Browsing by Issue Date, starting with "1992-05-20"

Now showing 1 - 20 of 90685

TRIANGULARLY PHASE-MODULATED OPTICAL FIBER RING RESONATOR SENSOR
(1992-05-20)
CHIEN, PY
;
PAN, CL
;
交大名義發表
;
光電工程學系
;
National Chiao Tung University
An optical fiber ring resonatory sensor system has been demonstrated by applying a triangular phase modulation signal to a fiber loop. The dynamic range for detection of optical phase change is 2-pi. The sensor is linear within this range. The minimum detectable optical phase change is 1. 5 x 10(-4) rad/Hz1/2.
GROWTH OF DIAMOND FROM CO2-(C2H2, CH4) GAS SYSTEMS, WITHOUT SUPPLYING ADDITIONAL HYDROGEN GAS
(1992-05-20)
CHEN, CF
;
LIN, CL
;
HONG, TM
;
交大名義發表
;
材料科學與工程學系
Recently, we have been searching for a new reaction path in diamond growth and discovered that diamond films can grow without adding extra hydrogen gas to the hydrocarbon feed. Using high pressure microwave chemical vapour deposition and gas systems of CH4-CO2 or C2H2-CO2, icosahedral polycrystalline particles were deposited and the surface of the films revealed a polycrystalline morphology with smooth (111) facets. The growth rate of the film is 3.11-mu-m h-1 in the CH4-CO2 system and 0.98-mu-m h-1 in the C2H2-CO2 system. The carbon ratio range of the former, which grows crystalline diamond, is broader than the latter ratio range. Furthermore, the growth rate of films in the C2H2-CO2 system is twice that of the CH4-H2 System, and similar to the maximum rate in the CO-H2 system. In the CH4-CO2 system, the maximum rate which resulted in crystalline diamond films is about 2-mu-m h-1, which is about four times that of the CH4-H2 system and twice that of the CO-H2 system. We believe that the crystallinity and quality of the film as determined by X-ray diffraction and cathodoluminescence spectral analysis are comparable with those of naturally occurring type IIa diamond.
DIRECT OBSERVATION OF SI DELTA-DOPED GAAS BY TRANSMISSION ELECTRON-MICROSCOPY
(1992-05-25)
LIU, DG
;
FAN, JC
;
LEE, CP
;
TSAI, CM
;
CHANG, KH
Direct observation of the Si delta-doped layer in GaAs has been achieved by high resolution transmission electron microscopy. Samples with different Si doses, from half a monolayer to two monolayers, were studied. The observed spreading of the delta-doped layer showed that Si atoms are largely confined in five monolayers at most (in the highest dose case), indicating excellent confinements of dopants in GaAs. From the images, the Si atoms were uniformly distributed in the doped layer, no cluster formation was observed. For delta-doped GaAs grown at low temperature (480-degrees-C), stacking faults originated from the doped layers were observed. These faults were thought to be caused by the large unrelaxed strain in the low-temperature grown GaAs.
MODEL-REDUCTION VIA FREQUENCY WEIGHTED BALANCED REALIZATION
(1992-06-01)
LIN, CA
;
CHIU, TY
;
電控工程研究所
;
Institute of Electrical and Control Engineering
We propose a model reduction method for multi-input multi-output linear time-invariant systems based on frequency weighting and balanced realization. We show that with both input-weighting and output-weighting, the method yields stable reduced order model, provided that the full order model is stable. Examples show the effectiveness of frequency weighting. Finally, our results are compared with those obtained by other frequency weighted model reduction methods.
A THEORETICAL-STUDY OF END EFFECTS FOR RECTANGULAR RESONATORS ON NARROW CHANNELS
(1992-06-01)
WU, YC
;
交大名義發表
;
土木工程學系
;
National Chiao Tung University
;
Department of Civil Engineering
In order to understand the end effects of rectangular resonators on narrow channels a linearized potential theory has been applied to study the deformation of waves by rectangular resonators in narrow channels, when the width of the channels is less than one half the incident wavelength. The "exact" solution is obtained by matched eigenfunction expansions; the resulting matrix equation is solved numerically. The numerical solution of the boundary integral equation method (BIEM) has also been obtained for comparison. The present two-dimensional mathematical model complements the previous laboratory investigation of James (J. Fluid Mech. 1970, 44, 615-621). The numerical results of the present analytical and numerical models are compared to James' laboratory measurements. Results show that both analytical solutions and BIEM solutions for the fundamental resonance mode are in reasonably good agreement with the laboratory experimental data. Therefore the considerable effect of main channel width on optimum resonator width and resonator length is further confirmed by theoretical analysis. The usual resonance theory, which predicts complete reflection on the incident wave train when the length of the branch canal is one quarter of a wavelength, was found incorrect by James' laboratory study. James' finding is proved theoretically by rigorous analytical and numerical models in the present study. In addition, higher resonance modes of extreme shape as well as intermediate shape rectangular resonators, not discovered in either James' laboratory investigation or his one-dimensional approximate analysis, were also discovered in the present study.
