Search

A rapid measurement of thermal deterioration of oil-immersed papers by mass spectrometer and gas chromatograph is described. A special assembly was developed using a mercury diffusion pump and automatic Topler pump to extract a very small quantity of generated gas from the oil and to collect it in a reservoir. The gas collected by the above assembly is analyzed by mass spectrometer and gas chromatograph. Logarithm of the generation rate of (CO + CO2) [log (CO + CO2)] is proportional to the reciprocal of absolute temperature (1/T). This linear relationship can be explained by chemical reaction-rate theory and it is shown that the thermal endurance and the thermal life of insulating papers in oil can be estimated by the generation rate of (CO + CO2) in a few weeks as was reported previously for air and vacuum. But the change of the slope of the line log (CO + CO2) versus 1/T at the temperature near 130°C was observed. This is a new mechanism of thermal deterioration of insulating paper in oil, which is apparently unexplained.

This paper describes the characteristics and properties of a new insulation system currently used to manufacture high-voltage bushings at voltage ratings of 15 through 69 kV for transformers and switchgear apparatus. The system is compared with other forms of insulation such as phenolic-paper, oil-paper, mineral-filled epoxy, and electrical porcelain. Its application in high-voltage condenser bushings is also discussed. With this new insulation, epoxy resin is used to totally impregnate an electrical grade of kraft paper resulting in a resin-rich (approximately 70 percent) material having unusual properties. Its predominant characteristic is the complete absence of voids resulting in high-corona-inception voltages and high-dielectric strength. It also has excellent physical properties which, coupled with a flexible manufacturing process. open up numerous new design possibilities in high-voltage engineering.

The objective of the work reported in this paper was to study the influence of the thickness of the paper mass and of the inner temperature distribution on the vacuum drying process; particularly to study the ways in which these two factors affect the quantitative distribution of residual moisture perpendicular to the layers of the paper. The quantitative distributions of residual moisture perpendicular to the layers in soft paper masses 0.3 to 2.4 cm thick were determined. The results show that both the magnitude and distribution of temperature throughout the paper affect the degree of drying. Neither a low vacuum nor a prolonged drying time will compensate for an unfavorable temperature distribution. The magnitude and distribution of temperature predominate over the effect of paper thickness. With unfavorable temperature conditions, paper only 0.6 cm thick will retain an excessive amount of moisture. With favorable temperature conditions, paper 2.4 cm thick can be dried under vacuum to an acceptable moisture content. The upper limit of residual moisture content for transformer insulating paper that can reasonably be expected to be achieved is about 0.5 percent, not 0.1 percent or less.

Computer-aided design techniques are now used extensively for the production of masks in the microelectronic industries. In these methods the microcircuit patterns are broken down into simpler component parts, such as rectangles, whose co-ordinates are punched onto paper tape. A logic system has been developed to control an electron probe from coordinates read in on paper tape. Successive rectangles specified on the tape are scanned by the probe on the specimen, exposing an electron-sensitive material. A microcircuit pattern is thus built up from these basic rectangular elements. The use of an electron probe allows the generation of patterns with submicron dimensions within a reasonable time. The system outlined above has been used to produce microcircuit patterns using two techniques. In one, electron-sensitive resist is used to define a pattern etched in underlying material; in the other, metallic patterns are deposited directly by the decomposition of a metallic compound.

This paper takes the reader through the steps of a control system design using both optimal control theory and sensitivity analysis. The problem is to control color on a paper machine. The physical aspects of color are briefly described and a model for the paper machine is given. The model responses are compared with responses obtained experimentally on a small-scale machine. The retention turns out to be an unknown and variable parameter. A sampled-data design is carried out for the nominal case where the retention is 100 percent using the linear-regulator theory. A parameter \rho weighing the control effort in the quadratic cost function is then used as a design parameter.

This short paper presents a design procedure for the widely used three-port waveguide circulator that has a partial height ferrite post in the junction region. Design formulas and curves are derived for two configurations of partial height circulators: one has a short circuit at one end of the ferrite post, while the other has dielectric spacers at both ends. The design method was used to build two circulators for operation at 38 GHz and 60 GHz, respectively. Excellent agreement between theory and experiment was obtained in predicting the center frequency and required matching structure of these devices.

