Head-media Separation Research Articles

Hard Disk Drive Reliability Modeling and Failure Prediction

A reliability model for the hard disk drive (HDD) is developed, focusing on head-disk separation as the primary independent variable. The model is structured to incorporate the theoretical effects of environmental factors, plus empirical dependence on the product operating mode. An experimental method based on magnetic spacing loss theory is used to characterize the head-media separation as a function of temperature, altitude, humidity, and HDD operating mode. A statistical model based on these empirical data is developed to predict HDD reliability for various operating conditions. The predictions of the model are verified experimentally through comparison with HDD product reliability test data.

Effect of Intermolecular Forces on the Static and Dynamic Performance of Air Bearing Sliders: Part I—Effect of Initial Excitations and Slider Form Factor on the Stability

The mechanical spacing between the slider and the disk has to be reduced to less than 5 nm in order to achieve an areal density of 1Tbit∕in2. Certain physical phenomena, such as those that can be caused by intermolecular and surface forces, which do not have a significant effect at higher flying heights, become more important at such low head-media separations. These forces are attractive for head-media separation as low as 0.5 nm, which causes a reduction in the mechanical spacing as compared to what would be the case without them. Single degree of freedom models have been used in the past to model these forces, and these models have predicted unstable flying in the sub-5-nm flying height range. Changes in the pitch and the roll angles were not accounted for in such models. A 3-DOF air bearing dynamic simulator model is used in this study to investigate the effect of the intermolecular forces on the static and dynamic performance of the air bearing sliders. It is seen that the intermolecular forces increase the level of flying height modulations at low flying heights, which in turn results in dynamic instability of the system similar to what has also been observed in experiments. The effect of initial vertical, pitch, and roll excitations on the static and dynamic flying characteristics of the slider in the presence of the intermolecular forces has also been investigated. A stiffness matrix is defined to characterize the stability in the vertical, pitch, and roll directions. The fly height diagrams are used to examine the multiple equilibriums that exist for low flying heights. Finally, a study was carried out to compare the performance of pico and femto designs based on the hysteresis observed during the touchdown-takeoff simulations.

Analysis and optimization of terabit magnetic recording systems limited by down-track and cross-track write field gradients

In this work, a simple head field is analyzed to suggest optimal perpendicular recording system configurations in the 1Tbit/in2 areal density range. The head field examined is that arising from a brick of magnetic material having a saturation induction, Bs, of 2.45T, perfectly magnetized in the perpendicular direction and perfectly imaged in a soft underlayer (SUL). An algorithm is presented to identify the maximum density achievable for a particular signal-to-noise ratio (SNR), under conditions of (i) constant medium thermal stability (KuV/kT=50); (ii) constant adjacent track erasure of 20% magnetization loss after 106 adjacent track writes; (iii) and a head medium separation of 5nm, and (iv) no separation between the medium and the SUL. The SNR is estimated from the number of grains across the read track width within the region where the transition is sharp enough to avoid percolation. Using this algorithm, we conclude that the optimum medium thickness is 15nm, and the optimum pole width is 70nm. A uniform background field of 150kA/m added to the pole field in a direction perpendicular to the medium offered a slight improvement (5%) in achievable density. Interestingly, the same analysis also suggests that tilting the medium anisotropy orientation does not improve areal density at fixed SNR, although the grain volume at the optimum recording condition on tilted medium is about 25% smaller. While the head field considered here is idealized, it captures the variation of down-track field gradient with cross-track position, which is frequently neglected in such analysis, but appears to be a first-order consideration.

A Novel Wear-In-Pad Approach to Minimizing Spacing at the Head/Disk Interface

Here we demonstrate a new approach to bring the disk drive recording head close to the disk media, while achieving tight control of the final head-media separation. The slider in this approach has a miniature Wear-In-Pad (WIP) at its trailing edge, encapsulating the read/write elements. As the slider flies over the disk, the slider overcoat and head recession are quickly worn off, resulting in the bottom of read and write elements clearing the disk roughness by only a few nanometers. We show that the Wear-In-Pad is a viable way to achieve the sub 7 nm head-media separations believed to be necessary for Tb/in/sup 2/ recording.

