CA1257390A - Optical disk tracking and seeking systems and specific track formats (wobbled pits) - Google Patents

Abstract

ABSTRACT
.
This invention relates to the field of data recording by optical or radiation means and in particular to optical disk technology. It provides for media configurations which can be read by systems such as those described within. The combinations described can provide accurate track crossing count and a more accurate track following signal than previously available. The media employs substantially parallel tracks to act as a diffraction grating, creating a push-pull signal which varies in a sine-wave fashion as the radiation spot used for reading or writing informa-tion in the tracks moves substantially perpendicularly across a group of track lengths. This creates the track count. The same detector and electronics which produce the count signal from the radiation beam returned from the media may be used to keep the beam on the track centerline (i.e. for following). The push-pull signal is used to indicate the spot's deviation from the center-line of the track. The low frequency content of the push-pull signal is corrected by combining it with a signal from a blank area. Another method of correcting it is with information obtain-ed from a data-type signal which is produced in response to off-centered wobbled spots.

Description

~5~39() 6082-217 This invention relates to systems for maintaining read/
write or read beams on the center of a track in which optically readable information is, or is to be, recorded or erased; and more particularly to such systems where the tracking information for centering and counting tracks is made present in the record medium.

BACKGROUND OF INVENTION
Several systems exist to center a radiation beam's incident spots on information track centers, but these other systems have several limitations, and for commercially feasible systems which provide optical recording densities on the order of lOOOM bytes to a 30cm disk, highly accurate system must be deviced.
In the preferred form of this invention, a single beam is directed at an angle to the disk or recording medium surface, and a detector receives the reflected beam which has been modulated by the disk surface. Parsing the signal generated by the detector means yields accurate informat~on concerning the location of the beam relative to the track center and concerning the data in the track. In another "seek" mode, the number of tracks crossed can be parsed from the same signal.
The concepts disclosed herein may be applied to the use of a separate beam for writing, multiple detector beams, or splitting the reflected beam between a multiplicity of detectors, provided that the format limitations on the configuration of track sector headers on the recording medium surface are adhered to and/or that such format limitations are used in accord with the methods described for track following or track counting.

~25~;~9() The preferred embodiments employ a reflective media surface, physically responsive to radiation (of laser light for these embodiments). However, the invention can apply to various forms of media, including reflective and transmissive, and physi-cally or chemically radiation responsive media so long as the novel and useful structure described herein is employed.
Generally, as is the case in the preferred embodiments described herein, disk media is "mastered," created with data tracks (grooves), non-data areas (generally lands surrounding the grooves) and servo or adjustment areas (called headers, situate at spaced locations in and about the centerline of the grooves).
These are all in the information layer or plane of the disk.
Disk media may be "mastered" with data too, and completely blank disk media may at some future date be sold for use in optical drives which do the mastering themselves, using the same equipment which reads and writes data.
The inventive concepts described herein may apply to each of these media forms, but in the Detailed Description of Preferred Embodiments only one form is described and the informa-tion bearing layer 15 detectable modulations due to writing aretherein referred to as "pits" although these pits may be bumps or other manifestations when a different media is employed.
One existing type of system for centering a beam of radiation in a track is shown in United States Patent No. 4,271,334 which provides for the dithering or wobbling of the beam (or related beam) within or across the width of the track as the track 1~5~391) passes. An error signal is produced using the increase of the reflected beam's average intensity (which increases as the beam gets farther off-center), and the fact that the reflected beam's intensity variation produces a phase angle with the dither signal on only one side of the track. The amount of increased intensity reflects the magnitude, and direction of the off-center error is found in the existence or nonexistence of the phase angle between the dither signal and the reflected intensity variation signal.
United States Patent Nos. 4,236,105 and 4,234,837 describe a dither system which finds "switching lines" to signal the servo mechanism to change direction. Dithering, or active wobbling, has inherent design problems however, which prevent its easy imple-mentation in write/read systems.
In United States Patent No. 4,243,850, the tracking error signal is generated by the use of three read beams' spots in which the outer two spots' reflections gain or lose intensity when they come in contact with the information pits or hills of adjacent tracks. This signal is a differential signal generated by paired photodetectors which read these outer reflected beams, the absolute value of the difference showing the magnitude of the error and the fact of à positive or negative difference indicating the direction of the error.
Other systems employ the diffraction of light by track edges themselves to generate a track following error signal called a push-pull signal described in United States Patent Nos.
4,232,337; 4,209,804 and 4,100,557. Difficulties with these - 125~90 systems are discussed in more detail below but basically inaccura-cies in beam alignment cause undiscoverable flaws in the push-pull signal, making it an inaccurate measure of tracking.
Other systems use an error signal generated by the disk track's surface structure wobbling with respect to the line of information pits embedded in the center of the track. This error signal may be generated by the sinusoidal variations caused by a wobbling groove in which the data pits lie on a straight path as described by United States Patent No. 4,135,083 (at the top of column 8), or by a series of off-center prewritten data pits spac-ed continuously around the track on either side of the data path center line at predetermined intervals as in Netherland's Patents Nos. 8,000,121; 8,000,122; 8,103,117 and 8,102,621. In using continuous "passive" wobbling techniques such as these, while they do eliminate the problems associated with active wobble or dither-ing techniques, the retrieval or parsing-out of the tracking signal (given at the wobble frequency~ may be difficult, primarily because the relevant beam spot must first be in track to get a phase lock onto the wobble frequency, and also because of poor signal to noise ratios. In those where many pits are required for timing or track following, rather than wobbling the groove itself, as is required by United States Patent No. 4,456,981, the amount of disk space available for data may be reduced because data cannot be written in the groove adjacent to such wobbled pits. The only abbreviated wobble pit pattern found in extant art was in United States Patent No. 4,428,069 which did not provide a means for ~5739~) correcting its inaccuracies nor does it in any way indicate use of a push-pull signal nor many of the improvements found herein.
(The use of wobbled pits in headers for centering has been found in the magnetic recording art too, see for example, United States Patent No. 4,472,750).
One system described a corrected error signal; United States Patent No. 4,476,555. In that patent a "traverse" signal which may roughly correspond to the "central aperture signal" here-in is used with a counter and RAM to correct the tracking error signal at a rate of one time per disk rotation, whereas this inven-tion corrects the tracking error signal continuously at each header. Even assuming that the "traverse" signal is a central aperture signal, there is no indication of how it is derived. In the present application, the limitations are taught and claimed which provide for a correcting signal to be generated in the cen-tral aperture signal, as well as how to decode the signal to get the corrected tracking information.
Another system for correcting tracking signal is describ-ed in European Patent Application No. EP0099576A2. That system uses a discontinuity or flat mirror area in a track groove, and the push-pull signal dèrived therefrom to correct the push-pull signal. It does not address the problems with signal strength variation caused by written data and reflected light level varia-tions. Neither does it address how to handle errors in location of the blank or "mirror" areas, nor defects around such "mirror"
areas.

