Method for magneto-resistive head electrostatic popping detection

[0002] 1. Field

[0003] The inventor's disclosure generally relates to disk drives and more particularly to a method and system for detecting electrostatic noise of magneto-resistive heads in hard disk drive assemblies.

[0004] 2. Description of the Related Art

[0005] Disk drives are magnetic recording devices used for the storage of information. The information is recorded on concentric tracks on either surface of one or more magnetic recording disks. The disks are rotatably mounted to a spin motor and information is accessed by means of read/write heads that are mounted to actuator arms which are rotated by a voice coil motor. The voice coil motor is excited with a current to rotate the actuator and move the heads. The read/write heads must be accurately aligned with the storage tracks on the disk to ensure proper reading and writing of information.

[0006] Dual element transducers are increasingly utilized in hard disk drives because they are lighter than single element transducers. Dual element transducers include a single write element and a separate read element which is constructed from a magneto-resistive material. Such dual element transducers are commonly referred to as magneto-resistive (“MR”) heads and the read elements of such heads are often referred to as GMR/MR sensors. More recently, heads have been constructed using alternating magnetic and nonmagnetic films thereby drastically increasing their resistivity. Such heads are generally known as giant magneto-resistive (“GMR”) heads.

[0007] Along with the desire to improve head design, there is also a desire to decrease the failure rate of the hard disk drives. While such failures occur for a variety of reasons, one cause is GMR/MR head instability, particularly head instability associated with electrostatic discharge. Such electrostatic discharge manifests as popping noise and may similarly be caused by a variety of conditions, such as junction permanent-magnet insufficiency or dielectric breakdown of the insulting layer between the conducting sensor stripes.

[0008] One problem with detecting popping noise in a GMR/MR head is that different causes of popping noise manifest as different types of popping noise. For example, popping noise manifested as amplitude spikes may be caused by GMR/MR internal design flaws, while popping noise manifested as an unstable baseline not correlated to a signal may be caused by dielectric breakdown. Finally, heads which exhibit unstable baselines correlated to a signal may be caused by junction-induced instability.

[0009] One problem in the art is that it is difficult to detect these different types of popping noise using a single simple method. Thus, there is a need in the art for a single improved method that is capable of detecting different types of popping noise.

[0010] The inventor discloses a method and system for detecting defective heads. In one embodiment, the method comprising erasing a track on a disk, said disk including a plurality of tracks, reading said track with a head, determining a first signal amplitude at a first frequency using a first current supplied to said head, and determining a second signal amplitude at said first frequency using a second current supplied to said head. The method further comprises determining a first noise measure, said first noise measure to be a function of said first signal amplitude and said second signal amplitude, and identifying said head as having a first defect type where said first noise measure is greater than a first amplitude threshold.

[0011] FIG. 1A is a perspective view of a read/write MR head according to one embodiment.

[0012] FIG. 1B is an exemplary block diagram of a system for testing a GMR/MR head, according to one embodiment.

[0013] FIG. 2A depicts a time-domain noise plot of a head exhibiting an amplitude spike defect.

[0014] FIG. 2B depicts a time-domain noise plot of a head exhibiting an unstable baseline defect.

[0015] FIG. 2C depicts graphs of the frequency spectrums of heads exhibiting amplitude spike and unstable baseline defects, respectively.

[0016] FIG. 2D depicts graphs of frequency spectrums of heads exhibiting acceptable popping noise characteristics.

[0017] FIG. 3 is a flow diagram of a method for implementing one embodiment.

[0018] FIG. 4A is a flow diagram of another method for implementing one embodiment.

[0019] FIG. 4B is a continuation of the flow diagram of FIG. 4A depicting one embodiment of a method being disclosed by the inventor.

[0020] FIG. 5 is a continuation of the flow diagram of FIG. 4A depicting another embodiment of a method being disclosed by the inventor.

[0021] FIG. 6 is a flow diagram of another method disclosed by the inventor.

[0022] The inventor discloses a method and system for detecting popping noise caused by electrostatic discharge across a GMR/MR head. One embodiment is to detect GMR/MR head popping noise manifesting as an amplitude spike. In this embodiment, a two-spot test method is used wherein the noise generated at a given spot frequency for a first current level is compared to the noise generated at the same spot frequency at a second current level. Thereafter, the result of this comparison may be used to determine whether or not the head being tested (hereinafter the “Test Head”) suffers from an amplitude spike defect. Another embodiment is to detect GMR/MR popping noise manifested as an unstable baseline in the frequency domain which is not correlated to a signal. Again, a two-spot method is used wherein the noise generated at a given spot frequency for a first current level is compared to the noise generated at the same spot frequency at a second current. As with the previous embodiment, the result of this comparison is used to determine whether a head suffers from an unstable baseline defect. In yet another embodiment, heads having either amplitude spike or unstable baseline defects are detected using a single two-spot frequency method.

