Search

This research presents a finite-element analysis of the electromechanical field of a BLDC motor considering speed control and mechanical flexibility. The magnetic field is analyzed by the nonlinear time-stepping finite-element method considering the switching action of the pulse width modulation (PWM) inverter. Magnetic force and torque are calculated by the Maxwell stress tensor. Mechanical motion of a rotor is determined by a time-stepping finite-element method considering the flexibility of shaft, rotor, and bearing. Both magnetic and mechanical finite-element equations are combined in the closed loop to control the speed using PWM. Simulation results are verified by the experiments, and they are in good agreement with the experimental results.

Small effective air gap is one of the major design considerations to improve performance in brushless DC motors. In axial gap motors, however, it is limited by thickness of armature coil as well as the friction loss in a bearing due to the axial magnetic force. This paper investigated the magnetic field, force, torque and loss in an axial gap motor using finite element method (FEM) and Maxwell stress tensor, and developed a novel shape of dual air gap printed coil motor which increases the performance as well as the manufacturing flexibility.

This paper presents a method to analyze radial force densities acting on each stator tooth of an induction motor with skewed slots. A two-dimensional finite element method is used for electromagnetic field analysis of an induction motor, and skew effects are considered by coupling several disks cut by planes perpendicular to the shaft. Radial force densities as a source of vibration are calculated along the surface elements of each stator tooth and its time harmonics are examined by discrete Fourier decomposition.

Summary form only given. This paper introduces the design and development of an axial-gap spindle motor using printed circuit boards (PCBs) and dual air gaps, which meet the mechanical rigidity, high efficiency and zero cogging torque required in a computer hard disk drive. Performance of the developed motor is experimentally compared with the conventional radial-gap spindle motor. Experimental results show the possibility that the developed motor can be used effectively for various disk drives with manufacturing flexibility.

The metallization of aluminum nitride substrates by electroless copper plating is investigated. The AlN substrate is etched by 4% NaOH to study the correlation between the adhesion strength and the surface roughness of etched AlN substrate. Both the as-received nonpolished and the polished AlN are studied. For the nonpolished substrate, the adhesion strength increases from 130 kg/cm/sup 2/ for the sample with an average surface roughness of 0.2 mu m to 230 kgf/cm/sup 2/ for one with an average surface roughness of 0.82 mu m. For the polished substrate, the adhesion strength reaches 271 kg/cm/sup 2/ with a surface roughness of 0.19 mu m. Mechanical interlocking is the major cause for the adhesion strength between the Cu and AlN substrates. The polished substrate that was subsequently etched could form fine cavities on the AlN surface, and the microetching effect resulted in a stronger mechanical interlocking, which increases the adhesion strength. >

A hybrid method combining finite and boundary elements (the FE-BE method) is presented to analyze the transient electromagnetic/mechanical behavior of coaxial induction launchers (i.e., the coilgun). The corresponding initial/boundary value problem is formulated in terms of the hybrid potentials, which mixes the vector and scalar magnetic potential functions. The problem is assumed to be axisymmetric and the forcing currents are circumferential. The thermal effect due to ohmic losses is considered during the launching processes. Simulations of a single-stage induction launcher and of a five-stage launcher are presented. >

A study of field-induced coupled quantum-well oscillators due to the interwell coherent oscillation is presented. The evolution of an electron wave packet in an asymmetric coupled quantum-well structure is described by the application of the time-development operator of the time-dependent Schrodinger equation. By adjusting the applied field, various resonant conditions giving rise to different oscillation frequencies can be obtained in a single device. Direct frequency tuning, amplitude variation, and dipole matrix element due to the external field are also investigated. >

We show that hydrogenated amorphous silicon thin-film transistors (TFT's) with active layer thickness less than 50 nm have improved performance for display applications. Using two-dimensional (2-D) modeling, we find previously observed degradation for thin active layers is due to electric field effects in the contact regions of staggered inverted devices and affects only the saturation characteristics; linear region performance actually improves with decreasing thickness. We have fabricated devices with extremely thin active layer (10 nm), and indeed find excellent linear region characteristics. In addition, direct tunneling across the undoped regions at device contacts reduces electric field effects, resulting in excellent saturation region characteristics, and gate-induced channel accumulation reduces the Schottky barrier width at direct metal contacts so that even devices without doped contact regions (i.e., tunneling contacts) are possible.

Systolic tree architectures are presented for data structures such as stacks, queues, dequeues, priority queues, and dictionary machines. The stack, queue, and dequeue have a unit response time and a unit pipeline interval. The priority queue also has a unit response time, but the pipeline interval is 2. The response time and pipeline interval for the dictionary machine are O(log n) and O(1), respectively, where n is the number of data elements currently residing in the tree. In each node of the tree, the mechanism for controlling the transmission and distribution of data is finite state. This feature makes the designs presented here suitable for VLSI. If there are n data elements in the data structure, the depth of the tree is O(log n). >