Track Misregistration Research Articles

Modeling of narrow track thin film write head fields

A thin film inductive write head suitable for 10 Gbit/in/sup 2/ recording density is simulated using boundary element modeling. The material was assumed to have the non-linear B-H properties of FeAlN-SiO/sub 2/ multilayers-an initial permeability of 3400 and saturation flux density of 1.9 Tesla. The simulations show that with a trackpitch of 1 /spl mu/m and a gap of 0.18 /spl mu/m, it is necessary to use trimmed poles and a trackwidth of 0.6 /spl mu/m to achieve a written trackwidth of 0.8 /spl mu/m and allow for an erased-band of 0.05 /spl mu/m at each track edge. This allows 0.1 /spl mu/m for track misregistration, which is within tolerance achieved by current optical servos. If focused ion beam etching is used to trim the poles from the air bearing surface (ABS), to prevent fields from the etched shoulders from disturbing data in adjacent tracks, the shoulders should be etched 0.5 to 1.0 /spl mu/m from the ABS. >

Array heads for estimating track misregistration

The authors examine the use of array heads for estimating the data and for providing servo information. A mathematical representation for array heads is developed, and a formula for the maximum likelihood estimate of the track misregistration is found. The case where the offtrack perturbation operator is dependent on the offtrack error is considered. This case requires a joint minimization of two branches of quadratic forms. Simulation results for a specific head configuration are given which validate the estimate and agree with the theoretical convergence rates. The estimate converges quickly enough for practical purposes in a real storage system. >

Recording asymmetries at large skew angles

The effect of head skew on the read, write, and erase properties of thin-film inductive recording heads is explored. Using novel track profiling methods, combinations of large skew angles and small recorded wavelengths were found to cause offtrack displacement of written transitions and asymmetric erase bands. The maximum offset measured was 0.3 mu m at 35 degrees skew and a 1.7 mu m wavelength. Increased side reading in the opposite direction to the write process offset was also observed. These asymmetries are expected to increase the likelihood of track misregistration errors in a drive environment.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Exponential series statistical modeling of track misregistration in magnetic storage channels

The authors apply the analytic method of exponential series expansion to empirical frequency distributions obtained from measurements of track misregistration (TMR) phenomena in magnetic storage head-disk-assembly (HDA) devices. Analytical expressions that represent the deviation of the empirical TMR frequency distributions from the Gaussian distribution are computed. The latter becomes the first term in a series of orthogonal functions with coefficients determined by the moments of the empirical distribution. This allows for adequately matching the tails of the empirical distributions where the Gaussian fit is poor and where HDA performance predictions are most interesting. The authors describe the computation of the exponential series coefficients for TMR empirical frequency distributions as well as the algebraic composition of these coefficients used to obtain models for the combined effect of several TMR phenomena.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Track density constraints in the application of MR head technology

A disk drive track density model has been developed for dual-element recording heads using a magnetoresistive (MR) reader and an inductive writer. In addition to the usual track misregistration issues, it is shown that the head fabrication tolerances and the spacing between the read and write elements limit track density with a rotary actuator due to skew misalignment. A systems solution of micropositioning that minimizes such problems is proposed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

TMR and squeeze at gigabit areal densities

The track misregistration (TMR) required to achieve gigabit areal densities is calculated for a set of components based on the average bit error rate for a file. The average error rate is a statistical average, based on the TMR of the file, of the on-track and off-track error rate. The on-track and off-track error rates are dependent on squeeze from adjacent tracks. In this analysis, the average error rate is calculated using measured error rate profiles of the center track. For a file average soft error rate of 1*10/sup -6/, the results show, for a particular head/disk combination designed for gigabit recording and operating at a linear density of 158 kbpi, that a track pitch of 4.0 mu m (3.35 ktpi) can be achieved with a TMR of 0.635 mu m. Obtaining this TMR value in a file is one of the challenges to recording of 1 Gb/in/sup 2/ in magnetic storage products. >

Analysis of viscous dissipation in disk storage systems and similar flow configurations

