accordance with well understood convention, a "cycle" is equal to two transitions, which also may be referred to as a "di-bit". The graphical representation of the sector layout of FIG. 3 is not intended to imply that the invention is limited to "waveforms". The graphical representation is provided herein only for purposes of visual convenience and to assist in an understanding of the true scope of the invention.

As can be seen in the exemplary pattern of FIG. 3, each sector comprises a first, servo field portion, which is conventionally followed by a second, data field portion. In accordance with practice of the present invention, the servo field comprises a fully embedded servo pattern which defines each servo sector. Defining the time period during which a transition would be expected as "T", the servo sector pattern may be described as follows.

Each servo pattern suitably comprises an AGC field 52, servo detect pattern 54, a synchronization field 56, a 2 T's isolation zone 57, a 2 T's code bit 58, a 2 T's second isolation zone 59, a 2 T's index bit (optional) 60, a gray coded 24 T's track address 62 followed by servo burst fields, servo A, B, C and D (63, 64, 65 and 66 respectively) in quadrature. In order to ensure separation of the servo field 50 from the following data field (not shown), a pad field (not shown) is typically included following the servo burst fields.

In accordance with the invention, the AGC field 52 is typically 4 microseconds in duration and suitably comprises 32 flux transitions written at a transition frequency of about 8 Megatransitions per second. The AGC field provides a constant and reproducible signal to the servo control electronics for purposes of automatic gain control, in a manner well known by those having skill in the art. Following the AGC field, a servo detect pattern comprises a lower transition frequency (equally spaced transitions at 0.25 microseconds) signal to uniquely identify the servo field to a disk drive's control electronics from formatted data header and user data fields. Preferably, the servo detect field has twelve transitions for a duration of about 3 microseconds.

A synchronization field 56 is provided to generate signals by which a conventional servo decoder circuit may be synchronized. The synchronization field is approximately 0.5 microseconds long and suitably comprises four transitions spaced at 0.125 microsecond intervals. In order to indicate the end of the synchronization field, an isolation zone is provided between the preceding sync field and the following code bit field. The isolation zone 57 is preferably a region of no transitions, approximately 0.25 microseconds in duration. The duration of the isolation zone will be chosen to provide a time interval of sufficient length for the disk drive's micro processor to shut off the servo decoder circuit such that the servo decoder will not receive, and be thereby confused by, any following information. A two transition region, commonly termed a code bit, 58 is provided to identify the type of servo information which is to be demodulated. For example, in well known fashion, the absence of the code bit indicates that phase information is to be demodulated. The code bit region has a duration of approximately 0.25 microseconds and is followed by a further isolation zone 59, comprising a 0.25 microsecond region of no transitions. Following the second isolation zone, an index bit 60 is provided in the servo field of the first three sectors of each data track. The index 60 may suitably comprise a pair of transitions (as exemplified in FIG. 3) in a particular timing window location in the servo sector, or alternatively, may be provided as a specific signal, such as a repetitive numeral, or the like. In the exemplary embodiment of FIG. 3, the index 60 is a pair of transitions comprising a pattern having a duration of 0.25 microseconds. Absolute track address, and, optionally, absolute sector address, is directly encoded as a Gray code pattern in the Gray code field 62 from the decimal representation of the track number. The Gray code pattern corresponds to a 12 bit binary representation of the decimal track number and is incremented (or decremented) by only one bit for each next concentrically inward data track on the disk (40 of FIG. 2). In accordance with practice of principles of the invention, a 12 bit Gray code field has 24 possible transitions and will have a duration of about 3 microseconds.

In a manner well known in the field of magnetic recording of Gray code patterns, each Gray code "one" bit is written with two transitions, while each Gray code "zero" bit is represented by a similar time interval during which there are no transitions. Following the Gray code pattern, a third isolation zone 67, comprising no transitions and lasting 0.5 microseconds appears, followed by a servo burst field 68 comprising a multiplicity of laterally spaced-apart sequential servo bursts, each having a duration of 2.5 microseconds and each comprising 20 transitions written at a transition frequency of about 8 Megatransitions per second. The servo burst field 68 is followed by a pad field (not shown) and a user data field (also not shown) capable of holding 512 bytes of user data and additional overhead information, such as error correction codes and speed tolerance filler gaps. Further details of the pre-recorded servo burst patterns are discussed below in the section entitled "Servo Burst Layout".