This short paper describes a proposed structure of a broad-band electrooptical modulator and an application of existing microstrip analysis programs to determine dimensions of the structure for a broad-band property. Results of numerical computations indicate that it is possible to obtain a broad-band modulator by using LiNbO/sub 3/ or LiTaO/sub 3/.

The WR15 thermal noise standard that is used as the national reference standard of noise power in the frequency range from 56 to 64 GHz is described in this short paper. The source forms a basis for both the noise-power comparison service and noise-figure service offered by the National Bureau of Standards in this frequency range.

The electromagnetic boundary value problem of two waveguides coupled by an aperture or an aperture in a waveguide radiating into free space may be described by an integral equation. An analytical solution to this integral equation cannot be readily found due to the complexity of the kernel. However, extremely useful results may be obtained if the method of moments is employed to reduce the integral equation to a matrix equation which can be solved by known methods. In this short paper, series and shunt slots in a rectangular waveguide are analyzed using this technique.

Based on measurements, the strength of synthetic oil impregnated capacitor paper (Askarel) to dc and ac voltage 1.2/50-?s voltage surges and switching surges of various wavefront duration are discussed. The results of measurements are compared with the performance figures of mineral oil impregnated capacitor paper. The diminishing effect of the number of lightning and switching surges and their amplitude upon electric strength and of initial voltage partial discharge is investigated.

The Q factor of an open-circuited resonator is influenced by dielectric, conductor, and radiation losses. This short paper discusses these losses and shows that insight into the radiation loss can be obtained by an extension to the analysis given by Lewirt. This shows that the radiation loss is a maximum for the dominant mode and that provided the permittivity is not less than 9.0 the radiation losses are at a minimum for the second mode. It is suggested that these results may be applied to the design of filters based on dielectric resonators. The application of the results to stripline resonators in which the dielectric extends past the termination of the upper conducting strip is more difficult, and it is suggested that experimental work is required to investigate postulated improvements. Finally, some of the radiation patterns of the open-circuited dielectric resonators, obtained in this paper, show interesting directional properties which may be applied to the design of antenna systems.

The dielectric response of an oil-impregnated-kraft paper system has been examined in the temperature range from 25-85°C at frequencies down to 10-2 Hz. Single and double sheet oil-paper specimens were studied. With single sheet paper specimens the long range charge carrier migration process was found to follow the Eyring behavior over the main test temperature region and was characterized by dc and ac activation energies ranging from 0.40 to 0.47 eV and from 0.37 to 0.48 eV, respectively. The involvement of mobile ions, possibly hydrogen, appeared to be a likely cause for the observed conduction process. Due to appreciable interfacial polarization effects, the dielectric behavior of the two paper sheet specimens exhibited strong deviations from the Eyring behavior. No definite physical significance could thus be attached to the derived values of activation energy over the restricted linear regions of the ln Rac versus 1/T plots.

FOR a number of years the standard low-tension direct-current feeder cable in Chicago was 1,000,-000-cir. mil two-conductor concentric paper-insulated cable with three pressure wires laid up with the outer conductor. Apparently there has been the greatest tendency to overload this particular size of cable more than any other as, with both single and concentric cables in service, it was difficult to make a proper distinction between the carrying capacity of the two kinds of cables. About ten years ago we had two cases in which the cables of this kind were called upon to carry loads somewhat larger than had been customary, and steadily throughout the day instead of a short peak in the evening. The two customers in these cases, in different portions of the city, were both manufacturing customers with an eight-hour load, and on account of some increase in business they increased their load on short notice when no cable was available for another feeder to their premises. In each case, the cable has to be ordered from the factory so that this unusual load was carried by these cables for some three or four months. In the first case of this kind we were so disturbed about the temperature that we made a rather extended series of temperature measurements, using one of the pressure wires imbedded in the outer conductor as a resistance, and determining the temperature by the rise in resistance. This, of course, would give us the temperature of the outer conductor which would be probably 8 or 10 deg. lower than the inner conductor. The temperature measured in this manner was slightly above 100 deg. cent., and as there was some doubt regarding the accuracy of the determination made in this manner, it was checked by cutting a small hole in the lead sheath of the cable and in the outer insulation, and measuring the temperature of the outer copper with a thermometer. The temperature taken in this way checked very closely with the temperature as determined by the rise in resistance of one of the pressure wires. The insulation removed in this manner was carefully preserved and compared with the insulation in a piece of new cable, and it was impossible, by careful examination, to discover any appreciable difference.

