Fig 1 - uploaded by B.J. Shi
Content may be subject to copyright.
Schematic diagram of an HDD without a base.

Schematic diagram of an HDD without a base.

Source publication
Article
Full-text available
This work aims to develop a method for predicting the displacement and failure of the Head Actuator Assembly (HAA) during a drop test. When a Hard Disk Drive (HDD) is dropped from a certain height, it will accelerate due to gravity until it hits the ground with a certain speed, and the head suspension system may lift off the disk and land onto it i...

Contexts in source publication

Context 1
... of disk-suspension-slider air-bearing systems was proposed [3]. More recently, an FE model of an HDD was developed [4] to investigate its response to a linear shock and a rotary shock. Comparison of the simulation results obtained under the two types of conditions led to the development of a correlation between the linear and rotary shock tests. Fig. 1 shows a schematic diagram of an HDD without a base, with major components annotated. The disk is mounted onto a spindle. A special electromagnetic reading/writing device called a magnetic head is fabricated on a slider. The slider is mounted onto a suspension and an actuator arm, which is in turn fixed to a base plate via a pivot. The ...
Context 2
... acceleration) and different pulse durations (pulse widths) from 0.04 to 4 ms were inputted into the numerical model. The acceleration level of 600g is close to the design shock limit of HDDs. Fig. 9 shows three cases of the input acceleration pulse loadings in which the pulse widths are 0.5 1 and 2 ms, respectively, shown as the abscissae. Fig. 10 shows time history of the relative displacement for the three different pulse widths of a single halfsine acceleration pulse loading with a 600g amplitude, corresponding to Fig. 9. From Fig. 10, it can be seen that the maximum relative displacement occurs at different times for different pulse widths with the same amplitude. Depending ...
Context 3
... limit of HDDs. Fig. 9 shows three cases of the input acceleration pulse loadings in which the pulse widths are 0.5 1 and 2 ms, respectively, shown as the abscissae. Fig. 10 shows time history of the relative displacement for the three different pulse widths of a single halfsine acceleration pulse loading with a 600g amplitude, corresponding to Fig. 9. From Fig. 10, it can be seen that the maximum relative displacement occurs at different times for different pulse widths with the same amplitude. Depending on the pulse width, the maximum relative displacement can occur at the first, second, or third oscillation of ...
Context 4
... variation of the maximum relative displacement (y max ) with different pulse widths for 600g pulses is shown in Fig. 11. It can be noted that, as the pulse width increases, the maximum relative displacement increases sharply for pulse widths less than 0.5 ms, reaches a peak value at a pulse width of 0.6 ms, decreases quickly to 0.2 mm at a pulse width of 1 ms, and, then, decreases slowly and seems to approach a constant value after about a 2 ms pulse ...
Context 5
... further investigate the pseudo-resonance phenomenon, three types of acceleration shocks different in shape (the half-sine, triangular and dual-quadratic waveforms) with an amplitude of 600g and with different pulse widths from 0.1 to 1.0 ms were applied (0.1 ms case shown in Fig. 12). Note that the dual-quadratic waveform consists of a rising portion in the form of bt 2 , where t is time and b a constant, and a falling portion as a mirror image of the foregoing. Fig. 13 shows the variation of the maximum relative displacements (y max ) of the actuator arm subjected to these three types of acceleration shocks with ...
Context 6
... triangular and dual-quadratic waveforms) with an amplitude of 600g and with different pulse widths from 0.1 to 1.0 ms were applied (0.1 ms case shown in Fig. 12). Note that the dual-quadratic waveform consists of a rising portion in the form of bt 2 , where t is time and b a constant, and a falling portion as a mirror image of the foregoing. Fig. 13 shows the variation of the maximum relative displacements (y max ) of the actuator arm subjected to these three types of acceleration shocks with different pulse widths of 0.1, 0.2, 0.4, 0.6, 0.8, and 1.0 ms. It can be noticed that, the maximum relative displacement experiences a peak value with an increase of pulse width, i.e., a ...
