Disk drive employing feed-forward compensation and phase shift compensation during seek settling

Disk drives comprise a disk and a head connected to a distal end of an actuator arm which is rotated about a pivot by a voice coil motor (VCM) to position the head radially over the disk. The disk comprises a plurality of radially spaced, concentric tracks for recording user data sectors and servo sectors. The servo sectors comprise head positioning information (e.g., a track address) which is read by the head and processed by a servo control system to control the actuator arm as it seeks from track to track.

FIG. 1 shows a prior art disk format 2 as comprising a number of servo tracks 6 defined by servo sectors 4 0 - 4 N recorded around the circumference of each data track. Each servo sector 4 , comprises a preamble 8 for storing a periodic pattern, which allows proper gain adjustment and timing synchronization of the read signal, and a sync mark 10 for storing a special pattern used to symbol synchronize to a servo data field 12 . The servo data field 12 stores coarse head positioning information, such as a track address, used to position the head over a target data track during a seek operation. Each servo sector 4 , further comprises groups of servo bursts 14 (e.g., A, B, C and D bursts), which comprise a number of consecutive transitions recorded at precise intervals and offsets with respect to a servo track centerline. The groups of servo bursts 14 provide fine head position information used for centerline tracking while accessing a servo track during write/read operations.

FIG. 1 shows a prior art disk format comprising a plurality of servo tracks defined by servo sectors.

FIG. 2A shows a disk drive according to an embodiment of the present invention comprising a head actuated over a disk.

FIG. 2B shows a closed loop servo system for actuating the head over the disk according to an embodiment of the present invention.

FIG. 2C is a flow diagram according to an embodiment of the present invention wherein the servo system compensates for a phase shift between a measured and estimated servo system state.

FIG. 3 shows an embodiment of the present invention wherein the servo system compensates for a phase shift caused by one or more notch filters by generating feed-forward compensation in response to a phase lead.

FIG. 4 shows an embodiment of the present invention wherein a feedback control signal is generated in response to a phase lag according to an embodiment of the present invention.

FIG. 5A shows an embodiment of the present invention wherein a model based feed-forward compensator operates according to a state trajectory, and the state trajectory is adjusted in response to the phase shift.

FIG. 5B illustrates different state trajectories based on a different initial state of the servo system prior to switching to a settle servo loop according to an embodiment of the present invention.

FIG. 6 shows an embodiment of the present invention wherein the position error signal generated by the model based feed-forward compensator is adjusted by a phase lag in order to compensate for the phase shift between the measured and estimated servo system state.

FIG. 2A shows a disk drive comprising a head 16 actuated over a disk 18 comprising a plurality of tracks, and control 20 circuitry comprising a closed loop servo system comprising a seek servo loop and a settle servo loop (e.g., FIG. 2B ). The control circuitry 20 is operable to execute the flow diagram shown in FIG. 2C , wherein the seek servo loop is used to seek the head toward a target track (block 22 ), and when the head approaches the target track (block 24 ), the control circuitry switches to the settle servo loop (block 26 ). A feedback control signal is generated based on a measured state of the servo system (block 28 ), and the feedback control signal is adjusted with a feed-forward compensation (block 30 ). A phase shift between the measured state of the servo system and a corresponding estimated state of the servo system is compensated (block 32 ), and the head is settled on the target track in response to the feed-forward compensation (block 34 ).

In the embodiment of FIG. 2A , a plurality of concentric servo tracks 36 are defined by embedded servo sectors 38 0 - 38 N , wherein a plurality of concentric data tracks are defined relative to the servo tracks 36 . The control circuitry 20 processes a read signal 40 emanating from the head 16 to demodulate the servo sectors and generate a position error signal (PES) representing an error between the actual position of the head and a target position relative to a target track. The control circuitry 20 filters the PES using a suitable compensation filter to generate a control signal 42 applied to a voice coil motor (VCM) 44 which rotates an actuator arm 46 about a pivot in order to actuate the head 16 radially over the disk 18 in a direction that reduces the PES. The servo sectors may comprise any suitable head position information, such as a track address for coarse positioning and servo bursts for fine positioning. The servo bursts may comprise any suitable pattern, such as an amplitude based servo pattern ( FIG. 1 ) or a phase based servo pattern.