A convenient form of graph paper is described that allows the frequency constant (f c ) and the MTF index ( n ) of many devices to be easily determined.

This paper is about the applications of an epoxy-impregnated paper insulation to ultra-high-voltage systems. An epoxy-impregnated paper insulation is a resin-rich void-free structure, and therefore has extremely superior voltage withstand characteristics. It also has outstanding resistances to thermal cycle and mechanical vibration. Consequently, such insulation systems can be applied easily to condenser bushings. From the standpoint of the production facilities, techniques, and economical considerations, in items with an overall length of more than 3 m, "prefab"-type condenser bushings, consisting of several condenser cores, are desirable. In cases when two-unit condenser cores are assembled leaving a simple conical-shaped gap between them, it is extremely effective to adopt a "potential-expanding"-type gap designed so that the equipotential surface in the gap will be in the vertical position to the cone surfaces as far as possible. The prefab-type condenser bushing of the 500-kV class was experimentally manufactured using this arrangement; it displayed all the desired properties. This paper also describes the other examples of applications of epoxy impregnated paper insulation: CT bushings and pulse transformers in the magnetic firing system for high-voltage dcconverters using thyristors, and draw-out-type condenser bushing for disconnecting devices used in the compact-type substation.

In modern paper making, accurate methods for measuring speed are of remarkable importance. The use of a time-to-speed converter for speed measurement leads to considerably shorter measuring times and more accurate results than are achieved with conventional methods. In this paper, algorithms usable in digital time-to-speed converters are analyzed, classified, and compared. Finally, the design of a time-to-speed converter based on the algorithm most suitable for paper speed measurement is described.

Research has shown that the determination of absorbed microwave energy as well as the control of environmental parameters are important in relating biological-effect data to, radiation protection. This short paper describes the development of an environmentally controlled waveguide irradiation facility for the exposure of small animals to 2450-MHz CW microwave energy. Integral dose rate is determined without perturbing the microwave field interacting with the irradiated animal.

The effect of ohmic and dielectric losses, conductor thickness, and frequency dispersion on the performance of edge-coupled microstrip directional couplers and interdigital filters have been determined in this short paper. The odd- and even-mode attenuation constants due to ohmic losses in the conductor have been calculated using Wheeler's inductance formula. The theoretical results for the characteristic impedance and propagation constants are in good agreement with the experimental results of Napoli and Hughes. Among the parameters that can be calculated from this theory are the isolation, directivity, and coupling coefficients of lossy directional couplers and the midband insertion loss of interdigital filters.

Design and performance of a low-noise down converter and a high-efficiency up converter for transmitter-receiver applications in the frequency range of 60-86 GHz are described in this short paper. The receiver and the transmitter are used in a guided millimeter-wave transmission system "W-40G" A wafer-type GaAs Schottky-barrier diode is used in both units. The diode mount consists of a semiconductor wafer and a whisker which are mounted in a waveguide circuit. The IF terminal is a coaxial line section which is connected to the whisker. The conversion loss of the up and down converters using the new wafer is 7.5 and 5.5 dB, respectively. The bandwidth of the frequency converters is approximately /spl plusmn/10 percent in the frequency region of 60-86 GHz.

Magnetostatic surface waves propagating along the ferrite slab adjacent to a semiconductor are discussed in this paper. Our numerical results indicate that the conductivity of the semiconductor plays an important role in the determination of the dispersion relation in the case of nondrifting carriers. The backward wave appears for a finite value of the conductivity.

A practical system for translation of information coded on punched paper tape to a matrix on a photograph is presented. The information here consisted of stereo random brightness fields; however, the system should be applicable to any three-dimensional information. The matrix is developed by slow digital scan of an intensity (ten levels) modulated oscilloscope and is photographed by a camera on time exposure. The system requires 166 seconds to produce each 10 000 element matrix and hence cannot be used for on-line work. However, for the researcher who can tolerate this limitation, the system's low cost, simplicity, and good resolution (better than 100 elements per inch) should prove attractive.