TEMPERATURE-DEPENDENCE OF THE ELECTRICAL-CONDUCTION IN RUO2-BASED THICK-FILM RESISTORS
(1992-06-01)
CHIOU, BS
;
SHEU, JY
;
電子工程學系及電子研究所
;
Department of Electronics Engineering and Institute of Electronics
In this paper, the temperature dependence of resistance of two generic RUO2-based resistors is investigated. The resistor compositions studied are 80 wt.% glass (63 wt.% PbO - 25 wt.% B2O3 - 12 wt.% SiO2, designated as G1) - 20 wt.% RuO2 and 80 wt.% glass (55.5 wt.% PbO - 22 wt.% B2O3 - 10.5 wt.% SiO2 - 12 wt.% Al2O3, designated as G2) - 20 wt.% RuO2. The sheet resistance of resistor 80 wt.% Gl - 20 wt.% RuO2 fired at 850-degrees-C decreases as the temperature is increased from 100 K to approximately 400 K, remains a minimum value at temperatures 400 K approximately 690 K, and then increases as the temperature is further raised. A negative temperature coefficient of resistance (TCR) of approximately -480 ppm/degrees-C is obtained from 100 K to 500 K. The TCR becomes less negative when temperature increases. Three models for conduction mechanism of thick film resistors are employed to explain the experimental results. A modified model, consisting of both tunneling and parallel conduction approaches, is proposed to elucidate the change in slope in the resistance-inverse temperature curve as well as the temperature dependence of the resistance. In addition, an equivalent circuit model is proposed to describe the electrical behavior of the thick film resistors.
E-POLARIZED SCATTERING FROM A CONDUCTING RECTANGULAR CYLINDER WITH AN INFINITE AXIAL SLOT FILLED BY A RESISTIVELY COATED DIELECTRIC STRIP
(1992-06-01)
WU, LK
;
HAN, LT
;
交大名義發表
;
傳播研究所
;
National Chiao Tung University
The potential use of resistive films for damping the resonance spikes observed in the radar cross section (RCS) spectrum of a partially open rectangular cavity is investigated using a recently developed finite-difference time-domain (FDTD) method that utilizes the resistive sheet boundary condition for the modeling of resistive films. Backscattering data obtained in the first resonant region for an E-polarized plane wave normally incident into the slotted side of the cavity are presented. It is shown that resonance behaviors can be eliminated completely with a low-resistance film (e.g., R(S) = 188.5-OMEGA) which attenuates significantly the impinging wave. Poorer resonance damping performance is observed as the film resistance increases to 754-OMEGA because more of the field is allowed to penetrate into the cavity. For the latter case, the presence of the resistive film lowers the Q-factor of the slotted cavity such thal the resultant resonance spectrum is lower in strength and broader in bandwidth.
ADAPTIVE-CONTROL OF NONLINEAR-SYSTEMS USING NEURAL NETWORKS
(1992-06-01)
CHEN, FC
;
KHALIL, HK
;
電控工程研究所
;
Institute of Electrical and Control Engineering
Layered networks are used in a nonlinear adaptive control problem. The plant is an unknown feedback-linearizable discrete-time system, represented by an input-output model. A state space model of the plant is obtained to define the zero dynamics, which are assumed to be stable. A linearizing feedback control is derived in terms of some unknown nonlinear functions. To identify these functions, it is assumed that they can be modelled by layered neural networks. The weights of the networks are updated and used to generate the control. A local convergence result is given. Computer simulations verify the theoretical result.
