Dissertations / Theses on the topic 'Disc height'

This thesis presents a study into the use of microgels to restore degenerated intervertebral discs (IVDs). This was undertaken using a pH-triggered microgel which was able to form self-supporting gels through an increase in pH. The microgels were based on the poly(A/MAA/X) formulation, where MAA is methacrylic acid and A is the structural and X is the crosslinking monomer. The microgel particles were also used to construct composite poly(ethylene glycol) dimethacrylate (PEGD) hydrogels. The properties of the microgels and the composite hydrogels were then investigated. Microgel particles were synthesised based on poly(EA/MAA/BDD) (poly(ethyl acrylate/MAA/butanediol diacrylate). These were able to swell on increasing pH. Concentrated dispersions formed a gel with a high elastic modulus. The EA and BDD were replaced with related monomers and gave gels with different properties. Using monomers with high glass transition temperatures reduced the rate of swelling, and using monomers with similar reactivity ratios appeared to produce more uniformly crosslinked particles. It is proposed from the data presented that those with a large difference in reactivity ratio resulted in microgel particles with a change in crosslinking gradient through the radius of the particle. In some cases this produced microgels which appeared to fragment on increasing pH. The microgels investigated were based on poly(EA/MAA/X), (E-X) with BDD, EGD (ethylene glycol dimethacrylate) and PEGD. The EGD and PEGD microgels were shown to fragment with increasing pH. Poly(EA/MAA/PEGD) dispersions were able to form a gel at a pH below the pKa which appeared to be an electrostatically repulsive gel. Following this work, it became apparent that the E-BDD microgel was the most ideal of all the microgels with gels giving low values tanδ and frequency dependence of tanδ (tanδ = G"/G', where G" is the viscous modulus and G' is the elastic modulus). It also appeared to give physical gels with the highest elastic modulus. This microgel was therefore used for the composite gels. The poly(EA/MAA/BDD) microgel was then used to form covalently-linked composite hydrogels with PEGD of different molecular weights. PEGD with a molecular weight less than 550 formed a hydrogel-linked microgel, with interpenetrating polymer chains. These composites had high G' values and swelling ratios. Using PEGD with molecular weights higher than 550 produced microgel-filled hydrogels which had high values for G' and swelling ratios. Furthermore, due to osmotic deswelling of the microgel particles, the dispersions underwent a gel-to-fluid transition prior to being heated with initiator and crosslinked. This meant that some of the formulations were injectable. The mixture of high molecular weight PEGD and microgel was therefore combined with an accelerator which enabled gel formation and crosslinking at physiological temperature. Composites formed under physiological conditions were then tested for their ability to support biomechanically meaningful loads using degenerated IVDs. The discs were then compressed and the compressive strain to measured. The results showed that the composite was able to restore the mechanical modulus and height of the degenerated disc, showing favourable results for future research.

A relatively new fracture toughness testing method called Straight Notched Disc Bending (SNDB) was used before for fracture testing of Ankara Andesite and Afyon Marble cores. In this work to investigate the applicability of the new method to other rock types. With a preliminary notch of 10 mm, straight notched disc type specimens with a diameter of 75 mm were loaded by three-point bending loads. Investigation of effect of specimen height on the stress intensity factor and fracture toughness was carried out. Specimen heights (B) between 18 &ndash
67 mm were tried for andesite and marble cylindrical specimens. Loading span, that is span/radius (S/R) ratio was changed between 0.6 - 0.9 for andesite specimens. Stress intensity factor for specimens was computed with ABAQUS program. Stress intensity factor was found to increase with increasing specimen diameter for a fixed span/radius ratio. Stress intensity factor decreased with increasing specimen height. Changing span was found to have no significant effect on fracture toughness of andesite. Fracture toughness was significantly lower for specimens with smaller height. The suggested testing height interval for this type of specimens was between height/diameter ratios of 0.49 &ndash
0.64. Results were compared to the results obtained by a well-known specimen geometry named semi-circular bend specimens (SCB) under three-point bending. SCB tests produced lower values for fracture toughness for both rock types. Fracture toughness was 0.99 MPa&
#8730
m for Ankara Andesite and 0.70 MPa&
#8730
m for Afyon Marble.