FIG. 2B shows components of a closed loop servo system for actuating the head 16 over the disk 18 according to an embodiment of the present invention. A measured position 48 of the head 16 is generated from the position information recorded in the servo sectors 38 0 - 38 N , wherein the measured position 48 is subtracted from an estimated position 50 of a target trajectory to generate a position error signal (PES) 52 . A feedback compensator 54 processes the PES 52 to generate a feedback control signal 56 . During the settle mode of a seek operation, a feed-forward compensator 58 generates feed-forward compensation 60 and a phase shift compensator 62 adjusts the feed-forward compensation 60 to compensate for a phase shift between the measured position 48 of the head and the corresponding estimated position 50 of the head. The feedback control signal 56 is adjusted by adding the phase adjusted feed-forward compensation 64 to generate a control signal 66 applied to a head actuator 68 (such as the VCM 44 of FIG. 2A ).

In one embodiment, the phase shift between the measured servo state and the corresponding estimated servo state is predetermined based on a characteristic of the servo system. FIG. 3 shows an example of this embodiment wherein the closed loop servo system comprises one or more notch filters 70 (fixed and/or adaptive) that filter the control signal 66 in order to suppress resonances of the servo system. The notch filters 70 introduce a phase lag between the estimated position 50 of the head due to the feed-forward compensation 60 and the measured position of the head 48 . In one embodiment, the head actuator 68 may also exhibit a phase lag that is additive to the phase lag of the notch filters 70 . In order to compensate for this phase lag in the embodiment of FIG. 3 , the phase shift compensator 62 induces a phase lead into the feed-forward compensation 60 .

The phase lag induced by the notch filters 70 and/or the head actuator 68 may be estimated in any suitable manner. In one embodiment, the effect of the phase shift is measured at a specific frequency W c . For the fixed notch filter phase shift, since the notch coefficients are known in advance, the phase shift of each notch filter can be calculated for each head. Assume for head i the phase lag of all the fixed notch filters together has a phase lag θ (i,fix) at frequency W c . For each adaptive notch filter, there are several profiles one of which is chosen depending on the process calibration for each disk drive. For the adaptive notch phase shift for head i a phase lag θ (j,k) is calculated based on the calibrated adaptive notch filter profile j for each adaptive notch k, the sum of which corresponds to the total phase lag due to all adaptive notch filters for the head:

θ ( i , adaptive ) = ∑ j , k ⁢ θ j , k , i ⁢ ⁢ at ⁢ ⁢ frequency ⁢ ⁢ W c .

For the head actuator 68 phase lag, the phase shift difference among different heads can be measured over multiple drives and a phase shift value can be specified for each head i as θ (i,ac) at frequency W c , where there may be a significant phase shift difference among inner heads and outer heads of the head stack assembly. At the beginning of each seek, the phase lag at W c for each head i is calculated as:

θ (i) =θ (i,ac) +θ (i,fix) +θ (i,adaptive)

In one embodiment, a scaled phase shift value is computed according to:

τ i = θ ( i ) W c * 2 Q lag

where τ i represents a phase lag if positive and phase lead if negative. At current time t, a new phase adjusted feed-forward compensation ACT_FF 64 is computed according to:

FIG. 4 shows an alternative embodiment of the present invention wherein a phase shift compensator 72 induces a phase lag into the estimated position 50 of the head to compensate for the phase shift after switching to the settle servo loop. In effect, the phase shift compensator 72 shifts the target trajectory so that the estimated position 50 of the head is delayed to generate a phase shifted version 74 of the estimated position 50 . Accordingly, in this embodiment the feedback control signal 56 is generated in response to the phase lag instead of adjusting the feed-forward compensation 60 with a phase lead as in FIG. 3 .

FIG. 5A shows an embodiment of the present invention wherein the feed-forward compensator 58 generates an estimated state based on a model 78 of the closed loop servo system. In the embodiment of FIG. 5A , the estimated state generated by the feed-forward compensator 58 comprises an estimated position 80 of the head relative to a target track, and an estimated velocity 82 of the head. The estimated states are multiplied by respective scalars −k1 and −k2, and the results summed to generate the feed-forward compensation 84 . The estimated position 80 and the estimated velocity 82 of the head are initialized based on a measured state of the servo system at the beginning of the settle operation (when the head nears the target track during a seek operation). In one embodiment, the estimated states in the feed-forward compensator 58 are initialized based on a measured position 48 of the head (derived from reading the servo sectors 38 0 - 38 N ) and a measured velocity of the head (e.g., based on the derivative of the measured position 48 ). The scalars k1 and k2 are also initialized based on the measured state of the servo system in order to initialize a state trajectory of the feed-forward compensator 58 . After initializing the estimated states and the scalars, the model 78 of the closed loop servo system will generate the feed-forward compensation 84 so that the estimated states (estimated position 80 and estimated velocity 82 ) follow the state trajectory, thereby causing the actual states of the servo system to substantially follow the state trajectory. In the embodiment of FIG. 5A , a phase shift compensator 86 adjusts the feed-forward compensation 84 to compensate for the phase shift between the measured position 48 of the head and the estimated position 80 as described above.