1~57;~9~:) It should be noted that the diffraction patterns generat-ed by a beam wandering to one side or the other of a track or groove (found in the low frequency push-pull signal) have proven to be unreliable for measuring track following when uncorrected.
This is due to shifts in the position of the reflected beam relative to the center of the photodetecting means and the inabil-ity to detect what caused the shift. These shifts may be caused by instability in the optics, mechanical displacements, or laser beam intensity distribution itself. This invention solves these problems because the track following signal is a combination of the push-pull signal and a correction signal. In both embodiments the correction signal is derived from the return beam modulated by the header structure of the information track being followed.
One branch of embodiments of the present invention uses a short pattern of wobbled or off-center-line pits or holes com-bined with the push-pull signal to produce a corrected tracking error signal. The modulation due to the off-center-line pits is found in the central aperture signal, which is derived from the full reflected beam. It also uses the push-pull signal to count track crossings.
The second branch of embodiments of this invention uses the discontinuities in the groove of a track sector header in order to correct the push-pull signal rather than the wobbled pits just mentioned. In this embodiment too, the counting of the changes in the push-pull signal which occur due to the crossing of the beam spot over each track may also be employed to determine relative 1~5~90 track address. However, where the invention employs continuous grooves (as in the first mentioned embodiment branch if used with-out discontinuities) there is no theoretical limitation to track crossing (or seek) speed, whereas there are seek speed limits beyond which accurate track count may not be possible where tracks are supplied with discontinuities.
A decision on which embodiment to use may depend on var-ious considerations including those just described, and extrinsics, such as the cost to produce the system. Of course, the sets of electronics described which decode either the first or the second media embodiment may be included in one system which could work with either of the two basic high data density media structures described.
SUMMARY OF THE INVENTION
Basically, the present invention provides for formatted surfaces on a recording disk medium so arranged and disposed that a beam reflected from this surface can be employed to generate and correct a track following signal and also to generate track count-ing signals. It also provides a method to generate these signals.
These signals can be used to direct servo mechanisms to dynamically adjust and readjust positioning of the beam relative to the infor-mation track center.
In some of the preferred embodiments, a clock signal is provided by reflections from a sine wave floor in the data track groove. In those embodiments employing servo areas within sectors, a timing pit is used for clocking. Pits representing data may be written in the center of the track groove and pits to provide servo information may be written "wobbled" to each side or on the center line in the header areas. These servo pits are written in a known format (or pattern) and receipt of that format is monitored. The track-following correction given by the push-pull signal is not used to the extent it does not conform to the expected pattern.
In the second branch of embodiments, discontinuities in the header-area are used to correct the track following signal and two dis-tinct patterns are taught which may be used for different disk sizes or applications.
In the invention, convenient track counting signals are found in the sinewave type changes in push-pull signal as a track seeking operation moves the radiation beam radially over the disk surface.
The push-pull signal is described in the United States Patent Nos. 4,232,337; 4,209,804 and 4,100,557 cited above, but basically it is a measure of the different strengths on two sides of the center of the reflected beam. These different strengths are due to changes in the diffraction patterns in the reflected beam.
These changes depend on the transv~rse location of the beam spot relative to the track center line. This diffraction pattern is due to the dif~erences in phase between that portion of the beam which is reflected from a land on the side of a track and that por-tion which is reflected from the track floor.
BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 is a representation of the top view of the disk 1~5~39~) recording medium depicting the track grooves and track lands greatly exaggerated in size.
Figure 2 is a cross-sectional view of the surface of the recording medium, taken at line 2-2 of Figure 1.
Figure 3 is a cross-section of the surface of the record-ing medium taken at a perpendicular to the cross-section of Figure

2, at line 3-3.
Figure 4 is a top view two tracks, are for each of the two preferred media embodiments and includes an enlargement of the section header areas of each.
Figures 4A(i), 4B(i), 4B(ii) and 4B(iii) show different header patterns. Figure 4A(ii) depicts the sum signal generated in response to the pattern of 4A(i) and Figure 4A(iii) depicts the location of the headers of Figures 4A(i), 43(i), 4B(ii) and 4s(iii) vis-a-vis data areas on a typical track.
Figure 5 is a schematic diagram depicting a laser light path to the surface of recording medium and reflected therefrom in a typical structure employing this invention.
Figure 6 represents the diffracted light spots which are generated by the reflected and diffracted beam returning from the surface of the recording medium as they strike the quad detector or split diode photodetector.
Figure 7 depicts the light spot as it appears on the photodetector in the path of the reflected light beam.
Figure 7A depicts the position of the light spot on the surface of the recording medium which produces the diffraction -~ 125~90 pattern of Figure 7.
Figure 8 is a block diagram depicting logical circuitry which may be used to parse the photodetector signals in one pre-ferred embodiment.
Figure 9 is a block diagram depicting logical circuitry which may be used to parse photodetector signals in another pre-ferred embodiment.
Figure 10 depicts adjacent track sector header areas in an area of the record media.

