GB2273794A - Magnetic disk drive unit - Google Patents

Abstract

The system comprises a disk drive cartridge and an electronics unit. The disk drive cartridge comprises at least one magnetic hard disk in a hermetically sealed, self-contained unit suitable for use in extreme environments of vibration, shock and temperature. The electronics unit comprises a circuit in which temperature induced variations are counteracted. The circuit comprises a precision rectifier 394 for rectifying a servo burst. The rectified signal is then integrated by an integrator 414. The integrated signal passes through a multiplexer 446 and is converted to an eight bit word by a high speed A/D converter 456. <IMAGE>

Description

MAGNETIC DISK DRIVE UNIT This invention relates to the field of magnetic disc memory units. More particularly, this invention relates to a ruggedized magnetic disc memory unit suitable for use in hostile or severe environments and capable of withstanding loads or stresses, such as shock, vibration, and temperature extremes and which is capable of operation at any attitude. The unit of the present invention is particularly suitable for use in military applications, but it may also find use in other demanding environments, such as, for example, oil and gas well drilling and seismic exploration.

Both magnetic tape recorder units and magnetic disc memory units have been known in the art for many years.

Tape units suitable for use in demanding applications are available, but a need exists for an improved disc unit.

Examples of such prior art disk units are disclosed in US Patent Nos. 4,791,508, 4,870,703 and 4,965,686.

The object of the present invention is to provide a magnetic disk drive memory unit capable of withstanding loads and stresses such as shock, vibrations and temperature extremes.

In accordance with this object, the invention provides : a disk drive assembly having at least one disk, said disk having a plurality of encoded servos stored thereon, said disk having a plurality of tracks thereon, at least one head positioned relative to said disk for reading said servos, means for providing corresponding servo signals from said read servos, said servo signals being used for positioning said head at a selected track of said disk, wherein a single peak detector means for detecting each of said plurality of servo signals is provided.

In accordance with the same object, the invention also provides : a method of detecting a plurality of encoded servos stored on a disk, said disk having a plurality of tracks thereon, the method comprising the steps of (1) reading one of said servos using a head positioned relative to said disk; (2) providing a servo signal from said read servo; (3) detecting said servo signal using only a single peak detector; (4) reading another one of said servos using said head; (5) providing another servo signal from said other read servo; (6) detecting said other servo signal using said single peak detector; and (7) repeating steps (4) - (6) for each of said servos.

In accordance with the same object, the invention also provides : a disk drive assembly having at least one disk, said disk having a plurality of encoded servos stored thereon, said disk having a plurality of tracks thereon, at least one head positioned relative to said disk for reading said servos, means for providing corresponding servo signals from said read servos, said servo signals being used for positioning said head at a selected track of said disk, wherein only single peak detector means for detecting each of said plurality of servos are provided.

The system of the present invention preferably comprises a disk drive cartridge and an electronic unit. The disk drive cartridge comprises five magnetic hard disks having 300 megabytes of usable capacity housed in a hermetically sealed, self-contained unit suitable for use in extreme environments of vibration, shock and temperature. The electronics unit comprises a metal structure wherein the electronics for the system are housed.

The disk drive cartridge has a stationary spindle to which a motor stator assembly is attached. A motor rotor assembly is constituted by a rotary hub mounted around the stator, the hub having a motor magnet mounted therein.

Upper and lower bearing assemblies are positioned between the stator and the rotor. This assembly is preloaded by a pair of canted springs which are disposed about the spindle and bear against the floating race of the upper bearing assembly. Preloading the assembly in this manner avoids abrupt contact between the bearing and the spindle shaft during vibrations. These abrupt contacts have been found to generate head position errors during vibration. These head position errors are outside the bandwidth of typical prior art closed loop servos. However, due to the increased capacity of the present invention a higher bandwidth is utilized. Accordingly, the canted springs of the present invention prohibit the abrupt contacts, thereby eliminating the head errors that would otherwise result.

The disk drive cartridge also includes a temperature sensor mounted on a circuit board of the spindle assembly.

The temperature sensor measures internal temperature of the disk drive cartridge. This temperature information is used to modify a seek algorithm. More specifically, to modify drive current levels to the voice coil (since voice coil resistance changes with temperature), to modify head load/unload operation, to modify temperature drift for readings of Hall effect devices, and to better account for flex bias of capton flex circuits.

The disk drive cartridge further includes a head loading sensor assembly. A magnetic flux sensor is disposed on a circuit board which extends over a portion of an arm assembly. The arm assembly has a permanent magnet therein position to pass under the magnetic flux sensor (e.g., a Hall effect sensor). Accordingly, as the arm assembly moves, the magnetic flux from the permanent magnet detected by the flux sensor will vary. This has been determined to be a generally linear relationship. Therefore, the position of the arm assembly (more particularly, the permanent magnet in the arm assembly) is tracked by the flux sensor.

Further, the arm assembly as does prior art arm assemblies has a park or stop position. However, unlike the prior art arm assemblies, the arm assembly of the present invention is completely moved away (fully retracted) from the disk in its park position. In this fully retracted position the heads and disk are immune to induced shock and vibration.

The electronics unit has a novel peak detection circuit. In the prior art individual peak detectors were used to recover each servo burst (for head positioning, i.e., track location and centre). The use of a single peak detector, as in the present invention, requires high speed operation. With a single circuit, temperature induced variations are consistent for each servo. The circuit comprises a precision rectifier for rectifying a servo burst. The rectified signal is then integrated by an integrator. The integrated signal passes through a multiplexer and is converted to an eight bit word by a high speed A/D converter.

The above-discussed and other features and advantages of the present invention will be appreciated and understood by those skilled in the art from the following detailed description and drawings, which are given by way of example only.