An ultra-high capacity flexible media recording disk in accordance with the invention, typically includes a multiplicity of servo sector portions, with the particular number being defined by the number of sectors per track which is, in turn, determined by whether or not some form of "zoned" recording technique is implemented on the disk. In the exemplary embodiment of the invention, two data zones are preferably implemented on the disk with the first, outer zone comprising 89 sectors per track with approximately 650 tracks in the zone. The second, inner zone comprises 71 sectors per track and includes approximately 356 concentric tracks in the zone. Each sector is offset from a preceding sector by a predetermined angular distance such that when the flexible disk is rotated at a constant angular velocity, each sector, including the servo sector portion, passes under an active read/write head's transducer in a predetermined angular phase relationship with the other sectors. Both servo sector phase timing and the length of a servo burst field within a servo sector phase are conventionally determined during the design process of an embedded servo system. Sector phase timing depends on the choice of rotational speed of the flexible recording disk and the number of servo sectors intended for an optimum servo loop sampling rate. The size of the servo burst fields within a servo sector are also a function of design choice. Servo sector phase timing and servo burst field length, are, therefore, determinable quantities whose values are typically stored by conventional software means in timing control registers comprising conventional servo timing and control circuitry provided with all disk drives having embedded servo systems. Accordingly, the servo timing and control circuitry defines a timing signal, synchronized with the servo sector phases, to define a succession of windows in time, such that each window opens while a servo burst is moving under the active read/write head's transducer.

SERVO BURST LAYOUT

In FIG. 4, there is depicted a diagrammatic, block-level representation of 5 contiguous concentric data tracks showing the arrangement of servo burst patterns on the flexible disk in accordance with practice of principles of the invention. As can be seen in FIG. 4, there are 4 servo bursts, burst A, burst B, burst C, and burst D, which are arranged both circumferentially consecutively, and radially in a particular pattern. As is illustrated in the pattern of FIG. 4. The A and B bursts are radially separated from each other by the distance of one full track pitch or width, as are the C and D bursts. Moreover, the C and D bursts are separated radially from the A and B burst pattern by one half of a full track pitch or width.

As is also evident from FIG. 4, the servo burst pattern, e.g. the radial and circumferentially sequential arrangements of the 4 servo bursts, preferably repeats itself in 2 track intervals. In particular, the arrangement of servo bursts as sensed by a read/write head's transducer at data track 0, are the same as sensed on data tracks 2, 4, 6 and etc. (even data tracks). The servo bursts as sensed at data track 1 are the same as sensed at data tracks 3, 5, 7 and etc. (odd data tracks) respectively.

Since the actual width of the gap of a read/write head's transducer in practice is typically two thirds that of a track pitch or width, approximately one sixth of a track pitch or width nominally separates the lateral edges of the head's recording gap from the boundary of an adjacent track. Preferably, when aligned along the centerline of a given odd data track, the illustrated servo burst pattern comprises two alternating half maximum read-back amplitude bursts followed by a full maximum read-back amplitude burst as sensed by the read/write head's transducer. In particular, each of the A bursts is written so as to straddle the boundary between data tracks and radially extend from the centerline of one data track to the centerline of an adjacent data track (in the example of FIG. 4, from the centerline of track 0 to the centerline of track 1, the centerline of track 2 to the centerline of track 3, and the like).

Likewise, the B bursts are also written so as to straddle data track boundaries and extend from the centerline of one specific data track to the centerline of another, but the B bursts are written a full track pitch or width offset in the radial direction from the A bursts, such that were the A and B bursts not circumferentially consecutive, they would interleave. Specifically, in the exemplary embodiment of FIG. 4, the B burst straddles the boundary between data track 1 and data track 2, and radially extends from the centerline of track 1 to the centerline of track 2 straddling the data track 1 to data track 2 boundary. The next B burst radially extends from the centerline of track 3 to the centerline of track 4 straddling the boundary between tracks 3 and 4. Following the A and B bursts, the C burst is written a half track pitch or width from the A and B bursts, and spans the centerline of the odd numbered data tracks. Next, the D burst is also written a halftrack pitch on width from the A and B bursts, but straddles the centerline of the even numbered data tracks, thus alternating with the C bursts.

Accordingly, as a read/write head's transducer radially travels across the centerline of a given data track (for example, data track 2 of the illustrated embodiment of FIG. 4), the active head's transducer will sense the A burst at approximately one half amplitude followed in time by sensing the B burst at approximately one half amplitude, and finally by sensing the D burst at full amplitude. This particular condition is depicted in the upper portion of the exemplary read back signal diagram of FIG. 5. It should be noted that in the preceding example, the read/write head's transducer will not sense any signal from the C burst, when reading servo information on even numbered tracks. In contrast, when a head is reading servo information, while aligned with the centerline on an odd numbered data track, it will sense or read an approximately half amplitude A burst, an approximately half amplitude B burst, and a full amplitude C burst, as depicted in the lower portion of the exemplary read back signal diagram of FIG. 5. Due to the radial position location of the D burst, it is not sensed or read on odd numbered tracks. Thus, for each track, one or the other of the C and D bursts is entirely within the width of the read/write head's transducer, while the other is entirely outside. This arrangement provides signals in quadrature, such that the relative amplitudes of servo bursts read by the read/write transducer head at a known position will provide valid direction feedback information during track seek operations as well as during track following operations.