In the embodiment of FIG. 5A , a virtual position error signal (V_PES) is generated based on the difference between the estimated position 80 and the measured position 48 so that the feedback compensator 54 compensates for the error between the estimated and measured states (i.e., the target trajectory in FIG. 2B is generated by the model 78 in FIG. 5A as the estimated position 80 ).

In the embodiment of FIG. 5A , the model 78 of the closed loop servo system comprises a second order system (a double integrator) that can be normalized and represented by the following continuous time state space equation:

The above continuous time state space equation can be transformed into a discrete-time representation:

Assuming the feedback for the above internal model is:

u ⁡ ( k ) = - [ k ⁢ ⁢ 1 k ⁢ ⁢ 2 ] ⁡ [ x 1 ⁡ ( k - 1 ) x 2 ⁡ ( k - 1 ) ]

then the model can be rewritten as:

[ x 1 ⁡ ( k ) x 2 ⁡ ( k ) ] = [ 1 - 0.5 · k ⁢ ⁢ 1 1 - 0.5 · k ⁢ ⁢ 2 - k ⁢ ⁢ 1 1 - k ⁢ ⁢ 2 ] ⁡ [ x 1 ⁡ ( k - 1 ) x 2 ⁡ ( k - 1 ) ]

If the state trajectory of the feed-forward compensator 58 is a straight line approaching in the phase plane (VEL vs POS) as shown in FIG. 5B , then the above equation simplifies to k1=0, k2=R=VEL(0)/POS(0), where POS(0) represents the measured position 48 and VEL(0) represents the measured velocity of the head at the beginning of the settle operation (when the head nears the target track during a seek operation) and the scalar k2 represents the slope (R) of the straight line. As shown in FIG. 5B , the slope of the linear trajectory will vary based on the measured position and velocity at the beginning of the settle operation.

Referring again to the embodiment shown in FIG. 5A , at the beginning of the settle operation a control signal 90 closes switches 92 A and 92 B to enable generation of the V_PES and to enable the feed-forward compensation 88 . The control signal 90 also signals block 94 to initialize the estimated states (POS_FF and VEL_FF) based on the measured states (e.g., the measured position 48 of the head) as well as initialize the scalars k1 and k2 (such as initializing k2 to establish the slope of the above-described linear trajectory).

In the embodiment of FIG. 5A , the feed-forward compensator 58 generates estimated states for the position 80 and velocity 82 of the head. In other embodiments, the feed-forward compensator 58 may generate other states, or additional states such as acceleration or jerk. The feed-forward compensator 58 may also comprise a more sophisticated model 78 and/or a more sophisticated state trajectory (other than the above-described linear trajectory).

Referring again to the above discrete-time representation of the servo system, when the state trajectory of the feed-forward compensator 58 is a straight line in the phase plane (VEL vs POS) as shown in FIG. 5B , and when the phase shift is taken into account, the above equation may be represented as:

[ x 1 ⁡ ( k ) x 2 ⁡ ( k ) ] = [ 1 1 - p * R _ 0 1 - R _ ] ⁡ [ x 1 ⁡ ( k - 1 ) x 2 ⁡ ( k - 1 ) ] u ( k )=− R x 2 ( k− 1)

where p represents the phase shift, and the phase shift p is used to modify the slope R to generated a modified slope R that accounts for the phase shift p. The approaching slope in terms of VEL and PES at each servo wedge is:

R = V ⁡ ( k - 1 ) PES ⁡ ( k - 1 ) .

The approaching slope in terms of the average VEL is:

R _ = V _ ⁡ ( k - 1 ) PES ⁡ ( k - 1 ) .

The VEL at the next servo wedge can be calculated as:

V ( k )= V ( k− 1)− R·V ( k− 1)

lim k→∞ V ( k )= V ( k− 1)− R ·V ( k− 1).

The average VEL can be calculated as:

V ( k )=(1 −p ) V ( k− 1)+ pV ( k ).

From the above equations:

V ( k )= V ( k− 1)− p R V ( k− 1)

and

R _ ⁡ ( k ) = V ⁡ ( k - 1 ) - p ⁢ R _ ⁢ V ⁡ ( k - 1 ) PES ⁡ ( k - 1 ) .