Referring now to the drawings wherein like elements are numbered alike in the several figures FIGURE 1 is a perspective view of a disk drive cartridge and associated electronics unit in accordance with the present invention; FIGURE 2 is a block diagram illustrating interconnection of the disk drive cartridge and the electronics unit of FIGURE 1; FIGURE 3 is an exploded perspective view of the disk drive cartridge of FIGURE 1 with the cover removed; FIGURE 4A is a top view of the disk drive cartridge of FIGURE 1 with the cover removed; FIGURE 4B is a side elevational view partly cut away of the disk drive cartridge of FIGURE 1; FIGURE 4C is a rear elevational view partly cut away of the disk drive cartridge of FIGURE 1; FIGURE 4D is a view taken along the line 4D-4D with the arm and sensor removed for clarity;; FIGURE 4E is a top diagrammatic view of flexible circuitry used in the disk drive cartridge of FIGURE 1; FIGURE 5 is a side elevational view partially in crosssection of the spindle motor assembly used in the disk drive of FIGURE 1; FIGURE 6A-B are views of a canted spring used in the spindle motor assembly of FIGURE 5 wherein, FIGURE 6A is a top view thereof and FIGURE 6B is a cross-sectional view taken along the line 6B-6B in FIGURE 6A; FIGURES 7A-B are views of a circuit board assembly used in the spindle assembly of the disk drive cartridge of FIGURE 1 wherein, FIGURE 7A is a top view thereof and FIGURE 7B is a cross-sectional view taken along the line 7B-7B in FIGURE 7A; FIGURE 8 is a block diagram of the spindle servo circuit used in the electronics unit of FIGURE 1;; FIGURES 9A-B are views of the arm assembly used in the disk drive cartridge of FIGURE 1 wherein, FIGURE 9A is a top view thereof and FIGURE 9B is a side elevational view thereof; FIGURES lOA-B are views of the magnet assembly in the arm assembly of FIGURES 9A-B wherein, FIGURE lOA is a top view thereof and FIGURE lOB is an elevational view thereof; FIGURES llA-D are views of the sensor assembly used in the disk drive cartridge of FIGURE 1 wherein, FIGURE 11A is a bottom view thereof, FIGURE llB is an end view thereof, FIGURE lIC is a top view thereof and FIGURE 11D is a view taken along the line llD-llD in FIGURE ilA; FIGURE 12 is a block diagram of a closed loop system used by the electronics unit of FIGURE 1;; FIGURE 13 is a block diagram of the circuitry of the electronics unit of FIGURE 1; FIGURE 14 is a diagram of embedded servo burst information used for track centring and location in accordance with the present invention; FIGURE 15 is a timing diagram for the embedded servo burst information of FIGURE 14; FIGURES 16A-C are a schematic diagram of the servo recovery circuit in accordance with the present invention; and FIGURE 17 is a top level structure chart of the software in accordance with the present invention.

Referring to Figures 1 and 2, a disk memory system is shown generally at 50. Disk memory system is a high capacity, high speed memory system which is designed to operate in severe environments (e.g., a military aircraft).

Disk memory system 50 comprises a disk drive cartridge 52 and an electronics unit 54. Disk drive cartridge 52 comprises five magnetic hard disks providing 300 megabytes of usable capacity. The disks utilize Winchester disk technology and are housed in a hermetically sealed, selfcontained unit which is configured as a plug-in module.

Accordingly, disk drive cartridge 52 is fully interchangeable and capable of rough handling, transportation and operation in extreme environments of vibration, shock and temperature.

Disk drive cartridge 52 comprises a housing 58 which has a main housing section 60 with a removable cover plate 62 which is fastened to main housing section 60 with a plurality of screw fasteners 64. Housing section 60 and cover plate 62 are of metal, preferably aluminium. A front cover 66 has a release latch mechanism 68 for locking the housing into or releasing it from a vibration cradle (not shown) equipped with shock mounts in which the unit would be mounted. Housing section 60 has a mounting slot 70 along the length of its two opposite sides, and these grooves mate with corresponding runners or guides in the cradle. To ensure that the unit is properly mounted in the cradle (and to ensure proper alignment with electrical connectors on the back end of the unit), the grooves are cffset on the two sides of the unit (as are the mating guides in the cradle).Since the grooves and mating guides are asymmetric, the unit can only be mounted in one position (i.e., the proper one) in the cradle.

Electronics unit 54 comprises a sealed metal structure 72 wherein the electronics for the system are housed. Structure 72 comprises a main housing section 74 with a removable cover plate 76 which is fastened to housing section 74 with a plurality of screw fasteners 73.

Housing section 74 and cover plate 76 are of metal, preferably aluminium. A plurality of connectors 78 are mounted at a front end of housing section 74 to provide a means for interfacing electronics unit 54 with disk drive cartridge 52, a host computer 85 (Figure 2) and a power source. Electronics unit 54 also includes an elapsed time indicator 75, a bit initiate switch 77 and bit indicator 79.

Disk drive cartridge 52 is interconnected with electronics unit 54, more particularly disk drive control electronics 80, by signal lines 82 (i.e., cables). Disk drive control electronics 80 communicates with interface electronics 84 (described hereinafter) within electronics unit 54 by signal lines 86. Electronics unit 54, more particularly interface electronics 84, communicates with host computer 85 via a small ' computer system interface (SCSI) bus 88 (i.e., cables).

Referring to Figures 3 and 4A-E, disk drive cartridge 52 includes a spindle assembly 90 (described hereinafter). Spindle assembly 90 includes a circuit board 96 at its lower end. An actuator plate 100 is mounted within housing section 60 adjacent spindle assembly 90. An arm pivot shaft 102 extends upwardly from actuator plate 100. A field assembly 104 having a head loading assembly 106 mounted thereon is mounted on actuator plate 100. Field assembly comprises a core 104a, an upper field 104b and a lower field 104c. A pair of adjustable arm stroke limiters 108 are mounted at opposing ends of field assembly 104 by fasteners 109.

