METHOD AND APPARATUS FOR ALIGNING AN ILLUMINATION UNIT TO A SLIDER FOR A MAGNETIC RECORDING DEVICE

Abstract

An alignment method for bonding a first component to a second component is described. A plurality of sliders having alignment markers are substantially aligned to the alignment markers of a plurality of illumination units and then positioned into alignment by moving the slider with respect to the illumination unit by an offset. The offset is calculated by scanning a light source around the waveguide to determine the distance that the alignment markers of the slider are separated from the alignment markers on the illumination unit.

Description

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application is a divisional of U.S. patent application Ser. No. 13/969,782, filed on Aug. 19, 2013, which claims priority to U.S. Provisional Patent Application No. 61/846,868 entitled "METHOD AND APPARATUS FOR ALIGNING AN ILLUMINATION UNIT TO A SLIDER FOR A MAGNETIC RECORDING DEVICE, filed on Jul. 16, 2013 for Chee Kheng Lim, both of which are incorporated herein by reference.

BACKGROUND

[0002] Heat assisted magnetic recording (HAMR) typically uses a laser source to provide additional energy to magnetic media during the data writing process. The laser source may include a submount assembly, which together is referred to as a Chip-On-Submount-Assembly (COSA). The COSA is attached to the back of a conventional magnetic head slider and light energy from a laser diode chip is guided to the air bearing surface through a waveguide to heat the magnetic media.

[0003] To ensure that the laser diode output is efficiently coupled to the waveguide on a slider it is desirable to accurately bond the slider to the laser source. Aligning the laser source and the slider to the desired position is time consuming and prone to error. The manufacturing yield for fabricating magnetic devices may be adversely affected due to misalignment of the laser with respect to the slider. Therefore, a need exists for a simplified alignment method that accurately aligns a slider to an illumination unit and that improves manufacturing yield.

BRIEF DESCRIPTION OF THE DRAWINGS

[0004] FIG. 1 is a flow diagram that illustrates an alignment method according to one embodiment.

[0005] FIG. 2 is a perspective view of a slider with alignment marks and a waveguide.

[0006] FIG. 2A is an enlarged view of section W of FIG. 2.

[0007] FIG. 3 illustrates an alignment apparatus that includes an active alignment station and a passive alignment station according to one embodiment.

[0008] FIG. 4 illustrates a laser irradiating the area of a waveguide in a cross-track direction.

[0009] FIG. 5 illustrates a COSA chip irradiating the area of a waveguide in a down-track direction.

[0010] FIG. 6 is a perspective view of a slider being bonded to a chip.

[0011] FIG. 7 illustrates a system according to one embodiment.

DETAILED DESCRIPTION

[0012] A hybrid active and passive alignment method is provided for bonding a first and second component together for use in a magnetic recording device. In certain embodiments, the method achieves high throughput without compromising the alignment accuracy and alignment yield of the components. In one embodiment, the first component comprises a slider and the second component comprises an illumination unit.

[0013] FIG. 1 is a flow diagram illustrating a method for aligning a first component to a second component according to one embodiment. The process starts by establishing an initial alignment position of a slider via block 110. This involves identifying the location of the alignment markers on a reference slider, and then substantially aligning the slider's alignment markers to alignment markers on an illumination unit. The initial alignment position will later be used to calculate an offset for passive alignment.

[0014] The process proceeds by irradiating a waveguide on a reference slider with a laser or an illumination unit according to block 120. Either a collimated or uncollimated illumination unit may be used. A detector monitors the light intensity output by the laser via block 130 to locate an optical alignment position. Upon establishing optical alignment, an offset may be calculated via block 140. Each slider is subsequently moved from its initial alignment position by the offset prior to bonding to the illumination unit in accordance with block 150. Then the sliders are separately bonded to a corresponding illumination unit via block 160.

[0015] The method of FIG. 1 will be discussed in further detail in association with FIGS. 2-7. FIG. 2 is a simplified drawing of a slider 200 having a waveguide 250 on top surface 210. The waveguide 250 is located between alignment markers 220 on the slider 200, and extends from top surface 210 to bottom surface 215. Although in other embodiments, the waveguide 250 is not necessarily located between the alignment markers 220.

