December 2003, Volume 3, Number 1
Summary: 
Ivo DOLEŽEL  Daniel MAYER  Bohuš ULRYCH Modelling of Resistance and Inductance of Nonferromagnetic Conductor with Respecting Its Direct Heating [full paper] The paper deals with simulation of electromagnetic and thermal phenomena in a nonferromagnetic cylindrical conductor of circular crosssection fed from a source of general timedependent voltage. The task is formulated as a hardcoupled electromagneticthermal problem that is solved in common with the computation of the corresponding time evolution of the current. Its mathematical model consists of two partial differential equations describing both the fields and an ordinary differential equation for the time dependence of the current. These equations supplemented with correct initial and boundary conditions are solved by the FDM using an appropriate explicitimplicit approximation. The methodology is illustrated on a task whose results are discussed.

Roland HOLCER  Linus MICHAELI  Ján ŠALIGA Histogram Testing of ADC by the Exponential Signal [full paper] Testing of ADC’s differential nonlinearity (DNL) by histogram method requires signal generator with extremely low distortion and high stability of the parameters. The new type of stimulus signal has been proposed, which achieves mentioned requirement. The studied testing signal has exponential form and is generated by discharging of the capacitor through known resistance. Besides precise signal shape the generating circuit could be ideally connected at the input of the ADC under test without interfering coupling with other instruments. The acquired digital samples from the output of ADC under test allow to determine the best fitted exponential signal by proposed three parametric method. The histogram from registered samples and that for the best fitted exponential shape allow to determine the differential nonlinearity DNL(k) for any code level k. In order to test each code level with similar number of samples the exponential shape was split into subparts. The acquired DNL(k) for each subranges are being fanfolded over the whole ADC input range. The proposed method has been experimentally verified for the ADC embedded on microcontroller ADuC812 where the testing results were compared by standardized methods.

Daniel MAYER  Jan HANÁK Contribution to the Compensation of a Nonlinear Load [full paper] In the article the method of optimised design of the passive LC type compensator for uniform line is described. The load at the end of the line can be linear, or nonlinear, e.g. a semiconductor converter. In the first case the currents and voltage in the line are harmonic, in the second case harmonic high orders are introduced into the line. These harmonics of higher orders increase power losses in the line. The article shows that a long line can be significant for simulating and designing a compensator.

Václav ŘÍČNÝ SignaltoNoise Ratio Meter for VideoSignals [full paper] This paper deals with problems and principles of direct measurement of the signaltonoise ratio (SNR) for videosignals (in the baseband).Detailed block diagram of designed measuring device and properties of partial functional blocks of this device are described.

Dušan KOCUR  Jana ČÍŽOVÁ MultiUser Detection Techniques for CDMA: A Review of Basic Principles [full paper] Multiuser detection (MUD) refers to the process of demodulating one or more user data streams from a nonorthogonal multiplex [18]. Here, the great work has been done especially for the development of receivers using MUD principle (so called MUD receivers) for the transmission systems based on code division multiple access (CDMA). The possibility of the development of optimum multiuser detectors providing maximum likelihood sequence detection [26] was the motivation for that work. The key algorithmic structure of the optimum MUD is that of a bank of matched filters (BMF) followed by a dynamic programming algorithm. The performance gain of the optimum MUD receiver is achieved by its centralized implementation involving a high degree of complexity. For practical systems, implementation complexity needs to be reduced to a reasonable level even if the performance is degraded from the optimum one. Following this idea, a number of suboptimum receivers have been studied over the past decade (e.g. [6,11,26,27]). In this paper, a short review of the basic principles of optimum and suboptimum MUD techniques is presented.

Ján KOLLÁR Partial Monadic Approach in Process Functional Language [full paper] This paper is devoted to manipulating the state in process functional language (PFL) using monadic approach [13,19], preserving at the same time the visibility of memory cells. Instead of exploiting just pure functional expressions, as it is done in Haskell, balanced binding of functional and state aspects of computation in PFL using monads is presented. Monadic approach in PFL is partial in the sense that it may be exploited at any hierarchical level of computation, concentrating to the points in which state manipulation and pure functional evaluation are switched. From this point of view, this paper contributes to a systematic joining of function and state aspects of computation.