After the burst amplitude values have been acquired by the read/write head, the amplified and demodulated amplitude values (the two half amplitude and the full amplitude values) are analyzed, in conventional fashion, by a microprocessor which determines the actual position of the read/write transducer head in relation to the centerline of the particular track of interest, and derives a correction signal value which is commanded to the head- positioning actuator to maintain or adjust the radial alignment of the read/write head to the data track centerline.

OFFSET AND RUNOUT CALIBRATION As is well known by those having skill in the art, position error signals (PES) derived by the analog circuitry prevalent in the field of disk drive servo demodulation systems, are subject to electronic and mechanical offsets. Such offset signals are typically superposed over the true position error signal and represents a degree of unwanted error in the PES. Accordingly, it is desirable to calibrate out these offsets in order to determine exactly where the centerline of a particular track lies, and to ensure that the position error signal provides a true indication of a read/write head's position with respect thereto. For any particular ultra- high capacity disk drive, its mechanical transport system and its electronic circuitry, will exhibit generally the same offset characteristics throughout its useful life. However, since the recording media (in this case a removable ultra-high capacity diskette cartridge) will often be transported from drive to drive, the electronic and mechanical offset characteristics experienced by any recording head/servo system combination will be different each time the media diskette cartridge is inserted into a different drive. In order to maintain the high track densities (small track pitches or widths) of ultra-high capacity recording media, tracking accuracy and performance must be maintained across various combinations of disk drive and diskette combinations.

In contrast to contemporary rigid disk drives, the servo patterns are not written on the media surface of ultra-high capacity recording disks by the disk drive's own read/write heads, capacity disks constrained to operate in combination with a single specific spindle motor assembly. In conventional rigid disk drive systems, the eccentricities due to repeatable run out are generally the same over the operating life of any given disk drive. Since the servo pattern for a rigid disk drive is written while the disk is rotated under the influence of the same mechanical systems, such as bearings, as during data transfer operations, resulting repeatable runout in rigid disk type systems, tend to be small and are easily handled by the head-positioning servo loop performance. In other words, it is not difficult to define the precise center of each data track for contemporary rigid disk type disk drive systems. All that is required, is that servo loop performance have sufficient bandwidth to take the resulting repeatable runout into account.

However, ultra-high capacity replaceable and/or interchangeable recording media is subject to considerably greater degrees of repeatable offset than that experienced by rigid disk drives. Particularly, temperature and especially humidity cause significant warping of a flexible media disk with a consequent large excursion amplitude for the offset. In addition, the read signal to noise ratio can easily change when different diskette cartridges are inserted into a particular floppy disk drive, and likewise when a particular diskette cartridge is inserted into a different floppy disk drive. Moreover, the absolute magnitude of a read signal may change over time and particularly with changes in ambient temperature and humidity. All of these factors combine in ultra-high capacity recording to require that the servo system is able to identify the centerline of a data track and be able to accurately position a magnetic recording or read/write head thereon.

In order to accomplish this objective, a servo system for ultra-high capacity flexible media recording must be able to compensate for the electronic offsets, introduced by the disk drive servo system electronics, thereby developing an accurate position error signal (PES) for accurately determining the centerline of the data track. In addition, once the centerline of a track is determined, such a servo system must be able to follow the defined centerline of each track by compensating for repeatable offsets such as those caused by changes in temperature and humidity. In accordance with practice of the invention, means are provided to minimize the effect of electronic and mechanical offsets in a particular diskette/disk drive combination by providing calibration tracks at particular radially spaced-apart intervals on the media disk surface.

With reference now to FIG. 6, which depicts the electrical input offset model of a preamp and demodulator section in accordance with the invention, the A,B,C and D servo bursts are read from the servo surface of a particular track and input into respective ones of a plurality of comparators 70a, 70b, 70c and 70d in a preamplifier and demodulator circuit 72. The servo burst amplitude signals are processed through a plurality of gain stages 74a, 74b, 74c and 74d which provide demodulated and amplified burst signals A', B', C, and D' respectively. In accordance with the invention, the demodulated and amplified burst signals

A' and B' are input to a summing amplifier 76 which evaluates the relative magnitudes of the A' and B' burst signals and develops a position error signal (PES) representative of the magnitude of the A' burst signal minus the magnitude of the B' burst signal. The sign of the PES signal indicates whether the A burst or the B burst was read with a larger relative magnitude, and thus indicates whether the active read/write head was positioned off the track centerline in the direction of the A burst or in the direction of the B burst.

