Showing papers in "Electronics and Communications in Japan Part Ii-electronics in 1989"

Slot—coupled directional couplers on a both—sided substrate MIC and their applications

Abstract: A slot-coupled directional coupler is proposed and analyzed. In the slot-coupled directional coupler, microstrip lines set on both sides of a common ground plane are coupled through a slot in the common ground plane. This coupler can easily realize a tight coupling such as 3 dB with a small physical size. This paper shows that it is possible to form a planar multiport directional coupler with 2n (n: integer) input or output ports without line crossings by combining slot-coupled directional couplers. Using such a planar multiport slot-coupled directional coupler, a multiport power amplifier and a BMN (Butler matrix network) can easily be formed in a small physical size. The 3-dB slot-coupled directional coupler fabricated for trial has a coupling of 3.2 ± 0.2 dB, a return loss of more than 25 dB, and an isolation of more than 28 dB in the frequency range of fO ±20 percent (fO = 1.5 GHz). These measured values agreed well with computed data.

Analysis of the waveguide with loss or gain by the finite‐element method

Abstract: In the past the finite-element method with a variational formulation has been used for numerical analyses of the propagation characteristics of electromagnetic waveguides. However, this method based on the variational representation is not applicable to waveguides with losses or gains. This paper derives a generalized Galerkin equation which is established for waveguides with losses or gains. The finite-element method is proposed based on this Galerkin equation. As examples of numerical calculations, a waveguide loaded with a lossy dielectric, a helical waveguide and a microstrip line were analyzed.

Fabrication of a micropump for integrated chemical analyzing systems

Abstract: Utilizing micromachining based on the semiconductor device fabrication technique, a miniature pump (micropump) was fabricated on a silicon substrate. The purpose of this study is to integrate a conventional chemical analyzing system on an Si substrate by using microchemical sensors such as ISFET's and microvalves. The micropump was the diaphragm type. The pump was composed of an inlet and outlet having a one-way microvalve, a pressure chamber formed on an Si substrate and a miniature piezoactuator. The size of the pump was 8 mm x 10 mm x 10 mm and the flow rate on the order of μl/min could easily be controlled by the voltage signal applied to the actuator. To monitor the flow of the pump, a differential pressure sensor which could be formed on the Si substrate was developed. The sensor was composed of a oneway valve and a piezoresistive diaphragm-type pressure sensor, and had a high sensitivity to small flow.

Approximate theory and characteristics of laser doppler velocimeter using self—mixing effect of semiconductor laser diode

Abstract: To explain the experimental characteristics of a laser Doppler velocimeter using the self-mixing effect of a semiconductor laser diode (LDV) with a rotating disc as a target, a theoretical analysis of a Doppler beat waveform was carried out, and its approximate solution was obtained by using the perturbation method. The theoretical characteristics based on this solution agreed fairly well with the experimental ones. The operating conditions of the LDV required for forming the beat wave into a sawtooth waveform, and the reasons why the sawtooth wave is transformed into a mirror image depending on the sign of the Doppler frequency were found, i.e., the phase angles of the second and third higher harmonics of the beat wave are inverted by the inversion of the direction of rotation. The upper limit of the Doppler frequency and the range of R parameter of the LD were estimated theoretically.

Utilization of piezoelectric stiffened effects on impact printer heads

Abstract: The electromagnetic printer head has traditionally been used in the impact dot-matrix printer based on the electromagnetic attractions. The traditional technique, however, is approaching its limit. To find a break-through for this limit, efforts have been made to develop a printer head based on the piezoelectric effect. So far, however, there has not been realized a piezoelectric printer head with a better performance than that of the electromagnetic printer head. This paper discusses the direction to be aimed at by the high-performance printer head based on the piezoelectric effect. The reported direction can be summarized as the use of the longitudinal stiffened piezoelectric effect, instead of the transversal piezoelectric effect considered in the past, as well as the use of the lever mechanism which is isolated structurally from the piezoelectric element, instead of the traditional bimorph structure. The first piezoelectric printer head for some commercial printer products has already been successfully fabricated based on the essential idea in this paper, indicating its practical usefulness.

