Showing papers by "Nathan Ida published in 2004"

Dual/triple redundant computer system

Abstract: A dual/triple redundant computer system having in one of the preferred embodiments triple redundant I/O modules and dual redundant central processor modules (CPM) that operate in parallel executing the same application program. Each input module includes three input circuits operating in parallel. The first CPM receives input data from first and third input circuits and transmits input data of the first input circuit to the second CPM. The second CPM receives input data from second and third input circuits and transmits input data of the second input circuit to the first CPM. Each CPM then performs a two-out-of-two vote among input data produced by first, second, and third input circuits and utilizes an outvoted data as input to the application program to provide output data by execution of the application program. Each output module includes three microcontrollers operating in parallel. First and second microcontroller receives output data respectively from first and second CPM, while a third microcontroller receives at the same time output data from both first and second CPM. First microcontroller transmits output data to a first and a second output circuit. Second microcontroller transmits output data to second and third output circuit. The third microcontroller performs a selected logic operation among output data produced by first and second CPM and then transmits a result of this operation to third and first output circuits. Each output circuit generates a logical product of output data received from two associated microcontrollers. Outputs of first, second, and third output circuit are connected to each other for providing a two-out-of-three voting among output data produced by first, second, and third microcontroller, and for allowing the system to generate a system output as a result of a two-out-of-two voting of output data generated by first and second central processor modules.

Transmission line matrix model for detection of local changes in permeability using a microwave technique

Abstract: A three-dimensional transmission-line matrix model was developed to simulate the microwave detection of local changes in permeability. The technique can be used to map local nonuniformities in magnetization. Numerical modeling was carried out for frequencies that are commonly used in microwave nondestructive testing (0.8-1 GHz). A comparison between experimental and numerically generated curves is provided. This comparison validated the proposed numerical model.

Applications of the Transmission Line Matrix Method to Microwave Scanning Microscopy

Abstract: A three-dimensional transmission-line matrix (TLM) model was developed to simulate microwave-scanning microscopy. A TLM algorithm that allows the simulation of the scanning was developed. Numerical modeling was carried out for frequencies that are commonly used in microwave nondestructive testing (1GHz - 20GHz). Structures with local discontinuities in the electric permittivity are modeled numerically. The excitation parameters used in numerical modeling of scanning microwave microscopy were determined based on an initial frequency experimental response obtained from a plate with known permittivity. The numerical model developed in this paper is based on the symmetric condensed node. The description of the TLM algorithm is given in a Hilbert space using a three-index notation.

Development of Time-Domain Models for Nondestructive Testing

Abstract: The development of a Transmission-Line Matrix (TLM) for the simulation of ultrasound and microwave propagation in structures common in Nondestructiv e Testing (NDT) is described. The spatial resolution of the proposed model is better than a tenth of a wavelength. Numerical modeling was carried out for frequencies commonly used in ultrasound and microwave nondestructive testing (3.5MHz - 20GHz). The sample results provided here for ultrasonic and microwave testing show the applicability and accuracy of the model. Introduction Many of the methods used in Nondestructive Testing (NDT) are based on wave interaction with the materials under investigation. Elastic and electromagnetic waves are used to identify and characterize the local changes that occur in a materials in response to external excitation. To improve the results obtained in NDT of materials, considerable theoretical effort is involved in developing reliable mathematical models of wave propagation in different media. Due to the complexity of the problems, numerical methods have proven to be an appropriate approach. The Transmission Line Matrix (TLM) is a time domain numerical technique which was found to be particularly suitable for modeling of complex geometries encountered in testing. The TLM method dates back to 1971 (1) and as such is one of the newest numerical methods available yet it has proven both reliable and flexible enough for the demands of many applications including those in NDT. The method is considered to be "a modeling process" rather than a numerical method for solving differential equations (2). The method is a direct numerical implementation of the Huygens principle (3). An appropriate field propagator (Green function) is first identified (4). Then the wave front at each iteration (instant in time) for each point in space is a result produced by the waveforms generated at neighboring points in the previous iteration. The TLM is a physical discretization approach and this method does not require the solution of a differential equation. The TLM requires division of the solution region into a rectangular mesh of transmission lines in ahich the nodes of the mesh are points of discontinuity for impedances. In addition, to solve a problem using the TLM, a set of boundary conditions and material parameters must be provided as well as an initial excitation. Then the impulsess are propagated throughout the mesh using scattering theory on the transmission lines. There is no limitation regarding the frequency of interest, but the size of the mesh imposes an upper limit on the frequency response analysis. Unlike some other numerical techniques, the TLM algorithm does not involve an explicit convergence criterion, a property that makes it an inherently stable method. This stability is reflected in the flexibility of the TLM method when dealing with various types of input signals and boundaries. The present work describes the development of models for microwave and ultrasound NDT and shows that the models are essentially the same for both NDT methods in spite of the inherent differences between acoustic and microwave applications including obvious differences in wavelength, material properties and interpretation of results. The purpose of this general model is to show the applicability, accuracy and flexibility of the method in modeling NDT environments which previously were difficult to model. Results from various configurations including conducting and dielectric objects and practical testing configurations are presented to demonstrate the method's applicability and flexibility. Extraction of S parameters and prediction of resonant frequencies are also demonstrated.