A second summing amplifier 77 is provided, wherein the position error signal developed by the summing amplifier 76 is summed with a variable reference signal 78. The reference signal 78 is variable in that its value may be incremented or decremented by, for example, a microprocessor operating under software program control. The microprocessor would write a digital word, representing a first reference signal value to, for example, a D-to-

A converter (not shown) which, in turn, outputs an analog signal suitable for summing with the position error signal in the second summing amplifier 77. The microprocessor adjusts the value of the variable reference signal, in a manner defined in greater detail below, so as to cause a null signal value to appear at the output of the second summing amplifier 77 during offset calibration of the servo system.

Following preamplification and demodulation, the processed servo burst signals are provided to an analog to digital converter, from whose output, for example, microprocessor is able to determine command signals for moving the linear actuator, thus repositioning the active read/write head over the centerline of the data track. However, since the A', B', C, and D' servo burst signals result from processing through analog circuitry, it is well known that there will be some unwanted offset associated with these signals. In addition, the offset will be added prior to the gain stages 74A through 74D and are, in turn, amplified along with the servo burst information. In order to measure the magnitude and sign of the offsets and calibrate out their effects from the servo burst information, three calibration and reference track areas are written in particular locations on the media disk surface with specific servo burst patterns as depicted in FIG. 7. With respect to the embodiment illustrated in FIG. 7, it can be seen that the calibration and reference track areas comprise three calibration tracks: a calibration track 0 centered between a calibration track -1 and calibration track +1. Each of the tracks comprise the same track pitch or width as a nominal data track and further comprise an alternating arrangement of servo burst sectors arranged in odd sector, even sector periodicity. It will be noted that each even sector comprises a servo burst pattern provided on the media surface in the same manner as the servo burst pattern of a non calibration-type track such as described above in connection with FIG. 4. For each odd numbered sector, the A and B servo bursts are written with full amplitude values over the width of, and spatially aligned with, the three radially consecutive calibration tracks. In other words, the A and B servo bursts span the full width of the calibration track region. A read/write head's transducer, reading the odd sector servo burst information on calibration track 0 will read three sequential full amplitude servo bursts followed by a null in the nominal D burst position. Likewise, a read/write head's transducer, reading the odd sector servo burst information on calibration tracks -1 or +1 will read two sequential full amplitude bursts followed by a null in the C burst position, followed in turn by a full amplitude D burst.

In this fashion, the quadrature arrangement is maintained for the calibration track regions, while also providing an active track-following servo burst arrangement for the even sector servo fields, and providing a calibration region for the odd sector servo fields.

Calibration is performed by evaluating the A' - B' demodulated burst signals for the odd numbered (i.e., calibration) sector servo burst patterns. For these calibration sectors, it is clear that the PES may be expressed (with regard to the input offset model of FIG. 6) as:

PES=A' -B' ₊(A _el+B _e^(A _el+B _el)A=B

Thus, it can be seen that the offset value for a particular calibration track can be determined from an evaluation of the full amplitude A' and B' bursts. Returning briefly to FIG. 6, evaluation of the full amplitude A' and B' bursts occurs in first summing amplifier 76 wherein a position error signal is developed which represents the effects of the electrical offset of the input circuitry. This unwanted position error signal is provided to the second summing amplifier 77 wherein it is summed with a variable reference signal which is, in turn, adjusted to cause a null output. The value of the variable reference signal, at this point, can be seen as representing a most nearly exact compensation to the electronic offset derived position error signal. Once the digital reference, representing the electrical offset, is determined, its value may be stored in memory where it may be easily recovered and applied to the servo system electronics during data track following operations.

In FIG. 8, there is depicted a high level flow diagram of a calibration technique in accordance with practice of the present invention. Initially, as is shown in the flow chart of FIG. 8, the calibration system performs a read on the calibration A burst, followed by a read of the calibration B burst. As has been described above, a demodulated and amplified A' and B' calibration burst signals are compared in a first summing amplifier (76 of FIG. 6) in order to define a position error signal which can be further expressed as "the absolute value of A' minus the absolute value of B'". Because the calibration A burst and calibration B burst are both recorded on the disk media surface in such manner as to give a full amplitude servo read back signal, in a perfect world the position error signal should have a null value. In other words, a full value A burst minus a full value B burst would equal zero. However, due to the physical properties of components used to construct an integrated circuit, there is always some degree of electronic offset, or other abnormality, contributed to a processed signal. In order to define and remove the electronic offsets introduced by upstream electronic components, the calibration burst pattern position error signal is combined with a variable reference signal in a second summing amplifier (77 of FIG. 6), which signal is indeed varied until the output of the second summing amplifier achieves a null value. In other words, "the absolute value of the position error signal minus the absolute value of the variable reference signal equals zero".