A detailed review of existing flying height or head disk spacing testing techniques has been made, and in-depth analyses of the working principles of the currently popular optical interferometry flying height testing methods are presented with simulation results. A new dual-beam normal incidence polarisation inteferometry method is then proposed. One advantage of this dual-beam polarisation interferometry is that it can be used for both the direct spacing measurement (DSM) method and relative displacement measuring (RDM) method. The RDM method is a good way to measure the head-disk spacing and the slider pitch or roll when a real magnetic disk is used for testing. The DSM method has the advantage that the absolute head-disk spacing can be observed and measured directly, especially in the case where it is difficult for the light beam to 'spot' the back surface of the head-slider. When used for the DSM method, the flying height can be measured down to contact without losing sensitivity. Slider pitch or roll can also be measured using the phase information. Another advantage of this polarisation interferometry is that, when used for the DSM method, with the measured intensity and phase information, the optical constants of the slider material can be determined, which is necessary to determine the flying height. By investigation of the application limits and potential problems of the intensity interferometry method, an improved intensity interferometry method is also proposed by using phase-shifting technique to improve the sensitivity of this method when the head-disk spacing is below 10 nm and near contact. An experimental testing system has been built to test the capability and effectiveness of the proposed interferometry methods. Experimental results are presented which show good agreement with the results gained from theoretical analyses and simulation.

As magnetic computer disks are developed to ever-greater data storage densities, the accuracy required for head positioning is moving beyond the accuracy provided by present technology using single-stage voice-coil motors in hard disk drives. This thesis details work to develop a novel active suspension arm with 2-dimensional actuation for use in advanced hard disk drives. The arm developed is capable of high-bandwidth data tracking as well as precision head flying height control motion. High-bandwidth data tracking is facilitated by the use of piezoelectric stack actuator, positioned closer to the head. The suspension arm is also capable of motion in the orthogonal axis. This motion represents active flying height control to maintain the correct altitude during drive operation. To characterise the suspension arm's structural dynamics, a high-resolution measurement system based on the optical beam deflection technique has been developed. This has enabled the accurate measurement of minute end-deflections of the suspension arm in 2-dimensions, to sub-nanometre resolution above noise. The design process of the suspension arm has led into the development of novel piezoelectric-actuated arms. In the work involving lead zirconate titanate (PZT) thick films as actuators, work in this thesis shows that reinforcing the films with fibre improves the overall actuation characteristics of the thick films. This discovery benefits applications such as structural health monitoring. The final suspension arm design has been adopted because it is simple in design, easier to integrate within current hard disk drive environment and easier to fabricate in mass. Closed-loop control algorithms based on proportional, integral and derivative (PID) controller techniques have been developed and implemented to demonstrate high bandwidths that have been achieved. The suspension arm developed presents an important solution in head-positioning technology in that it offers much higher bandwidths for data tracking and flying height control; both very essential in achieving even higher data storage densities on magnetic disks at much reduced head flying heights, compared to those in existing hard disk drives.

In high density magnetic hard disk drives, both fast track seeking and extremely accurate positioning of the read/write head are required. A new piezoelectric microactuator with transverse and lateral control of the head positioning system for high density hard disk drives is proposed. First, the structure of the new piezoelectric microactuator is illustrated. Design of the new microactuator is based on the axial deformation of piezoelectric elements for lateral motion and the bimorph actuation of piezoelectric elements for transverse motion. Next, a mathematical model of the microactuator system is defined. Static properties associated with the displacement of the system are evaluated and then dynamic system equations of the system are evaluated. Frequency response of the system is studied based on the dynamic system equations of the actuator system. Dynamic properties of the system with a variety of system parameters are evaluated. Finally, the controller design for the actuator is presented. Simulation results show that the new actuator achieves a maximum stroke of displacement of more than 0.2m with servo bandwidth of more than 5 kHz in the lateral direction and the flying height is decreased to less than 6 nm with resonance frequency of more than 100 kHz under the 0.5 % damping assumption. The new piezoelectric microactuator improves performance of high density hard disk drives by increasing servo bandwidth and decreasing flying height.