Figures 10A, 10B, 10C, and 10D are highly schematic tim-ing diagrams, of a triangular representation of what in reality should be a sinewave push-pull signal generated by the beam spot track crossings of Figure 10.
Figure 11 and llA depict track embodiments using discon-tinuities and the push-pull signal generated thereby, respectively.
Figure 12 is a circuit diagram which depicts an approach to the AGC function for use of discontinuities in header areas only.
Figure 13 is another preferred embodiment block diagram depicting logical circuitry to parse photodetector signals.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Figure 1 shows a typical recording disk 10, employing so called "ablative" media, a segment of which is enlarged to show detail. As discussed previously, this is only representative media and various types may be employed to take advantage of the concepts disclosed herein. The enlargement shows a data pit 30 and a servo :~.

~25~9t) pit 40 situated as they would appear in tracks 13 and 15 respec-tively. A discontinuity 39 is shown in track 13 also. The infor-mational or data tracks, before being written are known as pre-grooves. Between each track on the information surface or layer of the recording medium, is a land, depicted in Figure l as, for example, land 14 between tracks 13 and 15. Information tracks may be arranged in a concentric or spiral pattern over the infor-mation plane of the recording medium 10. (It is possible that the tracks might be oriented in a parallel series on a slidable planar medium or possibly on a tape, but currently such media are not in use for radiation based data storage.) Generally, pits 30 and 40 are representative of the pits to be written on the surface of the recording medium of the pre-ferred embodiments. "Pits" may vary in nature as required by the particular media employed. Minimally, it is required that there be a "change", modulation, or transformation which allows, alters or disallows the transmission or reflection of an incident radiation beam from or through the media. In other words, the manifestation of a modulation (which is a pit in this patent), whatever the media, must affect the incident radiation beam differ-ently than does the rest of the media's information layer and this difference must be detectable. Note that in various media, the information layer may be at the surface or at some plane within the disk. The form of the data tracks and non-data areas may vary in structure too, to accommodate other media, without deviating from the teachings of this invention. In short, this invention may be 1~57~391) used with numerous media forms.
In the preferred embodiments, the pits are nonreflective, and the disk at every other point in the information plane is optically reflective. The pits thus create a modulation in the resultant signal level when the beam spot passes over them.
Figure 2, a section view of the enlarged section of in-formation layer 19 in Figure 1, shows pits written into the sur-face of the recording medium and the lands and tracks in the preferred embodiments. The information layer 19 is seen covered by a clear photopolymer lacquer layer 18. Above layer 18 in the preferred embodiments may be a substantially clear substrate layer 18a such as glass, for structural strength, through which the laser or other radiation may easily pass. Data pit 30, may typi-cally be approximately 0.9 micrometers wide and, written properly in the center line of the track, it extends beyond both edges of the track 13. Likewise, wobble pit 40 partially obliterates land 16 and is of the same order of magnitude in size as data pit 30.
Lands such as lands 12 and 14 are approximately 1 micrometer wide, tracks are approximately 0.6 micrometers wide. For the preferred embodiments, the information layer 19 (the ablative surface) is a reflective tellurium or rhodium layer which is deposited on the entire surface of the recording medium structure and it ablates to form a hole or a pit (such as data pit 30) where a radiation or laser beam of appropriate power strikes the recording surface, lay-er 19, of the media. The average depth of the track groove, for example track 11, is equal to one-eighth the wave length of the radiation beam. This depth is known to generate, by causing dif-fraction of the beam, the most pronounced "push-pull" signal. The use of this push-pull signal to generate track following informa-tion is well-known in the art. See, for example, United States Patent No. 4,209,804.
A push-pull signal with higher frequency content is generated during seek (track crossing) because of the rapid push-pull signal change as the beam moves radially over a number of tracks. A sine wave signal is thus created, each repetition pro-viding one track crossing. Data pits written by the user mayaffect the push-pull signal quality at higher frequencies, and may prohibit proper track counting at higher seek velocities. This can be partially avoided by selecting a proper data modulation code with a so-called "D.C. free" frequency spectrum as for example that shown in United States Patent No. 4,464,714.
For track following, the push-pull signal's low frequency part may not be accurate. Abberations in the validity of the push-pull signal are caused by a shift of the beam returned from the disk surface with respect to the center line of the photo-detector. These beam shifts may be caused by optomechanical in-stability, laser pointing instability, tilt of the disk itself, displacements of the actuator, and intensity shifts in the laser beam distribution and perhaps other causes.
In data read/write systems used with this invention, an optical system such as that depicted in Figure 5 is likely to be employed. A laser 50 generates a beam 52 which is directed 1257~90 toward a generic optical path 57, routed by polarizing beam split-ter 51 through quarterwave plate 4, through the objective lens 55 so that beam 52 is reflected and modulated by disk 10 at beam spot 70 and reflected back into objective lens 55 becoming reflect-ed beam 53 which follows a different optical path through quarter-wave plate 4 and polarizing beam splitter 51 to its impingement on quad detector 80. The entire generic optical pathway 57 may be movable radially with respect to the disk responsive to actuator-servo means 56.
Figure 6 depicts beam 52 reflecting from track 11 at beam spot 70 to form reflected beam 53. The outline of a quad photodetector 80 is shown in the far field of reflected beam 53, positioned so as to receive zeroth order diffraction area 53' and first order diffraction area 53" and 53''' which create an inter-ference pattern on quad detector 80. The interference pattern drawn in Figure 6 is representative of a properly aligned and centered beam 52. The evenness of destructive interference fringes 72 and 73 cause an even signal to be generated by both sides of quad detector 80.
An off-track interference pattern is depicted in Figure 7. Zeroth order beam spot 71 is still located centrally on the quad detector 80 because reflected beam 53 is directed that way.
An interference between the zeroth order beam 53 and the plus first order diffraction area 53" is shown in the interference spot 72, while the destructive interference (between the zeroth and minus first order diffracted beams) is seen in area 74, and therefore, 1~57390 a weaker resultant signal is found on that side of the quad detec-tor 80. Figure 7A depicts the position of beam spot 70 which, in a properly aligned system, would create the interference pattern in Figure 7 on quad detector 80, relative to track 11.
The push-pull tracking error signal is the difference between the signal strength generated by one side of the quad detector and that generated by the other. The direction and the magnitude of error is given by the sign and value of the difference between the signals.
Where the reflected beam and its diffraction and inter-ference patterns are not correctly aligned with the photodetector diode (quad detector 80), or were the beam intensity is not proper-ly distributed across the entire spot 70, it can be easily under-stood that false tracking error signals will be generated using an uncorrected push-pull signal reliant method. This invention provides means to self-correct these errors in a continuous and automatic fashion with several media configurations.
Variations in the format of the media which may be em-ployed to correct track following signals will next be described, then descriptions of how these are employed by the preferred embodiments will follow.
In general, the relevant variations occur in sector head-er areas, the information track being divided into these headers ; and associated adjacent data bearing sectors.
A typical sector header is illustrated in Figure 4. The wobbled off-center pits provide an expected pattern which produces !