Growth of oxide films on the surface of Cu contact and its effect on the contact resistance property

Abstract: From the viewpoints of contact reliability for small-sized contact components, growth processes of the oxide film on the Cu contact surface were analyzed by studying the film thickness and formation of films. This analysis of the film was conducted in detail by ellipsometry which scarcely applied to the contact phenomena, in combination with an electrochemical reduction method. Moreover, the effect of the growth of the film on the contact resistance property was studied. As a result, in the oxidation by heating in air, Cu2O film grows mainly at the initial stage of the film growth. As the film grows around 1000 A, CuO appears on the surface of Cu2O. In this stage, the contact resistance increases sharply and contact failure occurs. Furthermore, it was found that the velocity of growth of the film is slowed which corresponds Co each stage of the oxidation: the initial, the middle, and the latter stage. By application of ellipsometry to the analysis of the contact surface, the measurement of the film thickness together with the possibility of discrimination for the formation of the film by using changes of the optical constants of the film were clarified.

Drawing condition dependence of pure‐silica‐core single‐mode fibers

Abstract: The conditions have been investigated for drawing the SiO2 core/F-SiO2 cladding single-mode optical fiber which is expected to have the lowest optical loss among the silica optical fibers. In this type of optical fiber, the softening temperature of the core is higher than that of the cladding, and the softening temperature distribution is different from that of the GeO2-SiO2 core/SiO2 cladding-type optical fibers. In this study, the residual stress in the fiber during drawing was investigated theoretically and experimentally, considering the softening temperature difference between the core and cladding. The following results were obtained. In the SiO2 core single-mode optical fiber, the tensile stress remains in the core where it is proportional to the drawing tension and inversely proportional to the area of the cross section. The residual stress influences the refractive index of the core and the cut-off wavelength, but degrades the propagation loss characteristic. Finally, this paper will show that: (1) this optical fiber belongs to the F-SiO2 core/F-SiO2 cladding type from the viewpoint of the material, and (2) the depressed-type structure is the most suitable.

Finite‐element solution of three‐dimensional periodic waveguide problems

Abstract: With a view to developing a method enabling derivation of the propagation characteristics of a three-dimensional (3-D) periodic waveguide with an arbitrary shape, an analysis technique based on the finite-element method with tetrahedral elements is proposed and its formulation developed. For the finite-element analysis of the electromagnetic field in a 3-D inhomogeneous region including a dielectric material, the variational expression in terms of three components of magnetic field is convenient. However, if the conventional variational expression is used directly for solving eigenvalue problems, spurious solutions can appear. In this paper, to suppress and eliminate these spurious solutions, various countermeasures to spurious solutions, developed for the finite-element analysis of a waveguide uniform in the propagation direction that can equivalently be treated as a 2-D problem. These include: the penalty function method, the transverse magnetic-field component method and the reduction method, all of which are extended to the finite-element analysis of 3-D periodic waveguide problems. Specifically, by way of the numerical examples for a rectangular waveguide loaded with dielectric material with a periodic structure, the validity of the proposed method and the adaptability of each method for suppressing and eliminating spurious solutions are investigated.

Analysis of a bending dielectric slab waveguide by integral equation formulation

Abstract: The waveguide bend is an indispensable circuit element in realizing optical integrated circuits. In this paper, the bend in a dielectric slab waveguide is analyzed numerically by means of the integral equation formulation. When a numerical solution of an integral equation is attempted by the boundary-element method, two problems exist: 1) how one can connect the field distributions expressed by the boundary-element method with those by the eigenmodes of the waveguide. In this paper, instead of representing the field at the input/output boundary of the bend in the slab waveguide in terms of the boundary-element method, a new method is proposed in which the field distributions expressed in terms of the eigenmodes are substituted directly into the integral equation; and 2) the treatment of the radiation mode. In this paper, a discretization method based on the sampling theorem is used. As an example, numerical calculations have been carried out for a conventional bend and a reflection bend in a dielectric slab waveguide. It is confirmed that the results agree well with data obtained by other methods. It is pointed out that good results can be obtained for such cases as a structure with large difference in index of refraction between the core and cladding and the one with a large bend angle, which were difficult to analyze by the conventional methods. Further, it is shown that the bending loss of the reflection bend is substantially smaller than that of the conventional bend.