There is large variability between individuals on many measurable parameters of the optic disc. The retinal fundus is most commonly measured indirectly via slit-lamp biomicroscopic examination with a fundus spherical lens. Three optical methods of measuring the vertical disc height were compared: – Optical Coherence Tomographic (OCT) method – Calculation using a modified Littman’s equation: t = pqs – Measurement using five different double aspheric Volk fundus lenses on two different slit lamps (Haag-Streit and CSO) The aims of the study were: – To calculate a conversion equation of optic disc height measurement by clinical assessment using any one of five double aspheric fundus lenses at the slit-lamp – To compare the control mechanism of optic disc height measurement on OCT versus calculation using a modified Littman’s equation – To compare measurements between the Haag-Streit BM900 and CSO SL990 slit lamps – To compare accuracy and agreement of Volk conversion factors based on manufacturer magnification values to the control mechanism DESIGN & METHOD: A cross-sectional study was performed on patients presenting to the St John Eye Hospital from August ’11 to July ‘12. 100 eyes were measured for: – The optic nerve head vertical disc height on OCT – Vertical disc height (VDH) on Haag-Streit and CSO slit lamps with Volk super 66D, 78D, Superfield, Digital Wide Field, Digital High Magnification lenses – Axial length – Refraction Subsequently, 44 new eyes were measured and used to test agreement of the formulated equations. RESULTS: – Axial length was not found to have a correlation with VDH (p=0.64). – The modified Littman’s equation: t=pqs where p varies for each lens and q = 0.01306(axial length – 1.82), had very low correlation and poor agreement – Comparing formulated equations to the OCT VDH, the formulated Digital High Magnification and super 66D lenses were found to have the most accuracy – Measurements with the Volk Digital Wide Field lens were highly inaccurate and are therefore not recommended for vertical disc height measurement in future – Comparison of the fundus lens measurements using the two slit lamps showed alarmingly different results. There was no agreement between slit lamp measurements using four of the lenses CONCLUSIONS: – The use of axial length and other quantitative factors attained by measures of a schematic eye cannot be used accurately in clinical practice to determine fundus object sizes. – The slit lamp technique of measuring optic disc size with application of a simple equation yields useful results in close agreement to those yielded by expensive technology. It cannot replace accurate analysis using an OCT but is sufficient in aiding clinical decisions – Two important factors need to be taken into account when applying slit lamp fundus measurements: the type of lens used, and the slit lamp being used – as measurement standards are inconsistent between slip lamps

碩士
國立成功大學
機械工程學系碩博士班
97
An innovative magnetic module which is concurrently capable of performing as an actuator and a sensor is proposed and illustrated. The magnetic module is basically similar to a micro-scale solenoid coil which is employed to actively adjust the height of a seismic rotating disc used in a gyroscope so that the induced pitch angle by Coriolis effect, purely due to exerted angular excitation, can be accurately measured. That is, the micro-magnetic module acts like an actuator for height regulation on the seismic rotating disc by applying appropriate electric current to the primary coil of the Micro-Magnetic Height Adjuster (MMHA). On the other hand, the secondary coil pair of the MMHA acts as a sensor which can detect the gap change between the seismic rotating disc doped by metal material, and the micro-magnetic module.

碩士
國立臺灣大學
機械工程研究所
81
This research addresses the problem of the read/write head fly height control in a computer hard disk drive. The computer hard disk read/write head flies above the disk at a very small distance in order to pick up small magnetization changes on the recording surface. The distance between the head and the disk thus determines the available recording density of the media. Modern computer disk drive achieves minimum fly height by using the air-bearing technology without feedback control. The air- bearing is able to maintain the read/write head at a sub-micron distance above the disk surface. As the need for storage capacity increases, it is necessary to futher reduce the fly height, and certain form of feedback control becomes necessary. In this research, we develop a mean to measure the read/write head fly height in a commercially available hard disk drive. In order to perform controller design, we also develop a system model that includes both the slider dynamics and the air- bearing squeeze effect. The popular $H_\infty$ optimization procedure is then used for controller synthesis. Experimental results on the fly height measurement and simulation results on the fly height control confirm the feasibility of the approach.