TRACKING A MANEUVERING TARGET USING INPUT ESTIMATION AT HIGH MEASUREMENT FREQUENCY
(1992-06-01)
GUU, JA
;
WEI, CH
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
In tracking a manoeuvring target by radar system, the measurement noise is significantly correlated when the measurement frequency is high. Bogler's input estimation method with decorrelation algorithm is used to track the manoeuvring target. It is found that the decorrelation process may improve the system performance significantly when the target is in a quiescent period or in steady state of the accelerating period. The transient performance may be improved effectively by a modified method denoted as 'modified CHP algorithm'. The performance in the quiescent period can also be improved by adding a quiescent detector to the system.
IMPLEMENTATION AND APPLICATION OF RESISTIVE SHEET BOUNDARY-CONDITION IN THE FINITE-DIFFERENCE TIME-DOMAIN METHOD
(1992-06-01)
WU, LK
;
HAN, LT
;
交大名義發表
;
傳播研究所
;
National Chiao Tung University
Use of resistive sheet boundary condition in the finite-difference time-domain (FDTD) analysis of scattering problems involving resistively coated dielectric object is described. The algorithm is introduced through an analysis of E-polarized scattering from a thin resistive strip. For the sheet resistance R = eta-Z0/2 with Z0 being the intrinsic impedance of vacuum, numerical experiments indicate that algorithm stability is ensured in all cases if delta-t is chosen according to c-eta-delta-t less-than-or-equal-to delta. Validity of the resultant FDTD method is verified in a comparison of computed E-polarized scattering data for several resistive strips with existing data. New results on the E-polarized scattering behavior of resistively coated dielectric strip as a function of surface resistances and angle of incidence are also presented. Finally, extension to the H-polarized case and application of the present method to pulsed problems are briefly discussed.
AN IDENTIFICATION METHOD FOR ESTIMATING THE INERTIA PARAMETERS OF A MANIPULATOR
(1992-06-01)
LIN, SK
;
交大名義發表
;
電控工程研究所
;
National Chiao Tung University
;
Institute of Electrical and Control Engineering
This article presents an off-line identification method to estimate the minimal knowledge of the inertia parameters for determining the dynamic model of a manipulator. A new approach is proposed to find a set of the minimal knowledge of the inertia parameters. This set is recursively estimated by moving one joint at a time. The off-line identification procedure also provides a sufficient condition for a persistently exciting trajectory. A simulation example of Stanford arm illustrates the validity and simplicity of the identification procedure.
TOUGHENING BEHAVIOR OF ELASTOMER-MODIFIED POLYCARBONATES BASED ON THE J-INTEGRAL
(1992-06-01)
LEE, CB
;
CHANG, FC
;
應用化學系
;
Department of Applied Chemistry
The J-integral method to determine the fracture toughness of tough and ductile polymeric materials previously developed has been applied to the elastomer-modified polycarbonates. This investigation compares three different methods to obtain Jc: the conventional crack growth length, the stress whitening zone, and the newly developed hysteresis method. Jc values obtained from these three comparative methods are fairly close. The hysteresis method has the advantage over the other two methods of obtaining Jc without requiring the measurement of the crack growth length or the stress whitening zone, therefore avoiding the controversy in defining crack blunting. Results also indicate that the effect of elastomer quantity in polycarbonate on Jc is insignificant as long as the crack is in a stable condition. Higher elastomer contents in polycarbonate result in higher dJ/d-DELTA-a, dJ/d-DELTA-l, and tearing modulus (Tm). This indicates that the elastomer toughening mechanism is due to the increase of the energy required for crack growth extension. The hysteresis loss energy is directly related to the size of the crack tip plastic zone, and the presence of more elastomer indeed increases the crack tip plastic zone, thus making the polycarbonate tougher. Besides, the presence of elastomer tends to increase the crack initiation displacement and shift the failure modes from an unstable fracture. Jc and the criterion for crack initiation based on rate change of hysteresis energy are discussed in detail.
VECTORIZABLE WAVE-PROPAGATION FORTRAN CODE FOR CALCULATIONS OF MULTIPHOTON DISSOCIATION
(1992-06-01)
TING, JJL
;
YUAN, JM
;
JIANG, TF
;
電子物理學系
;
Department of Electrophysics
A vectorizable FORTRAN code for the ETA10 or other vector machines for numerical calculation of the time-dependent Schrodinger equation is presented. Preliminary results for the multiphoton dissociation of nitrogen oxide are shown. The applied electric field was approximated by a classical sinusoidal wave; both sinusoidal and cosinusoidal wave forms were tested, and the internuclear potential was taken to be a Morse function. Dissociation histories of three types were found.