A first bearing assembly 110 is disposed on shaft 102 and is supported by plate 100. An arm assembly 111 comprises a centre arm 112 having upper and lower arms 114, 116 attached thereto by fasteners 117. Arm assembly 111 is disposed on shaft 102 and is supported by bearing 110 to allow the arm assembly 111 to rotate relative to shaft 102.

Centre arm assembly 112 includes a housing portion 118 wherein a voice coil unit 120 is housed. Voice coil 120 includes a central opening 120a wherein motor core 104a is disposed. The current through voice coil unit 120 in cooperation with field assembly 104 drives the arm assembly in a linear fashion. A landing zone flag 117 and landed position flag 119 are mounted to housing 118. Flags 117 and 119 are detected by an optical assembly 115. Each arm 112, 114, 116 includes a plurality of flexures 121.

Flexures 121 have a raised or button feature 122 on at least one surface thereof and a head assembly 124 disposed at one end thereof. A second bearing assembly 126 is disposed on shaft 102 above the arm assembly. Nut and preload washers 128 are secured to shaft 102 to retain and provide a preloading force on the arm assembly. A clock plate 130, preferably comprised of metal, is secured to housing section 60, spindle assembly 90 and shaft 102 by a.

plurality of screw fasteners 132.

A landing pad and ramp bracket 134 is mounted to actuator plate 100. A landing pad and ramp slide 136 is mounted to bracket 134 by fasteners 137. Landing pad and ramp slide 136 include a plurality of landing ramps 138 and landing pads 142. Landing ramps 138 include recesses 140 which are configured to accept buttons 122 on flexures 121.

A voice coil flex circuit 144 is secured to a flex mounting bracket 146 by a pair of fasteners 148.

Bracket 146 is mounted to plate 100 by fasteners 150.

Circuit 144 provides connection to voice coil 120.

A hybrid circuit board 152 is mounted on a mounting bracket 153 which is secured to the inside surface of housing section 60. A head flex circuit 151 is connected to board 152 and is retained by a flex retainer 154 attached to board 152 by fasteners 156. Circuit 151 is connected at the other end to the arm assembly. Fasteners 156 also secure one end of board 152 to bracket 153. The other end of board 152 is secured to bracket 154 by fasteners 158.

Flex circuit 151 includes a stiffener 160. Another flex circuit 162 provides connection between board 152 and a connector 164 through a rear end of housing section 60. A spindle motor flex circuit assembly 166 provides connection between board 96 and connector 164.

Spindle assembly 90 comprises a plurality of magnetic recording disks 168 (e.g., five disks) disposed about a spindle motor assembly 170. Disks 168 are spaced apart by a plurality of disk spacers 172. Disk 168 and spacers 172 are retained on assembly 170 by a disk retainer 174.

Spindle motor assembly 170 spins the disks 168 in a precise rotary path so that the magnetic tracks on the disks remain in an established concentric location through the operating environment of extreme temperature and vibration. This degree of accuracy ultimately limits the minimum track width that can be used and the related maximum tracks per cm which is directly related to data capacity (described hereinafter).

Prior art commercial disk drives are susceptible to vibration and temperature. For example, during exposure to low levels of vibration, the disk platters begin to wobble as the spindle bearings are overloaded by g forces and, in some instances, the shaft itself can bend and contribute to this effect. Displacements as small as 7.6 * 10 2 mm (300 clinches) will cause track mispositioning and data errors.

Further, when exposed to the temperature range of -55 C to +71 C, severe track mispositioning occurs due to both the design of the disk clamp and the disk media itself. In the case of the disk clamp, a mismatch of temperature coefficients resulting from a steel spindle shaft (required for the bearing load) and an aluminium platter causes the platter to expand and contract around one of the clamp fastening screws, rather than the centre line of the spindle. This causes a "one time around" wobble of the platter resulting in track mispositioning and data errors.

Also, a second temperature induced effect is caused by the non-uniform coefficient of thermal expansion around the circumference of the platter. This lack of concentric expansion and contraction over temperature of anisotropy also caused track mispositioning and data errors. The spindle assembly 90 of the present invention has resolved these vibration problems.

Referring now to Figure 5, spindle motor assembly 170 comprises a hub 176 with a shaft 178 at about the centre thereof. Shaft 178 (preferably a steel shaft) is mounted on a support plate 180. Spindle shaft 178 is attached to support plate 180 by screw threads 182, and spindle shaft 178 has a hollow interior defining a central passageway 185. Support plate 180 is, in turn, fastened to the bottom of outer case 60 by a plurality of screws (not shown). The upper end of spindle shaft 178 is held by plate 130 via screw 132' (Figure 4B). Thus, both the upper and lower ends of spindle shaft 178 are rigidly held.

An annular motor winding 184, which forms the stator of an electric motor is mounted on shaft 178 via upper and lower sleeves 186, 188. Sleeves 186 and 188 are secured about shaft 178. Sleeve 186 is secured to shaft 178 by a locking bar 189 and set screw 190. A washer 191 sits on a shoulder 192 of shaft 178. Washer 191 and shoulder 192 support sleeve 188 and stop it from downward movement along the axis of spindle shaft 178.

The rotor structure of the motor has an annular magnet 194 which is adhesively bonded to a steel sleeve 196 which, in turn, is locked to rotary outer steel hub 176. An annular cap 198 is fit within the inside diameter of hub 176 at the upper end thereof.

Magnet 194, sleeve 196, hub 176 and cap 198 constitute a rotor structure which rotates around the stator structure mounted on the spindle shaft 178, so this rotor structure must be mounted on bearings. According, the rotor structure is mounted on lower and upper bearings 200 and 202. The inner race of lower bearing 200 is trapped and locked against rotation between a shoulder 204 on spindle shaft 178 and a shoulder 206 on base plate 180. A radially inwardly extending flange 208 from hub 176 engages the rotating outer race of bearing 200.