~:25739~) an error signal in the following way. The presence of a pattern of off-center pits in a certain format is expected and confirmed by reading them. This confirmation is critical when a small number of headers is used per unit of track length.
Where the signal which depicts the read pattern is missing the pits expected on one side only, or where the signal is weaker when the expected pattern on one side is passed and confirm-ed than when the expected pattern on the other side is passed and confirmed, it is apparent that the beam spot is following more to the side of the track with the stronger signal.
Discontinuities in the track provide for a similar func-tion in a dissimilar way. Their presence gives a signal which directly reflects the beam misalignment or anomoly in the push-pull signal, because the discontinuity is a flat reflective surface.
Accordingly the electronics to incorporate this information into the track following servo signal must be different than those employing the wobbled pit patterns suggested above. Also there are differences in the systems required to use the several discontinu-ity patterns described, the main differences being between those where the discontinuities are only in the headers and those where the discontinuities are present in special servo bytes in the data area.
Figure 4 shows two tracks 11 and 13 a segment of each of which illustrates the two embodiments just described which have servo information (modulations) in the header only. Track 13 has the wobbled pits 29, and 11 the discontinuities 41. The en-l'Z5~39() larged segments depict headers 24 from track 13 and 44 from track11. The use of synchronization pits 25 and 45, qualifier pits 26 and 46 and address pits 27 and 47 are enhancements described in detail below. It should be noted that track 13 may be discon-tinuous at the point 28 and where each wobbled pit segment exists without affecting the ability to derive a corrected tracking signal from such a header. A two pit to each side pattern as shown will work as will any wobbled pattern which is recognizable, balanced and known or expected will work.
If the spot is moving along the track centerline, the average sum signal will be equal when passing the left and right wobbled pit locations. If the loss is greater during the passing of the right side then it is clear that the spot is centered toward the right of the track centerline.
More detail regarding the track design in preferred embodiments can be seen in Figure 3, depicting a section down the center of a representative track. The sinewave floor 21 of the track extends along the entire length of the track and the size and orientation of this floor can be understood by noting that the point 22 represents a peak and the point 23 represents the next peak. The use of a clocking sinewave floor to generate synchroni-zation information so that the electronics "know" at what rate to expect data, is well-known in the art of optical disk storage.
The choice of whether to use an embodiment which employs a discontinuous track or one in which a track is continuous, or ; how many wobbled pits to employ to each side of the track, will 12S7~90 be based on many factors including the user's ability to create such patterns on the recording media. In embodiments using dis-continuities to correct the tracking signal, the wobble pits may be eliminated altogether, and the corrective signal may then be generated by "zeroing", that is, finding and applying the off-center value of the push-pull signal where there is no interference grating (at the discontinuity), as the beam spot 70 passes over the discontinuities. These discontinuities are plain reflective surfaces set in the track sector headers, also called blank areas.
In all embodiments such as those represented by Figure 4, track sector headers such as are written at a predetermined interval from each other along the length of each track. The known dis-tance, or number of sinewave clock peaks between each sector header location, together with the configuration of and pattern in each sector header provides the demodulation means with the ability to parse or separate out, the push-pull signal, a corrective signal, a track counting signal and a data signal as well as the clock signal from the tracks on the disk surface. This invention com-prehends embodiments which do not use sinewave clock floors in track grooves and these will be explained later.
To illustrate track crossing count determination, refer-ence should be had to Figure 10, in which a wobble pattern using 2 wobble pits 31 and 32, a spacer 33 and 2 wobble pits 34 and 35 on the other side of track 13 continued over the course of three track sector headers on three tracks. The wobbling pattern for this is shown to vary the timing patterns of Figures lOA, B, C and 125~390 D as centered lines A, B, C and D of the path of the reading spot would cross the edges of a track in track crossing (seek) mode.
It can be easily seen that counting from peak to peak of the resul-tant signal, or any one period measure of the signal shows that the spot has crossed from one track to the next.
The pattern of two pair of double wobble spots on each side with empty (nonwritten) track segments between them in a left, space, right, space, left, space, right, space sequence com-prises one successful pattern. Track crossing count will not be missed by the push-pull signal with such a sequence even at a track crossing speed which allows a track to be crossed radially at the longitudinal length of four timing spots. (A completely continuous track with the same pattern of wobbling will eliminate even this restriction on track crossing speed.) Assuming the implementation of Figures 10 and lOA to illustrate track crossing count generally, note that actual peak 3 of Figure lOA occurs at the crossing of line A of Figure 10 with the location 5, and also note that trough 2 is found at the crossing of line A with the left edge location 6 of track 15. A similar function is observed for lines B, C and D in Figures lOB, lOC and lOD, respectively.
The dotted line figures of lines A, B, C and D represent the signal generated by the same crossing taken over tracks without such discontinuities. Thus, in either case, as each center line crosses one track, one sinewave signal is generated, and counting these sinewaves gives the number of tracks crossed. Remember that the timing diagrams are triangularized for clarity and that the push-573gO