Microstrip—line split—ring resonators and their application to bandpass filters

Abstract: The author has already reported a much more miniaturized split-ring resonator compared with the existing one-wavelength ring resonators by using microstrip lines and a prototype two-stage tunable bandpass filter. In the present paper, basic characteristics and salient features of split-ring resonators are summarized. Using a general-purpose simulation program for analysis of microwave circuits, calculating methods of filter coupling parameters, derived by using generalized expression of inverters, were described. The design chart was constructed indicating the concrete steps for designing. Finally, a three-stage filter for a 600-MHz band based on the forementioned results was designed and fabricated, and its operational characteristics were presented. The experimental results agreed quite well with the results of a computer simulation. The validity of the present design method has been confirmed. The bandpass filters using split-ring resonators have a suitable structure for a high-frequency hybrid IC. They are considered to be applicable over a wide frequency band in UHF to SHF bands.

Fiber misalignment measurement method using one—directionally observed image

Abstract: This paper investigates the measurement method of optical fiber offset in two directions by one-directionally observed image. This method employs the transmitted illumination and the bright lines of the intensity distribution due to the lens effect of the fiber are used to observe the position. The method is advantageous because of a simple mechanism for its realization. The observation principle of fiber position is investigated based on computer simulation and experimental results to clarify characteristics of the method. The real measurement of the fiber offset and its precision is investigated for fusion mass-splicing, to which the present method is very suitable. Employing this method, the standard deviation of 1.3 μm is achieved in precision of the measurements of the fiber offset. The screening characteristics for a given offset are also evaluated.

An integrated AlGaAs two—beam laser diode—photodiode array fabricated with reactive ion beam etching

Abstract: An integrated A1GaAs two-beam laser diode-photodiode array for optical disk applications was fabricated monolithically by the reactive ion beam etching technique. The laser cavity was formed by etched and cleaved facets. The threshold current and the external differential quantum efficiency are 18 mA and 28 percent, respectively. These parameters are as good as those of a laser with two cleaved facets. The detection efficiency of the photodetector (PD) was as high as 32.5 percent. The crosstalk due to the neighboring laser diode (LD) was minimized by forming an optical barrier between two PD and LD pairs. The cross-talk was as low as −20 dB. In the dual cavity-type laser using the light receiving surface of PD as the external mirror, the optical noise was lower than that in the laser without the external mirror.

Measurements on the influence of metallic phase arc on electrode surface for breaking Ag contacts

Abstract: The influence of metallic phase are discharge on electrode surface was studied for breaking Ag contacts. A whole are duration, metallic phase are duration, contact resistance and electrode material mass change were measured when both conditions of air pressure and circuit current were varied. The result showed that both the material loss and transfer had a close relationship to the metallic phase are duration. The contact resistance increased as the are duration grew in the metallic phase are region, and it became high and almost constant when the are sustained until it transferred to the gaseous phase.

Preparation and magnetic properties of anodic oxide magnetic films

Abstract: The M-H curve, coercive force and magnetic anisotropy of magnetic anodic oxide films containing electro-deposited iron particles in pores were measured for the various pore diameters between 150 and 1200 A, packing densities between 0.13 - 0.48 and film thicknesses between 0.08 and 1.3 μm. As a result, it was found that the coercive force perpendicular to the film surface depends only on the pore diameter when the length of the iron particle is 15 times larger than its diameter, and that it decreases in inverse proportion to the square of the pore diameter when the pore diameter is larger than 350 A. This indicates that the mechanism of magnetization reversal in these films follows the curling mode. The coercive force parallel to the film surface is affected by the shape of the initially deposited iron particles. The magnetic anisotropy depends strongly on the packing density of the iron particles. When the packing density is less than 0.32, the direction perpendicular to the film surface is the direction of easy magnetization. When the packing density is higher than 0.32, the direction parallel to the film surface is the direction of easy magnetization.