Zbynek MARTÍNEK  Ivan TŮMA Reliability Assessment of Complex Networks in the Power System [full paper] This paper deals with the analysis and operability assessment of the power system during nonstandard operation regimes. Increasing demands on high quality of power delivery and high reliability of the power system put growing pressure on power utilities that have to cope with these matters. The questions of reliability are important in the field of both power generation and power transfer. In order to meet sometimes very tough conditions on operation readiness of the power system, it is necessary to analyze the situation thoroughly. Several methods using different approaches for power system analysis have been worked out. Each of them has its advantages and drawbacks that determine their use in reliability assessment under real operating conditions. Some of these methods are relatively simple but have only limited use due to their lower accuracy. Anyway, they are still usable for some kinds of calculation. On the other hand, much more complex methods are needed in order to achieve higher accuracy of results. But calculations using these methods are usually much longer and more demanding, requiring iterations. Three different methods for analysis of power transfer networks are described in the following text, including evaluation of their limitations.

Jerzy SZKUTNIK The Level of Energy Losses in the Network of a Distribution Company As Multiparameter Correlation Function [full paper] In the report, the software package designed by the team of the Technical University of Częstochowa has been presented. The software is commonly used in distribution companies for calculation of efficiency of energy distribution in the networks. Calculations are conducted for current and forecasted periods of the network functioning. The report depicts the impact of energy saving transformers on the level of losses in distribution networks.

Jaroslava ŽILKOVÁ  Jaroslav TIMKO OnLine Identification of a Rotor Resistor of an Induction Motor Using the Neural Network [full paper] This paper deals with the advanced method for identification a rotor resistor of an induction motor, presents a new method to online identification on a base of artificial neural networks (ANNs). ANNs are recently showing good promise for application in power electronics and motion control system. They have been applied mainly in control of converters and drives. The purpose of this paper is to show that such a technology can be applied for identification of parameters in an area of electrical drives.

Slavomír KARDOŠ  Miloš SOMORA  Karol FLACHBART  Slavomír GABÁNI  Vladimír PAVLÍK  Jarmila TRPČEVSKÁ Properties of RuO_{2} Resistors Prepared by Modified Technology [full paper] The paper deals with properties of RuO2 based thickfilm resistors prepared by a modified (different to standard) manufacturing technology. It was shown that the received temperature dependence of the resistance R(T) of these resistors can be strongly influenced by the manufacturing baking profile. Both the increase of baking temperature and baking time shift the resistance of RuO2 resistors to lower values. Moreover, under these conditions the R(T) dependence of resistors changes and its typical resistance increase towards low temperatures is less expressive. The modification of manufacturing technology thus enables to adapt and “tune” the resistance and the temperature dependence of RuO2 based resistors, and prepare them as low temperature sensors with appropriate or desired properties.

Alena PIETRIKOVÁ  Juraj BANSKÝ  Magdaléna BUJALOBOKOVÁ  Ján URBANČÍK Accelarated Aging Thermal Stress of Thick Film Bending Conductors in Multilayer Ceramic Modules [full paper] This paper deals with quality and reliability of bent multilayer modules for electronics based on LTCC (Low Temperature Cofired Ceramic). It refers to the method of the bent multilayer ceramic modules preparation with builtin thick film conductors for electronics with number of layers 7, 9, and 11 and with bent angle 0°, 20° and 40°. It is oriented on measurement of basic electrical parameters of thick film conductors in bent multilayer ceramic modules, which contain resistivity (R), conductivity (G), capacitance (C) and inductance (L). The experiments have been concentrated on the effects of reliability tests and on the measurements of electrical properties of bent multilayer modules. The achieved results of accelerated aging stress confirm the influence of temperature as an important factor to the technical life and reliability of electronic systems in working conditions. These tests show the expressive influence of temperature, which limits operation life and reliability of systems applied in electronics. Along with these analyses of characteristics is demonstrated the impact of frequency on the stability of electrical properties. The higher frequency more increases the stability of properties of bent modules.