The floating inner race 209 of upper bearing 202 is held at spindle shaft 178 by a pair of canted springs 210 which are disposed within a recess of groove 211 about shaft 178. Canted springs 210 are preferably comprised of 302 stainless steel wire per ASTM-A580. Canted springs 210 eliminate abrupt metal to metal contact (due to external vibration/shock environments) between race 209 of bearing 202 and spindle shaft 178. The abrupt contact generates head position errors which are outside the bandwidth of typical closed loop servos (e.g., a normal bandwidth for a closed loop servo is 250 Hz). However, the present invention has track densities of about 817 tracks per cm (2075 tracks per inch) with centre spacing of about 1,2 * 10 2 cm (476 pinches). This is a much higher capacity than taught by the prior art (e.g., about 344 tracks per cm (875 tracks per inch)).The abrupt contact does cause head errors within the bandwidth of the closed loop servo of the present invention. Accordingly, the canted springs 210 avoid the abrupt contact and, therefore, eliminate head position errors caused by such contact. The canted springs 210 are an important feature of the present invention, since they provide a high axial and radial stiffness in the spindle motor assembly which, in turn, provides high vibration/shock/temperature resistant performance. The canted springs 210 also provide adequate axial movement allowing for bearing preload. This movement is required for a constant bearing preload to be applied as materials expand and contract with temperature variations. One of the two identical canted springs 210 is shown in Figures 6A-B.

Spring 210 comprises a plurality of continuous coils 212 configured to form a ring shape having an inner diameter for engaging the recessed portion of shaft 178. The inner race 209 of bearing 202 is also held by a washer 213.

Washer 213 is held in place by a spring washer 214, a washer 215 and a nut 216 which is threaded onto the top of spindle shaft 178. The rotary outer race of bearing 202 engages a radially inwardly projecting flange 218 on cap 198. Thus, the rotor structure of magnet 194, sleeve 196, hub 176 and cap 198 is supported for rotation by bearings 200 and 202.

A preload force also is imposed on bearings 202 by nut 216 acting through washer 210, spring washer 214 and washer 213 to impose a load on the inner race which, in turn, is transferred through the balls of bearing 202 to load the outer race against the flange 218 on cap 198. The preload is also imposed on bearing 200 by being transmitted from cap 198 through sleeve 196 and outer hub 176 to load the outer race of bearing 200, and thence through the balls of bearing 200 to react against plate 180. The preload operates to maintain contact throughout the bearing pack to reduce the effects of shock load and vibration.

Wires 220 from the windings of stator 184 pass into central passage 185 in spindle shaft 178 through a pair of opposed openings 222 in spindle shaft 178. The wires 220 pass through a passage 224 in plate 180 and are connected to printed circuit board 96. Board 96 is spaced from and attached to plate 180 by a plurality of spacers 226 and screws 228 which thread into plate 180. Current is delivered through wires 220 to the windings of stator 184 to drive the rotor and disks 168.

Referring to Figures 7A-B, spindle motor circuit board 96 includes a plurality of Hall effect circuits 230, 232 and 234, each comprising a Hall effect sensor 236, resistors 238, 240 and a concentrator 241. Circuit 234 also includes a capacitor 242. Sensors 236 and capacitor 242 are secured to board 96 by a suitable adhesive. Circuit board 96 is generally round and has an outside diameter about the same as disks 168. Board 96 also has an opening 244 at about the centre thereof where shaft 178 passes. Hall effect sensors 236 are used to provide rotating magnet position information which is fed into a spindle servo circuit 245 (Figure 8). The disk spindle rotation speed is controlled by the closed-loop velocity servo circuit 245 and the motor (described hereinbefore). A phase-locked loop 246 on the spindle servo circuit uses the Hall signals to lock the spindle speed.The spindle servo circuit 245 includes a predriver 247 which regulates the amount of drive current set regardless of temperature induced gain variations. This results in predictable and repeatable spindle start times throughout the operating temperature extremes.

A temperature sensor 247 is mounted on circuit board 96. A mounting pad 248 is disposed between sensor 247 and board 96. Temperature sensor 246 is an important feature of the present invention. The output of temperature sensor 247 is used to modify the seek algorithm. More specifically, drive current levels to the voice coil 120 are adjusted for temperature variations (since voice coil resistance changes with temperature), to adjust head load/unload operation, to modify temperature drift for readings of Hall effect devices, and to better account for flex bias of capton flex circuits.

The output of temperature sensor 247 is used to modify servo parameters during head load/unload and seeking operations. The output of temperature sensor 247 is converted by an A/D converter 456 (Figure 13 described hereinafter in conjunction with the novel peak detection circuitry) which is controlled by the drive controller 350.

The corresponding temperature dependent servo parameters are stored in a ROM lookup table. The following servo parameters are temperature dependent for head load/unload operation (1) Hall effect position sensor gain; and (2) driving current necessary to drive the actuator arm in an open loop fashion at a constant acceleration.

The following servo parameters are temperature dependent for seeking operation (1) flexible head cable compliance; and (2) driving current necessary to accelerate actuator arm at constant acceleration.

Typical commercial disk drives are susceptible to what is commonly known as "head crashes" when exposed to vibration. The flying head is supported by an air bearing which is stable and stiff enough to resist contact with the disk surface due to the exponential nature of the air bearing force vs. displacement curve. It has been determined that head crashes and media damage generally occur when the disk drives are exposed to vibration during the following operations. Rapidly positioning the head laterally track to track which tends to tilt or roll the head causing the air bearing to become unstable, particularly under vibration. By optimizing the head flexure attachment point to the centre of gravity of the head the tendency to tilt or roll the head under rapid positioning is eliminated, whereby air bearing stability is increased and tolerance to operating vibration is improved.

