Data processing: measuring – calibrating – or testing – Calibration or correction system – Sensor or transducer
Reexamination Certificate
2002-04-22
2004-01-27
Bui, Bryan (Department: 2863)
Data processing: measuring, calibrating, or testing
Calibration or correction system
Sensor or transducer
C702S057000, C451S005000, C451S010000
Reexamination Certificate
active
06684171
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
The present invention relates generally to magneto resistive elements and in particular to stripe height calibration of magneto resistive elements. Still more particularly, the present invention relates to a method and system for performing accurate, result-directed/predictive stripe height versus resistance calibration of magneto resistive sensors during lapping operation.
2. Description of the Related Art
Many direct access storage device manufacturers employ thin film magnetic recording heads. Conventional thin film read/write heads in data storage systems generally include an inductive write head in combination with either an inductive or magneto resistive (MR) read head. One type of MR/inductive head includes an inductive write head formed adjacent to a MR read head. In manufacturing such heads, rows of magnetic recording transducers are deposited simultaneously on wafer substrates using semiconductor type process methods. Subsequent to these depositions, the wafers are fabricated into rows of single element heads called slider rows. When separated from the slider rows, each slider contains magnetic read/write components and an air-bearing surface configured to aerodynamically “fly” over the surface of a spinning magnetic disk medium. The rows are separated by kerfs that facilitate subsequent slicing into individual sliders.
Commonly assigned U.S. Pat. No. 5,531,017 describes the process by which a wafer consisting of multiple slider rows is divided into quads of 29 slider rows prior to completing the lapping process. A number of such rows of sliders are deposited together onto a single semiconductor-type wafer, which is then cut into pieces commonly termed “wafer quadrants” (or just “quadrants”). A wafer quadrant is then bonded onto an extender tool (also sometimes known as a row tool, transfer tool, or support bar) and the foremost slider row is lapped as a unit on an abrasive surface, such as a plate coated with an appropriate slurry mix. The slider row is then cut from the wafer quadrant, so that lapping of a new foremost slider row may commence. The sliced off row of sliders is ready for additional manufacturing steps, dicing into individual sliders, and then the final steps which ultimately produce working disk drive heads.
As a further enhancement to this process, commonly assigned U.S. Pat. No. 6,174,218 describes the manner in which the quads are placed on an extender tool that is bendable so that the slider rows may be straightened out while the lapping operation is being completed. This process of bending the quad while lapping is also referred to as a bow compensated lapping (BCL) process. Extender tools provide a mechanism for holding the row of sliders while lapping or grinding operations are performed to produce an air bearing surface. Typically the slider rows distort from a co-linear line according to the internal stress of the wafer material and the surface stresses developed when reducing the wafers to slider rows. Further distortion of the rows of sliders from a co-linear line can occur as a result of the tool bonding operation. The combined stress distortion and bonding distortion of slider rows results in a total distortion or curvature condition called row bow.
Row bow may cause a row of sliders to be non-uniformly lapped during the lapping process. As such, this row bow condition can detrimentally affect critical head performance parameters, such as stripe height in MR heads, and throat height in inductive heads. To achieve optimum performance of MR/inductive heads, both the stripe height and throat height must be tightly controlled.
In order to control the amount of lapping performed on a slider row and to accurately determine the final MR element height (at the conclusion of lapping), the resistance must be known. Thus, the lapping process is controlled by the measured resistance of the MR elements in a slider row. The measured resistances are used for controlling the degree of lapping for each of the MR elements in a slider row to compensate for row bow. The electrical resistance is related to the desired MR element height (also referred to as stripe height), and the lapping process is terminated when the desired MR element height is reached.
FIGS. 1 and 2
illustrate two current configurations of lapping control systems, which both utilize resistance measurements to control the lapping process. In
FIG. 1
, a dual element, wire bonded electrical lapping guide (ELG)
103
(with both long or short elements) is placed in each kerf between MR elements
101
in a slider row. The MR elements are wire bonded to electrical contacts so that the resistance can be measured. This configuration is primarily utilized with wafers having a density of
36
slider rows and relatively large kerfs.
With the introduction of higher density wafer designs (e.g., the 44 slider row per wafer designed by International Business Machines), the increased row density resulted in narrower kerfs and restricted the placement of the dual element ELG studs in the kerfs. The dual element ELGs were therefore replaced with alternating long and short ELGs placed in adjacent kerfs. Thus, as shown in
FIG. 2
, the long ELGs
204
and short ELGs
203
were placed within the kerfs of MR elements
201
and utilized in the calibration process.
Further development in calibration systems led to the introduction of row level kiss lap (or flatness control lapping), which made it necessary to utilize element predicted stripe height for process control. However, at this juncture, it was discovered that due to lead current crowding and other physical characteristics in current MR devices, simple linear calibration methods no longer produced valid and/or accurate results.
In response, a higher order method of calibration called (abc) (i.e., calibration in which the constants of a quadratic equation are first determined) was introduced, which utilizes wafer resistance data (from MR elements) and resistance and stripe height data after a first BCL operation. One problem with this technique is that it is greatly compromised by the lack of MR resistance sensitivity to stripe height at the wafer level.
Another problem is that the technique mixes data from unlike structures. Thus, wafer element data utilized has unlapped and undisturbed edges as illustrated in
FIG. 3B
, while the same element measured after lapping (shown in
FIG. 3C
) has a lower stripe edge that provides completely different data from the wafer data structures of FIG.
3
B. This difference is depicted by the graph on FIG.
3
A. Thus, a non-linearity exists, which affects the results of the lapping operation.
Still another problem with using post-BCL data to calibrate MR elements is that post-BCL calibration can only be determined after first lapping the rows. Since, for accurate results, it is preferred to complete lapping based on measure resistance and stripe height (i.e., result-directed/predictive lapping), element calibration after BCL is too late.
Because of the above stated issues/problems with current (abc) lapping processes, the (abc) method does not provide adequate methods for result-directed/predictive lapping and is not an adequate calibration method for carrier stripe height control to the 0.05 micron 3 sigma regime required for the newer products being produced in 2002 and beyond.
The present invention thus realizes that it would be desirable to provide a method and lapping control system/process that provides more accurate responses to and/or representation of the relationship between resistance and stripe height of magneto resistive elements being lapped. A method and lapping control system that enables in-situ (predictive) calibration of the lapping operation on MR elements utilizing accurate, predicted relationship data between stripe height and resistance would be a welcomed improvement. It would be further desirable to provide a calibration system design that enables collection of more accurate resistance data without wire bonding for utilization in result-directed/
Church Mark A.
Desouches Alain Michel Louis
Krebs Richard E.
Bracewell & Patterson L.L.P.
Bui Bryan
International Business Machines - Corporation
Martin Robert B.
LandOfFree
In-situ stripe height calibration of magneto resistive sensors does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with In-situ stripe height calibration of magneto resistive sensors, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and In-situ stripe height calibration of magneto resistive sensors will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3218914