Systems and methods for correcting fabrication error in magnetic recording heads using magnetic write width measurements

Description

FIELD

The present invention relates generally to manufacturing components for magnetic storage devices, and more specifically to systems and methods for correcting fabrication error in magnetic recording heads using magnetic write width (MWW) measurements.

BACKGROUND

Magnetic storage devices such as hard disk drives use magnetic media to store data and a movable slider having magnetic transducers (e.g., read/write heads) positioned over the magnetic media to selectively read data from and write data to the magnetic media. Electronic lapping guides (ELGs) are used for precisely controlling a degree of lapping applied to an air bearing surface (ABS) of the sliders for achieving a particular stripe height, or distance from the ABS, for the magnetic transducers located on the sliders. U.S. Pat. No. 8,165,709 to Rudy and U.S. Pat. No. 8,151,441 to Rudy et al., the entire content of each document is hereby incorporated by reference, provide a comprehensive description of ELGs used in manufacturing sliders for hard drives.

As the design of magnetic transducers becomes more and more intricate, their fabrication processes become increasingly complex as well. Such complex fabrication processes inherently include some imperfections that ultimately manifest as undesirable variations in the final product. By observing certain performance parameters of the final product (e.g., sliders including one or more magnetic transducers), these undesirable variations can be measured and quantified. A system and method for reducing or eliminating these undesirable variations in the performance of magnetic transducers is therefore needed.

SUMMARY

Aspects of the invention relate to systems and methods for correcting fabrication error in magnetic recording heads using magnetic write width (MWW) measurements. In one embodiment, the invention relates to a method of correcting for fabrication error in magnetic recording heads, the method including separating a wafer into a plurality of sections, each section containing a plurality of row bars, each row bar including a plurality of magnetic recording heads, selecting a first row bar from a plurality of row bars of a first section of the plurality of sections, lapping the first row bar to form a plurality of sliders, performing a test of a magnetic write width (MWW) on each of the plurality of sliders, calculating a first error profile for the first row bar based on results of the magnetic write width tests, generating a second error profile for a stripe height of a component of the plurality of sliders based on the first error profile, where the component is selected from the group consisting of a magnetic read head and a magnetic write head, and lapping a second row bar from the plurality of row bars of the first section using the second error profile.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flowchart of a process for correcting fabrication error in magnetic recording heads using magnetic write width (MWW) measurements in accordance with one embodiment of the invention.

FIGS. 2a to 2l illustrate a sequence of views of a wafer, row bars, sliders, and corresponding MWW test data of the sliders in a process for correcting fabrication error in magnetic recording heads using magnetic write width (MWW) measurements in accordance with one embodiment of the invention.

DETAILED DESCRIPTION

As discussed above, a system and method for reducing or eliminating undesirable variations in the performance of magnetic transducers is needed. Such variations can be observed in the measured magnetic write width (MWW) of current magnetic heads. Current lapping algorithms are designed to achieve preselected reader or writer stripe heights (SHs) on a slider without consideration to MWW variations within a particular wafer.

The MWW measurements are measurements of variations in actual recording performance. Such variations may be caused by variations in the recording pole geometry, in the material properties, in yoke magnetic structures, and defects and misalignment associated with the write coil, lapping variations, etcetera. While multiple methods for performing MWW measurements are well known in the art, one exemplary method will be discussed. In the exemplary MWW test method, a test region of a magnetic medium is identified and pre-conditioned (e.g., by erasing the test region area). A data pattern is written to the test region at a given track center, where the data pattern can be a pseudo-random bit sequence that mimics actual recorded data or another suitable data pattern. In some cases, the data pattern is a single frequency square wave data pattern at about 50 percent of a maximum data rate for simplicity. The method then measures the read-back amplitude dependence on the offset from the track center. The MWW is then calculated as the width of the track profile at 50 percent amplitude. In several embodiments, the MWW measurements are made using a spin-stand device. The MWW measurements are indicative of variations from intended write-field parameters, recording pole geometry, or other parameters, where the variations are often caused by the slider fabrication process.