Crashes also occur when the disk is spinning up or down, whereby the head air bearing is weakened and, under vibration, the head may come into contact with the disk.

Also when the disk is stopped, the head will rest on the disk surface. Normally, during spin down, most commercial disks allow the head to come into contact with the disk surface until the disk is spun up again and the air bearing re-established. This condition is not suitable in a military type product which is subject to operational shock, vibration and rough handling.

Accordingly, at the first onset of spin down, the head must be removed away from the disk surface. The arm assembly 111 of the present invention provides for the heads to be completely removed away from the disk.

Referring to Figures 9A-B, the actuator arm assembly is shown generally at 111. Arm assembly comprises centre arm assembly 112, upper arm 114 and lower arm 116. Centre arm assembly includes housing portion 118. Arm assembly 111 is dynamically balanced around its pivot axis so that inputs of shock, vibration and load factors to the cartridge do not induce torsional effects in the arm. This results in the head staying on track during high acceleration input disturbances. Housing 118 has a cylindrical opening 250 therethrough having a diameter sufficient for accepting bearings 110 and 126. Housing 118 also has opposing pairs of rectangular opening 252, 254 which define a cavity within housing 118 wherein voice coil 120 is mounted.

Housing 118 further includes yet another opening 256 wherein a magnet assembly 258 is mounted by fasteners 260 to housing 118. Also, referring to Figures 10A-B magnet assembly 258 comprises a block assembly 262 having an accurate front surface 264, opposing side surfaces 266, 268, a rear surface 270 including extensions 272 at each side 266, 268 and opposing top and bottom surfaces 274, 276. A mounting hole 278 is provided at each extension 272-.

Fasteners 260 pass through mounting holes 278. Surface 274 includes a platform 280 having a channel or slot 282 therein. A permanent magnet 284 is secured in channel 282 by a suitable adhesive. Magnet 284 includes opposing accurate front and rear surfaces 286, 287, side surfaces 288, 290, and top and bottom surfaces 292, 294.

Magnet 284 has a north pole 296, designated (N),-at end 288 and an opposing south pole 298 designated (S), at end 290.

Referring to Figures 4A and 11A-D, head loading sensor assembly is shown at 106. Assembly 106 comprises a printed wiring board 300 having opposing bottom and top surfaces 302 and 304. A magnetic flux sensor 306 (i.e., a Hall effect sensor) having leads 308 is mounted at surface 302 by a suitable adhesive. Sensor 306 is positioned over a Hall effect concentrator 310 disposed within an opening 312 through board 300. A plurality of contact pins 314 extend through board 300. Pins 314 provide connection means with spindle motor flex circuit assembly 166. Resistors 316, 318 and a capacitor 320 are mounted at surface 304 of board 300. Sensor 306 is preferably a Hall effect type magnetic flux sensor having a ceramic substrate with a hermetically sealed element.

Sensor 306 senses magnetic flux from magnet 284 in the actuator arm assembly 111. Assembly 111 is shown in the parked or fully retracted position in Figure 4A, whereby buttons 122 on flexures 121 are disposed within recesses 140 of landing ramps 138 with heads 124 disposed between corresponding landing pads 142. In this position the heads and disks are immune to induced shock and vibration. This fully retracted position of the arm/head assembly is an important feature of the present invention.

In this fully retracted position the magnetic flux from magnet 284 sensed by sensor 306 is greatest. As the arm assembly 111 is rotated to position heads 124 over corresponding disks 168, magnet 284 will move away from sensor 306 reducing the magnetic flux from magnet 284 sensed by sensor 306. It has been determined that the output of sensor 306 relates to arm assembly 111 position (i.e., head position) in a linear manner. Accordingly, sensor 306 senses magnetic flux from magnet 284 to determine head position.

Also, automatic head retraction is accomplished in both the normal and abnormal operating situations. For normal operation, the heads are retracted off the media whenever the spindle is turned off. Retraction is accomplished in less than 100 m seconds so it does not add appreciable time to an orderly shut-down. The abnormal operations occur when the primary system power is lost or when the operator inadvertently removes the cartridge while the spindle is still spinning. In response to a primary power failure which would cause the disk to stop rotating, electronics senses power failure and utilizes the back EMF of the spinning spindle motor to generate the power required for retracting the heads off the disk and onto the resting pads.Thus the energy in the rotating inertia of the motor and disk assembly is converted to retraction motion unloading the heads, and protecting both them and the media from exposure to shock and vibration. The second abnormal sequence occurs if the operator removes the cartridge while the spindle is spinning. In this case a sensor in the cartridge handle senses that the handle release has been activated. The electronics respond to this signal by rapidly retracting the heads to the resting pad. Here again, this is accomplished in less than 100 m seconds, which is before the cartridge has disengaged the connector.

The use of magnetic flux sensor 308 and magnet 284 as a position sensor is an important feature of the present invention. This position information can be used to correct position errors.

Referring to Figure 12, an example of such a closedloop system is shown generally at 322. A commanded position signal is presented on a line 324 to a summing junction 326. The output of junction 326 on a line 328 is a position error and is utilized by actuator arm dynamics 330 to position the heads. The position of the heads is sensed by a position sensor (i.e., sensor 308 and magnet 284) at 332. A signal indicative of head position is presented on a line 334 to junction 326. The actuator arm position 322 determined by the Hall effect sensor 332 (e.g.

sensor 306) is fed back at 334 and subtracted from the commanded position at 324 to provide the position error signal at line 328.