Study of BiCMOS cache memory

Abstract: To improve the speed of the processors for CPU of minicomputers and other systems, it is required to improve the machine cycle and the memory cycle, which are the performance indexes of the system. The general idea to realize a high-speed memory cycle is to apply the cache memory, which provides a hierarchical structure for the memory. This paper considers the realization of the cache memory by Hi-BiCMOS (high-performance bipolar CMOS) technology, which aims at a high-speed, highly integrated LSI by combining the microstructured CMOS and high-speed bipolar devices at the element level of the circuits. As the basic circuit, a RAM-combined Hi-BiCMOS circuit comparator is proposed, where the comparator logic is introduced into the sense amplifier of RAM. Applying the proposed circuit, TLB (Translation Lookaside Buffer) is designed and constructed, which performs the address transformation by cache memory. A satisfactory operation of the circuit was verified with the address conversion time of 13.3 ns. An evaluation by simulation was made for applying the proposed circuit to a 4K Byte cache memory, and it is indicated that the cache access through an address conversion can be made in 17.5 ns.

Self-injection locked gunn oscillator using nonradiative dielectric waveguide

Abstract: A Gunn oscillator is developed using a nonradiative dielectric waveguide which is employed in millimeter-wave integrated circuits. First, the structure and operation mechanism of the oscillator are described. The test of the oscillating properties shows the possibility of determining its frequency, and the maximum output power can be compared to that of the waveguide oscillators. Then a self-injection locked Gunn oscillator using a dielectric resonator is developed for frequency stability and noise reduction. It is found that a TEQ2δ mode high-dielectric ceramic resonator gives the best results when installed, and the oscillator was proved to be practical by measured performances which show the pushing figure and temperature coefficient as 1 MHz/V and -5 ppm/°C, respectively.

Biopolar-voltage-controlled optical switch fabricated on Z-cut Ti. LiNbO3 intersecting waveguides

Abstract: A new optical switch is proposed in which attention is paid to the behavior of the guided wave at the crossing of an intersecting waveguide. Design and fabrication of an optimal configuration were carried out at a practical wavelength of 1.3 μm. For the analysis of the operating mechanism and the optimum design, a waveguide analysis method based on the Fourier transform developed originally has been used. By means of the electrodes formed via a buffer layer on a Z-cut Ti:LiNbO3, a bipolar optical switch was fabricated in which switching is possible in both the positive and negative polarities. This switch has a low loss and low crosstalk if the intersection length is short, although the operating voltage is high. If the device length is increased for reduction of the operating voltage, other characteristics are degraded. Therefore, it is necessary to diffuse the double Ti layers below the electrodes and to improve the electrode structure. In the device with an intersection length of 8 mm thus obtained, the operating voltage of 14 V was realized. By way of theory and experiment, the characteristics of this switch were understood and the guideline for the future applications was obtained.

Electromagnetic fields in circular bends of slab waveguides

Abstract: A method for rigorous analysis of transmission in and radiation from a uniformly bent finite-length section of a slab wave length is proposed. The eigenmodes with continuous eigenvalues, which are similar to the radiation modes of the straight waveguide, are defined first for expressing fields in circular bends of slab waveguides, and a kind of orthogonality relation for these eigenmodes is derived. The spectral integral of these eigenmodes can give a rigorous expression of fields in circularly bent sections, which is proved by using the radiation condition and the Fourier transformation technique. The theory is applied to the analysis of transient phenomena of fields in the uniformly bent section connected to the straight section. Numerical examples of transition of field distributions and radiation patterns are presented and the results are discussed in detail.

Automatic measurement for investigation on the contact surface using nonlinear characteristics

Abstract: To improve the reliability of an electrical signal transmission through a contact, not only the contact resistance, but also the nonlinear characteristic must be evaluated, especially in the presence of a surface film. This method is effective in the separation of the film resistance and the constriction resistance, and the measurement of film thickness. To investigate the distribution of the oxide film thickness, the system was fully automated to measure the film thickness at various points on a sample surface. The measured properties were the contact resistance and the dc current which is one of the nonlinear components and is needed to cancel the rectification characteristic. To form a contact, copper, platinum and nickel wires were used. This paper demonstrates three-dimen-sionally and statistically the characteristics at various spots on a plane.