Specifically, with reference to the flow diagram of FIG. 8, an arbitrary value is initially selected for the value of the variable reference signal by, for example, a control processor. This initial value may be a value which has been previously defined and stored during an earlier calibration process on a previously inserted diskette. This initial value is converted to an analog signal by a D-to-A converter, and applied to the second summing amplifier (77 of FIG. 6) whence it is combined with the position error signal. The output of the second summing amplifier is evaluated to determine whether it has a null value. If not, the variable reference signal is incremented, or decremented, depending on the magnitude and polarity of the second summing amplifier's output. This new variable reference signal value is again applied to the second summing amplifier and the output of the second summing amplifier is again evaluated for a null value.

This process is repeated until the output of the second summing amplifier achieves an output null, in which case the magnitude and polarity of the variable reference signal, which achieves the null, is saved for future reference in the disk drive's memory. Thus, it will be evident to one having skill in the art that the variable reference signal thus determined and stored, describes a mirror-image of the electronic offset introduced by the up-stream circuitry to the position error signal. Once the position error signal offset values are removed therefrom, the position error signal can be seen as representing an accurate indication of the position of a read/write head with respect to the centerline of a data track. Additionally, the position error signal (PES) offset values can be further refined in order to minimize the effects of noise disturbance, by evaluating the offsets for a multiplicity of sequential odd sector calibration bursts in accordance with the following formula:

Since the servo burst information on even sectors of the calibration track area is the same as the burst pattern for normal sectors, the servo burst information on even sectors may be used to obtain both magnitude and phase information of the PES signal. Conventional feed forward servo techniques may then be used to correct the actuator position due to, for example, repeatable runout error.

In accordance with practice of the invention, such feed forward servo techniques include dividing the surface of a flexible media disk into seven arbitrary zones and causing the servo system to attempt to follow a particular data track for several revolutions of the disk. During this time, the magnitude and direction of a position error signal derived for each servo sector, which may be averaged with the PES readings for the same servo sector during subsequent revolutions, is stored in memory in a data table, which represents an excursion map of the position error signal for a data track in each of the arbitrary zones. The PES values for each zone may be stored directly in the data table, or alternatively, the PES values for a particular data track may be processed by, for example, fast Fourier transform, in order to define a repeatable magnitude and phase value of the PES excursion due to repeatable offset caused by, for example temperature and humidity.

Calibration, both for electronic offset and the repeatable offsets, is performed in accordance with practice of principles of the invention, at least following disk spin-up after insertion of the new diskette cartridge into a disk drive. Compensation values are determined and stored in memory for subsequent use by the servo system during track following operations. In addition, the calibration routine may be invoked by firmware resident in the disk drive electronics, to perform a calibration sequence following a seek, read, or write failure, or preferably following the second retry after such a fault indication is returned by the drive. If the disk drive is being used in combination with conventional power saving features, calibration may be easily programmed to be performed following any sequence of power- down and recovery wherein the flexible media disk is spun down (the disk ceases revolving). Notwithstanding the foregoing, it will be evident to one having skill in the art that the calibration procedure may be performed at anytime during operation of the disk drive, including during periodic intervals during read and write operations. Such intervals may easily be programmed into firmware which controls operation of the drive, so as to provide an on-the-fly calibration capability.

The invention, as described above, provides for a system and method for track formatting and head servoing for ultra-high capacity removable flexible media, and in particular provides a system and method for determining and calibrating out electronic and mechanically caused offsets from the head servoing system. It should be recognized that the embodiment of the invention described above is exemplary and that there are numerous ways in which the various calibration tracks and servo burst patterns described may be rearranged in order to provide the required functions. For example, rather than alternating calibration burst patterns with nominal servo burst patterns in a calibration track area, the calibration burst sectors may be provided at any integral multiple of servo sectors along a particular band of tracks. Calibration sectors may be provided at every third sector, fourth sector, fifth sector, or the like, so long as their location is periodic and determinable such that conventional control electronics are able to generate an appropriate timing signal to identify their occurrence. In addition, the location of the calibration track areas are a matter of design choice and need not be positioned in the exemplary inner region, outer region, or center region, but may instead be positioned anywhere on the surface of the media disk that the designer chooses. A greater number of calibration tracks, in a greater number of radial locations, would provide additional calibration information and improve the tracking accuracy and performance of the system at the expense of losing recording area on the disk that could be used for customers storage capacity. A person having ordinary skill in the art may easily imagine alternative configurations, not only of track locations, and servo burst patterns, but also of the circuit elements comprising the present invention. Although treated individually as circuit elements, the electrical components of the present invention may be implemented as a single chip integrated circuit or, alternatively, various portions of the circuit may be combined with other circuits, such as, but not limited to, an ADC/DAC circuit in a multifunction VLSI integrated circuit.