[0016] FIG. 2A is an enlarged view of section W of FIG. 2. Distance d represents an offset, which corresponds to the manufacturing tolerance or variance of slider 200, and may include other manufacturing tolerances. Thus, offset d reflects the misalignment of waveguide 250 from alignment markers 220 on slider 200. Offset d is used to precisely align other sliders 200 to illumination units 285 (discussed in reference to FIG. 5) based on the distance separating the initial alignment position and the optical alignment position of a reference slider. The reference slider resembles slider 200 and is selected from sliders manufactured from the same wafer or wafer section. The manner by which various elements on slider 200 are located is described below in reference to FIG. 3.

[0017] The alignment apparatus 300 of FIG. 3 includes two stations: a measurement station 305 and a bonding station 310. Measurement station 305 may utilize a high resolution vision recognition camera to locate alignment markers 220 on a reference slider. The alignment markers 220 are then used to approximately determine the waveguide position, since waveguide 250 is not generally visible even with a high resolution optical system. The corrected alignment position is later determined by the active alignment portion of alignment apparatus 300.

[0018] The waveguide position coordinate data determined at measurement station 305 may be dynamically transmitted to the bonding station 310 to ensure that slider 200 is adjusted by the offset d prior to bonding. In certain embodiments, this adjustment substantially increases the likelihood that the output of a laser is correctly aligned to waveguide 250. Slider 200 may then be eutectic bonded to illumination unit 285 using a conventional soldering process or other adhesion method.

[0019] FIGS. 4 and 5 illustrate two different alternative illumination units (245 and 285) that are suitable for implementing one embodiment of the present alignment method. FIG. 4 illustrates a laser 245 that irradiates the slider 200 while scanning top surface 210 in a down-track direction. In some embodiments, the beam may be scanned in both a cross-track and a down-track direction. During irradiation, the slider may be incrementally moved with respect to the illumination unit from the initial alignment position. In addition, while slider 200 is being irradiated around waveguide 250, a detector 212 measures the light intensity exiting surface 215, which is the bottom surface of slider 200 (opposite top surface 210) in accordance with one embodiment of block 130. In some embodiments, lasers suitable for scanning via block 120 operate at a wavelength of approximately 830 nm. Although the laser 245 is illustrated as emitting a collimated beam, it will be appreciated that an uncollimated source may be used as well. If a collimated source is used to scan the light beam, then the distance from the source to the waveguide 250 may not be critical. However, when an uncollimated source is used, in certain embodiments, the uncollimated source may be operated at a power level between 1-10 milliwatts.

[0020] In alternative embodiments, illumination unit can be a COSA chip 285 as shown in FIG. 5. When a COSA chip 285 is used as the illumination unit, the distance between alignment markers 240 and the waveguide 250 can be less than 100 micrometers, due to the divergence of the beam from the laser diode chip. In one embodiment, COSA chip 285 is scanned in a down-track direction relative to slider 200. In other embodiments, COSA chip 285 is scanned in both cross-track and down-track directions relative to slider 200. In several embodiments, the waveguide position is identified when detector 212, such as a photodiode, detects the maximum light output from the slider 200.

[0021] Once the waveguide 250 is precisely located, the offset d of waveguide 250 from alignment markers 220 is calculated. It is desirable to compute offset d because of process variations during manufacturing that tend to affect alignment accuracy. Since the waveguide 250 and alignment markers 220 are fabricated separately, and a number of layers separate waveguide 250 from alignment markers 220, component variations are often inevitable. Although the exact position of waveguide 250 varies, the location of the waveguide 250 is not measured on every slider 200. With the offset d calculated, slider 200 is ready for bonding.

[0022] Bonding station 310 resembles a passive alignment system due to the absence of an illumination source. In block 150, a pick and place robot moves the alignment markers 240 of illumination unit 285 into substantial alignment to alignment markers 250 as shown in FIG. 6. Next the position of slider 200 is adjusted by the previously calculated offset via block 150. Then the slider and illumination unit 285 are bonded together via block 160.

[0023] The measurements performed at measurement station 305 can be used to compensate for any offset d needed to accommodate component variation. For example, some sliders may vary in the thickness of layers separating waveguide 250 and markers 220, or have waveguides that are at a distance of .gtoreq.0.02 microns from alignment markers 220. Such variations can be detected before or after bonding at bonding station 310. For example, to detect a component variation prior to bonding, an alignment marker on slider 200 is used as a reference point (reference marker). Then the intensity of the light beam exiting the waveguide 250 is evaluated by a detector 212. When the maximum light intensity is located, optical alignment is established. Upon determining the optical alignment position, the distance from the alignment marker 220 is measured to ascertain if any large component variations are present. If variations exceeding 0.1 microns from one slider to another slider, then the sampling frequency of the waveguide 250 is increased. Alternatively, a post-bond optical test system can be used to determine the amount of light that is coupled into the waveguide. Variations among different parts can lead to significant reduction in the light intensity output from the waveguide, to thereby indicate that such variations are affecting the alignment accuracy. At which point, the next slider would be evaluated in measurement station 305 prior to bonding any further parts in bonding station 310.