[0023] Referring to the drawings more particularly by reference numbers, FIG. 1 illustrates an MR read/write head 100. In particular, the head is comprises of an inductive head 105 and a GMR/MR sensor 110. The function of the inductive head 105 is to write data to a disk, while the function of the GMR/MR sensor 110 is to read signals from a magnetized surface, such as the surface of a hard disk in a hard disk drive. The inductive head 105 consists of a top pole 115 and a bottom pole 120. The poles, separated by some gap 125, are magnetized by passing a current through the coils 130. The resulting magnetic field across the gap 125 is capable of magnetizing the surface of a disk when passed sufficiently close to a disk's surface, thus enabling digitized data to be written to a disk.

[0024] The GMR/MR sensor 110 is typically positioned near the bottom pole 120 of the inductive head 105 and is typically narrower than the inductive head 105. Furthermore, the layer nearest the inductive head 105 may be a laminate of a bias magnetic film 135. Thereafter, an insulting layer 140 and a magneto-resistive layer 145 may follow.

[0025] FIG. 1B is an exemplary block diagram of a head test system. Spin stand 150 contains the Test Head and the media to be used in conjunction with the Test Head. While in one embodiment spin stand 150 is a Guzik Micro Positioning Spinstand, it should be appreciated that any device, including a traditional disk drive assembly, which is capable of passing a GMR/MR head over the surface of a magetizable media while supplying a current to the head may be used. Moreover, it should be appreciated that a plurality of Test Heads may be simultaneously located and tested at spin stand 150.

[0026] Once the Test Head at spin stand 150 generates a signal, the signal may then be passed to a universal amplifier 155 which amplifies the signal. Thereafter, an over-write filter 160, such as a controllable band pass filter, may be used to narrow the range of frequencies which are to be passed to the read/write analyzer (“RWA”) 165. In one embodiment, the over-write filter 160 is set to pass a 100 kilohertz band centered on 3 megahertz. It is to be appreciated that the over-write filter 160 may be set to pass any frequency range.

[0027] In one embodiment, the RWA 165 is a Guzik RWA analyzer. However, any analyzer which converts the head signal from an analog to a digital form and which supports frequency domain analysis may be used. Once the head signal is analyzed at 165, computer 170 may be used to store and/or display the results. Computer 170 may include a processor or central processing unit (CPU). The CPU may, in turn, include an Arithmetic Logic Unit (ALU) for performing computations, a collection of registers for temporary storage of data and instructions, and a control unit for controlling operation of the computer 170. The CPU may be a microprocessor as is commonly known in the art, or may take on other forms such as microcontrollers, digital signal processors, reduced instruction set computers (RISC), application specific circuits, and the like. The computer 170 may also include system memory (RAM), input/output (I/O) devices, and a mass storage device. The mass storage device may include a hard disk, floppy disk, CD-ROM, DVD-ROM, tape, high capacity removable media, solid state memory device, etc. The computer may further include system firmware such as system BIOS, ROM and flash memory.

[0028] Moreover, computer 170 should include at least one application program capable of controlling spin stand 150 and the current to be supplied to the Test Head. Computer 170 may also be configured to run software which automates the process of adjusting the current supplied to the Test Head, the frequency to be passed to the RWA 165, and the specifics of what read/write operations to have the Test Head perform.

[0029] FIG. 2A is a time-domain plot 200 in which y-axis is a measure of noise generated when a Test Head reads a recorded signal as a function of time. When the Test Head suffers from certain defects, such defects may manifest as random signal spikes, such as amplitude spikes A1 and A2. Similarly, FIG. 2B is a time-domain plot 202 in which head noise is again plotted as a function of time. In this case, the Test Head suffers from-a defect which is manifesting itself as an unstable baseline. The inventor further discloses using a spectrum analysis (using an spectrum analyzer such as RWA 165) in the frequency domain of the noise generated by a head at a given current to identify heads that have amplitude spike and/or unstable baseline defects. Referring in particular to FIG. 2C, graph 204 contains plots of popping noise (dB) versus frequency (MHz) for a Test Head having defects manifesting as amplitude spikes. As mentioned previously, amplitude spikes may be caused by different conditions, such as internal head design flaws. Thus, for convenience, a head exhibiting such characteristics will be referred to herein as a head having amplitude spike defects.