Media tribology design to improve flyability at low flying height

As the head media separation (HMS) continues to decrease with increasing areal density, maintaining good flyability performance is crucial to ensure the robustness of the head-disk-interface. Carbon wear and lubricant pickup on the slider surface are found to be the main contributors to flyability degradation. The effects of carbon overcoat and lubricant on load/unload flyability were investigated at the near-contact flying height by either reducing the environmental pressure or slowing down the disk rotating speed. Changing lubricant process and X-1P additive amount can affect lubricant pickup. In order to reduce the head wear, a thin layer of softer functional carbon layer was applied on the top of plasma-enhanced chemical vapor deposition carbon. However, carbon and lubricant modification has some limitations on improving flyability as HMS continues to drop. As an alternative approach to minimizing slider pickup at low flying height, landing the head in the laser textured contact-start-stop zone serves the function of cleaning the head and removing head smear.

Switching speed limitations in perpendicular magnetic recording media

A method to study switching speed limitations in high-frequency magnetic recording is presented. The characteristic parameter is the minimum irreversible switching time that is evaluated by numerical simulation of the writing of a dibit of two transitions using Landau–Lifshitz dynamics. The dibit is written on a perpendicular polycrystalline thin film medium using a single-pole head with a soft underlayer. We study the physical factors that determine the ultimate data rate and consider the problem of optimization of switching speed by changing the relative head-medium velocity, the magnetocrystalline anisotropy, the rise time of the head field, the head-medium separation, the exchange intergranular coupling, and the dispersion in the direction and magnitude of the local anisotropy.

Optimization of heads for perpendicular recording

The effects of gap length, head-underlayer separation, and return pole potential on the magnitude and direction of the field of a gapped head are investigated using an analytic two-dimensional model. Criteria derived from a micromagnetic model of a perpendicular recording medium are used to optimize the head geometry for various media parameters. A head with a gap length which is at most 2.5 times the head-medium separation can write on higher coercivity media than a single pole.

Interface tribology for 100 Gb/in 2

Critical head/disk interface requirements for 100 Gb/in 2 recording systems are examined with an emphasis on head design, head/disk protective overcoats and disk lubrication. A magnetic spacing of 10 nm that is necessary to achieve the areal density requires implementation of new technologies in each of these areas. To support the 100 Gb/in 2 areal density, novel air bearing surface (ABS) designs conceived to meet the low flying height requirement are presented. Numerical evaluation of various parameters necessary for optimum flying performance is discussed. These simulation studies provide guidance on the tolerance values and other aspects of the fabrication process that are needed to produce heads with low flying standard deviation. In an ultra-high density recording (100 Gb/in 2) system, the media property and the head–media separation distance become extremely important. To minimize the risk of head–disk interactions at low flying heights, super-smooth media with ultra-low glide heights will be needed. Reducing disk flutter at high rotational speeds will be a key issue affecting both the reliability of the interface and the electrical performance. Surface topography of such media will need to be on the order of atomic dimensions. Alternative substrate materials with higher stiffness and damping capacity are discussed as a solution to the above issues. The DLC coating on the air bearing and the transducer is another critical spacing parameter and alternate processing techniques of depositing ultra-thin carbon film are being explored. Lubricant thickness on the order of 1 nm will be applied to accommodate the decrease in the head–media spacing. Tribological measurements will be more crucial in ensuring a reliable head–disk interface for 100 Gb/in 2. These measurements are made even more difficult owing to the shrinking size of the sliders.

Wavelength, frequency and interface effects in magnetic recording system replay noise spectra

A method allowing the identification of wavelength, frequency and head-medium interface effects in the replay process in magnetic recording is described. It involves the measurement of replay bulk erased noise spectra at a number of replay head-medium speeds. As an example, a particulate medium is examined using a fixed head instrumentation tape recorder. A set of parameters is derived from the spectra obtained in order to characterize the medium noise process. Increased head-medium speeds have been shown to result in an increased head-medium separation parameter on replay.

Media noise improvement through head-disk spacing reduction

Experimental and theoretical results show that media noise decreases with reduced head-media separation. Media with the same magnetic layer but with Cr overlayers of varying thickness were fabricated as a means to adjust head-media separation. Noise power vs. recording density curves exhibit an initial linear region followed by a supralinear region indicative of appreciable bit interaction. The noise maximum for the lower flying cases occurs at a higher density. The slope of the noise curve is also lower for lower fly heights, indicating a lower noise level per transition. Simulations of transition pairs show a clearly defined zig-zag structure for large head-media separations. When flying near contact, the zig-zag amplitude is greatly diminished.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Head sensitivity function determined from recording measurements