pull signal would appear to be more sinewave like in reality.
Figures 11 and llA are referenced here to illustrate thenature of D.C. or low frequency push-pull signal variance used for track following. Line E of Figure llA is the push-pull signal which would be derived from an accurately aligned system in using a beam spot whose center travels along line e of Figure 11. Figure 11 shows an example configuration of three radially adjacent header areas 201, 202 and 203 with Qualifier pits 204, address pits 205, discontinuities 206 and synchronization pits 207 and data areas 208 where they may be located in accord with the teachings herein.
In all embodiments using discontinuities, it can be easily seen (in the use of refective media) that the reflection of a beam from the discontinuity will be of a greater intensity than its reflection during the time the beam spot is over the grooved areas, and greater still than the intensity of the reflec-tion from the written areas.
Because of this fact, an automatic gain control is re-quired to compensate for the resultant variance in signal intensity so that the signals from the three described areas can be mixed.
To illustrate, an adaptation of the discontinuity based embodiment which relies on numerous headers is described with re-ference to Figures 4A(i), (ii), (iii) and 4B(i), (ii) and (iii).
The embodiment shown in these figures eliminates the need for a gain control to compensate for the presence of data because employ-ing it eliminates the need to use the push-pull signal generated ~ ~25~7~90 over the data areas for part of the tracking signal.
Specifically, the blank area or discontinuity 314 is seen at positions 2 through 11 of the second servo byte 311. The clear indication of its termination is given by the pit written at position 12 as seen in the Sum Signal 299 of Figure 4A(i). The sum signal also shows the difference between the reflectivity of the pregrooved area at position 1 and position 2 (the first position of full blank area). From this observation, it is clear that the push-pull signal, a measure of the D.C. offset in the sum signal, would be lost amidst the data signal without continuous automatic gain control. Where only the signal generated by the unwritten pregroove area and the blank (or discontinuity) are used for track-ing, the difference in gain between the signals of these two areas will be substantially constant.
As is known in the art, use of high read power levels may damage the data areas. The blank area (discontinuity) signal is used to check read power levels and the laser level is adjusted downwardly if it is too high, thus a secondary gain control circuit is established.
This invention teaches that data area need not be used to generate the push-pull signal and how adequate tracking can still be accomplished without such use, but it also demonstrates that continuous push-pull signal can be used for tracking with correction and describes means to do so.
While providing for simplified gain control, the formats ~ suggested in the Figures 4A and 4B affect the systems employing ; - 21 -' them in a number of ways and there are a number of other points which should be considered for those formats. First a complete explanation of the formats themselves should be had, starting with an overview of Figure 4A(ii). Shown in this figure is a segment of a data area in one of the track sectors which may be conside~ed the equivalent to the track sectors of Figure 4. The headers areas are assumed to contain normal track address, sector and synchroni-zation information. Each sector data area, usually following a sector heàder area, would be divided up into a number of segments such as segment 1 and segment 2 each shown with two servo bytes 319 at the head of each segment. If each segment is 16 bytes long, the last 14 bytes of such segment would be considered available for user data. The size of the segment chosen will depend on a number of factors including the size of the sector, the size of the modulation code used for writing data, and the amount of servo byte pair repetitions required so that there can be enough sampling of the servo areas per unit of time so that the phase lag of the servo responses is not great enough to impair servo function.
Another factor in determining the amount of servo byte area required is the need to write user data or have user data space available.
This servo byte format also eliminates the need for a sinewave floor in the user data areas and groove areas of the track for clocking. However, if the method of clocking described beiow is employed then it is advisable to keep these servo bytes in line from track to adjacent track. Doing so will enable a system -5739() employing such media to have accurate clocking during track seek-ing as well as during track following.
A simple configuration of the use of the blank area or discontinuity in servo bytes which are not in the header areas may be described with reference to Figure 4A(i).
Note first that each "byte" (including servo byte 310 and servo byte 311) of the servo area is divided into 15 spaces. This is representative of one data modulation code in which data can only be written in a certain number of the 15 spaces available for holes for each byte, i.e. 8 bits of infcrmation. Note also that the servo type modulation in the servo bytes cannot be one of the 16 recognizable modulations forms under whatever modulation code is used. This unique signal can be used a qualifier signal and the update which normally occurs at the passing of each pair of servo bytes for all sampled signals (including for instance clock-ing, focusing, tracking, the level of the sum signal, and the laser "read" level) can be disabled by this qualifier signal.
Since a pit is not allowed in the modulation code used in this example at position 4 as is illustrated in Figure 4A(i), a pre-groove 315 after it (pit 317) and the blank area 314 after that are sampled to set the levels of the sample signals just mentioned.
Detection of the pit 316 at position 12 of the second servo byte 311 updates the clock thus allowing the clock to be constantly re-synchronized at the occurrence of every pair of servo bytes in a data area. Note that the clocking hole at position 12 is placed directly after the blank area (or discontinuity) in order to avoid ~5~39~

transients in the read signal which might generate a false trigger in the clocking circuit. This allows for very exact synchroniza-tion, without the use of the sinewave clocking floor with reference to the embodiments whose only servo is in header areas.
There are several advantages to not using a sinewave clocking floor. secause this invention does not use a sinewave clocking floor, user data holes do not destroy any part of the clock and therefore bit density is not limited by that type of destruction. Therefore, unevenly distributed user data holes do not introduce any signal components which might create clock pull.
Further, the bit density limitation imposed on the user data by the sinewave clock floor frequency does not exist. The sinewave clock floor may be written at or near the limits of optical disk drive resolution capabilities, thus increasing bit frequency may make it difficult to resolve the clock in such systems. The sampl-ed servo byte does not have this problem. Nor does the user data modulation code have to provide a zero signal power content in the frequency spectrum at the clock frequency. Nor need any com-promise be struck between the amplitude of the clocking sinewave in the pregroove and direct read during write detection.
In the use of the servo byte embodiment and the design of the format for these servo bytes, other considerations may still be pointed out with reference to Figure 4A(i). Note that the two pits used must be chosen to be in locations such that user data holes which are written adjacent to the servo byte cannot introduce clock pull. It is felt that three free positions in a ~5~390 15 position code will provide enough free space to eliminate any potential for clock pull. Since sampling of most signals is done in the blank area, it must be made of sufficient length to allow for accurate detection yet it must also be short enough not to interfere with reliable track counting as was explained above with reference to Figure lO.
Also, the free pregroove area in the first servo byte must be chosen long enough to register the D.C. offset in the push-pull signal since with the servo byte embodiments, no other area besides the blank area is used for generating a tracking signal.
Note however, that where the "qualifying" pattern of the servo byte area is not properly detected, and enough track length has gone by so that proper tracks centering may be lost, the uncorrected continuous push-pull signal might be temporarily employed for tracking. The same use of uncorrected push-pull signal would be employed with the header located discontinuity but in that case, where there are far fewer discontinuities to check the push-pull signal this use of uncorrected signal is critical, and without employing qualifier modulations tracking may be lost completely.
Figure 4B(i) and Figure 4B(ii) illustrate variations of the pattern described in Figure 4A(i) which use wobbled pits 318 as a part of the qualifier signal and which may also be used for track following as is described above with reference to wobbled pits in header areas. However in these embodiments, clocking would be accomplished in the manner described with reference to Figure 4A(i) and the blank area may be used for track correcting and .