[0030] Referring still to graph 204, plot 205 represents the frequency spectrum where the Test Head reads a DC-erased track while being supplied with a current of 2.1 mA (the low base bias current). Similarly, plot 210 represents the frequency spectrum resulting from supplying the Test Head with a 5 mA and plot 215 is based on a 6 mA current (the high test bias current), both also reading a DC-erased track. In contrast, plot 220 represents the resulting frequency spectrum when the head reads a signal with a 6 mA current which was written at 10 MHz. It should further be appreciated that, as seen in graph 204, amplitude spike defects manifest as a large spectrum gap between the low base bias and high test bias curves for both low frequencies and high frequencies.

[0031] Referring now to graph 230, the various depicted plots were generated using the same currents as those in graph 204 (i.e., plot 235 used 2.1 mA, plot 240 used 5 mA, plot 245 used 6 mA, and plot 250 used 6 mA on a 10 MHz signal). While the plots of graph 230 also exhibit a large spectrum gap between the low base bias and high test bias plots, the spectrum gap is appreciably smaller at higher frequencies. Such a decreasing gap is exhibited by heads having defects manifesting as an unstable baseline. Again, for convenience, a head exhibiting such characteristics will be referred to herein as a head having unstable baseline defects.

[0032] FIG. 2D contains graphs of two non-defective heads. The individual plots of FIG. 2D were generated using the same current levels as used in FIG. 2C (i.e., 2.1 mA on a DC-erased track, 5 mA on a DC-erased track, 6 mA on a DC-erased track, and 6 mA on a 10 MHz signal).

[0033] Comparing FIG. 2C with 2D illustrates that the spectrum gap between the low base bias plots (205, 235, 260 and 285) and high test bias plots (215, 245, 270 and 293) is much smaller for the non-defective heads of FIG. 2D than for the defective heads of FIG. 2C. Thus, both amplitude spike and unstable baseline defects can be detected by comparing signal amplitude as a function of frequency over a plurality of currents. Moreover, a comparison of the frequency spectrum graphs of FIGS. 2C and 2D illustrates the fact that the delta between signal spikes in plot 220 is much greater than the delta for such spikes in plot 275.

[0034] It should further be appreciated that the unstable baseline defect seen in graph 230 is an unstable baseline defect not correlated to a signal. That being the case, a single-current defect test system, which compares the signal plot baseline against the DC-erased baseline, would not detect such a defect since the baselines for plots 245 and 250 are so similar. Thus, analyzing popping noise at the high test bias current and low base bias current, as in FIGS. 2C-2D, also enables detection of unstable baseline defects which are not correlated to a signal.

[0035] Referring now to FIG. 3, a block diagram of one embodiment is depicted. The process of FIG. 3 begins by DC-erasing at least one track on a disk (block 302) which is to be read by a Test Head (i.e., a GMR/MR head to be tested for defects). It should be appreciated that a plurality of tracks may similarly be DC-erased and used in the detection process. In one embodiment, the disk tracks to be erased may be erased using a GMR/MR head and spin stand 150.

[0036] At block 303, a First Spot Frequency is set. In one embodiment, the First Spot Frequency is between 1 and 10 megahertz. At block 304, a Second Spot Frequency is set which, in one embodiment, is between 20 and 40 megahertz. It should be appreciated that the desirable spot frequencies and spot frequency ranges may be different for different heads. Next, at block 305, the Low Base Bias current is set. The Low Base Bias is a relatively low current to be supplied to the Test Head. In one embodiment, the Low Base Bias is between 1 and 3 milliamps, though the Low Base Bias may be less than 1 mA or greater than 3 mA.

[0037] At block 306, the track average amplitude (TAA) is measured for a given track, or plurality of tracks. In particular, the value TAA1-1 represents the average noise amplitude at the First Spot Frequency generated by the Test Head when the Low Base Bias current is being supplied to the Test Head. The Test Head, which in one embodiment is connected to the spin stand 150, is directed to read a plurality of sample points along one of the DC-erased tracks on a disk. The TAA may then be computed by dividing the total signal amplitude by the number of sample points.

