Adaptive Feedforward Repetitive Run Out Tracking in Bit Patterned - - PowerPoint PPT Presentation

UC Berkeley

Adaptive Feedforward Repetitive Run‐Out Tracking in Bit Patterned Recording

Behrooz Shahsavari, Ehsan Keikha, Fu Zhang, Omid Bagherieh, Roberto Horowitz, CML Sponsors Meeting

• Conventional media: data is written on concentric circular tracks
• Bit patterned media: data should be written on tracks with

predetermined shapes, which are created by lithography on the disk.

Repeatable Runout in Bit Patterned Recording

Data tracks Bit-patterned media Conventional media Servo tracks

• The goal of this project is to control the voice coil motor (VCM) such

that the read/write head follows unknown repeatable runout (RRO).

Objective

Bit-patterned media

• RRO frequency spectrum is unknown
• RRO frequency spectrum can spread up

to very high frequencies; therefore, will be amplified by the servo controller

• System dynamics is changing from

drive to drive, and by temperature variation.

• RRO is changing in both circumferential

and radial direction

Issues

• Feedback controller, , attenuates the following noises

– NRRO: – Meas. Noise: – Windage:

Controller Architecture

FB

C

w

G

nrro

G

n

G

Head position Error signal

FB

C

VCM

G

w

G

nrro

G

n

G

rro

d

n

nrro

d

h

y

e

w

u

FB

u

• Feedback controller,

– NRRO: – Meas. Noise: – Windage:

Controller Architecture

FB

C

w

G

nrro

G

n

G

A

C

FB

C

VCM

G

w

G

nrro

G

n

G

rro

d

n

nrro

d

h

y

e

w

u

FB

u

A

u

Head position Error signal

• Adaptive controller, , is added

in a “Plug‐in” fashion to track

– RRO:

A

C

rro

d

• We aim to design an adaptive controller, , such that the error

signal, , is minimized. In other words, the head position is following

the RRO, .

Controller Architecture

A

C

e

h

y

rro

d

A

C

FB

C

VCM

G

w

G

nrro

G

n

G

rro

d

n

nrro

d

h

y

e

w

u

FB

u

A

u

Head position Error

• We replace the unknown RRO, , by another unknown periodic

sequence, , that is added to .

• We assume that the noises are attenuated by the feed‐back

controller ; therefore, can be ignored in feedforward control design.

Controller Architecture

Head position Error

A

C

FB

C

VCM

G

w

G

nrro

G

n

G

rro

d

n

nrro

d

h

y

e

w

u

FB

u

A

u

rro

d

rro

d

u

rro

d

FB

C

• We replace the unknown RRO, , by another unknown periodic

sequence, , that is added to .

• We assume that the noises are attenuated by the feed‐back

controller ; therefore, can be ignored in feedforward control design.

Controller Architecture

rro

d

u

rro

d

FB

C

e

u

FB

u

A

u

rro

d

VCM

G

FB

C

Error

e

A

u

rro

d

R

1

VCM FB VCM

R G C G  

, T rro k k

d   

uA  k

Tk

฀

known regressor unknown parameters estimated parameters

1. The PES is converted to an auxiliary error variable, in order to use in the parameter adaptation algorithm. 2. A novel adaptive step size is proposed to increase the convergence rate and boost the steady state performance.

Adaptive Control Algorithm

Estimate the unknown parameters adaptively

We propose a new adaptive control algorithm based

• n the following two key ideas:

, T rro k k

d   

uA  k

Tk

฀

measurable known known regressor unknown parameters estimated parameters

Adaptive Control Algorithm

We want to estimate the unknown parameters adaptively

e

A

u

rro

d



R

R

 

1

ˆ ˆ

k k k k k

R e    

 



• Parameters update
• Auxiliary error

Adaptive Step Size – Key ideas

• The step size in adaptation is a function of “Auxiliary Error”

convergence.

• As we get closer to the real parameters, the step size becomes

smaller.

• Finally, the algorithm stops when a certain performance is

achieved.

• The adaptation starts again whenever the error becomes large

(e.g. we move to another track with a different RRO).

 

1

ˆ ˆ

k k k k k

R e    

 



 

 

 

2 1 1

1 min( , ) 0 and 0 or 0 otherwise

k h k i i k h h d k k ub k allowed k k k k

V e h V V         

   

           



Mean squared error approximation Desired performance

Adaptive Step Size

 

 

 

2 1 1

: scalar gain : desired PES variance 1 : aprox. variance of aux. error min( , ) 0 and 0 or

h d d k k k h h k i k i k h ub k allowed k k k k

V V V V e V h          

   

      



• therwise

    

: maximum step size to guarantee convergence : design parameter defining the dead-zone width

allowed ub

 

k



ub



+ + + + + + Positive step size Zero step size +

A

C

FB

C

VCM

G

w

G

nrro

G

n

G

rro

d

n

nrro

d

h

y

e

w

u

FB

u

A

u

• ,

, and are modeled based on the real PES, and and are modeled based on real frequency responses, all from a drive provided by HGST, a Western Digital company.

• Artificial RRO, that contains frequencies up to the 90th multiple of

fundamental (spindle) frequency, is added to the real PES (from HGST and Seagate).

• RRO harmonics are divided into low, mid, and high frequency

regions and their adaptation is scheduled in time.

Simulation Results

VCM

G

FB

C

w

G

nrro

G

n

G

rro

d

• Tracking Performance

Simulation Results

Simulation Results

Nyquist Freq.

Simulation Results – Adaptive Step size

Drive mech Card R/W channel VCM driver X

Two jump wire for read-back signal Soldering on card

PES demodulation electronics DSP (Servo controller) Current amplifier Additional hardware

Read-back signal Track ID and PES VCM control signal VCM Current

Experimental setup

• Power up
• Initialization
• Reset
• Seek to ID/OD/MD
• Digital PES available
• External controller signal can

be injected

HDD Toolkit

• A real time embedded system has been developed on the DSP

hardware using CCS.

• This system includes: initialization of system’s interrupt, memory

management and peripherals IO ( timer, SPI module, PLL module etc.)

• The system successfully reads the digital PES from the HDD and

sends a control signal through DAC.

• The interaction between the HDD, DSP system and DAC has been
• tested. The RRO following controller, which only relies on the PES

signal, is ready to be implemented.

• Code has been optimized significantly to save computation time.

Controller implementation

• Develop adaptive algorithms for 2D RRO

variation.

• Adaptive compensation for mismatch

(temperature and manufacturing)

• Implement and evaluate the algorithms

Future Work

e

A

u

rro

d



R

R