碩士
元智大學
機械工程學系
91
The present objective of the research is to get the high density recording system. However, a higher recording density requires smaller head-disk spacing, and therefore lowering-flying heads and smoother disk surfaces are needed. Due to the lowered fly height of the slider, glide avalanche requirements of the disks are consequently being tightened while the stiction requirement remain unchanged. In this article, the limitations of padded slider to meet stiction requirement on low glide height media tribology issue were investigate. In order to meet the glide avalanche requirement, to change the laser energy and its size of the laser texturing machine to get a concaved laser bump geometry, it can be seen that such design can get lower glide height but higher stiction. Therefore , to get the optimal design of such head/disk interface to get lower stiction is necessary. In the research, we observed that smaller bump spacing, bigger bump diameter, lower concave depth and fewer lubricant apply can reduce the contact area and meniscus of the head/disk interface, and exhibits better CSS performance in the test. In terms of the stiction model, it can be seen that more uniform asperity height distribution and lower lubricant tensor can get lower stiction force., Hence the better understood CSS technology is expected to be further extended into future high-performance disk drives.

碩士
國立交通大學
機械工程系
90
Based on near-field optics, a flying pickup head is employed in optical disk drives, so as to increase data storage capacity and density. However, it remains to enhance flying height control performance of the pickup head. This study attaches a piezoelectric bender to the pickup head. System identification is carried out to obtain a transfer function as the plant model. For real-time control, this study uses a DSP motion control card to execute control algorithm and uses a double-beam laser doppler interferometer for sensing head-disk spacing. A PID control method deals with the PZT bender motion to maintain a stable flying height, in spite of optical disk deformation and flying head vibration during near-field disk rotation at 5400rpm.