1~5~39~) sampling of sampled signals tfocusing, laser level, variation in intensity of ~he four quads, and quad sum level). Figure 4B(ii) may be written with a simple, single laser mastering machine as can the formats of Figure 4A (i) and 4B(iii). With reference to any of the 4B Figures, by virtue of the fact of there being so many servo bytes per unit track length, tracking signal may be entirely derived by the use of the wobble signal. Figure 4B(iii) illustrates a two servo byte format which might be used if track counting at high speeds is not required.
For the various format types described, differing cir-cuitry must be employed to parse the relevant signals from the detector.
Where the invention employs wobbled or off-center pits, track following signals are created generally (with reference to Figure 8) as follows:
A positive voltage is applied across resis-tor Rl to both cathodes of the detector 80.
Detector 80 is shown here as a split photo diode or pair of incident radiation detectors because this is the minimum con-20 figuration necessary for the function of such a detector with thisinvention. (Ordinarily, a quad detector is used to optimize focusing ability and for other reasons unrelated to this invention.
Merely tying the two inputs and two outputs of both the left and of the right sides of the quad together would yield the equivalent to what is shown here.) The sum of the modulations detected by detector 80 appears at the input to amplifier 108, whose output is the "central -` 1257~90 aperture" signal, a reflection of the overall modulation in the beam returned from the media. Directing this signal through circuit 111 which has phase locked loop and associated data demodulating logic brings the incoming signal into phase with the system clock.
The data signal and timing confirmation signals will appear on line 97. Timing logic 102 will pass the part of the now in phase central aperture signal occurring during a sector header to Qualifier 101. Timing logic 102 checks a predetermined pattern in phase with Circuit 111. Therefore it could simply signal the Qualifier 101 on line 102a during a sector header and Qualifier 101 could just read that part of signal which it could receive across a line 97b during such a sector header. The preferred embodiment, however, passes this signal from line 97a through Timing 102 to the Qualifier 101 across line 102a during such sector headers. Either method would work. Qualifier 101 is enabled during a nonseeking mode, i.e. track following condition by line 103. When enabled it detects whether the qualifier code embedded in the sector header (described above; Refer to Figures 4, 11) matches what is expected, and only if so, it enables Sample and Hold gate 90 to pass the wobble signal to low pass filter 99.
The central aperture signal from amplifier 108 is also supplied to left and right peak detectors 84 and 85. These peak detector logic circuits are enabled by the timing logic 102 which enables left peak detector 85 through input 88, but only during a left window (time a left wobble signal would be present) and also enables right peak detector 84 by line 86 when a right window is lX5~390 present. Outputs of these peak detectors are supplied to differ-ential amplifier 89 whose output is supplied to Sample and Hold gate 90, and allowed to pass as the "wobble signal" when "qualified"
as explained in the previous paragraph. This wobble signal cor-rects the push-pull signal.
To find the push-pull signal, the anodes of detector 80 are tied to the inputs of a differential amplifier 94, whose out-put at line 95 is the push-pull signal. High pass filter 109 and low-pass filter 99 have the same "corner" (also called "break") frequencies thus allowing the components of the push-pull and wobble signals to merge, at that corner frequency. This resultant signal is a corrected tracking signal. This signal is then ampli-fied by amplifier 91 and supplied to track following logic 117 and servo control logic 115 to cause servo mechanisms 150 to accurately follow tracks. Mixing of the signals may be achieved via connec-tion of lines 99a with lO9a or by bringing line 99b and lO9a directly to amplifier 91 as shown.
When seek status line 103 disables this Qualifier 101, only the push-pull signal passes to amplifier 91, providing the sinewave push-pull signal (Figure 10 et al) used for track count-ing logic 116.
Qualifier 101 may be eliminated from the circuit of Fig-ure 8 altogether if one is willing to sacrifice the redundancy it provides. A Sample and Hold gate, enabled on the occurrence of a header indicated by timing logic 102 during a non-seek status indicated by line 103, could substitute. Use of a Qualifier is ;