[0038] At block 307, the TAA at the second frequency is measured (TAA1-2) while the Low Base Bias current is supplied to the Test Head. Next, the current to be supplied to the Test Head is set to the High Test Bias (block 308). In one embodiment, the High Test Bias is between 3 and 7 milliamps, although it may be less than 3 milliamps or greater than 7 milliamps. At blocks 309 and 310, the TAA is measured at the First Spot Frequency (TAA2-1) and the Second Spot Frequency (TAA2-2), respectively, while the High Test Bias current is supplied to the Test Head.

[0039] At block 311, a first measure of popping noise is calculated (BLPN_1), which represents a quantification of the popping noise generated by the Test Head in question. In one embodiment, the following equation is used to compute BLPN_1: $\begin{array}{cc}\mathrm{BLPN\_}1=20*{\log }_{10}\left(\frac{\mathrm{TAA2}-1}{\mathrm{TAA1}-1}\right).& \left(1\right)\end{array}$

[0040] At block 312, a second measure of popping noise is calculated (BLPN_2), which represents a second measure of the popping noise generated by the Test Head. In one embodiment, BLPN_2 is computed as follows: $\begin{array}{cc}\mathrm{BLPN\_}2=20*{\log }_{10}\left(\frac{\mathrm{TAA2}-2}{\mathrm{TAA1}-2}\right).& \left(2\right)\end{array}$

[0041] At block decision 313, BLPN_1 is compared with a first reference threshold value (e.g., Threshold,). In one embodiment, this threshold value is 10 dB. In another embodiment, Threshold, is in the range of 7 to 15 dB. In yet another embodiment, Threshold, may be any other value. Where BLPN_1 exceeds Threshold, the head is identified as having an unstable baseline defect (block 315). Next, at decision block 314, BLPN_2 is compared with a second threshold value (e.g., Threshold₂). In one embodiment Threshold₂is 8 dB, while in another embodiment Threshold₂is between 6 and 10 dB. If BLPN_2 exceeds Threshold₂, the Test Head is identified as having an amplitude spike defect (block 316).

[0042] It should be appreciated that while equations (1) and (2) have the logarithm-portion of the equation being multiplied by the value of 20, this value may be any constant. Where the constant is something other than 20, the values for Threshold, and Threshold₂may be correspondingly adjusted.

[0043] The flow diagrams of FIGS. 4A and 4B illustrate another embodiment. In this embodiment, an iteration variable is set to i=1 at block 402. As with the process of FIG. 3, at least one track to be read by the Test Head is DC-erased at block 403. The process through blocks 404-410 correlates to the process through respective blocks 303-312 discussed above. However, in the process of FIG. 4, the First Spot Frequency and Second Spot Frequency set at block 404 is set for the current iteration i, which in one embodiment is different for each iterations. In another embodiment the spot frequencies remain the same for each iteration.

[0044] The corresponding TAAs measured at blocks 406 and 408 relate to the present iteration i. Thus, for the first iteration, the TAAs measured at block 406 for the Low Base Bias with be TAA1-1₁and TAA1-2₁for the First Spot Frequency and the Second Spot Frequency, respectively. Similarly, the TAAs for the High Test Bias current would be TAA2-1₁and TAA2-2₁.

[0045] At blocks 409 and 410, BLPN_1_iand BLPN_2_iare computed using equations (1) and (2), respectively. At decision block 411, a determination is made as to whether or not a predetermined number of iterations have been completed. In the illustrated embodiment, the number of iterations to be completed is 5. However, the predetermined number of iterations may be any number, depending on how many data points are desired.

[0046] After the first iteration, the process will increment i by 1 at block 412 and return to block 404 to obtain new TAA values to be stored as TAA1-1₂, TAA1-2₂, TAA2-1₂, and TAA2-2₂. From these values, BLPN_1₂and BLPN_2₂may then be calculated using equations (1) and (2).

[0047] Once all of the predetermined number of iterations have been completed, the process of FIG. 4A proceeds to the process of FIG. 4B. At blocks 413 and 414, the numerical averages for both BLPN_1_iand BLPN_2_iare computed using the value obtained for all completed iterations. It should be appreciated that other measures of amplitude may be used in place of the numerical average.

[0048] At block 415, the average for BLPN_1_i(BLPN_1_avg) is compared to a threshold value for amplitude (Threshold,). In one embodiment, Threshold₁is between 7 and 15 dB. If Threshold, is exceeded, then the Test Head is identified as having an unstable baseline defect at block 418. If, on the other hand, Threshold, is not exceeded, the process of FIG. 4B continues to decision block 416. At decision block 416 a determination is made as to whether the average amplitude for BLPN_2 (BLPN_2_avg) exceeds the a second threshold value (Threshold₂). In one embodiment, Threshold₂is in the range of 6 to 10 dB. Where Threshold₂is not exceeded, the Test Head is identified as not having an amplitude spike defect, nor an unstable baseline defect. Where Threshold₂is exceeded at block 416, however, the Test Head is identified as having an amplitude spike defect at block 419.