INFLUENCE OF CONTACT TREATMENTS ON THE ELECTRICAL CHARACTERISTICS OF SHALLOW-JUNCTION TITANIUM-BASED CONTACTS
(1992-06-01)
LIAUH, HR
;
TSENG, MF
;
CHEN, MC
;
CHEN, LJ
;
電子工程學系及電子研究所
A comparison of the electrical characteristics of Al-Si/TiW/TiSi2/Si, Al-Si/TiW/Ti/Si and Al-Si/Si contact structures has been carried out. The influence of a contact implantation and anneal on the electrical characteristics of the TiSi2 contacted shallow junctions has also been investigated. Compared with the Al-Si/TiW/TiSi2/Si contact system, the Al-Si/TiW/Ti/Si contact was shown to be a more suitable metalization scheme for the shallow junction contact technology, due to its lower contact resistance and junction leakage as well as its simplicity of fabrication process. The contact resistance of the Ti/n+-Si system was found to be lower than that of the TiSi2/n+-Si system. The low contact resistance of Ti/n+-Si is attributed to the presence of a uniform amorphous interlayer as well as significantly less influence of As on Ti/a-alloy/n+-Si compared to TiSi2/n+-Si. A contact anneal was found to reduce contact resistance for both TiSi2/n+-Si and TiSi2/p+-Si contact systems. With an additional P+ implantation and anneal, the contact resistances are drastically reduced for the n+-Si contacts. On the other hand, a B+ contact implantation and anneal did not significantly improve the contact performance of the TiSi2/p+-S system. The difference is thought to arise from the lower degree of recovery of B+ implantation damage resulted from the contact implantation of TiSi2/p+-Si contact compared to that of TiSi2/n+-Si contact. The addition of the contact P+ implantation increased the total dopant concentration underneath the surface to such an extent that it leads to a reduction in contact resistance.
AN O(KN) ALGORITHM FOR A CIRCULAR CONSECUTIVE-K-OUT-OF-N-F SYSTEM
(1992-06-01)
WU, JS
;
CHEN, RJ
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
An O(k.n) algorithm is described for evaluating the reliability of a circular consecutive-k-out-of-n:F system.
OPTIMIZATION OF DEPOSITING Y1BA2CU3O7-DELTA SUPERCONDUCTING THIN-FILMS BY EXCIMER LASER ABLATION WITH CO2 LASER-HEATED SUBSTRATES
(1992-06-01)
WU, KH
;
JUANG, JY
;
LEE, CL
;
LAI, TC
;
UEN, TM
The optimum deposition conditions for growing Y1Ba2Cu3O7-delta superconducting thin films in situ by using xenon, Nd:YAG, and KrF excimer lasers with CO2 laser-heated substrates were studied. The dependences of the T(c0) and microstructure of the films on the deposition parameters such as substrate temperature, oxygen partial pressure, laser energy density and repetition rates have been investigated systematically. It was found that both the surface mobility on the substrate and the reaction temperatures for the particulates arriving at the substrates are the key factors for obtaining high-quality films. High-quality Y1Ba2Cu3O7-delta thin films with mirror-like surface morphology, T(c0) almost-equal-to 90 K, and nearly perfect c-axis oriented normal to the film surface were routinely grown in situ on SrTiO3 substrates with the following deposition conditions: substrate temperature at 670-degrees-C, oxygen partial pressure of 0.1 torr, laser energy density of 4 J/cm2 per pulse, and repetition rate of 10 Hz. More significantly, it took only about 3 min to grow a 0.3-mu-m thick film and less than 50 s to cool down the film to room temperature after deposition by the present process. Subsequent processes, such as slow cooling in oxygen atmosphere or any further heat treatments, were unnecessary for the present process. The unique advantage of the present process further offered us good opportunities to delineate the effects of each deposition parameter on the detailed film growth mechanisms without having to deal with the complications introduced by prolonged post-deposition treatments. The details of all the interesting features observed will be presented and their physical implications will be discussed.