~5~390 preferred. It enables the system to shut down focusing, laser level and other adjustments in the event of a misreading as pre-viously explained with reference to Figure 4 et sec.
To employ the signals generated by discontinuities in the track headers for the corrected tracking signal, circuitry such as that described in Figure 9 should be employed. In Figure 9, quad photodetector 80 is represented by a split anel, but the same general considerations as to its form apply to this circuit as they do to detector 80 of Figure 8. In the circuit of Figure 13 the push-pull signal is also generated by a differential amplifier 60, each of the two inputs to this amplifier being responsive to a signal generated by either half of detector 80.
Lines 64 and 65 may be amplified jointly by amplifier 61 to produce an amplified sum signal on line 63 representative of the central aperture signal. This central aperture signal is sup-plied to synchronization and data demodulation circuitry 67 which in turn enables (via line 1) sample and Hold Gate 1 (68) on the occurrence of and for approximately the duration of each discon-tinuity. Synchronization and data demodulation cir~uitry 67 anticipates the passage of a discontinuity by looking at the number of clock pulses between headers, as well as synchronization infor-mation which may be contained at the beginning of a track sector header as shown in Figure 11, for example. Synchronization and demodulation circuitry 67, may pass (on line 3) either the segment of the signal received by it which it interprets as a qualifier signal, or some larger portion of the synchronized signal to the ~5~390 qualifier 69 for a comparison. Qualifier 69 will not disable the output of AND gate 75 unless there is a mismatch between the qualifier signal received from circuit 67 and the expected quali-fier signal in the comparator of qualifier 69. (This is a similar function to that of qualifier 101 in Figure 8.) Circuit 67 will send an enabling pulse on line 2 to AND gate 75 for a length of duration equal to the size of the sector header length in the pre-ferred embodiment, or at some other length sufficient to allow the push-pull signal from Sample and Hold Gate 1 (68) to pass through Sample and Hold Gate 2 (77), which is limited by this pulse on line 2. Of course, AND gate 7~, will be prevented from deliver-ing an affirmative signal on line 76 to sample and hold gate 2 (77) if line 78 is low indicating that the drive unit is in track seeking mode rather than track following mode.
Radial push-pull signal RPP from differential amplifier 60, is supplied to an Automatic Gain Control Circuit 62 to provide a usable signal on line 63. Without some kind of gain control the variance in RPP signal caused by the presence of user data areas, non-written areas of track groove and blank areas would not pro-vide a usable signal. Therefore, in the embodiments requiring it,Automatic Gain Control (AGC) may be accomplished in a number of ways as is known in the art, but for its application to this inven-tion several limitations must be observed. No. D.C. offsets can be introduced by the AGC and the variance in input voltages must be handled without distortion. (These limits on AGC are not necessary for its application to the two servo byte embodiments.) By way of ~ r~7390 illustration only, Figure 12 is provided, showing a circuit 300 having matched pair of FETs (Field Effect Transistors) with posi-tive voltage input to circuit 300 at point D. It is stressed, however, that the form of AGC will vary considerably with the requirements of the system in which it is employed, but it is be-lieved that the description just provided illustrates succinctly the best approach to the AGC problem.
The signal on line 63, (the normalized push-pull signal) is a low frequency signal, held up by Sample and Hold gate 1 (68) unless circuit 67 allows Sample and Hold gate 68 to open by an enabling pulse on line 1. This signal from Sample and Hold gate 1 (68), if qualified by qualifier 69, occurs during the time of passing of a sector header as determined by synchronization and data demodulation logic 67. If the machine is in a track follow-ing mode this signal may then be passed by Sample and Hold gate 2 (77). That is, Sample and Hold gate 2 (77) is enabled by AND
gate 75.
Resistors Rl and R2 can be used to adjust the signal appearing at line 79 so as to compensate for voltage differences between this signal and the signal at line 63 which may result from incomplete gain normalization by circuit 62 and the track groove geometry. The inputs from lines 79 and 63 when fed into a differ-ential amplifier 92 produce the offset corrected tracking signal on line 93. Note that a third Sample and Hold gate, S&H 3 (in ghost), could be added as an enhancement to limit the portion of the radial push-pull signal which reaches op-amp 93 on line 63.

f 1~5~390 This could be used to equalize the duration of the signals received by op-amp 93 over lines 79 and 63 by making the "pass through" type operation of S&H3depend on pulses from the Synchronization and data demodulation circuitry 67. These pulses (on dotted line 4) might, for instance be issued between written data pits.
Where the device is used to count track crossing, (during track seeking) no output is permitted through Sample and Hold gate 2 (77) and therefore no offset occurs in differential amplifier 92. Hence, the output of the circuit of Figure 9's differential amplifier 92 may be used in a manner identical to the output of amplifier 91 of Figure 8. If synchronization is correct during seek, output can be permitted through Sample and Hold gate 2 (77), and the track count may still be found at the output of different-ial amplifier 92.
To describe the finding of the tracking signal in the radial push-pull signal from the servo byte formats described in Figures 4A(i), 4B(i) and 4B(ii), reference should be had to Figure 13 and 4A(i). Based on either timing or the detection of the servo pits in the first servo byte, signals Sl and S2, developed by synchronization and data demodulation circuit 309, are provided to Sample and Hold gates 302 and 303 respectively, so that each may pass that portion of the radial push-pull signal to be sampled.
Sample and Hold gate 302 will pass the portion of the radial push-pull signal represented by the pregroove which, for example, would correspond to the section length Sl of Figure 4A(i). Sample and Hold Gate 303 responding to signal S2 would pass that portion of 125739(~

the radial push-pull signal representative of the blank or discon-tinuity portion of the second servo byte, by way of example sec-tion S2 in Figure 4A(i). Because there is no indeterminate gain variance, that is, the difference in gain between the pregroove area and the blank area will always be substantially the same, a fixed gain reduction 304 may be applied to the radial push-pull signal generated by the blank area. The two sampled portions are provided as input to differential amplifier 305 generating a cor-rected signal at line 306. If the proper qualifier does not appear in the data signal, signal Ql developed by qualifier 310 will not allow Sample and Hold Gate 307 to pass a tracking signal 308.
Note that the use of the same detector means 80, summing amplifier 61, and differential amplifier 60 of Figure 13 is similar to their use in Figure 9.
Synchronization and Data Demodulation Logic circuit 309 receives the sum signal from summing amplifier 61 and works similarly to the analogous circuit 67 of Figure 9, i.e. counting internal clock pulses between servo areas to provide said signals Sl and S2 and (through qualifier 310) Ql. Its (309's) clock pulses are updated by timing pits rather than by a sinewave varying floor height.
The circuit diagrams of Figures 8, 9 and 13 described the best mode known to the inventors herein for producing offset corrected tracking signals from the preferred record media describ-ed, and all the elements included are to enable one of ordinary skill in the art to make or use the invention. Still, they are l~S~39U

drawn in general terms to avoid unnecessary limitations. Note, for example, that resistors Rl and R2 of Figure 9, which provide gain compensation, should be regarded only as an exemplary circuit element pair. Also the output of synchronization circuit 67 could for example, be input to a second AND gate instead of AND gate 75 and the output of AND gate 75 could also be directed to the input of that second AND gate and the result would produce the same signal on line 76. The constraints of the particular device employed as well as limitations of the media employed will suggest alternatives or enhancements to the circuits described without exceeding the scope of this invention.