Referring to Figure 13, controller interface electronics 84 comprises an I/O controller 336, e.g., an Intel 80C186 microprocessor (operating at 12.5 Mhz) as the main control element. The Intel 80C186 is a highly integrated 16 bit microprocessor with chip select logic, two channel DMA controllers, three timers and an interrupt controller contained on-board. A control program is stored in memory 338 of electronics 84, e.g., a 64K x 16 bit EPROM or E2 PROM. There are also two 32K x 8 static RAMs (SRAM) 340 for scratch pad and variable storage. Further, twelve discrete system input signals and sixteen discrete output signals are used for various functions from control of the ready light, to I/O controller, to drive controller communications. A combination I/O and disk data controller 342 (e.g. an 8496 from National Semiconductor) is in electronics 84.The I/O section 344 of this device is implemented as a state machine sequencer that receives commands from the processor. These commands execute various functions such as data in phase and status phase. When an operation is complete an interrupt to the I/O controller is activated. Various control status registers are used to store error status conditions. The disk data controller 346 section is also a state machine sequencer that executes commands issued by the I/O controller. The commands control the reading and writing to the disk. -Again, various control, status and error registers give the ability to correct errors or determine if a retry is necessary. The combination I/O and disk data controller utilizes its own dedicated data buffer which consists of two high speed 32 K x 8 SRAMs. Three disk channels into the data buffer are controlled by the I/O and disk data controller.The three channels are I/O, disk and processor. The disk channel has the highest priority and the processor channel the lowest.

This scheme allows for concurrent data transfer between the I/O and disk interfaces. Thirty-one analog voltage channels monitor various system functions for self test. An eight bit AID 348 measures these channels and the BIT software determines if there is a fault in the system.

Disk drive cartridge 52 utilizes both a dedicated surface and embedded servo system. This means that one of the five media platters is totally dedicated to track servo information. Servo information is also embedded on all other surfaces at the beginning of every sector on every track. This servo system provides the best combination of coarse and fine positioning necessary for high speed operations in extreme environments.

Disk drive circuitry 80 implements the critical drive level functions necessary for high speed disk drive operations. Disk drive circuitry provides read/write head positioning (control of the rotary actuator motor, i.e., movement of voice coil 120 in field assembly 104), read/write head selection, spindle motor start/stop, BIT functions, and power interruption and transient recovery control.

A drive controller 350 is the main controlling element of the disk drive circuitry 80. The drive controller utilizes a high speed digital signal processor (e.g., Texas Instruments TMS 320C30 state-to-the-art 32 bit digital signal processor) to control the position of the heads through the implementation of a closed-loop digital control algorithm.

The disk drive control scheme utilizes a complimentary system of dedicated and embedded servo information to insure optimum control of read/write head position during track seek, read and write operations.

Track acquisition is accomplished via closed loop control 352 of the rotary actuator. Coarse position information is written on a dedicated surface (i.e., one surface of one of the five disks 168) and is comprised of alternating tracks 0.70 Mhz and 1.00 Mhz signals. The mid point between each of these tracks defines the centre of each cylinder and consequently the data tracks located on the remaining disk surfaces. When a seek command is received, the drive controller utilizes the coarse position information to control head actuator acceleration and velocity during the seek operation.

Once track acquisition is complete, the embedded servo 354 information on the acquired track is utilized to perform track centring and following. Referring to the Figures 14 and 15 a novel peak detector (described hereinafter) is utilized to recover amplitude information for each of four servo bursts, designated A, B, C, D respectively. This information is processed by the drive controller 350 to determine exact head position and the amount of current to apply to the rotary actuator (i.e., voice coil 120) to correct and maintain head position.

Tracks A1, B1, A2, B2, A3 and B3 are shown in Figure 14. The embedded servo information is designated A, B, C and D respectively. The amplitude peak information indicated at 356 is detected by the novel peak detector circuitry described hereinafter. This servo information is a quad burst of time displaced pulses (e.g., a burst may compromise a 2.5 Mhz sine wave for 4 g sec.). The heads can be maintained on track centre by equalizing the amplitude between overlapping servo burst. For example, burst 358 designated A and burst 360 designated B overlap track centre line 362 of track A1. Line 362 corresponds to equal amplitudes for both A and B as designated by VA and VB (i.e., VA=VB). Accordingly, the heads can be maintained on track centre by adjusting head position so that amplitudes between overlapping servo bursts are equal (e.g., VA=VB).

The quad burst provides for linear track-to-track positioning with a tolerance of +/- one track. Also, these embedded servo burst are used for track to track positioning and fine positioning, as - described hereinbefore.

Each sector (0-54, Figure 15) includes a data segment 364 which is preceded by the embedded servo burst A, B, C and D. The timing sequence for the embedded servo/sector is provided in Figure 15. This includes timing or the following operations : zero; dedicated surface index; sector mark detection window; read signal, data head; sector mark detected; multiplexers Ag, A1, and A2; servos CS, WR, INT and RD; and embedded servo reset.

A rotary actuator drive circuit 365 comprises a twelve bit D to A converter 366 and an amplifier 368. The actuator drive 365 receives positioning data from the drive controller 350 which is converted to analog voltage for the rotary actuator motor drive. velocity profiles and rotary actuator drive currents are kept constant regardless of variation in resistance to provide optimum performance throughout the severe environments.

The drive controller 350 is supported by a drive controller memory which utilizes both RAM 370 and PROM 372.

The RAM 370 is a 64 Kbit CMOS static RAM organized as 2K x 32 and is on-board the drive controller. It is used to temporarily store and manipulate program data elements. The PROM 372 is a 1 Mbit CMOS EPROM organized as 32K x 32 which stores the drive controller operating program.

The drive controller 350 communicates with the disk data processor through a parallel communications port.

Input/output ports for discrete signals are also provided.

The discrete signals come from other parts of the disk electronics 80 and cartridge 52 and include "heads retracted", "handle closed", communication port flag, A/D converter status and power status. Output signals are also incorporated and include spindle motor enable, rotary actuator enable, head selection, A/D and D/A converters.

