Browse Prior Art Database

Iterative Feedforward Error Sum Method for ACRMS Reduction in Direct Access Storage Device Files

IP.com Disclosure Number: IPCOM000109131D
Original Publication Date: 1992-Jul-01
Included in the Prior Art Database: 2005-Mar-23
Document File: 3 page(s) / 172K

Publishing Venue

IBM

Related People

Erickson, KJ: AUTHOR [+2]

Abstract

A method is described that can be used to improve the tracking performance of high track density direct access storage device (DASD) files by significantly reducing the repeatable runout components of the position error signal, which considerably reduces the track mis-registration (TMR) caused by these components. This method is based on an iterative feedforward process that results in a better compensation signal than that obtained from direct amplitude and phase measurements of the repeatable components. Two different implementations of this method are described.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 43% of the total text.

Iterative Feedforward Error Sum Method for ACRMS Reduction in Direct Access Storage Device Files

      A method is described that can be used to improve the tracking
performance of high track density direct access storage device (DASD)
files by significantly reducing the repeatable runout components of
the position error signal, which considerably reduces the track
mis-registration (TMR) caused by these components.  This method is
based on an iterative feedforward process that results in a better
compensation signal than that obtained from direct amplitude and
phase measurements of the repeatable components.  Two different
implementations of this method are described.

      A relatively straightforward, yet very accurate, method can be
used to determine the feedforward signal to apply in a DASD file
servo control system in order to significantly reduce the repeatable
components of TMR.  Because servo control systems exhibit gain and
phase characteristics that are dependent upon frequency (among other
conditions such as temperature, component tolerances, etc...), it is
difficult to directly determine how to compensate for these
characteristics such that the feedforward signal contains the proper
amplitude and phase attributes to provide for the best possible
repeatable component reduction.  The virtue of the method discussed
herein is that direct measurements and/or estimations of the gain and
phase characteristics of the closed-loop system are unnecessary as
they are automatically determined during the process of iteratively
developing the feedforward compensation signal.

      The figure diagrams a closed-loop servo system similar to that
of a traditional closed-loop system, with the addition of a
feedforward algorithm and ACRMS monitor, which are simply functions
added to the existing servo control microprocessor and typically
require no additional hardware.

      NOTE: ACRMS is an RMS (Root Mean Square) measurement of the
repeatable (AC) runout components in a DASD actuator servo system and
is also used as a generic term to refer to the repeatable runout
components in general.

      In this system, an additional summing node exists such that a
feedforward signal, F, can be added to the error signal, E, such that
the repeatable portion of the reference signal, R, is compensated
for, thereby no longer producing a repeatable component in the error
signal and hence providing better tracking.  The output of the
summing node is a compensated error signal, E bar, that is fed to the
compensator, which responds to the error by generating a corrective
output signal, U, which is then fed to the plant.

      The iterative method can be described as follows: First define
      E = ENR + ER
where
      ENR = Non-Repeatable Component of Error
      ER = Repeatable Component of Error
The goal is to drive ER = 0, such that E = ENR.

      This can be accomplished by determining a feedforward signal,
F,...