Context 7
... explain the phenomena observed above, a power spectrum analysis was conducted for the different acceleration shocks. Fig. 14 shows the acceleration power spectrum curves of these acceleration shocks different in shape with a pulse width of 0.1 ms. It can be found that, at the resonant frequency of 1.3 kHz of the actuator arm, the acceleration power magnitude of the half-sine acceleration waveform is the largest, followed by those of triangular and ...
Context 8
... different in shape with a pulse width of 0.1 ms. It can be found that, at the resonant frequency of 1.3 kHz of the actuator arm, the acceleration power magnitude of the half-sine acceleration waveform is the largest, followed by those of triangular and dual-quadratic waveforms. This relative relation of the acceleration powers is also shown in Fig. 15 and is consistent with that of the corresponding maximum relative displacements as shown in Fig. 13 for the case of 0.1-ms pulse ...
Context 9
... of 1.3 kHz of the actuator arm, the acceleration power magnitude of the half-sine acceleration waveform is the largest, followed by those of triangular and dual-quadratic waveforms. This relative relation of the acceleration powers is also shown in Fig. 15 and is consistent with that of the corresponding maximum relative displacements as shown in Fig. 13 for the case of 0.1-ms pulse ...
Context 10
... is also noticed, from Fig. 14, that there is a cross-over point (or small interval) for the three different power spectrum curves at a frequency value (location) of 1/T, where T is the pulse width. To prove the ...
Context 11
... When the characteristic frequencies of a group of acceleration shocks with different pulse shapes are very close to the resonant frequency of the dynamic system, these acceleration shocks will have nearly equal acceleration powers at the resonant frequency; consequently, they will produce nearly equal shock responses. Fig. 15 shows the variations of acceleration power at the resonant frequency of 1.3 kHz for these three types of acceleration shocks with pulse widths of 0.1, 0.2, 0.4, 0.6, 0.8, and 1 ms. It can be found that, for the 1-ms pulse width and a resonant frequency of 1.3 kHz, the acceleration power magnitude of the dual-quadratic acceleration ...
Context 12
... dual-quadratic acceleration waveform is the largest, followed by those of triangular and half-sine waveforms. This relative relation of the acceleration powers for the 1-ms pulse width is in a reverse sequence of that for the case of 0.1-ms pulse width, but it is consistent with that of the corresponding maximum relative displacements as shown in Fig. 13 for the case of 1-ms pulse ...
Context 13
... can also be noticed from Fig. 15 that, for both the half-sine and the triangular acceleration shocks, the maximum acceleration powers, for a resonant frequency of 1.3 kHz, are reached at the approximately same pulse width of about 0.6-ms. However, for the dual-quadratic acceleration shock, it is reached at about 0.8-ms pulse width. This observation of the acceleration ...
Context 14
... maximum acceleration powers, for a resonant frequency of 1.3 kHz, are reached at the approximately same pulse width of about 0.6-ms. However, for the dual-quadratic acceleration shock, it is reached at about 0.8-ms pulse width. This observation of the acceleration powers gives a reasonable explanation to the pseudoresonance phenomena observed in Fig. ...
Context 15
... it can also be found that there is a narrow cross-over interval for the three curves in Fig. 15 around a pulse width of 0.8 ms, and a similar narrow cross-over interval also exists in Fig. 13 around the same pulse width. This is because when the pulse width is approximately equal to 0.8 ms, the characteristic frequencies (1=0:8 ¼ 1:25 kHz) of these acceleration shocks are very close to the resonant frequency (1.3 kHz) of the ...
Context 16
... it can also be found that there is a narrow cross-over interval for the three curves in Fig. 15 around a pulse width of 0.8 ms, and a similar narrow cross-over interval also exists in Fig. 13 around the same pulse width. This is because when the pulse width is approximately equal to 0.8 ms, the characteristic frequencies (1=0:8 ¼ 1:25 kHz) of these acceleration shocks are very close to the resonant frequency (1.3 kHz) of the actuator arm. These observations confirm the prediction of the corollary stated above. ...