A system employing RF substitution techniques for automatic phase and intensity measurement of microwave fields is described. The system uses homodyne detection to generate error signals for servomechanisms which balance the attenuation and phase secondary standards. Accuracy and dynamic range of this method of measurement can be strongly dependent on the method of modulation employed. This paper analyzes errors from this cause and indicates what means are available to minimize them. This method of measurement is less expensive than some alternate techniques, and, at the present stage in the evolution of microwave near field measurements, its errors are less serious than uncertainties introduced by absorber reflections, probe characteristics, and mechanical positioning errors.

In spite of instant intercontinental communication by satellite and computer, there are still barriers to efficient information transfer and attainment of comprehensive awareness between scientists/technologists in different countries. The tremendous growth rare of technology has, in fact, brought new problems in its wake. These are grouped under seven headings: language, overpopulation, pollution, jargon, time, economics, and migration. Each of these is briefly examined and its consequence discussed. The language problem is dynamic and keeps changing with time. There is increasing overpopulation of papers; rehash and abstracts all contributing to pollution and diffusion of data. Noise is getting worse due to redundant data, wrong indexing, etc. For years, a jargon explosion has been going on without control. There are too many built-in delays in the conventional publication-translation-processing-search-feedback chain. Information search is hampered by overpopulation and pollution. Most methods of communication have become more expensive with passage of time, while budgets at both company and national level have become tighter. In this age of rapid innovation, scientists and engineers keep changing their field of work and job location, delaying and making less effective attempts at direct communication.

The purpose of this short paper is to give an exact theory of quarter-wave coupled circulators with Chebyshev characteristics. The synthesis starts by replacing the lumped-element equivalent shunt resonator of the circulator by a distributed one that has the same susceptance slope parameter as the original circuit. In this way the overall network involves commensurate transmission lines only. The bandwidth over which the assumed form of the equivalent circuit applies is carefully discussed in terms of the two split frequencies of the magnetized junction. Tables for the required circulator parameters and transformer admittances for one and two transformer sections as a function of VSWR and bandwidth are included. The realizable solution for the latter arrangement is severely restricted by the equivalent circuit of the basic junction. Experimental results on an octave-band stripline circulator, with a two-section transformer, are also included.

The purpose of this short paper is to construct the dissipation and scattering matrices of lossy junctions in terms of the eigenvalues of the dissipation matrix. This removes the need to rely on inequality relations between the scattering parameters of lossy circulators. The eigenvalues of the dissipation, scattering, and admittance matrices are related. The eigenvalues of the dissipation matrix give the dissipation associated with each possible way of exciting the junction. The ones of the scattering matrix give the reflection coefficients associated with these different excitations. The admittance eigenvalues define in each instance the eigennetworks of the junction. This leads to the definition of the entries of the dissipation matrix in terms of the loaded and unloaded Q-factors of the junction eigennetworks. The scattering matrices of a number of lossy 3-port junctions are also constructed directly in terms of the elements of the eigennetworks.

Using the Frankel-Young method, fast estimation of the potential distribution for a microstrip structure is obtained when an accelerating factor /spl omega/ is introduced in the finite-differences (relaxation) method. It is possible to calculate such a factor by an iterative technique, but the time of computation needed to find /spl omega/ annihilates the theoretical gain. In this short paper, the authors present a method which gives an analytical expression for /spl omega/. The realistic case examined here, as an illustration, is that of the suspended microstrip couplers for which odd and even impedances are the interesting parameters. Given an analytical expression for /spl omega/, the overrelaxation method appears as a very powerful and attractive method for finding the solution of any type of microstrip structure.

To have a full understanding of the experimental behavior of a junction circulator it is necessary to be able to measure the gyrator conductance, the susceptance slope parameter, and the frequencies of the two split resonant modes of the junction. It is also useful to be able to construct mode charts to determine the geometry of the device. The purpose of this short paper is to give simple methods by which each of these parameters may be measured in the case of the below-resonance stripline circulator. The techniques developed here all stem from the scattering and immittance matrices of the device. They are therefore quite general. They also require no phase information which makes them ideally suitable for reflectometer-type measurement. All measurements described here are made in the input transmission line of the device with the other two ports connected to similar transmission lines terminated in their characteristic impedance. The results obtained here apply to Iossless circulators for which the two resonant modes are symmetrically split by the magnetic field, and for which the frequency variation of the third mode can be omitted.