Tese de mestrado em Física, apresentada à Universidade de Lisboa, através da Faculdade de Ciências, em 2018
As galáxias são os blocos fundamentais constituintes do Universo e podem ser observadas no céu com diferentes formas, tamanho, brilho, inclinação, cor e composição química. Estes parâmetros permitem classificá-las em diferentes classes de acordo com o esquema de Hubble: elíticas, espirais (barradas) e irregulares. Estudar a Via Láctea tem muito interesse porque é a única galáxia gigante que podemos ver e estudar em detalhe, embora estudar a sua estrutura apresente os seus desafios, uma vez que apenas é possível estudá-la a partir do seu interior. Assim, as suas propriedades globais apenas podem ser determinadas das observações. Hoje em dia, a Via Láctea é estabelecida como sendo uma galáxia espiral, com uma barra central e contabiliza no total quatro braços espirais de grande desenho. Nesta é possível distinguir três estruturas principais: o bojo central, o halo, onde se encontram as estrelas de menor metalicidade, juntamente com os enxames de estrelas globulares e o disco, o cerne da formação estelar e é principalmente composto por gás, estrelas e enxames de estrelas abertos. Temos ainda no seu centro a existência de um buraco negro supermassivo. Todavia, as galáxias passam por mudanças morfológicas durante a sua evolução. Observar estas mudanças e identificar os mecanismos responsáveis são o cerne para entender como estas evoluem. Uma destas propriedades é a espessura do disco, definida através da escala de altura da distribuição de objetos acima e abaixo do plano Galático formal, que corresponde à linha imaginária que define o zero da altura. É de salientar que observacionalmente diferentes populações de objetos se distribuem com escalas de altura aparentes distintas: enxames abertos definem escalas de altura de cerca de 57 pc, nebulas planetárias entre 200 e 250 pc, estrelas do tipo OB aproximadamente 40 pc e nuvens moleculares gigantes entre 30 e 50 pc. Para comparação, os valores determinados para a altura do Sol em relação ao plano Galático variam entre 15 e 35 pc acima do plano Galático formal. O estudo da evolução da escala de altura pode ser feito com diferentes populações de objetos da Via Láctea. Em particular, os enxames de estrelas abertos são escolhidos por ser possível determinar as suas idades e distâncias com maior precisão do que estrelas individuais. Em contrapartida, com o aumento da distancia, o número de aglomerados mais velhos observados tem tendência a diminuir. Este efeito deve-se ao facto que os aglomerados mais velhos são constituídos por estrelas mais velhas e por isso menos brilhantes, o que resulta com que estes sejam mais desvanecidos quando comparados com aglomerados mais jovens. Assim, tornam-se mais difíceis de encontrar nas regiões de baixa latitude, onde existe maior absorção do meio interstelar e também maior confusão com as estrelas de campo. Como consequência, a escala de altura observada do disco é artificialmente maior quando se consideram amostras de aglomerados com distâncias maiores. Observa-se que as populações de aglomerados abertos definem escalas de altura que aumentam com a idade da amostra considerada. Tipicamente, isto é associado ao engrossamento do disco Galático, um fenómeno descrito por uma série de mecanismos que acontecem na Galáxia, responsáveis pelo aumento da altura dos aglomerados estelares em relação ao plano Galáctico formal. Porém, reparamos que as observações parecem mostrar que isto poderá ser uma consequência da destruição seletiva de aglomerados que se encontram mais perto do plano Galático formal, onde os efeitos disruptivos têm uma maior amplitude, fazendo com que a distribuição de aglomerados se torne mais larga e consequentemente a sua escala de altura aumente. Em suma, os aglomerados abertos estão sujeitos a diferentes efeitos de dissolução, sendo estes a perda de gás primordial, evolução secular, forças de maré, encontros com braços espirais, com a barra e com nuvens moleculares gigantes, sendo o último o mecanismo dominante. Este projeto propõe que o principal mecanismo responsável pelo aumento da escala de altura do disco Galático seja explicado pela colisão de aglomerados abertos com nuvens moleculares gigantes, tendo como consequência a perda de estrelas ou mesmo a disrupção total do aglomerado. O mecanismo proposto para explicar o aumento da escala de altura do disco é testado com um modelo computacional que se baseia na dinâmica vertical dos aglomerados abertos em torno do plano Galático. Este consiste em gerar aglomerados abertos distribuídos acima e abaixo do plano Galático que seguem uma certa distribuição de alturas inicial, num potencial realista da Via Láctea. Durante as suas órbitas, os aglomerados abertos estarão sujeitos a processos disruptivos, que apenas dependem da distância ao plano Galático e será tanto maior quanto mais próximos se encontrarem deste. A taxa de nascimento de aglomerados abertos é mantida constante durante toda a simulação, com um número suficientemente grande de forma a que os resultados estatísticos sejam robustos. O modelo computacional apresenta três parâmetros livres que são necessários calibrar: a escala de altura de nascimento, responsável pela distribuição inicial de alturas dos aglomerados abertos, a escala de altura de dissolução, associada à combinação de todos os mecanismos de dissolução de aglomerados estelares, e uma amplitude de dissolução, que controla a probabilidade global de um aglomerado aberto se desfazer à medida que se desloca ao longo da sua ´orbita. Para tal, utiliza-se um catálogo observacional de aglomerados abertos, dividido em três amostras delimitadas por cortes cilíndricos de distâncias heliocêntricas de 1000, 1750 e 2500 pc. Por sua vez, cada amostra é dividida em quatro grupos de idade distintos: aglomerados jovens, aqueles que têm idade inferior a 200 Myr, aglomerados intermédios-jovens, com idades compreendidas entre 200 e 450 Myr, aglomerados intermédios-velhos, com idades compreendidas entre 450 e 1000 Myr e aglomerados velhos, com idades superiores a 1 Gyr. A amostra de 1000 pc, uma vez que está confinada a um volume menor, apresenta um menor número de aglomerados abertos, porém estará mais completa, no sentido em que apresenta menos aglomerados não detetados. Por outro lado, as amostras com distâncias heliocêntricas maiores, apresentam menor completitude mas estará sujeita a menos flutuações e estruturas locais. Com as escalas de altura definidas pelos diferentes grupos de idade, para as diferentes amostras, são determinados os três parâmetros livres do modelo computacional de forma que este consiga reproduzir o melhor possível o catálogo observacional. Posto isto, normaliza-se também o número total e em cada grupo de idade de aglomerados abertos obtidos no fim das simulações, ao assumir que o grupo de aglomerados jovens encontra-se completo para as três amostras. Enquanto a escala de altura de nascimento consegue ser determinada de forma independente, a escala de altura de dissolução e a respetiva amplitude são determinados em simultâneo. Com o conjunto de parâmetros diferentes para cada amostra do catálogo, observa-se que o modelo computacional consegue reproduzir bem o valor da escala de altura definida por todos os grupos de idade, com a exceção dos aglomerados abertos mais velhos, em todos as amostras, pelo que se pode concluir que o mecanismo de dissolução de aglomerados abertos em encontros com nuvens moleculares gigantes é viável para explicar o aumento da escala de altura do disco Galático até 1 Gyr. Além disso, o modelo computacional consegue fazer estimativas da completitude dos aglomerados abertos em cada grupo de idades, para as diferentes amostras. Os resultados das simulações reproduzem a evolução da escala de altura do disco Galático até idades de aproximadamente 1 Gyr. Após esta idade, estas afastam-se das observações. Este efeito vai no sentido do que é esperado, tendo em conta que a incompletitude do catálogo observacional tende a aumentar com a idade dos objetos e que pode resultar num aumento artificial da escala de altura observada do disco. A complexidade do modelo computacional também desempenha um papel importante nestes resultados. Crê-se que a utilização de uma distribuição inicial de massas para os aglomerados abertos, juntamente com uma perda parcial de massa resultante da colisão com nuvens moleculares descreva melhor a realidade. Mais ainda, outros efeitos de disrupção, nomeadamente a evolução secular, que corresponde à perda de estrelas do aglomerado por evaporação, ajudará na redução do número de aglomerados obtidos na simulação, o que permite fazer previsões mais exatas na completitude do catálogo observacional considerado.
Throughout their evolution, galaxies experience morphological changes. Tracking these changes and identifying the mechanisms that drive them are thus at the heart of understanding how galaxies evolve. The Milky Way is important as it is the only giant galaxy that we can see and study in detail. However, studying the structure of the Milky Way is challenging, as we only able to study it from the inside and must derive its global properties from observations. One of these properties is the thickness of the disk, described by the scale height of the vertical distribution of objects above and below the Galactic plane. In particular, open clusters (OCs) are often used as probes because their distances and ages can be derived much more precisely than individual stars. The older populations of OCs define larger scale heights than those of the young clusters. This is often attributed to the thickening of the Galactic disk, due to mechanisms that take place in the Galaxy. However, we notice that observations also suggest that this simply a consequence of selective destruction of OCs at lower heights, where the disruptive effects have an higher intensity, making the vertical distribution wider and thus the observed scale height would increase. This project’s contribution is a computational model to test the proposed mechanism for increasing the scale height of the Galactic disk, using the vertical dynamics of OCs which may disrupt when passing through giant molecular clouds (GMCs). The results from the simulations explain the evolution of the scale height of the disk up to ages of 1 Gyr. Hereafter, these deviate from the observations. Hereafter, these deviate from the observations and it is needed to improve the complexity of the computational model. This is to be done in the future.