The authors present a novel approach to measuring the response of a recording head used with a hard disk to form a recording system. The technique can be implemented on any test stand used for recording measurements using standard instrumentation with minor modifications to the write electronics. The resolution across the track width is about 0.5 mu m, and the head-media separation is controlled in the usual way by the design of the air-bearing surface and the velocity of the disk. Data are given for a thin-film head to illustrate the utility of the technique. >

Comparison of the magnetic recording performance of medium-scale ring, double-sided pole, metal-in-gap heads on high-coercivity longitudinal media

The magnetic recording performance of medium-scale ring, double-sided pole, and metal-in-gap (MIG) heads were investigated on standard longitudinal high-coercivity floppy, video tape, and isotropic floppy media using a stretched-surface multi-head magnetic recording tester. The isolated pulses for the ring and MIG heads exhibited sharp peaks near tbe edges of the gap for a head-medium separation of less than 0.1 um. Low density maximum output voltage was independent of the type of the write head. The largest output voltage was achieved from the MIG head. The isolated outputs on video media agreed with calculated values for the ring head efficiency set to 52% and the pole head efficiency set to 33% using the Karlqvist head field approximation.

Transition lengths in magnetic recording on particulate disks

Magnetic recording parameters including output vs. Write current, output vs. recording density and overwrite modulation vs. Write current have been measured on eight experimental disk samples of cobalt- doped gamma iron oxide using manganese-zinc ferrite heads. Transition length parameter plus head-medium separation distance values were calculated from output vs. recording density curves. Transition length was found to increase with increasing write current and with increasing record gap length for gap lengths ranging from 0.4 to 1.25 µm. Transition lengths were also found to depend on the recording medium parameters of remanent moment, coercivity, S* and thickness. Minimum values of transition length were found to occur at write current levels insufficient to obtain adequate overwrite modulation.

On the transition length in perpendicular recording

Using the analytical model previously described by the authors [1],[2], magnetization transitions and transition lengths are calculated as a function of relevant physical parameters. When a probe head is used to record on a medium with a permeable under-layer, the readback voltage and the transition length are determined mostly by the head-medium separation. The calculated results agree with experimental observations of the variation of high frequency readback voltage with medium thickness.

On the perpendicular recording process

An analytical model of the recording process is used to examine the factors that influence the recorded magnetization. If the ‘‘switching’’ occurs at the top of the medium, the magnetizaton is influenced mostly by the head-medium separation. If the ‘‘switching’’ occurs at the middle of the medium, the magnetization is influenced by the head-medium separation, the slope loop of the M-H at M=0, and the medium thickness. The magnetization is asymmetrical in both cases and the asymmetry translates into asymmetrical readback waveforms.

Digital magnetic recording theory

A brief review of digital magnetic recording theory is given with emphasis on the analysis of ferrite, inductive thin film, and magnetoresistive heads. The utility of several approaches is demonstrated by three calculations: (1) resolution and signal amplitude of a shielded magnetoresistive head, (2) readback properties of a single sheet of permalloy with adjacent current conductor, and (3) analysis of spurious pulses generated by outside corners of ferrite heads. A discussion of high density digital recording is given, and it is concluded that the magnetoresistive head offers the greatest possible areal density of information. This density is conservatively estimated to be 1.6×107flux reversals per cm2(108per in2) at a head-medium separation of 0.125μm. Experimental data in support of this conclusion are given.

Gap Loss Function and Determination of Certain Critical Parameters in Magnetic Data Recording Instruments and Storage Systems

After a brief review of Westmijze's semi-infinite model for the reading of recorded signals in magnetic media by gapped ring heads, a program for the computation of the gap loss function S is described. By introducing the higher order sine and cosine integrals, this function can be evaluated to any degree of accuracy over its useful range. A four decimal place table is compiled and included in this report. Two experiments significant to the validity of the theory and the practical evaluation of the data system performance were carried out with the help of the table compiled. By autocorrelating the system output, the null points in the frequency response curve of tenths of micron (micro-inch) gaps were located with confidence. The first null point determines the effective gap length accurately, from which the signal amplitude reduction due to gap interference at arbitrary wavelengths can be calculated. Further measurement of output signals at three suitably chosen frequencies enables one to determine the effective thickness of magnetization and head medium separation. The good agreement of the theoretical response curve based on these parameters with the directly measured data indicates the general validity of Westmijze's model. From a practical point of view, the knowledge of these parameters makes it possible to separate the signal amplitude losses during reading operation and suggest proper approaches to the improvement of system resolution.