Moreover, although the servo patterns of the exemplary embodiment of the invention have been depicted and described as written at the beginning of each data sector along a data track, the servo sectors and data sectors may be easily interspersed with one another in a manner well known by those having skill in the art. "Splitting" data sectors so as to accommodate a servo sector between the segments, is a well known technique that has been established for rigid disk drives implementing a "zoned" recording architecture. Implementing the present invention in a disk drive and flexible recording media combination with a "zoned" recording architecture and split-field data sectors is well within the comprehension of one having skill in the art and may be put into practice without any undue experimentation.

Accordingly, a significant improvement has been brought to the art of ultra-high capacity removable flexible media disks and disk drives. Although the illustrated embodiment has been presented in conjunction with a 3.5" removable floppy disk drive, many varying embodiments and applications of practice of principles of the invention will be readily apparent to those skilled in the art. For example, removable floppy and rigid disks of larger or smaller diameter may be used with equal success and the performance of optical disks and linear media storage systems may be substantially improved and enhanced by application of the principles of the present invention. It will be understood by those skilled in the art that the presentation of the preferred embodiment is by way of illustration and example only and is not intended to be limiting in any sense. The scope of the invention is to be determined only with reference to the following claims.

Claims

WHAT IS CLAIMED IS:

1. In an ultra-high capacity removable media diskette data storage system of the type comprising a removable media disk drive having an actuator assembly including an active read/write transducer head for reading/recording data on the media surface of a removable diskette, and a motor assembly for rotating a media disk at a controlled angular velocity, the data storage system comprising: a media storage disk rotatable at a controlled angular velocity and defining a plurality of concentric, pre-formatted data tracks each defining a plurality of data sectors; a servo portion at the beginning of each data sector, each servo portion comprising a pattern of repeating, spaced-apart servo bursts, each burst having a predetermined radial offset relative to the other bursts, the active read/write transducer head being positionable relative to the tracks in response a position error signal derived from the bursts; a micro processor for controlling the position of the read/write transducer head during track following operations in response to servo burst information read by the read/write transducer head from each servo portion; first means for compensating for electronic offsets introduced by the circuit means to the position error signal; and second means for compensating for repeatable offsets introduced by at least the flexible media storage disk to the position error signal.

2. The data storage system of claim 1 , wherein each servo portion comprises four bursts in quadrature relationship and includes a first burst in each servo portion offset radially from a data track centerline by one half of a data track pitch increment, a second burst offset radially from the data track centerline in the opposite direction by one half of a data track pitch increment, a third burst straddling the data track centerline, and a read back null portion alternating between the third burst position and the fourth burst position on respective odd and even numbered tracks.

3. The data storage system of claim 1, the first means for compensating for electrical offsets further comprising: a plurality of calibration tracks, each track including a first series of servo sectors for providing position information defining the centerline of the calibration track and a second series of servo sectors, alternating with the first series, for providing calibration information; and electronic circuit means for defining a magnitude and sign of electronic offset introduced into the position error signal, in response to said calibration information.

4. The data storage system of claim 3, wherein the second series of servo bursts comprise a calibration servo burst pattern including four servo bursts, the first and second bursts straddling the calibration track centerline and written at a width of one full track pitch, the third servo burst spanning the centerline of each calibration track and written at a width of one full track pitch, and a read back null provided in alternating third and fourth burst locations for sequential ones of the calibration tracks.

5. The data storage system of claim 4, wherein the first and second calibration bursts are adapted to provide a read-back signal, the first calibration servo burst recorded to have a read-back signal magnitude nominally the same as the magnitude of the read-back signal of the second calibration servo burst.

6. The data storage system of claim 5, the electronic circuit means further comprising: first comparator means for reading a magnitude of the first and second bursts of the calibration servo burst pattern, the first comparator means further for generating an output signal having a value representing a difference in a magnitude and polarity between the first and second bursts; and second comparator means for comparing the output signal of the first comparator means with a variable reference signal so as to provide a null output, the variable reference signal value defining a magnitude and polarity of the electronic offset introduced to a position error signal by the circuit means.