An SOI/CMOS flash A/D converter

Abstract: It is important from the viewpoint of increasing the application range of the three-dimensional circuit to construct analog circuits such as A/D converters on a silicon substrate formed on an insulating oxide film, or silicon on insulator (SOI) devices. Analog devices on a silicon on sapphire (SOS) substrate have been reported. However, no work on analog circuits has been reported with conventional SOI technology because of the lack of uniformity of the transistor characteristics. Due to the substrate floating effects represented by the kink phenomena characteristic to the configuration, the characteristics of the analog device are degraded. The laser recrystallization technique is improved and the chopper-type comparator is used which is less susceptible to the variations of the device parameters. Furthermore, a transistor structure that suppresses the kink phenomena is introduced in the amplifier inverter. With these approaches, a 4-bit flash A/D converter is fabricated successfully with the SOI/CMOS process. As a result, the basic operation is confirmed up to the power supply voltage of 5 V, input signal of 10 kHz, and conversion speed of 30 MS/s. In this paper, the design and evaluation are described for the SOI/CMOS A/D converter for three-dimensional circuits.

Analysis of the light—beam scattering and the sum and differential signal output by arbitrarily shaped pits and bosses

Abstract: When the boundary element method is applied to an electromagnetic scattering problem in an open region, an infinite integral must be carried out if the scatterer has an infinite extent. If the incident wave is a beam wave, it exists in a narrow region centered around the illuminated region. However, although the surface current excited by this incident field decays quickly in the case of E-wave incidence since the position is separated from the illuminated region, it is distributed in a broad range in the H-wave incidence. In this paper, a method for evaluating such an integral over a broad region is applied to the problem of scattering by arbitrarily shaped pits and bosses. It is demonstrated that an analysis for the H-wave incidence can be done with the same accuracy for the E-wave. In conventional analysis methods, substantial numerical errors are unavoidable for the H-wave case. In addition, an analysis is carried out for the output signal characteristics for the hypothetical signal detection system of an optical disk. A detailed study is carried out as to how the sum and difference signal output by rectangular trapezoidal and V-shaped pits and bosses depend on the shape and dimensions of the pit and boss or the off-axis shift (tracking misalignment) of the incident beam and the shift of the focus.

Theoretical analysis of the cerenkov laser using a nonlinear dielectric waveguide

Abstract: Consider a Cerenkov laser, using for the slow-wave structure a dielectric waveguide whose substrate consists in a nonlinear dielectric presenting the self-focusing effect. The characteristics of the Cerenkov laser specified in the foregoing are investigated theoretically for the two-dimensional model which takes into account the field distribution in the transverse direction. First, the dispersion relations are considered for the electromagnetic wave modes (TM waves) propagated along a two-dimensional nonlinear dielectric waveguide. Then the dispersion relations are discussed for the coupled electromagnetic fields in the Cerenkov laser using the nonlinear dielectric waveguide. Finally, the spatial growth rate of the growing wave obtained from the dispersion relation for the coupled fields derived in the foregoing is investigated numerically in detail. From the forementioned discussion, it is found that the spatial growth rate for the Cerenkov laser treated in this paper is improved compared to the Cerenkov laser for which the slow-wave structure is composed of linear dielectrics only.

A measurement method for visualization of static electricity distribution

Abstract: The number of damaging incidents due to electrostatic discharges (ESD) on digital electronic equipment has been increasing with the increased use of integrated semiconductor elements with lower operation power. We proposed the idea that by visualizing the static electricity which caused ESD for realtime recognition and tracking, preventive measures could be applied before ESD occur. This paper proposes a measuring principle in which the electric potentials induced on an array of small square electrodes placed in parallel in the vicinity of a charged body are proportional to the amount of the charge of the body. A prototype based on this principle was built and attempts were made at visualization and tracking for the static electricity distribution of a charged body. As a result, we confirmed that we could visualize charge distributions of a charged body directly below the electrode array using the electric potential data from a finite number of measuring electrodes. Furthermore, we could observe a two-dimensional charge distribution of a charged body using an acrylic sheet.