碩士
國立交通大學
機械工程系
90
In order to overcome the diffraction limit of conventional optical disk drives, the flying pickup head in near-field optical disk drives enables the slider to fly at a stable spacing above the disk surface, thereby substantially increasing storage capacity and density. Hence, enhancing control performance to improve focusing speed and accuracy is required. To sense flying height in control experiments, a gap capacitance servo measures capacitance variation between the pickup head and disk surface and process voltage signals out of the capacitor by demodulation. Sliding mode control inherited with robust properties can deal with model uncertainty and disturbances. Therefore, a DSP control card executes a sliding mode control algorithm to compensate the vibration of the optical disk surface and achieve a stable flying height for the pickup head.

碩士
國立交通大學
機械工程系所
93
To realize near-field optics and hence increase storage capacity and density in near-field optical disk drives, the flying pickup head has to enable the slider to fly at a stable height above the disk surface. Since both the precision of track pitch and the flying height reach nanoscale, it remains to increase the motion accuracy of a pickup head. In this study, a PZT bender is used as an actuator of the pickup head while quadrant photodetectors are used as sensors. Based on an optical lever method and addition-and-subtraction operations in op-amp circuits, the relative position variation between a disk and a pickup head is measured. This study designs a sliding mode controller inherited with robust properties to deal with model uncertainty and external disturbances, so as to maintain stable flying height.

