• Tribology and Lubricants
• /
• 제16권5호
• /
• pp.395-402
• /
• 2000

The effect of particulate contamination on friction and wear between a negative-pressure picoslider and a laser-textured disk was studied. Particles of different concentrations, materials and sizes were injected to the head-disk interface (HDI), consisting of disks with various textures, at the same speed. In a contaminated environment, durability of head-disk interface gradually decreased as the particle concentration increased. Large particles caused HDI failure early and resulted in an extensive damage to the slider and disk surfaces. Hard particles also caused HDI failure earlier and damages more extensive than soft ones. Based on the test results, mechanisms of HDI failure with picoslider were presented.

• Journal of Mechanical Science and Technology
• /
• 제15권3호
• /
• pp.281-290
• /
• 2001

Particle contamination has been an ongoing problem affecting the reliability of the magnetic hard disk drives. Especially the recent use of MR head requires much tighter control of particle contamination due to thermal asperity (TA) phenomenon. In this study, the effects of slider air bearing surface design of TA reduction capability were investigated by manufacturing two types of sliders. Numerical methods were used to simulate the motion of particles in the head/media interface. Experiments were conducted to verify the results predicted by the numerical simulation. Drives were built and exposed to particle contamination using a particle injection chamber, which turned out to be a very simple and reliable particle generation method over conventional aerosol technique. Then the number of TA generated in the drives was recorded and compared. Also the contacts between slider and particles were investigated by acoustic emission study. It was found that a new ABS design, which has aerodynamic U-shaped rail and central flow passage, was beneficial in reducing the particle contamination on the slider.

• 대한기계학회논문집B
• /
• 제26권8호
• /
• pp.1069-1076
• /
• 2002

Contamination particles in a hard disk drive can cause serious problems including slider crash and thermal asperities. A recirculation filter is typically installed in the hard disk drive to remove the particles. Measurements and theoretical predictions of particle concentration decay with the filter are carried out for a commercially available HDD. Especially, the effect of disk rotational speed on the particle capture efficiency is investigated. Results show that filter efficiency is higher for higher disk rotational speed.

• 한국정밀공학회지
• /
• 제19권8호
• /
• pp.26-33
• /
• 2002

• 대한기계학회논문집B
• /
• 제24권3호
• /
• pp.477-484
• /
• 2000

Microcontamination caused by particle deposition on the head disk interface threatens the reliability of hard disk drive. Design of slider rail to control contamination becomes an important issue in magnetic recording. In this paper, how particles adhere to the slider and the disk is examined. To investigate accumulation mechanism of the particles, trajectory of the particles in a slider/disk interface is simulated with considering various forces including drag force, gravitational force, Saffman lift force, and electrostatic force. It is found that the charged particles can easily adhere to the slider or disk surface, if an electric field exists between the slider and the disk. It is supposed that the vertical motion of the particles should be related with not only Saffman force but also electrostatic force.

• Tribology and Lubricants
• /
• 제16권6호
• /
• pp.469-476
• /
• 2000

The effect of particulate contamination on the friction and wear between a negative-pressure picoslider/tri-pad nanoslider and laser-textured disk was studied. Particles of different concentration were injected at the head-disk interface consisting of disks with various textures and slider types at different speed. Durability increased and coefficient of friction decreased as the disk speed increased in a contaminated environment. Frictional characteristics and durability in the data Bone were better than those in the laser-textured zone. It was also found that durability of head-disk interface (HDI) decreased as the particle concentration increased. The interface durability with a picoslider was better than that with a nanoslider at any condition in a contaminated environment. Based on the test results, mechanisms were proposed to explain the reasons why durability with a picoslider was superior to that with a nanoslider.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2005년도 추계학술대회 논문집
• /
• pp.484-487
• /
• 2005

In order to get the surface characteristics of the HDI of HDD, the surface damage mechanisms must be totally understood. Particle contamination in hard disk drives is a big concern in today's magnetic recording industry since they are major sources of reliability problems. Namely upon contact with the slider or a contaminant particle, the disk may be scratched or the particles may be embedded into the disk surface. In this work, comparison of scratches was made between those found on actual hard disks and those created using a scratch tester. It was found that ramp loading method is an effective way to make similar scratches as the actual ones. From the ramp loading condition, the relationship between the pressure and the scratch track width could be identified.

• Journal of information and communication convergence engineering
• /
• 제21권3호
• /
• pp.208-215
• /
• 2023

Deep learning techniques provide powerful solutions to several pattern-recognition problems, including Raman spectral classification. However, these networks require large amounts of labeled data to perform well. Labeled data, which are typically obtained in a laboratory, can potentially be alleviated by data augmentation. This study investigated various data augmentation techniques and applied multiple deep learning methods to Raman spectral classification. Raman spectra yield fingerprint-like information about chemical compositions, but are prone to noise when the particles of the material are small. Five augmentation models were investigated to build robust deep learning classifiers: weighted sums of spectral signals, imitated chemical backgrounds, extended multiplicative signal augmentation, and generated Gaussian and Poisson-distributed noise. We compared the performance of nine state-of-the-art convolutional neural networks with all the augmentation techniques. The LeNet5 models with background noise augmentation yielded the highest accuracy when tested on real-world Raman spectral classification at 88.33% accuracy. A class activation map of the model was generated to provide a qualitative observation of the results.