Finite element analysis of discontiuities in nonradiative dielectric waveguide

Abstract: Discontinuity problems in the nonradiative dielectric waveguide are treated by the finite element technique. Both computing time and memory can be reduced significantly by introducing infinite elements which take into account the evanescent nature of the fields outside the dielectric strip. The hybrid nature of the fields is also taken into account to improve numerical accuracy when the analysis expression for the far fields and the FEM solution for the near fields of discontinuities are matched. Various structures, such as gaps, high-permittivity dielectric sheet inserts, bandpass filters and even T-junctions have been analyzed. Some results are confirmed by experimental value.

Multinozzle drop generator for continuous ink jet printing

Abstract: Uniform, stable drop generation is the key technology for continuous ink jet printing. Previously, the method to produce the standing wave driven by the high polymer piezoelectric transducer had been utilized for acoustic excitation. However, the characteristics varied from assembly to assembly and were unstable to the environmental changes. As a new alternative approach, the Traveling-Wave Drop Generator (TWDG) was developed in which the acoustic wave is generated in one end, and then travels to the other end of the jet array. The characteristic results showed that this generator could fully solve the forementioned problems, and it has a simplified structure.

Analysis of reflective SAW gratings for SH-type modes using finite-element method

Abstract: A solution based on the finite-element method (FEM) is presented for the analysis of SH-type surface acoustic wave gratings. Although the FEM has already been applied to the discontinuities in elastic waveguides for SH-type modes, only the isotropic substrate has been treated. Also, it is difficult to apply the FEM directly to many discontinuities such as grating reflectors due to excessive computation time and computer memory requirements. This paper presents a numerical approach to the analysis of SH-type surface acoustic wave gratings on anisotropic substrates. To apply the FEM directly to many discontinuities, the substructure method is introduced. As a result, mutual interaction between surface waves and bulk waves is automatically taken into account over the entire discontinuity region. To show the validity and usefulness of this approach, examples are computed for various gratings on PZT-4, 41° Y-X LiNbO3, and 36° Y-X LiTaO3 substrates.

GaAs‐ALGaAs double‐heterostructure coupled‐waveguide optical modulators and their high‐frequency modulation characteristics

Abstract: In this paper, a GaAs-AlGaAs double-heterostructure (DH) coupled-waveguide optical modulator, which is important as a multifunction optical component and for possible monolithic integration, has been studied in detail. A DH device with pn-junction structure, which is expected to have the lowest switching voltage, has been fabricated and the optical modulation characteristics have been measured at 1.06 μm wavelength. At a reverse bias of 13 V, an extinction ratio of 11 dB has been obtained. However, the cutoff frequency has been limited to about 250 MHz, which was lower than the calculated value of 1.1 GHz obtained using a simple lumped-element circuit analysis. Therefore, the optical modulation characteristics have been analyzed assuming the device as a distributed-element circuit line, and also taking into account the external parasitic impedancees. Thus it has been concluded that a reduction of the top-electrode resistance and an increase of the conductance across the device cross-section are necessary to improve the high-frequency optical modulation characteristics. To verify these results, a DH Schottky contact structure, which essentially possesses no ohmic contact resistance with a thick topelectrode, has been fabricated on a MBE-grown wafer. At a reverse bias of 16 V, an extinction ratio of 12 dB and a cut-off frequency of 1.5 GHz have been obtained successfully. It has also been concluded that an even thicker top-electrode and an improved mounting method which minimizes the external parasitic elements are necessary for further improvement of the optical modulation frequency characteristics.

An analysis of scattered near field and induced current of a beam wave by pits on optical disk using boundary element method

Abstract: Numerical analysis of the scattering of a beam wave caused by pits which are assumed to be absolute conductors is carried out in order to obtain the optimum shape of pits and pregrooves and achieve high density on an optical disk. For analysis, the boundary element method (BEM) is used because it is adaptable to any shape of pits and pregrooves and any type of incident beam. For rectangular or trapezoidal concave and convex pits, scattered fields and induced currents are calculated as functions of the angle of the pit wall and track pitch, and comparison between the two is made. In the case of convex pits, it is found that the minimum track pitch without crosstalk for E-wave and H-wave are 2w and 1.8w (w: the spot size of the incident beam), respectively. As the pit shape is changed from rectangular to trapezoidal, the intensity change of the main lobe with or without the pit is small. Finally, the current density on the disk, which is induced by scattering, is calculated.