Abstracts from MMM-Intermag 96 held last November in Atlanta.
Note that the information in these abstracts may be dated already considering the following:
In the first quarter of 1997, Ampex obtained a significant further increase in capacity with inductive heads by utilizing a new generation of disk drive channel that is becoming widely used in the disk drive industry.
This sounds like to me that Ampex has been able to squeeze 30-40% more capacity out of KM+TFI using channel electronics (PRML, preamplifier) optimized for KM.
Journal of Applied Physics -- April 15, 1997
Volume 81, Issue 8, pp. 4693-4695
Magnetic recording performance of keepered media
T. M. Coughlin
Ampex Corporation, Redwood City, California 94063
Y. S. Tang and E. M. T. Velu
Western Digital Corporation, Santa Clara, California 95054
B. Lairson
Rice University, Houston, Texas 77005
Using low flying and proximity inductive heads, keepered media show
improved on and off ttrack performance leading us to conclude that a
greater than 20% areal density improvement is possible with a keeper
layer over the magnetic storage layer. For Sendust keeper layers there is an optimal range of thickness and an optimal bias point for best performance. There are both amplitude and timing asymmetries that are functions of the read-back bias. For a peak detect channel the best performance corresponds to the minimum timing asymmetry although this is not the point where the pulses are narrowest. Keepered media may have an advantage in total jitter and partial erasure. NLTS is almost identical for keepered versus unkeepered media. (Copyright) 1997 American Institute of Physics.
=============================================
Journal of Applied Physics -- April 15, 1997
Volume 81, Issue 8, pp. 4843-4845
High density recording with keepered media and planar heads
T. M. Coughlin, J. R. Hoinville, B. R. Gooch, W. G. Haines,
and A. Payne Maxtor Corporation, Longmont, Colorado 80501
Modeling and experimental results are presented for keepered longitudinal recording media and planar (undershoot-reduced) thin film recording heads with low flying heights for areal densities (greater-than)1 Gb/in.[sup 2]. The keeper layer is magnetically coupled to the medium magnetic transitions, reducing the transition demagnetization and narrowing the transition length by about 10% in the media after recording. The reproduced bias field and the
transition fields combine in the keeper to produce a partially saturated
region, thereby modifying the fields from the medium transitions at the head during playback. We present experimental data on the write and read process for keepered media. Boundary element model results are presented which explain the amplitude gain and pulse asymmetries observed experimentally. Use of a keepered medium allows areal density improvements (greater-than)20% through higher bits per inch. (Copyright) 1997 American Institute of Physics.
=============================================
Journal of Applied Physics -- April 15, 1997
Volume 81, Issue 8, pp. 3934-3936
Effects of Cr and magnetic bias on read/write and noise characteristics of
CoCrTa/Cr longitudinal thin-film media
Brij B. Lal and Michael A. Russak
HMT Technology, Fremont, California
The effects of Cr and magnetic bias on read/write and noise characteristics of CoCrTa/Cr longitudinal thin-film media were investigated. Samples of magnetic thin films were prepared by using dc magnetron sputtering at four different substrate-bias conditions: no bias (N), Cr bias only (C), magnetic bias only (M), and both Cr and magnetic bias (CM). Magnetic properties, parametrics, media noise, signal-to-noise ratio (SNR), nonlinear transition shift (NLTS), crystallography, and microstructures were studied. Comparison
of read/write characteristics measured with a magnetoresistive head at 72.4 kfci recording density showed improvement in resolution and pulse width for the media sputtered with M and CM bias. When the magnetic layers were sputtered with no bias, it was found that the Cr bias had almost no effect on media noise or SNR. However, the application of magnetic bias individually and substrate bias during both Cr and magnetic deposition reduced media noise about 20% and 30% and increased SNR about 5 and 7 dB, respectively, at 136.4
kfci recording density. NLTS was studied as a function of linear recording density for all these four samples. It was observed that the M and CM bias-sputtered media produce lower NLTS which correlates very well with the corresponding media noise. Co(11.0) and Co(10.0) peaks of magnetic films sputtered with M and CM bias showed broader peak width which produces smaller grain size of magnetic film for reduced transition noise. Transmission electron microscope images did not exhibit significant difference in
microstructure between the films prepared by sputtering with or without applying substrate bias. (Copyright) 1997 American Institute of Physics.
=============================================
Journal of Applied Physics -- April 15, 1997
Volume 81, Issue 8, pp. 4683-4685
Four discrete magnetic states of keepered media
W. B. Reed, U. Cohen, D. R. Hollars, and R. B. Zubeck
Velocidata, Incorporated, 3160 De La Cruz Boulevard, No. 106,
Santa Clara, California 95054
Keepered recording media includes a high permeability ( mu ) keeper layer overlying a conventional longitudinal thin-film magnetic layer. Vibrating sample magnetometer (VSM) hysteresis loops show four discrete magnetic states in the superposition of two exchange decoupled loops: one is the high coercivity storage layer and the other is the low coercivity keeper. The keeper layer provides an adjustable reluctance shunt across the head gap and a closure between recorded transitions. Keepered media output power
is attenuated more than 20 dB until a bias current is applied to the
reproduce head windings. Head bias current and recorded wavelength, lambda , have a direct linear effect on both head output level and mirror-symmetric peak pair offset within a defined variable reluctance zone. Head output level within the variable reluctance zone is 1 dB per mA turn of bias and normalized peak pair offset is (plus-minus)1% or (plus-minus)3(degree) per (plus-minus)mA turn of bias, with phase reversal at zero current. Variable reluctance amplitude and phase functions are linear until keeper layer
saturation at (plus-minus)17 mA turns bias. A method of encoding binary data into the four magnetic states is described. Eight user bits are coded into five flux changes and recorded at 90(degree) increments. Reproduce circuitry uses a 2F phase locked loop keeper head bias which decodes the four magnetic states into four voltage levels at 90(degree) spacing. The described (1,5) four level code requires one less flux change per 8 user bits than (1,7) code, and is stable only within the variable reluctance zone of keepered media. (Copyright) 1997 American Institute of Physics. |