Referring now to the drawings, embodiments of systems and methods for correcting fabrication error in magnetic recording heads using magnetic write width measurements are illustrated. In effect, the methods involve acquiring MWW test data for one or more sample sliders of a section of a wafer and then adjusting lapping stripe heights for the other sliders of the section to compensate for the measured MWW test data pattern across the section. As a result, the methods can reduce the measured MWW variation of the sliders and thereby provide significant yield improvement.

FIG. 1 is a flowchart of a process 100 for correcting fabrication error in magnetic recording heads using magnetic write width (MWW) measurements in accordance with one embodiment of the invention. The process first separates (102) a wafer into a number of sections, where each section contains a number of row bars and each row bar includes a preselected number of magnetic recording heads. The process then selects (104) a first row bar from a group of row bars in a first section of the wafer sections. The process then laps (106) the first row bar to form a preselected number of sliders. In several embodiments, the process laps the first row bar with an initial lapping profile. In some embodiments, the process selects two or more row bars and laps each of them to form the sliders.

The process then performs (108) a test of a magnetic write width (MWW) on each of the sliders. In several embodiments, the test of MWW is performed on a test machine (e.g., spin-stand) configured to test the performance characteristics of one or more sliders. The process then calculates (110) a first error profile for the first row bar based on results of the magnetic write width tests. In many embodiments, the first error profile includes calculation of an offset from a mean MWW value. In some embodiments, the mean value is for a particular group of sliders along the row bar (e.g., such as a first half and/or a second half of the sliders). In many embodiments, the first error profile includes an offset for each slider and a position of the respective slider along the row bar prior to the lapping.

The process then generates (112) a second error profile for a stripe height of a component of the sliders based on the first error profile, where the component is a magnetic read head and/or a magnetic write head. The second error profile can include a stripe height offset for each slider which can also be associated with a position of a respective slider. The process then laps (114) a second row bar from the row bars of the first section using the second error profile. In several embodiments, the process may lap all of the remaining row bars from the first section using the second error profile. In several embodiments, the process can be repeated for other sections on the wafer where each section has its own error profile based on the first row bar from the respective section that is processed to slider form and tested for MWW. In a number of embodiments, the process is repeated for each of the other sections on the wafer.

In one embodiment, the process can perform the sequence of actions in a different order. In another embodiment, the process can skip one or more of the actions. In other embodiments, one or more of the actions are performed simultaneously. In some embodiments, additional actions can be performed.

FIGS. 2a to 2l illustrate a sequence of views of a wafer, row bars, sliders, and MWW test data of the sliders in a process for correcting fabrication error in magnetic recording heads using magnetic write width measurements (MWW) in accordance with one embodiment of the invention. In FIG. 2a, the process provides (250) a wafer 200 on which a number of magnetic recording heads/transducers (not visible) have been formed in rows. In FIG. 2b, the process separates (252) the wafer 200 into sections (202a, 202b), where each section contains a preselected number of row bars and each row bar contains one or more magnetic recording heads/transducers. In several embodiments, the wafer 200 may be separated into about 25 sections. In FIG. 2c, the process selects (254) three row bars (204a, 204b, 204c) from one section 202a. In several embodiments, the process can select more than three row bars for better accuracy. In FIG. 2d, the process laps (256) the three row bars to form sliders (206a, 206b, 206c).

In FIG. 2e, the process performs (258) magnetic write width (MWW) tests on the sliders from the three selected row bars. In several embodiments, the MWW tests are performed on a test machine (e.g., spin-stand) configured to test the performance characteristics of one or more sliders. The MWW test results are illustrated in graph 208 of FIG. 2e showing the MWW profile (e.g., MWW measured in micro-inches or “uin”) across each of the three row bars based on the slider position along the respective row bar. In FIG. 2f, the process calculates (260) a MWW mean profile across the three row bars by slider position. FIG. 2f illustrates a graph 210 of the MWW mean profile across (e.g., MWW mean in micro-inches or “uin”) the three bars by slider position. In FIG. 2g, the process converts (262) the MWW mean from micro-inches or “uin” to nano-meters or “nm”. FIG. 2g illustrates a graph 212 of the MWW mean profile across (e.g., MWW mean in nm) the three bars by slider position.