A new microwave structure is proposed, consisting of a circular waveguide loaded with two ferrite toroids circumferentially magnetized at remanence in opposite direction. It is shown that nonreciprocal parameters such as differential phase shift can be doubled with respect to the single toroid configuration. A method for biasing the toroids at remanence in opposite directions by means of a single wire passing through the axis of the waveguide is proposed. Modal purity is taken into account in order to select dielectric Ioading parameters which ensure operation within the modal inversion window in which the TE/sub 01/ mode is dominant. The propagation factor and differential phase shift are computed under these conditions, and their variation with several parameters such as remanent magnetization, toroid location, and toroid thickness is shown.

Noise measurements have been made on an IMPATT pumped S-band parametric amplifier. For the properly adjusted pump, no significant increase in noise temperature was observed for interfering signal levels up to -40 dBm, 20 MHz from the noise measurement frequency.

The increase in TE/sub 11/-TM/sub 01/ mode bandwidth obtained by inhomogeneously loading (dielectric lining) a circular waveguide is systematically documented. Maximum bandwidth is about 31.83 percent of center frequency (up from about 26.54 percent for fully filled or empty circular waveguides). This makes circular waveguides competitive with square waveguides (bandwidth/spl aprox/ 34.3 percent) as radiators in wide-band dual-polarization arrays. Certain interesting symmetries involving the TE/sub 21/-TM/sub 01/ modal inversion are also examined.

Modifications to a normal microstrip transmission line are proposed, with the aim of reducing the attenuation constant of the line. The results of a computer analysis of a structure containing multilayers of dielectric show that significant reductions in attenuation appear possible.

The crossed-field and in-line Mason circuit models for interdigital surface-wave transducers give opposite predictions for the dependence of acoustic reflection coefficients on electric load resistance for purely resistive loads. Experiments described herein show that the crossed-field model correctly describes the reflections for YX quartz, ST-X quartz, and YZ lithium niobate substrates. A low-resistance load minimizes reflections for transducers with double electrodes operating at the fundamental synchronous frequency. For single electrode transducers, optimum reflection suppression may call for a load resistance comparable to the transducer impedance.

Computationally efficient numerical methods for determining the frequency response of uniform transmission lines consisting of a large number of mutually coupled conductors in homogeneous and inhomogeneous media, and illuminated by an electromagnetic (EM) field are presented.

The feasibility of a compact bandstop-filter geometry is demonstrated. The filter geometry is particularly suited for stripline and microwave-integrated-circuit (MIC) fabrications in that grounding is not required for any part of the filter. Hybrid geometries allow the filter designer increased flexibility in choosing a suitable shape factor without significantly affecting the filter's electrical characteristics. Folded-line filter geometries are suitable for narrow bandwidths (provided capacitive-coupled stubs are used) to wide-bandwidth applications. Experimental confirmation is presented.

A derivation is presented for the signal-to-noise ratio of negative-conductance amplifiers and oscillators in FM/frequency division multiplexing (FDM) communications applications. Results indicate the limiting value of signal-to-noise ratio depends on the semiconductor properties and channel loading only. This means circuit adjustments, such as Q, cannot increase the signal-to-noise ratio without bounds. Typical specifications are given. Limiting values of signal-to-noise ratio for Gunn and Si IMPATT devices are given in typical applications. Results indicate that Gunn devices have a clear advantage over Si IMPATT'S in a signal-to-noise sense.

A method of estimating quasi-static discontinuity inductances in microstrip lines is outlined. Numerical results for symmetric right-angle bends are presented and compare well with experimental results.

A theoretical study, backed by experimental verification, was undertaken to determine the modal characteristics of quadruple-ridged circular and square waveguides. Field lines for the first few important modes and cutoff frequencies were determined. It is shown that for square waveguides quadruple-ridge loading always decreases the TE/sub 10/-TE/sub 11/ bandwidth whereas for circular waveguides only a small amount of additional separation between the first two fundamental modes may be obtained over a limited parameter range. Symmetrical excitation will not excite the asymmetrical higher-order modes. This feature makes these waveguides acceptable as feeds for wide-band reflector antennas and for similar applications but raises a question mark regarding their use as radiators in wide-band phases arrays.