博士
國立交通大學
機械工程系所
93
This study aims to control the flying height of pickup heads in near-field optical disk drives. Near-field optical disk drives apply the near-field optics theory to replace conventional optical disk drives, thereby substantially increasing data storage capacity. In order to achieve near-field optical recoding, the distance between the objective lens and disk surface, i.e. the flying height must keep within the range of sub-wavelength. As a flying height actuator, a piezoelectric bender is used in this study to complement an air bearing at the head/disk interface to perform the feedback control in order to decrease flying height variation due to disk vibration and tilt. Firstly, this study aims to develop a sliding-mode based learning controller for eliminating repetitive error in flying height control. It incorporates characteristics of sliding mode control into learning control. The reason for using sliding mode control is attributed to its robust properties dealing with model uncertainty and disturbances. The learning algorithm utilizes shape functions to approximate influence functions in integral transforms and estimate the control input to reduce repetitive error. Thus, the repetitive component in the flying error can be compensated and a stable flying height can be maintained under a periodic disk vibration. In addition, this study presents a discrete frequency-shaped sliding mode control law for flying height control of near-field optical disk drives. Frequency shaping is conducted such that filtered state variables are penalized at the disk vibration frequency using inverse notch filters in LQR weighting functions. Hence, the controller can achieve stable flying height in the presence of certain frequency vibration of optical disks. Using the proposed methods, this study carries out experiments for flying height control with the pickup head. According to experimental results, the proposed methods are validated in comparison with conventional controllers.

博士
國立交通大學
機械工程系所
93
Conventional optical disk drives are faced with diffraction limit. To substantially increase data storage capacity and density, near-field optical disk drives remain to be realized. The slider of a flying pickup head in a near-field optical disk drive has to fly at a very low and stable height above the disk surface. To sense the slider flying height, a gap capacitance method is developed in this study to measure capacitance variation between the pickup head and disk surface. The capacitance varying with the flying height is modulated by a Colpitts oscillator. Subsequent demodulation accounts for height variation of the flying pickup head. Measurement results of this method are verified by using a laser Doppler interferometer. The interface between the slider and optical disk constitutes an air bearing. The air bearing force is affected by disk rotation speed and deformation, which represent low-frequency periodic disturbance to the slider. A repetitive control method is adopted to eliminate the periodic disturbance. H∞ control inherited with robust properties and sliding control can deal with model uncertainty and disturbances. Therefore, this study designs two controllers. One is a repetitive controller cascaded by an H∞ controller, while the other error-feedback sliding mode control. The proposed controllers are validated by simulation and experimental results.

碩士
國立臺灣大學
機械工程研究所
81
This research explores the possibility of using the piezo- electric actuator in the computer hard disk drive fly height control. As the demand reises on increased data storage capacity, people have investigated different methods for the hard disk fly height control. Even though most approaches face the difficulty of too much design change and lack of reliablility for a commercial product, the piezo-electric material still demonstrates the most desirable characteristics for this control. In this research, we study the characteristics of the piezo-electric material. We mount the actuator on the base of the suspension so that minimum change in the disk drive configuration is required. In this case the suspension dynamics has to be included as part of the actuator dynamics. An ANSYS finite element model is developed for suspension dynamics together with the piezo-electric material behavior. The simulation includes both static and dynamic characteristics. Experiments are then conducted to confirm the results.

碩士
國立交通大學
工學院精密與自動化工程學程
101
This thesis uses a PID controller to control the pickup head flying height in a near-field optical disk drive. Near-field optical theory applied to an optical pickup head can improve optical disk drive storage capacity. In order to achieve near-field optical recoding, the distance between the objective lens and disk surface, i.e. the flying height must keep within the range of sub-wavelength. In such small flying clearance, head disk crash risk will become higher and higher. This study uses National Instrument sbRIO-9602 embedded device using closed-loop control voice coil motor by the PID method to reduce the flying height variation induced by media waviness under critical clearance. The result successfully avoids the pickup head to crash media surface and reduces the flying height to 750nm.