[0049] Yet another embodiment is described in FIGS. 5. The process of FIG. 5 begins where the process of FIG. 4B would have in the previously described embodiment. Thus, once block 411 determines that the predetermined number of iterations has been reached, the process, according to this embodiment, continues to block 501. In particular, at blocks 501 and 502 the numerical averages and standard deviations for both BLPN_1_iand BLPN_2_iare computed using the value obtained for all completed iterations. It should be appreciated that other measures of amplitude and stability may be used in place of the numerical averages and standard deviations. However, for simplicity, the measures of amplitude and stability for the TAAs will be described herein as being the averages and standard deviations, respectively.

[0050] At block 503, the average and standard deviation of BLPN_1_iis compared to threshold values for amplitude and standard deviation. (Threshold, and SD₁). In one embodiment, Threshold, is between 7 and 15 dB, while SD₁is between 0.65 and 1.0. If both Threshold₁and SD₁are exceeded, then the Test Head is identified as having an unstable baseline defect at block 506. If, on the other hand, Threshold, and SD₁are not both exceeded, the process of FIG. 5 continues to decision block 504. At decision block 504 a determination is made as to whether the amplitude and the standard deviation for BLPN_2 exceed the corresponding threshold values (i.e., Threshold₂and SD₂). In one embodiment, Threshold₂is in the range of 6 to 10 dB, while SD₂is between 0.65 and 1.0. Where both Threshold₂and SD₂are not exceeded, the Test Head is identified as not having an amplitude spike defect. Where both threshold value are exceeded at block 504, however, the process continues to block 507 where a determination is made as to whether the standard deviation for BLPN_1 exceeds SD₁. If so, then the Test Head is identified as having an amplitude spike defect at block 508. If, on the other hand, SD₁is not exceeded then the Test Head is identified as not having an amplitude spike defect.

[0051] In yet another embodiment, rather than selecting a single High Test Bias current, an iterative process may be used to converge on the High Test Bias current which produces maximum values for TAA2-1 and TAA2-2. For example, FIG. 6 illustrates one embodiment which converges on the optimal test bias. At block 610, the outer points of a range of High Test Bias values is set (i.e., TB_low, and TB_high). Moreover, the size of the increment (TB_step) for each iteration is set, the counting variable i is to set to 0, and the initial test bias (TB_i) is set to TB_low.

[0052] As with processes of FIG. 3, the TAAs are measured for a First Spot Frequency and a Second Spot Frequency while providing the Test Head with a current of TB_i, yielding values for TAA2-1_iand TAA2-2_i, respectively (block 620). These values are stored for later use. In one embodiment, these values are stored in a lookup table on computer 170. In another embodiment, these values are stored at RWA 165. In one embodiment, the First Spot Frequency is between 1 and 10 megahertz, while the Second Spot Frequency is between 20 and 40 megahertz.

[0053] At decision block 640, a determination is made as to whether the present test basis (TB_i) is equal to TB_high. If so, loop 620-640 ends and the process continues to block 650. If, on the other hand, TB_iis still less than TB_high, then TB_iis incremented by TB_stepand the counting variable i is incremented by 1. The loop continues, with the TAA values being stored as they are measured, until TB_iequals TB_high. At that point, the stored TAA values may be analyzed to determine the maximum values. In this manner, the TB_ivalue which corresponds to the maximum TAA values may thereafter be used as the High Test Bias for the processes described in FIGS. 3-5.

[0054] The present invention may be implemented as a method, apparatus, system, computer program product, etc. When implemented in software, the elements of the present invention are essentially the code segments to perform the necessary tasks. The program or code segments can be stored in a processor readable medium, which includes any medium that can store or transfer information. Examples of the processor readable medium include an electronic circuit, a semiconductor memory device, a ROM, a flash memory, an erasable ROM (EROM), a floppy diskette, a CD-ROM, an optical disk, a hard disk, etc.

[0055] While certain exemplary embodiments have been described and shown in the accompanying drawings, it is to be understood that such embodiments are merely illustrative of and not restrictive on the broad invention, and that this invention not be limited to the specific constructions and arrangements shown and described, since various other modifications may occur to those ordinarily skilled in the art.