7. The data storage system of claim 6, wherein the microprocessor includes a memory for receiving and storing the variable reference signal value defining the electronic offset of the electronic circuit means.

8. The data storage system of claim 7, wherein the pre-formatted data tracks of the flexible media storage disk are divided into a plurality of zones, the second means for compensating for repeatable offsets including means for reading the servo portion at the beginning of each data sector, the second means further for defining a plurality of position error signals derived from servo bursts in each of the plurality of zones.

9. The data storage system of claim 8, wherein the plurality of position error signals are stored in memory, thus defining a map of repeatable offsets introduced by at least the flexible media storage disk.

10. The data storage system of claim 9, wherein the pre-formatted data tracks are divided into seven zones.

11. An ultra-high capacity removable media disk drive including signal processing circuitry for an embedded type head-positioning servo, in which spaced-apart prerecorded servo bursts move under a read/write head to provide a servo read signal that represents the position of the read/write head with regard to the centerlines of a data track, the disk drive comprising: an actuator assembly for supporting a read/write head adjacent a major surface of a flexible media disk, the read/write head having an output such that while servo bursts move under the head, the read/write head produces a servo read signal; control circuitry for controlling the position of the read/write head during track following operations in response to the servo read signal, the read/write head being positionable relative to the centerline of a data track in response to a position error signal derived from the servo read signal; and calibration circuit means, responsive to the servo read signal, for producing and temporarily storing a calibration signal such that the calibration signal represents a correction value which when combined with a position error signal defines the centerline of a data track.

12. The ultra-high capacity removable media disk drive of claim 11 , wherein the servo read signal comprises a plurality of servo burst signals, the calibration circuit means further comprising: first circuit means including a plurality of amplifier stages, each amplifier stage having an input and an output, each amplifier stage further for receiving and amplifying respective ones of the plurality of servo bursts; first summing means connected to the outputs of at least two of the amplifier stages, the summing means responsive to the outputs for producing a position error signal for use in correcting an error in the read/write head-position relative to a desired track centerline.

13. The ultra-high capacity removable media disk drive of claim 12, the calibration circuit means further comprising second summing means, coupled to the first summing means, for receiving the position error signal, the second summing means operatively responsive to a variable reference signal so as to null out the position error signal.

14. The ultra-high capacity removable media disk drive of claim 13 further comprising a memory for storing the variable reference signal value which causes the second summing means to null out the unwanted offset contained in the position error signal.

15. The ultra-high capacity removable media disk drive of claim 14 further comprising means for retrieving the variable reference signal value from the memory and further for applying the variable reference signal value to selected ones of the plurality of amplifier stages so as to compensate the amplifier stages for electronic offsets introduced to the position error signal.

16. A removable media diskette cartridge adapted for operation with an ultra-high capacity removable media disk drive, comprising: a housing defining an enclosed volume; a rotatable media disk disposed within the housing, the disk including first and second major surfaces, each surface for supporting a recording media; a plurality of concentric data tracks disposed on each major surface of the disk, each data track defining a plurality of data sectors; a plurality of spaced-apart servo sectors, each servo sector disposed in spatial relationship to a data sector, the servo sectors for providing a signal representing the position of a read/write head with respect to the centerline of a data track; and at least one calibration track disposed on at least one major surface of the disk, the calibration track including calibration means for defining the centerline of respective ones of the plurality of data tracks.

17. The removable media diskette cartridge of claim 16, each servo sector further comprising a pattern of repeating, spaced-apart servo bursts, each burst having a predetermined radial offset relative to the other bursts, the servo bursts moving under a read/write head to cause production of a position error signal, the read/write head being positionable relative to the centerline of the data tracks in response to the position error signal.

18. The removable media diskette cartridge of claim 17, wherein each servo burst pattern comprises four bursts in quadrature relationship and includes a first burst in each servo portion offset radially from a data track centerline by one half of a data track pitch increment, a second burst offset radially from said data track centerline in the opposite direction by one half of a data track pitch increment, a third burst straddling said data track centerline, and a read back null portion alternating between the third burst position and the fourth burst position on respective odd and even numbered tracks.

19. The removable media diskette cartridge of claim 17, wherein the at least one calibration track comprises a first series of servo sectors for providing position information for a read/write head and a second series of servo sectors, alternating with the first series, for providing calibration information for defining the centerline of the at least one calibration track.

20. The removable media diskette cartridge of claim 19, wherein the second series of servo bursts comprises a calibration servo burst pattern including four calibration servo bursts, the first and second calibration bursts straddling the calibration track centerline and written at a width of one full track pitch, the third calibration servo burst spanning the centerline of said calibration track and written at a width of one full track pitch, and a read back null provided in alternating third and fourth calibration burst locations for sequential ones of concentric tracks.