Citations

... The stiffness and damping ratio of the FDB motor were determined in a previous study (Rai and Bogy 2012). Also, the stiffness of the pivot bearing was cited by Shi et al. (2007). The operational shock simulation was carried out via the large-mass method, which is generally used to simulate shock conditions (Ao et al. 2009). ...
Article
A hard disk drive (HDD) is very sensitive to shock. Increasing portability demands have led to increased HDD exposure to unexpected shocks. Therefore, the dynamic characteristics of an HDD were utilized to investigate the relative behavior of the disk and head stack assembly (HSA) during operational shock. A finite element model of HDD was constructed to simulate operational shock. This model included the spindle system, base, HSA, and disk. The relative behavior of the disk and HSA was analyzed using different bases with different stiffness. A drop test was performed to verify the simulation results. A modified base design was then proposed to protect against contact between the disk and HSA in HDD.
... Shock response of the hard disk drive (HDD) was investigated in Refs. [152,153]. Dropped from a certain height, a HDD hit the ground with a certain speed. The impact during the head slap often led to failure of the Head Actuator Assembly (HAA). ...
Article
Full-text available
This paper reviews some topics related to the advances and applications of structural impact dynamics in recent years. Dynamic behavior of structural members including tubes, beams and plates under axial or transverse loading, and cellular materials and sandwich structures under impact or blast loading are summarized here. The research methodology involves experimental studies, theoretical modeling, as well as numerical simulations. However, as we mainly focus on the longer time dynamic responses of structures and cellular materials, studies of stress wave propagation and the material's strain-rate sensitivity are not included.
Article
This paper proposes a new technique to analyze the resonance vibration problem, which is a major problem in Hard Disk Drive Arm (HDD arm), by using a multi-grid, Finite Element Analysis (FEA) with PSO (Particle Swarm Optimization) based non-uniform refinement level of mesh partitioning. Summation of error between the measured values of resonance frequencies from Laser Doppler Vibrometer (LDV) and the simulated values from the developed FEM program is formulated as the objective function in the optimization problem. Particle Swarm Optimization (PSO) is adopted to find the optimal characteristic lengths to achieve both fast simulation time and acceptable accuracy of the result. The proposed technique can reduce the computational time in the analysis and is adopted to study the behavior of resonance vibration affected by the epoxy thickness. The results show that the developed program can effectively simulate the resonance behavior in HDD arm. In addition, the computational results by the proposed method are obtained faster than those by the uniform refinement level method while the accuracy of results by both methods is almost the same.
Article
Drop impact crashworthiness robust design method was presented based on Taguchi method. A 2.5 inch mobile hard disk finite element modal was built using FEM. The key parameters affecting the crashworthiness robustness were analyzed. The Taguchi method was used to make robust design. Optimal parameters with reducing influence of floor rigidity were acquired. The simulation results show that the crashworthiness of mobile hard disk has been obviously improved comparing with former general parameters. This research work provides theoretic evidence for improving crashworthiness of mobile hard disk.
Article
Mobile hard disk is very sensitive to vibration because of the unique working way of its mechanical structure and the complexity of its components. According to its inner structure and complying with mass equivalent principle, the finite element model of a 2.5 inch mobile hard disk in non-operational state was established based on ANSYS/LS-DYNA. The crashworthiness evaluation index was analyzed and set forward. The influences of the material and shape of the head actuator arm assembly as well as the drop angle were analyzed through simulation. The study provides a theoretical basis for crashworthiness analysis, evaluation and design of mobile hard disks.
Article
The finite element model of the head disk system (HDS) of the hard disk driver was established based on ANSYS/LS-DYNA. The dynamic characteristics of the HDS in operation state under shock were studied. The influence of dynamic response for the sliders key parameters was analysed with different disk constraints and suspension prestress. The results show that: the relative displacement response between the sliders and disk in the Z direction decreased with the increase of the disk freedom restriction. The significant influence of relative displacement between upper sliders and disk for suspension prestress was showed under the fully and simply supported constraints for the disk. The result showed significant influence of relative displacement between lower sliders and disk for suspension prestress in shock stage.
Article
In order to study the dynamic characteristics of a hard disk drive subjected to external shock, a dynamic model was established to describe the head/disk interface with slider suspension and the lift-tab/ramp interface. The dynamic performance of the system subjected to external shocks with half-sinusoidal shape for different durations and amplitudes were simulated. The dynamic response of the head slider under the shocks and the main factors affecting the air bearing force were discussed. The simulation results indicated that during shock, the squeezing effect of the thin film at the head/disk interface influences the air bearing force, and the variation of the shock duration and amplitude effects the stability of the slider. The simulation results provided a theoretical reference and analytical method for design of hard disk drives with good anti-shock performance.
Article
Hard disk drive (HDD) is the most important storage system in computers. Due to the importance and complexity of the problems of shock and vibration resistance in HDDs, many researchers have been investigating on these problems in various aspects. This paper begins with a brief introduction of the basic structure and the working principle of HDDs, followed by the static/dynamic characteristics of the magnetic head/disk system. The main factors and mechanisms that influence the properties of the head/disk system are summarized and analyzed. A literature review is presented and analyzed of the investigation on the shock and vibration robustness of HDDs. Finally, the investigation on controlling mechanisms on shock and vibration of HDDs, as well as the critical issue in HDDs and the possible solutions for future work are discussed.
Article
Crashworthiness is an increasingly important design consideration for operational hard disk drive (HDD). The simulation model of HDD is first investigated by the finite element method. Maximum flying height, pitch angle and roll angle, maximum stress are put forward as indexes for HDD response to drop impact. Sometime, these indexes are crashworthiness evaluation parameters. The drop impact response of the finite element model under different surface drops and different angle drops are presented. Simulation results are proven to be fundamental method of crashworthiness analysis, evaluation and design for HDD.
Article
This paper presents a modified design for a stamped base for a 2.5-in hard disk drive (HDD). A stamped base is a thinner thickness and lower stiffness than a die-casting base. To improve shock performance of the stamped base, the finite element (FE) model of the HDD is constructed. Then, the FE model was verified by modal analysis. Drop test was performed to confirm the shock simulation model. Parametric analysis was used to identify the key parameters for the shock performance which were the base thickness, rib shape, number of ribs, and reinforcement. The correlation between the relative displacement and natural frequency of the HDD was investigated for each parameter. The dominant parameter was identified and applied to the FE model of the stamped base. The modified stamped-base model was then evaluated under nonoperational shock conditions.