In FIG. 2h, the process calculates (264) the MWW mean across a right flash field (e.g., roughly half of the sliders of a given row bar) and a left flash field (e.g., roughly half of the sliders of a given row bar). In some embodiments, the row bars have about 54 sliders and the first half or left flash field corresponds to sliders 1 to 27 and the second half or right flash field corresponds to sliders 28 to 54. In one embodiment, such as the one depicted in FIG. 2i, the first slider and the last slider are not considered such that the left flash field includes sliders 2 to 27 and the right flash field includes sliders 28 to 53. In other embodiments, the row bars can be segmented into different groups for the flash fields in accordance with particular design goals. In several embodiments, each row bar may include about 50 to 60 sliders. FIG. 2h illustrates a graph of the MWW mean for the three row bars 214a, for the left flash field 214b, and for the right flash field 214c.

In FIG. 2i, the process performs (266) a first order line fit across the sliders of each flash field and determines a MWW slope and intercept for each flash field. FIG. 2i illustrates a table showing the MWW slope and intercept values for the right and left flash fields. In several embodiments, the process can perform a line fit that is greater than a first order line fit instead of the first order line fit. In FIG. 2j, the process generates (268) a fitted mean 216 for each slider using the slope and intercept values for the left and right flash fields. FIG. 2j illustrates a graph of the MWW values for the mean of the three row bars 214a, the mean of the left flash field 214b, the mean of the right flash field 214c, and the fitted mean 216.

In FIG. 2k, the process calculates (270) a MWW mean across the bars and across the right and left flash fields using the fitted mean values. FIG. 2k illustrates a table showing the MWW mean values across the bars and across the right and left flash fields using the fitted mean values. The process then calculates (272) a MWW offset for each slider by the slider position. In one embodiment, the MWW offset is calculated using the expression, (slider MWW−flash field MWW mean)+(flash field MWW mean−section mean). The process then converts (274) the calculated MWW offsets into stripe height offsets for an electronic lapping guide (ELG). In several embodiments, the ELG is for a magnetic read head of the slider. In some embodiments, the ELG is for a magnetic write head of the slider. In one embodiment, the stripe height offsets are calculated using the expression, (slider MWW offset/(MWW to stripe height sensitivity)), where the MWW to stripe height sensitivity is a known parameter of the sliders from a particular wafer.

In FIG. 2l, the process converts (276) the stripe height offsets into resistance offsets 218. FIG. 2l is a graph illustrating the MWW mean 214a, the MWW fitted mean 216, and the resistance offsets 218 where each of these parameters is shown by slider position. In one embodiment, the resistance offsets are calculated using the expression, (wafer resistance*MC slope)/(reader stripe height−MC intercept), where the MC or model curve is a transfer function that converts the calculated “stripe height offset” into its equivalent resistance value. The process then laps (278) one or more row bars of the section of the wafer using the resistance offsets. In one embodiment, the process laps all remaining row bars of the section from which the initial three row bars originated.

In several embodiments, the process can be repeated for other sections on the wafer where each section has its own error profile based on the first row bars that are processed to form the sliders tested for MWW. In a number of embodiments, the process is repeated for each of the other sections on the wafer. In some embodiments, the process laps (278) the one or more row bars of the section using the resistance offsets and a preselected limit (e.g., upper or lower boundary) for the stripe height of the component.

In several embodiments, the process laps (256) the three row bars to form the sliders using a first lapping profile (e.g., initial lapping profile). In such case, the process then laps (278) the other row bars using a second lapping profile (e.g., updated lapping profile) that takes into account the second error profile (e.g., first lapping profile modified by stripe height offsets or MWW offsets derived from MWW tests).

In several embodiments, the process can be executed on any general purpose type computer having a processor, memory, and other such components that are well known in the art. In one embodiment, the process can perform the sequence of actions in a different order. In another embodiment, the process can skip one or more of the actions. In other embodiments, one or more of the actions are performed simultaneously. In some embodiments, additional actions can be performed.

While the above description contains many specific embodiments of the invention, these should not be construed as limitations on the scope of the invention, but rather as examples of specific embodiments thereof. Accordingly, the scope of the invention should be determined not by the embodiments illustrated, but by the appended claims and their equivalents.