21. The removable media diskette cartridge of claim 19, wherein the first and second calibration bursts are adapted to cause production of a servo read signal, the first calibration servo burst recorded to have a servo read signal magnitude nominally the same as the magnitude of the servo read signal of the second calibration servo burst.

22. The removable media diskette cartridge of claim 21 , wherein the rotatable media disk is flexible.

23. The removable media diskette cartridge of claim 21 , wherein the plurality of spaced- apart servo sectors are pre-recorded on the first and second major surfaces of the rotatable media disk.

24. A removable media diskette cartridge adapted for operation with an ultra-high capacity removable media disk drive, comprising: a housing defining an enclosed volume; a rotatable media disk disposed within the housing, the disk including first and second major surfaces, each surface for supporting a recording media; a plurality of concentric data tracks disposed on each major surface of the disk, each data track defining a plurality of data sectors; and a plurality of signal patterns, each pattern disposed in spacial relationship to a data sector, the patterns causing production of a servo read signal when rotatably moved beneath a read/write head, wherein the patterns are configured such that the servo read signal comprises servo feedback information and track centerline calibration information.

25. The removable media diskette cartridge of claim 24, further comprising a plurality of spaced-apart servo sectors, each servo sector including a first pattern of repeating, spaced- apart servo bursts, each burst having a predetermined radial offset relative to the other bursts, the servo bursts moving under a read/write head to cause production of a position error signal.

26. The removable media diskette cartridge of claim 25, further comprising a plurality of spaced-apart calibration sectors, the calibration sectors including a second pattern of repeating, spaced-apart calibration bursts, the calibration bursts moving under a read/write head to cause production of a calibration signal.

27. The removable media diskette cartridge of claim 25, wherein each servo burst pattern comprises four sequential burst positions in quadrature relationship, the servo burst pattern further comprising: a first burst, in the first position, offset radially from a data track centerline by one-half of a data track pitch increment; a second burst, in the second position, offset radially from said data track centerline in the opposite direction by one-half of a data track pitch increment; a third burst straddling said data track centerline, the third burst alternating between the third burst position and the fourth burst position on respective even and odd numbered tracks; and a read back null alternating between the third burst position and the fourth burst position on respective odd and even numbered tracks.

28. The removable media diskette cartridge of claim 26, wherein the calibration burst pattern includes four sequential calibration burst positions, the calibration burst pattern further comprising: a first calibration burst, in the first position, straddling the track centerline and written at a width of one full track pitch; a second calibration burst, in the second position, straddling the track centerline and written at a width of one full track pitch; a third calibration burst straddling the track centerline and written at a width of one full track pitch, the third calibration burst provided in alternating third and fourth calibration burst positions for sequential ones of the calibration tracks; and a read back null provided in alternating fourth and third calibration burst positions for sequential ones of the calibration tracks, the read back null position alternating with the third calibration burst.

29. The removable media diskette cartridge of claim 28, wherein the first and second calibration bursts are adapted to cause production of the servo read signal, the first calibration servo burst recorded to have a servo read signal magnitude nominally the same as the magnitude of the servo read signal of the second calibration servo burst.

30. The removable media diskette cartridge of claim 28, further comprising a plurality of calibration tracks disposed on at least one major surface of the disk, the calibration servo burst patterns provided in spaced-apart relationship along respective ones of the plurality of calibration tracks.

31. A method for calibrating a servo system for controUably positioning a read/write head with respect to the centerline of concentric data tracks of a removable media diskette cartridge including a rotating media disk, the method comprising: providing servo bursts on respective ones of the concentric data tracks, the servo bursts for causing production of servo read signals, the servo read signals comprising head-positioning servo signals and servo calibration signals; reading said servo read signals; processing the head-positioning servo signals in a manner taking into account the calibration signals; and adjusting the servo system in operative response to the processed head- positioning servo signals.

32. The method of claim 31 , wherein the head-positioning servo signals cooperate to define a position error signal.

33. The method of claim 32, wherein the head-positioning servo signals alternate in time with the calibration signals, the method further comprising: reading respective ones of the calibration signals; and processing respective ones of the calibration signals to determine thereby a magnitude and polarity of an electronic offset of the servo system.

34. The method of claim 33, wherein the calibration signals are processed to define a reference signal, the reference signal representing magnitude and polarity of the electronic offset of the servo system.

35. The method of claim 34, wherein the reference signal is logically added to the position error signal so as to correct the position error signal for the electronic offset.

36. The method of claim 35, wherein the reference signal is stored in a memory, the reference signal being recovered from memory prior to being logically added to the position error signal.