The closed-loop spindle control circuit 374, described hereinbefore, utilizes a Hall signal feedback 376 to control spindle speed.

Read/write analog signal processing circuitry 378 provides the high density encoding/decoding for disk data read/write operations. The data is encoded from NRZ and clock to run length limited by an encoder 380.

During the write process, NRZ formatted data and clocking information is encoded and supplied to the write drivers. During the read operation, encoded data stored on the disk is recovered by analog amplification and detection with subsequent clock recovery and decoding via a data synchronizer, phase lock loop (PLL) 381 and a run length limited (RLL) to NRZ decoder 382.

The first stage of the analog circuitry includes an automatic gain control amplifier, which maintains a constant signal amplitude at the output of a Bessel filter in the read string. The Bessel filter band limits the read signal supplied to the equalizer and detector. The equalizer restores critical amplitude and phase information for detection of the data transitions by the detector circuitry.

The output of the data detector is supplied to the phase comparator of the phase-locked loop via a data synchronizer. The synchronizer is utilized to insure fast acquisition of phase lock by running the PLL on a reference clock when the system is not in the read mode. The synchronized run length limited data and phase-locked clock are supplied to a decoder 382 for conversion to NRZ data and clock.

Write interlocks and cartridge file protection capabilities are incorporated into the circuits at 384 to allow for user-selectable protection against loss of stored information due to an inadvertent overwrite.

Referring to Figures 16A-C, the novel peak detection circuit is shown generally at 386. Figures 16A-C are a schematic diagram of the track following servo electronics 354 and the course position servo 352.

Accordingly, the schematic also shows a dedicated servo detection circuit 388, an index detection circuit 390 and a sector mark detection circuit 392. The peak detection circuit 386 comprises a precision rectifier 394 having an input on line 396 to an amplifier 398. Amplifier 398 has a feed back loop comprising resistors 400, 402 and a diode 404. The output of amplifier 398 is connected to a diode 406. A capacitor 410 and an inductor 412 are also connected to amplifier 398. An output of rectifier 394 is presented on a line 408. This output is a precision rectified signal of the input signal (i.e., servo burst) presented at line 396. An integrator 414 receives the signal presented at line 408. This signal is presented through a resistor 416 to an input of an amplifier 418. A resistor 420 is connected between ground and the other input of amplifier 418.A capacitor 422 and an inductor 424 are also connected to amplifier 418. The output of amplifier 418 is presented on line 426. This output is an integrated signal of the signal input at line 408.

An integration dump 428 receives the signal presented at line 426. Dump 428 comprises a pair of high speed analog switches 430, 432 and a resistor 434. The integrated signal is then presented on a line 436 to an amplifier 438 having a feed back resistor 440. The other input of amplifier 438 is connected by a resistor 442 to ground. The output of amplifier 438 on line 444 is presented to a multiplexer 446. Capacitors 448 and inductors 450 are also connected to multiplexer 446. A selected input of multiplexer 446 is presented at its output on a line 452.

The signal at line 452 is presented to an amplifier 454.

The output of amplifier 454 is converted from an analog signal to a digital signal (i.e., an eight bit word) by an A/D converter 456. A voltage regulator 458 provides a reference voltage to converter 456. Capacitors 460, 462 and resistor 464 are connected to regulator 458.

Capacitors 466, 468 and inductor 470 are connected to converter 456.

This single high speed peak detection circuit 386 is utilized in conjunction with high speed A/D converter 456 to recover the magnetically recorded, time division multiplexed, spatially displaced servo information bursts.

This method of recovery is in contrast to prior art commercial practice where individual detectors are utilized for recovery of each servo burst. The benefit of a single recovery circuit is in operation over extreme temperatures, where parameter drift of individual circuits would have to be accounted for. The single recovery circuit parameter variations over temperature will be identical for each servo burst, and would therefore cancel any negative effects.

Referring to Figure 17, a top level structure chart of the software is provided. The software for the I/O control 336 performs the following functions : accepts commands from the I/O bus and generates the required control signals to execute the commands; controls data flow between the I/O bus and the disk drive; monitors drive status and reports abnormalities on the I/O bus; and generates and transmits disk status on command. The software for drive control 350 controls the following functions : read/write head positioning; spindle motor starting and stopping; head retraction on power loss or cartridge removal; read/write head selection; power interruption and recovery control for both controllers; built-in-test functions; and drive status monitoring and reporting to the I/O controller. The functions performed by the software are illustrated here by describing the read operation, the write operation and the Built-In Test (BIT) function.

A command is received on the I/O bus and is decoded as a read command by the I/O control. The status of the drive is check for cartridge installed and disk spinning. If these conditions are met, the command is determined to be valid and the read operation is started. The address information received with the read command is translated into a head and cylinder number. This information is transmitted in a seek command to the drive control which decodes the seek command and selects the required head. The drive controller 350 then positions the heads on the required cylinder by driving the rotary actuator. This motion may involve acceleration and deceleration depending upon the number of cylinders the heads must be moved. After completion of this open-loop move, the servo error voltage is read and used to centre the heads on the track by means of a closed-loop digital control algorithm.After the heads are centred a command complete response is returned to I/O control.

The I/O control loads the disk data controller 346 registers with the proper information to perform a read operation. This information includes the starting sector number on the track and the number of sectors to be read.