[0024] In certain embodiments, it is not essential to pre-measure the waveguide prior to passive alignment at bonding station 310 if the waveguide is known to be within +/-0.1 microns of alignment markers 220. Thus, in such embodiments, a throughput of up to 100 sliders is possible before the waveguide is measured again. In this manner, the waveguide position data from the reference slider can substantially minimize misalignment due to part-to-part variation.

[0025] Accordingly, in certain embodiments, high throughput may be achieved by measuring the waveguide position in a sampling manner. For example, the waveguide position can be measured in one of every ten sliders. In other embodiments, the waveguide position can be measured in one of every 100 sliders. The frequency of the waveguide measurement will depend on the offset variation from the waveguide 250 to the alignment markers 220. The waveguide coordinates pursuant to block 140 are then dynamically transmitted to the bonding station to perform the passive alignment process. In some embodiments, block 140 contributes to high throughput yield by periodically sampling sliders to compute an offset, rather than measuring optical alignment positions on every single slider before bonding each slider 200 to an illumination unit 285.

[0026] FIG. 7 summarizes yet another embodiment of the present disclosure. At the start of the system 700, a vision camera 710 can be used to identify the location of alignment markers 220 on slider 200. Once alignment markers 220 are identified, the slider 200 is transferred to measurement station 715 to measure the waveguide position. Measurement station 715 is analogous to station 305, discussed in reference to FIG. 3; however the two measurement stations are not the same. At measurement station 715, the degree of variation from part to part can be assessed. In certain embodiments, the offset d will be known after the waveguide position is measured at measurement station 715. Then a processor, or other calculating apparatus (not shown), can be used to calculate whether offset d exceeds an acceptable threshold (offset threshold), or otherwise determine if the sampling frequency should be increased. The waveguide sampling frequency can be adjusted depending on the level of component variation detected. For example, if large part-to-part variation is detected, more frequent measurement will be activated to account for the passive alignment offset. On the other hand, if a trivial variation in waveguide position is determined by the processor, then a less frequent measurement may be needed. In one embodiment, intelligence software can be used to automatically adjust the sampling frequency.

[0027] Once the processor determines whether a variation is acceptable or not, then the alignment process can proceed in either of two ways as indicated at decision block 725. If no variation or a trivial variation is detected, then the system will bond the slider to the illumination unit at bonding station 730. Subsequently, the bonded slider and illumination unit can then be integrated into a magnetic device via block 750. After product integration, bonding of additional sliders may continue based on the originally calculated offset d. On the other hand, a significant variation may be detected after bonding at testing station 740. If testing determines that the bonded slider varies substantially from the reference slider, then system 700 will proceed to begin measuring the alignment markers and waveguide position on a reference slider via block 735. In other words, if block 725 identifies a component variation that exceeds the offset threshold, then a new offset d is determined using a reference slider via block 735.

[0028] By performing several embodiments described above, alignment of components can be achieved with accuracy and high throughput. The method and system of the disclosure are primarily designed for submicron bonding accuracy. It is estimated that by adopting the aforementioned hybrid alignment system that the bonding cycle for aligning multiple components may be reduced by about 3-5 seconds. As a result, the disk drives and other devices fabricated with components aligned by the methods described herein may have improved performance.

Claims

1. A system for aligning a slider to an illumination unit comprising: an active alignment apparatus for substantially aligning a slider to an initial alignment position relative to the illumination unit, wherein the active alignment apparatus includes a light source, a detector, and a measurement station, and wherein the light source irradiates a reference slider while the detector evaluates light emitted from a bottom surface of the reference slider to determine a position of optical alignment; a measurement station that periodically measures an offset corresponding to the distance from the initial alignment position to the optical alignment position of a reference slider; a passive alignment apparatus that includes a bonding station for attaching a plurality of sliders to a plurality of illumination units, and a testing station that periodically checks for misalignment of sliders and illumination units by measuring the amount of light coupled into the waveguide of the slider after being bonded to the illumination unit.

2. The system of claim 1, wherein the light source is part of the illumination unit.

3. The system of claim 1, wherein the illumination unit is not part of the light source.