Flows in rotating magnetic disk storage systems and similar devices are prone to instabilities and the viscous generation of heat. Serious degradation in the performance of disk storage systems occurs when differential thermal expansion, due to the maldistribution of heat, and flow-induced vibrations result in track misregistration by the magnetic read/write heads. Submicron flying heights and micron track widths exacerbate such alignment difficulties. High rotation speeds, also characterizing the emerging disk storage technology, impose significant torque requirements. Experimental measurements of torque obtained by Daily and Nece [J. Basic Eng. Trans ASME 82, 217 (1960)] for a single disk, and by Hudson and Eibeck [J. Fluids Eng. 113, 648 (1991)] for a disk stack with and without obstructions, both in cylindrical enclosures, are found in this study to be correctly correlated by the analytically derived equation, TNo≡Ttotρ/Nμ2R2 = (2C1 + C2H/R2) × [Re/(1 − tL/HR2)]3/2+m, where TNo is a newly defined dimensionsless torque number, Re is the Reynolds number of the interdisk flow; H, R2, t, and L are geometrical length scales; and, C1, C2, and m are experimentally determined constants of order unity. Regressions to the experimental data of these authors yield values for C1, C2, and m in this equation which offer a combination of accuracy and universality not previously available. Moderate amounts of imposed radially directed flows in the spaces between unobstructed pairs of corotating disks are shown, via numerical calculation, to affect the interdisk flow fields quite notably. Analysis of results obtained solving the Navier–Stokes equations, assuming unsteady, axisymmetric, streamlined flow at low speeds of rotation shows that sucking air radially inward in the interdisk space, rather than blowing it radially outwards, substantially reduces the torque required to rotate the disks relative to the unventilated condition. Notwithstanding, depending on how the sucking condition is implemented, in obstructed geometries the reduction in torque may come at the expense of flow instabilities that could affect magnetic head read/write performance, a problem requiring close attention in the application of present findings to the improved design of disk storage systems.

A technique for the measurement of track misregistration in disk files

The authors describe a noninvasive technique for measuring track misregistration (TMR) in disk files using variations in read signal amplitude as an indicator of relative motion between the read head and the written track. The readback signal amplitude from a magnetic recording head can be used as a sensitive linear position transducer for measuring dynamic head-to-track misregistration in disk files. Subtracking the repeatable part of the signal amplitude eliminates virtually all non-TMR related modulation, leaving only small contributions from electronics noise and nonrepeatable fly height fluctuations. These error terms can be measured in isolation by locating the head directly on-track, where sensitivity to TMR vanishes. This technique can provide detailed information about the electromechanical performance of completely assembled and installed disk files. >

Digital servo control of a novel disk actuator

A novel disk actuator transported on parallel flexures is considered for digital servo control. The control scheme derives the state estimates from a position error signal which resembles a sawtooth wave form. The estimator is based on a second-order actuator model with two augmented states for both track-following and track-access modes. The structure of the track-following controller is of a proportional-integral-derivative type. The estimator's residue time history shows that the algorithms are sufficiently robust to accommodate the model uncertainty, measurement noise, and force disturbance that are associated with a disk drive mechanism. The performance characteristics of the actuator are presented in terms of transfer functions, a step input response, a track misregistration probability distribution function, and 1/3 stroke seek dynamics.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Demonstration of 500 megabits per square inch with digital magnetic recording

The authors describe the testing and modeling methods undertaken to demonstrate and verify an areal density of 500 Mb/in/sup 2/ with a combined inductive, thin-film write head and a magnetoresistive read head on an advanced, state-of-the-art thin-film disk. The linear density was 100000 bit/in (3937 bit/mm), the track density was 5000 tracks/in (196.9 tracks/mm), and the flying height was 2.5 mu in (64 nm). The components were characterized in great detail, including write, read, erased-band, and read-profile skirt widths. A combination of off-track measurements and the track density error rate model showed that the target density was obtained for a soft error rate of 10/sup -6/ and a TMRw/r (write-to-read track misregistration) of 30.8 mu in (782 nm).< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Design of a disk-file reference-velocity generator

The IBM 3370/3375 disk-file access servomechanism uses a nonlinear transconductance function to construct a velocity-reference generator that produces a smooth continuous phase-plane trajectory. Absence of trajectory discontinuities provides the potential for a disk-file access servomechanism to have faster access times and to have less track misregistration upon arrival at the target track center. An integrated circuit implementation of the reference-velocity generator is described, and a solution is provided for the generator's unique loop stability and bandwidth requirements.

Target and Equalizer Design for High-Density Bit-Patterned Media Recording

In bit-patterned media recording (BPMR) channels, the inter-track interference (ITI) is extremely severe at ultra high areal densities, which significantly degrades the system performance. The partial-response maximum-likelihood (PRML) technique that uses an one-dimensional (1D) partial response target might not be able to cope with this severe ITI, especially in the presence of media noise and track mis-registration (TMR). This paper describes the target and equalizer design for highdensity BPMR channels. Specifically, we proposes a two-dimensional (2D) cross-track asymmetric target, based on a minimum mean-squared error (MMSE) approach, to combat media noise and TMR. Results indicate that the proposed 2D target performs better than the previously proposed 2D targets, especially when media noise and TMR is severe.