SYNTHESIS OF SIDE-CHAIN LIQUID-CRYSTALLINE POLYACRYLATES, POLYMETHACRYLATES AND POLYSILOXANES CONTAINING 4-CYANO-BIPHENYL 4-ALKANYLOXYBENZYL ETHER SIDE GROUPS
(1992-06-01)
JUANG, TM
;
CHANG, CF
;
HSU, CS
;
交大名義發表
;
應用化學系
The synthesis and characterization of side-chain liquid crystalline polyacrylates, polymethacrylates and polysiloxanes containing 4-cyanobiphenyl 4-alkanyloxybenzyl ether side groups are presented. All polymers display respectively a smectic A mesophase. The influence of the polymer backbone flexibility on the phase transition temperatures of the synthesized side-chain liquid crystalline polymers is discussed. The results demonstrate that flexible backbones enhance the decoupling of the motions of the side chain and main chain. Therefore, among three kinds of polymer backbones which contain the same mesogenic side groups, the most flexible one, i.e., polysiloxane, reveals a highest isotropization temperature and a widest temperature range of mesophase.
ACOUSTIC-EMISSION CHARACTERIZATION OF A MEDIUM STRENGTH STEEL DURING HYDROGEN PERMEATION PROCESSES
(1992-06-01)
WENG, CC
;
LIN, GC
;
CHEN, RT
;
交大名義發表
;
土木工程學系
The characteristics of acoustic emission (AE) signals detected from the hydrogen permeation test of a medium strength steel were investigated. The aims of this research are to study the correlation between the AE signals and the severity of hydrogen-induced damages, and to investigate the relationship between the AE signals and the variation in hydrogen permeation current. In this study, an electrochemical approach using Devanathan's dual-cell method was adopted for the hydrogen permeation test. A total of five specimens were tested with a charging current varying from 10 to 60 mA. From the test results, good correlation was observed between the detected AE signals and the severity of the hydrogen-induced blistering and cracking. Also, it was found that the variation in the hydrogen permeation current showed a close correlation with the detected AE energy and events.
DESIGN CONSIDERATIONS OF METAL-FILM BOLOMETER WITH MICROMACHINED FLOATING MEMBRANE
(1992-06-01)
SHIE, JS
;
WENG, PK
;
交大名義發表
;
光電工程學系
;
National Chiao Tung University
A metal-film microbolometer formed on a glass membrane suspended on a silicon substrate has been fabricated by the anisotropic etching technique. The rectangular glass membrane is supported by its four leads connected to the corners of an etched V-groove cavity. This structure allows flexible thermal management and sensitivity optimization of the bolometer for different kinds of application, with a trade-off between lead thermal impedance and device active area. Optimizations of the structure applied to both a focal-plane array and a single detector are interpreted. A thermal impedance as high as 4 x 10(5)-degrees-C/W with a response time of several milliseconds can be achieved. For a 320-mu-m x 320-mu-m detector, theoretical predictions give a responsivity and normalized detectivity of over 700 V/W and 10(9) cm square-root Hz/W, respectively. For dimensions of 128-mu-m x 128-mu-m, suitable for a focal-plane-array detector, theory predicts a D* of 5.5 x 10(8). A discrepancy exists between the experiment and the theory, which is also discussed.
STATE-SPACE SEARCH FOR HIGH-LEVEL CONTROL OF MACHINE VISION
(1992-06-01)
HWANG, SY
;
交大名義發表
;
資訊工程學系
;
National Chiao Tung University
;
Department of Computer Science
Computer vision is a task of information processing that can be modeled as a sequence of subtasks. A complete vision process can be constructed by synthesizing individual operators performing the subtasks. Previous work in computer vision has emphasized the development of individual operators for a specific subtask. However, the lack of knowledge about other levels of processing, while developing the operators for a specific level, makes the development of a robust operator and thus a robust system unlikely. To obtain vision problem-solving methods that are robust in the face of variations in image lighting, arrangements of objects, viewing parameters, etc., we can simply incorporate all possible sequences of image-processing operators, each of which deals with a specific situation of input images; then an adaptive control mechanism such as a state-space search procedure can be built into the methods. Such a procedure dynamically determines an optimal sequence of image-processing operators to classify an image or to put its parts into correspondence with a model or set of models. One critical problem in solving vision problems with a state-space search model is how to decide the costs of paths. This paper details the state-space search model of computer vision as well as the design of cost functions in terms of information distortions. A vision system, VISTAS, has been constructed under the state-space search model and its parallel version has been simulated.