As part of the read operation, the I/O control checks the header information to determine if the heads were centred over the correct track by the open-loop move. If the heads are on an incorrect track, the I/O control calculates the correction required to position the heads. The I/O control sends another seek command to the drive control to position the heads correctly. Upon completion of this operation, a command complete is sent from the drive control to the I/O control again. Another read operation is performed by the I/O control. During the data transfer from disk to RAM, under the control of the disk data controller 346, error detection is active. If a data error occurs, the disk data controller 346 supplies syndrome bytes to the I/O control to enable it to correct the error.The I/O control performs an "exclusive or" operation on the data with these syndrome bytes to correct the data error. After a sector of data has been loaded into ram, the I/O control signals the drive control so that its disk controller can transfer data from RAM to the I/O bus. Thus, data transfers occur concurrently from disk to RAM and RAM to the I/O bus.

It should be noted that the drive control cannot determine actual head position. Upon receipt of a seek command, it loads a cylinder address register with the cylinder number and makes an open loop move. If the I/O control determines an incorrect positioning, it must send a special command to the drive control, which repositions the heads without changing the contents of its cylinder address register. This maintains synchronisation between the two controllers.

In the write operation, the same head positioning operations are performed as during the read operation. Data transfers from the I/O bus to RAM are performed concurrently with the initial head positioning operation.

During the execution of BIT some visual indication is provided to indicate progress through the test. Upon entering the test, both BIT indicators are set. Next the disk drive cartridge ready led is lit to indicate that both processors are operational. The led is turned off upon completing the ROM checksum tests regardless of the result of the tests. Upon completion of the test, the I/O control processor will clear only those BIT indicators representing a fully functional line replaceable unit (LRU).

The system incorporates a BIT capability to provide failure detection and location functions for both onaircraft and off-aircraft levels of maintenance. BIT features may be initiated by a combination of manual, automatic or remote commands; self-monitoring of normal operating modes is also included. Fault status indication is provided remotely via the bus or locally via the fault indicators.

A BIT power controller is used to control the BIT indicators. This controller powers the BIT indicator drivers directly from the +28V DC input and provides a power-up pulse to allow both BIT indicators to he set upon power-up despite a power supply failure. The power controller is power sequenced to prevent normal power-up and power-down routines from generating false BIT indications.

Two BIT indicators 79 are provided to give a visual indication of the functional condition of the LRU they represent. One indicator represents the electronics unit 54 and the other indicator represents the disk drive cartridge 52. Both indicators are magnetically latched so that the last tested functional condition will remain displayed after powering down. Once set, the indicators will remain set until the indicated LRU successfully passes execution of the BIT program.

The BIT switch 77 is provided to allow manual initiation of the BIT routine. This switch only initiates BIT when the system is not performing a command function.

Wrap functions are provided to test the data path in expanding loops. These wrap tests loop the data locally at the controller, at the encoder/decoder and finally at the disk itself.

The system contains the capability to monitor test points during the BIT routine. Built in test circuits consisting of an analog to digital converter with a multiplexed input performs discrete measurement of key circuit test points located in the read/write data channel, the spindle drive circuitry, the rotary actuator drive and the power supply. The measured values are compared to ROM stored values to determine signal validity.

While preferred embodiments have been shown and described, various modifications and substitutions may be made thereto without departing from the scope of the invention. Accordingly, it is to be understood that the present invention has been described by way of illustrations and not limitation.

Claims (15)

1. A disk drive assembly having at least one disk, said disk having a plurality of encoded servos stored thereon, said disk having a plurality of tracks thereon, at least one head positioned relative to said disk for reading said servos, means for providing corresponding servo signals from said read servos, said servo signals being used for positioning said head at a selected track of said disk, wherein a single peak detector means is provided for detecting each of said plurality of servo signals.

2. The disk drive assembly of claim 1 wherein said peak detector means comprises rectifier means for rectifying a servo signal to provide a rectified signal; integrator means for integrating said rectified signal to provide an integrated signal; and analog to digital converter means for converting said integrated signal to a digital signal.

3. The disk drive assembly of any of the claims 1 or 2 wherein said servos each comprise a plurality of time displaced pulses.

4. The disk drive assembly of any of the claims 1 to 3 wherein said digital signal comprises a digital word.

5. A method of detecting a plurality of encoded servos stored on a disk, said disk having a plurality of tracks thereon, the method comprising the steps of (1) reading one of said servos using a head positioned relative to said disk; (2) providing a servo signal from said read servo; (3) detecting said servo signal using only a single peak detector; (4) reading another one of said servos using said head; (5) providing another servo signal from said other read servo; (6) detecting said other servo signal using said single peak detector; and (7) repeating steps (4) - (6) for each of said servos.

6. The method of claim 5 further comprising the steps of positioning said head at a selected track of said disk using said detected servo signal.

7. The method of claim 5 or 6 wherein each of said steps of detecting comprises the following steps rectifying said servo signal to provide a rectified signal; integrating said rectified signal to provide an integrated signal; and converting said integrated signal to a digital signal.

8. The method of any of the claims 5 to 7 wherein said servos each comprise a plurality of time displaced pulses.

9. The method of any of the claims 5 to 8 wherein said digital signal comprises a digital word.

10. A disk drive assembly having at least one disk, said disk having a plurality of encoded servos stored thereon, said disk having a plurality of tracks thereon, at least one head positioned relative to said disk for reading said servos, means for providing corresponding servo signals from said read servos, said servo signals being used for positioning said head at a selected track of said disk, wherein only single peak detector means is provided for detecting each of said plurality of servos.

11. The disk drive assembly of claim 10 wherein said only peak detector means comprises rectifier means for rectifying a servo signal to provide a rectified signal; integrator means for integrating said rectified signal to provide an integrated signal; and analog to digital converter means for converting said integrated signal to a digital signal.

12. The disk drive assembly of claim 10 or 11 wherein said servos each comprise a plurality of time displaced pulses.

13. Thedisk drive assembly of claim 10, ll or 12 wherein said digital signal comprises a digital word.

14. Disk drive assembly substantially as described hereinbefore with reference to the accompanying drawings.

15. Method substantially as described hereinbefore with reference to the accompanying drawings.