Losses in microstrip on fused-silica and alumina substrates have been experimentally evaluated for various values of stripwidth. Radiation losses in circuits, commonly used for measuring losses in microstrips, depend on the dimensions of the circuit and affect the total losses considerably. Radiation and conduction losses of open-ended line resonators have been separately determined. These measurements have also drawn attention to discrepancies between published theories and our experiments. The effect of surface roughness upon conduction losses has been measured.

In terms of scattering parameters, the equation transducer power gain is shown to be capable of representation as a family of circles of constant gain from which the design of load and source terminations to achieve a restricted bandwidth can be obtained. This is an extension of an earlier approach which only allowed either load reflection coefficient or source reflection coefficient to be considered in a given design. Through the use of a specification statement of VSWR, it is shown how a marginal stability factor can be derived. From the study of the interaction between the input and output reflection coefficients, a detuning factor is analytically derived to correlate the interaction between the input and output reflection coefficients. Either of these factors can be chosen and used to select optimum input and output reflection coefficients which provide stable operation for an amplifying stage that is to employ a conditionally stable active element. An example using these factors is given.

Half-wavelength line resonators are employed to enlarge the bandwidth of a 50-GHz Y-circulator. The bandwidth is successfully increased from 3.5 to 5.5 GHz. The characteristics of the half-wavelength line resonator are described.

The modal and eigenvalue approaches of 4-port single-junction circulators are combined to describe the theory and construction, of a waveguide device and a stripline device. The three independent variables used in the case of the waveguide one are a pair of HE/sub spl plusmn1,1,1/ open dielectric resonances in a demagnetized ferrite disk, a TM/sub 0,1,1/ resonance on a metal post, and the magnitude of a direct field to remove the degeneracy between the HE/sub spl plusmn1,1,1/ modes. The eigenvalue approach is used to establish systematically each condition one at a time. The variables used in the construction of the stripline junction are a pair of radial n= /spl plusmn/3 degenerate resonances, a radial n=0 resonance, and the amplitude of a direct field to split the degeneracy between the former modes. In this case a circulation condition is found in which it is possible to omit one of the circulation adjustments.

A new nondestructive method has been developed for measuring the dielectric constant and the loss factor of a slab-type material using a microstrip cavity. The method, which uses a simple and rapid substitution procedure, yields accurate results and has a number of advantages over currently available techniques. Experimental details and the theoretical basis are explained and experimental data are presented.

An FM transmitter having 220-mW output power and 5-W total dc input power and operating in the 4-GHz band has been developed. This transmitter provides a dc-to-RF signal-conversion efficiency of more than 4 percent. Featuring low power consumption and high reliability, this transmitter is suitable for use as a transmitter or an exciter for radio relay of a maximum of 1380 channels.

Coax-slot and microstrip-slot transitions are discussed. Experimental VSWR and impedance data are presented and compared with values computed using equivalent circuits for these transitions. Thick-film chip terminations are also investigated.

A method for the measurement of the dynamic admittance and susceptance of "negative-resistance" diodes is described. The device under test is allowed to oscillate in a microwave cavity, and operated as an injection-locked oscillator. Injected locking signals of the same order as the free-running output power of the oscillator are used. The dynamic conductance and susceptance of the device are obtained from the phase and amplitude response of the system.

A stationarity property for the resonant frequency of a rectangular cavity as a function of the thickness of a low-loss dielectric slab inserted within is used for the accurate determination of the microwave permittivity of the sample. The accuracy estimated to a few promille in X band has been confirmed by experimentation on a standard material.

A technique for the optimization and sensitivity analysis of broad classes of electrical networks is illustrated. The method utilizes the exact algebraic partial derivatives of functions with respect to any desired independent variable. This completely automated technique has the obvious advantage that the derivatives of any circuit response function with respect to any desired component parameter may be obtained with no additional analytical effort on the part of the designer. Several examples are given to illustrate the procedure.

Carpenter and Livstone's experiments on beetle pupae are repeated and extended. In the experiments conducted, increased incidence of abnormal development occurred due to exposure to microwave energy, both CW and pulsed. This effect was observed at the power level of 8.6 mW/cm/sup 2/. Measurements are reported which specify the microwave environment encountered by the insect.