Therefore the modified slope R of the state trajectory can be calculated as:

In one embodiment, the phase shift p between the measured position 48 of the head and the estimated position 80 may be computed dynamically based on the measured position and the measured velocity prior to switching to the settle servo loop. Consider that:

PES( N )(1 + R ) N =PES(0)

then from the above equations:

PES ⁡ ( N ) ⁢ ( 1 + V ⁡ ( k - 1 ) PES ⁡ ( k - 1 ) + pV ⁡ ( k - 1 ) ) N = PES ⁡ ( 0 )

where PES(0) represents the PES prior to switching to the settle servo loop and PES(N) represents the PES at the end of the settle operation. Assuming that the modified slope R is fixed prior to switching to the settle servo loop, then:

In one embodiment, the phase shift p is modified to achieve a desired settle performance within a predetermined N number of servo sectors:

p = 1 PES ⁡ ( 0 ) PES ⁡ ( N ) N - 1 - PES ⁡ ( 0 ) V ⁡ ( 0 ) 1 PES ⁡ ( 0 ) _ PES ⁡ ( N ) _ N - 1 - PES ⁡ ( 0 ) _ V ⁡ ( 0 ) _ ⁢ p _

which can be simplified by substituting:

PES ⁡ ( 0 ) PES ⁡ ( N ) N ≈ PES ⁡ ( 0 ) _ PES ⁡ ( N ) _ N

such that:

p ≈ delt - PES ⁡ ( 0 ) V ⁡ ( 0 ) delt - PES ⁡ ( 0 ) _ V ⁡ ( 0 ) _ ⁢ p _

where:

p is a nominal phase shift;

PES(0) represents a measured position of the head prior to switching to the settle servo loop;

V(0) represents a measured velocity of the head prior to switching to the settle servo loop;

PES(0) represents a nominal position of the head prior to switching to the settle servo loop;

V(0) represents a nominal velocity of the head prior to switching to the settle servo loop;

N represents a nominal time (number of servo sectors) to settle the head on the target track; and

PES(N) represents a nominal position of the head at time N.

In one embodiment, after estimating the phase shift p based on the measured position and the measured velocity prior to switching to the settle servo loop as described above, the modified slope R of the state trajectory may be calculated using the above equation in response to the estimated phase shift p.

In another embodiment of the present invention, the phase shift p may be computed based on the measured servo states after switching to the settle servo loop and while settling the head on the target track in response to the feed-forward compensation. That is, the phase shift p may be updated at each servo sector after switching to the settle servo loop. In one embodiment, the phase shift p may be estimated according to:

p ( k )= T ·[Vel_est( k− 1)−Delta_PES( k− 1)]/[Vel_est( k− 1)−Vel_est( k )]

where:

p(k) represents the phase shift at time k; PES(k) represents the measured state comprising a measured position of the head at time k; Delta_PES(k)=PES(k)−PES(k−1); and Vel_est(k) represents an estimated velocity of the head at time k.

In the embodiment of FIG. 5A , the phase shift compensator 86 adjusts the feed-forward compensation 84 based on a phase shift p computed dynamically as a function of the measured servo system states as described above. In another embodiment, the feed-forward compensation 84 in FIG. 5A may be further adjusted based on a predetermined phase shift, such as a predetermined phase shift caused by the notch filter(s) or the phase shift caused by the head actuator 68 . In this embodiment, the phase shift p computed dynamically may compensate for an error in estimating the predetermined phase shift as well as other sources of phase shift.

The phase shift compensator 86 in the embodiment of FIG. 5A adjusts the feed-forward compensation 84 by inducing a phase lead (i.e., the phase shift p is positive). In another embodiment shown in FIG. 6 , a phase shift compensator 96 may adjust the estimated position 80 of the head by inducing a phase lag (i.e., the phase shift p is negative). In both embodiments, compensating for the phase shift between the measured position of the head 48 and the corresponding estimated position 80 of the head improves performance of the feed-forward compensator 58 .

Any suitable control circuitry may be employed to implement the flow diagrams in the embodiments of the present invention, such as any suitable integrated circuit or circuits. For example, the control circuitry may be implemented within a read channel integrated circuit, or in a component separate from the read channel, such as a disk controller, or certain operations described above may be performed by a read channel and others by a disk controller. In one embodiment, the read channel and disk controller are implemented as separate integrated circuits, and in an alternative embodiment they are fabricated into a single integrated circuit or system on a chip (SOC). In addition, the control circuitry may include a suitable preamp circuit implemented as a separate integrated circuit, integrated into the read channel or disk controller circuit, or integrated into a SOC.

In one embodiment, the control circuitry comprises a microprocessor executing instructions, the instructions being operable to cause the microprocessor to perform the flow diagrams described herein. The instructions may be stored in any computer-readable medium. In one embodiment, they may be stored on a non-volatile semiconductor memory external to the microprocessor, or integrated with the microprocessor in a SOC. In another embodiment, the instructions are stored on the disk and read into a volatile semiconductor memory when the disk drive is powered on. In yet another embodiment, the control circuitry comprises suitable logic circuitry, such as state machine circuitry.