Claims (14)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. An improved record carrier for a data storage apparatus wherein said record carrier has an information layer having tracks for storage of data, said tracks being so arranged and disposed to form a diffraction grating for an incident radiation beam so as to generate a push-pull signal usable as an uncorrected track following signal, said tracks being comprised of longitudinally disposed sectors each sector having an associated header and where-in said storage apparatus may produce and read detectable modula-tions in said record carrier by means of directing at least one incident radiation beam to a spot movable along the longitudinal direction of one of said tracks or perpendicularly across substan-tially parallel tracks and wherein said storage apparatus reads said detectable carrier modulations in a returned radiation beam and wherein the improvement comprises:
wobbled record carrier modulation means for correcting said push-pull signal located in said headers, and record carrier modulation clocking means.

2. An improved record carrier as set forth in claim 1 where-in said information layer is reflective, said wobbled carrier modulation means comprises at least one nonreflective pit in said reflective surface at a predetermined location in said header to each side of said track.

3. An improved record carrier as set forth in claim 2 where-in said clocking modulation means comprises a height variation in said track floor disposed in continuous sinewave manner along the length of said track floor.

4. An improved record carrier as set forth in claim 2 where-in said clocking modulation means varies regularly a predetermined number of times for a predetermined distance, said predetermined number of times being invariant from the start of any one of said headers to the start of the next such header.

5. An improved record carrier as set forth in claim 2 where-in said closing modulation means comprises at least one nonreflec-tive timing pit, said timing pit being disposed a predetermined distance prior to said first record carrier modulation means.

6. An improved record carrier as set forth in claim 2 where-in said header includes address modulation means so arranged and disposed to reveal track and sector address of said header.

7. An improved record carrier as set forth in claim 2 where-in said header further comprises qualifier means for providing confirmation of the timing of its associated header.

8. An improved record carrier as set forth in claim 3 where-in said header includes address modulation means so arranged and disposed to supply track and sector address of said header.

9. An improved record carrier as set forth in claim 3 where-in said header further comprises qualifier means for providing confirmation of the timing of its associated header.

10. An improved record carrier for optical information storage as set forth in claim 3 wherein said wobbled modulation means comprises a predetermined pattern of nonreflective modula-tions written to either side of said track in said header said pattern configuration being limited to those which have an equal number of modulations wobbled to each side of the track.

11. An improved record carrier for optical information stor-age as set forth in claim 2 wherein said wobbled modulation means comprises modulations of equal size and number on either side of said track.

12. An improved disk shaped record carrier as set forth in claim 2 wherein said wobbled modulations are in a pattern compris-ing at least:
a) an (a) portion comprising; at least one pit to one side of said track, b) thence, in a longitudinal direction following said (a) portion, a (b) portion comprising; a length of track groove with-out wobbled modulations, c) thence, in a longitudinal direction following said (b) portion, a (c) portion comprising; modulations to the other side of said track from those in (a) portion, of the same size as that in (a) portion, d) thence, in a longitudinal direction following said (c) portion, a (d) portion comprising; a length of track without wob-bled modulations, e) thence, in a longitudinal direction following said (d) portion, a (d) portion comprising; a number of modulations to the one side of said track as those in (a) portion, or the same number as those in (a) portion, f) thence, in a longitudinal direction following said (e) portion, an (f) portion comprising; a length of track without wobbled modulations, g) thence, in a longitudinal direction following said (f) portion, an (g) portion comprising; a number of modulations to the other side of said track from those in (a) portion, of the same number as those in (a) portion.

13. Tracking signal generator means for obtaining corrected track-following signals from a returned radiation beam from a carrier for use within a data storage device as set forth in claim 2 comprising:
a system clock means, for generating system clock input, detector means for detecting said returned radiation beam, said detector means having at least a first and second por-tion, each portion being so arranged and disposed to receive a longitudinally split one half of said returned beam for providing a first and second detector means signal, respectively, each of said detector means signals being representative of the intensity of radiation returned to each respective detector means portion, and wherein said detector means also provides a third detector means signal, and third detector means signal being representative of the entire modulation present in said returned beam, first differential means for generating a difference out-put signal representative of the differential strength between said first and second detector means signals, said first and second detector means signals being input into said first differential means, phase locked loop and data demodulation means for receiv-ing said third detector means output and for comparing the timing of signals in said third detector means output representative of said record carrier clocking modulation means with timing of input from said system clock means and for generating a clocked data signal representative of each of said record carrier data and wobbled modulations, in phase with the system clock means, timing means, with said clocked data signal and said system clock signal as input, for generation of input to said left and right peak detector means at the expected timing of said wobbled carrier modulation means: to the left side of said track for said left peak detector means timing input, and to the right side of said track for said right peak detector means timing input, left and right peak detector means which both have for input said third detector means output signal, and as enabling in-put said left and right peak detector timing input signals res-pectively, for passing said third detector means output signal at the timing of the passing of a left and right window, respectively, as left peak detector output and right peak detector output, respectively, second differential means for generating an output signal representative of the difference in signal strength between said left and right peak detector output signals (and having said left and right peak data output signals as input), Sample and Hold Gate means for receiving said difference signal from said second differential means, Qualifier means for comparing a portion of said data signal output (its input) with a predetermined pattern and for generating an enabling signal to said sample and hold gate means where said comparison is indicative of the passing of one of said headers and wherein said enabling signal is of a duration equival-ent to the duration of the passing of said one of said headers, filter means for joining appropriate bandwidth portions of said difference output from said first differential means and from said sample and hold gate means to produce a corrected track following signal as output.

14. A tracking signal generator means as set forth in claim 13 wherein said qualifier means has a seek status input which selectively prevents said qualifier means from sending said enabl-ing signal to said sample and hold gate means, thus permitting the output of said band pass filter means to represent the un-corrected push-pull signal.